 

. . ...z::. .7. c.
. . . . . .t:t...$i¢»i$ .111... .mnnzﬂ “v
. 4 . . . A 7-45 It .. x. 1.. ‘ .-
4... . .4.... 35.4...- a...
x. . use... Wililziiw {:5
. . .3 .. I
. 1...... 3.2!. , . _.. .-
. v .. 13...: .
.1... .23. . inn. 593...... . .
2. .. :3... 1523...... .vxxx .44..
. 41!... a..{.¢..v...2.....v........v 3.2...
,.i u. '..k.r..r...\q.:6...
i. .12....738
.7. 19!. .2: ..
....: 152.11.32.03
:. 1.1.6.. .. ..
:5 4..

 

i :

..

‘II. 7921315....4
3..

 

an...» “4.x. ”nu. F... . . , a ovum

 

 

 

 

o
1.3.2.2.... ,7.
a . , .
calf: .. .

.z .5 5.
kn?

 

T...
. 4 .4...
. 3142‘-
...2 4...: .
S .
i

.13..le
.vv . 31.

:e. f :3:
:14'5.

.

x :ﬁqﬁa
I s .

.‘I

‘
.1 . I9 .
ah». I 1

. .1 .

91,. 1...;

3.1.... v 5.

p..~i£~§0bh..$| 1. .
.-

 

..

 

you.
I 71.0".
:3... t

:1:- 315
5. 41...”. .6

a!

 

 

 

.15....
01.1}.
1.... 44.1.1.3... < V11
.1... H. I. -3327

 

 

. 1 A. 3. 4;.
>9 V . '3'...“th
I... .Ovu.Vvl..l..l .i It
. {v4.rsalft '45:».
a. 3:11.... 1"...” 1 2.: 1..
1 it: 1;. it 0..
t In: .311}...- If...

 

.v.
.I.

 

 

     
       

   

       

 

    

 

   
 

        

 

 
 
      

   

 

  
 

1.1.1: 0".o .c ottuong.cl:avaﬁn.. J.'.
In... - k .34}-.. .1}..- , .4. . ﬁrs”? §é05§§§§§aiﬁkg
.....;.....t....£..r.au.yi. .5325???..wwﬂ.l.u§#umur§ﬁmﬁ wi.ﬁdw1¢un$m....§ . . . .. r . . x 1. . 1.

 

[l'lnln'll‘

 

 

LIBRARY
Mlinhm

~ Gil-niln
uvuvn Haul I \JLQtC

University

 

 

 

This is to certify that the
dissertation entitled

JOINT VENTURE TERMINATION: FAILURE OR ADAPTATION?

presented by

ANNA SHAOJIE CUI

has been accepted towards fulﬁllment
of the requirements for the

 

Ph.D. degree in Marketing

 

 

n <7 i
(1/874, \_.AIOLé/ﬂ,“i& (“(4

U L/ Major Professor’s Si

Date

 

/'

 

MSU is an Afﬁrmative Action/Equal Opportunity Institution

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before' date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:/C|RC/Date0ue.indd-p.1

 

 

JOINT VENTURE TERMINATION: FAILURE OR ADAPTATION?
By

Anna Shaojie Cui

A DISSERTATION
Submitted to
Michigan State University

In partial fulﬁllment of the requirements
For the degree of

DOCTOR OF PHILOSOPHY

Department of Marketing and Supply Chain Management

2006

ABSTRACT

JOINT VENTURE TERMINATION: FAILURE OR ADAPTATION?
By

Anna Shaojie Cui

In the joint venture literature, termination is often considered analogous to
failure, and the reasons for termination attributed to initial characteristics of the
venture. This research argues that venture termination may arise due to adaptation
instead of failure, and changes during the joint venture process, rather than initial
formation conditions, are key drivers of joint venture termination. These changes
extend beyond the boundaries of the joint venture to include parent ﬁrm strategy and
the market environment.

Adopting a co-evolutionary theory view of joint ventures, this study
investigates the inﬂuences of external change on the propensity of joint venture (JV)
termination. It develops and tests a model of JV termination that incorporates JV
performance as well as external change, including changes in parent ﬁrm overall
strategy and in the market environment, as determinants of the propensity of JV
termination.

Event history analysis is used to empirically test the model and a longitudinal
study design is employed to examine the over time effects. The results conﬁrmed that
changes in parent ﬁrm overall investment strategy and governance strategy both
signiﬁcantly influence the propensity of JV termination. Different effects are also

found for two types of IV terminations, i.e., dissolution and acquisition.

Based on the model, this study identiﬁes two primary reasons for JV
termination: low JV performance and changes external to the joint venture. While
termination due to low performance is directly associated with failure, change-
induced termination is a result of ﬁrm strategic adaptation.

This dissertation contributes to previous research in ﬁve ways. First, it
provides an empirical test of co-evolutionary theory of alliances and further co-
evolutionary theory in the context of joint ventures. Further, it identiﬁes changes,
rather than the initial JV conditions, as key drivers of JV termination. Third, it
demonstrates the embeddedness of joint ventures in parent ﬁrm overall strategy and
how JV termination can result from parent ﬁrm strategic adjustment. Fourth, it
clariﬁes the relationship between JV termination and failure by showing termination
can be ﬁrm strategic adaptation rather than failure. Lastly, it addresses the need for
longitudinal studies in the literature by utilizing an event history methodology and

time series data.

Copyright by
ANNASHAOHECUI

2006

To my mother

ﬁé‘iﬁiﬂﬂﬁ

ACKNOWLEDGMENTS

I wish to acknowledge the assistance and support of several people in the
completion of this dissertation and my doctoral program:

Dr. Roger Calantone, my dissertation committee co-chair and research adviser,
for giving me not only guidance and assistance, but also inspiration and
encouragement that supported me through the program.

Dr. David Griffith, my dissertation committee co-chair and a long time co-
author, for his help on every aspect of my work.

Dr. Cornelia DrOge for her valuable advice and for showing me what is rigor
in research.

Dr. Jon Bohlmann for his support, and his sharp and constructive suggestions
on the dissertation.

My fellow doctoral students with whom I have become friends. Their
friendship is one of the most valuable things in my life. Especially I would like to
thank Chun Zhang for her help at my most difﬁcult time, and Kwong Chan, who, with
his endless support and encouragement, made the Ph.D. experience a lot more

enjoyable and fruitful.

vi

TABLE OF CONTENTS

 

 

 

 

 

 

 

TABLE OF CONTENTS ................................................................................ VII
LIST OF TABLES ........................................................................................... X
LIST OF FITURES ......................................................................................... XI
CHAPTER 1
INTRODUCTION ............................................................................................. 1
1.1. Overview of the problem - ......... 1
1.1.1. Joint venture termination and failure ............................................................... 1
1.1.2. The reasons for JV termination: JV process and JV outcome ......................... 2
1.1.3. The reasons for JV termination: exogenous factors versus factors under JV
management control ................................................................................................... 3
1.1.4. Types of JV termination ................................................................................... 3
1.2. The co-evolutionary theory view ........ . ...... - ............ 4
1.3. This dissertation study 5
1.3. 1 . Deﬁnitions ........................................................................................................ 5
1.3.2. Purpose of the study ......................................................................................... 6
1.3.3. Research questions ........................................................................................... 7
1.3.4. Research methodology ..................................................................................... 8
1.4. Overview of the dissertation ......... 8
CHAPTER 2
REVIEW OF JV TERMINATION RESEARCH WITHIN THE CO-
EVOLUTIONARY THEORY OF ALLIANCES ............................................... 10
2.1. Co—evolutionary theory of alliances 10
2.2. JV termination within a co-evolutionary perspective- -- -- 13
2.2.1. JV Termination is a result of the co-evolutionary process. ........................... 14
2.2.2. Change during the co-evolutionary process drives JV termination ............... 15
2.2.3. Parent ﬁrm overall strategy and the market environment have important
influences on JV termination ................................................................................... 15
2.2.4. JV Termination can be resultant from ﬁrm adaptation rather than failure 16
2.2.5. Commonalities and differences between JV dissolution and acquisition ...... 16
2.3. Previous research on JV termination - 19
2.3.1. JV termination and failure ............................................................................. 20
2.3.2. Focus on initial formation conditions ............................................................ 20
2.3.3. Neglection of exogenous factors .................................................................... 22
2.3.4. Different types of JV termination .................................................................. 26
2.3.5. Methods for empirical testing ........................................................................ 26

vii

CHAPTER 3

 

 

 

 

 

 

 

 

THE CONCEPTUAL MODEL AND HYPOTHESES ..................................... 29
3.1. The conceptual model 29
3.2. Hypotheses - - - ............ 34
3.2.1. Change in market competition ....................................................................... 34
Market growth ...................................................................................................... 34
Increase in market concentration ......................................................................... 36
3.2.2. Change in parent ﬁrm overall investment strategy ........................................ 3 7
Product line shiﬁ .................................................................................................. 3 8
Increase in parent ﬁrm resource ........................................................................... 39
Disruptive event ................................................................................................... 43
3.2.3. Change in parent ﬁrm overall governance strategy ....................................... 44
Increase in risk of appropriation .......................................................................... 46
Decrease in alternative partners ........................................................................... 47
Formation of competing partnerships .................................................................. 48
3.2.4. Joint venture performance .............................................................................. 49
3.2.5. Joint venture type ........................................................................................... 49
Related JV versus unrelated JV ........................................................................... 50
Direct-competitor JV versus non-direct-competitor JV ....................................... 53
3.3. Further statements -- ........... 55
CHAPTER 4
RESEARCH METHODOLOGY ..................................................................... 58
4.1. Model formulation -58
4.2. Data - - - -- 60
4.2.1. Sample ............................................................................................................ 60
4.2.2. Data collection method .................................................................................. 61
4.3. Measures -- . - ........ 62
CHAPTER 5
ANALYSIS RESULTS AND DISCUSSION ................................................... 70
5.1. Statistical results - -70
5.1.1. Change in market competition ....................................................................... 73
5.1.2. Change in parent ﬁrm overall investment strategy ........................................ 73
5.1.3. Change in parent ﬁrm overall governance strategy ....................................... 75
5.1.4. JV performance .............................................................................................. 76
5.1.5. The interaction effects with JV type variables ............................................... 76
5.1.6. JV termination in total ................................................................................... 77
5.2. Discussion of results -- ..81
5.2.1. The inﬂuence of change in market competition ............................................ 81
5.2.2. The inﬂuence of change in parent ﬁrm overall investment strategy ............. 83
5.2.3. The inﬂuence of change in parent ﬁrm overall governance strategy ............. 85

viii

5.2.4. The inﬂuence of JV sales performance .......................................................... 86

 

 

 

 

5.2.5. The moderating effects of JV type ................................................................. 87
5.2.6. A comparison of JV dissolution and acquisition ........................................... 90
CHAPTER 6
CONCLUSION .............................................................................................. 92
6.1. Synthesis of signiﬁcant ﬁndings - 92
6.2. Contributions - _ - _ - ........ 94
6.2.1. Theoretical contribution ................................................................................. 94
6.2.2. Empirical contribution ................................................................................... 96
6.2.3. Managerial implications ................................................................................. 97
6.3. Limitations and directions for future research -- - 98
6.4. Conclusion -- - -- - 100
REFERENCES ............................................................................................ 1 01

ix

LIST OF TABLES
Table 1. Previous studies on joint venture termination ................................................ 21

Table 2. A classiﬁcation of explanatory variables in previous JV termination studies

.............................................................................................................................. 25
Table 3. Variables, measures and data sources ............................................................ 63
Table 4. Mean, standard deviation, and correlations of covariates .............................. 69
Table 5. Analysis results for the base model without interaction effects. ................... 71
Table 6. Analysis results for the full model with interaction effects ........................... 72

Table 7. Comparison of test results for JV termination in total and JV dissolution and
acquisition: without interaction effects .......................................................... 78

Table 8. Comparison of test results for JV termination in total and JV dissolution and
acquisition: with interaction effects ............................................................... 79

LIST OF FIGURES

Figure l. A model of joint venture termination ........................................................... 30

xi

CHAPTER 1
INTRODUCTION
1.1. Overview of the problem
1.1.1. Joint venture termination and failure
Globalization and increasingly intense competition have brought a dramatic
acceleration in joint venture formation. Nevertheless, most joint ventures are short-
lived, with estimated termination rates reaching 50 percent or higher (Harrigan 1988:
Parkhe 1993; Porter 1987). The high termination rate has been generally viewed as
an indication of high failure rate (Park and Ungson 2001 ). Previous studies have used
duration and survival as measures of joint venture performance, with termination
indicating low performance (Chowdhury 1992; Harrigan 1988).

However, termination and failure are not the same because ajoint venture
may be terminated because it no longer ﬁts into the parent ﬁrm’s strategy and a long—
lasting joint venture might be suffering from low performance (Yan and Zeng 1999).
In business failure studies, termination is used as a measure of failure, because for an
independent ﬁrm going out of business indicates failure. For joint ventures, however,
the relationship between termination and failure is more complex. The existence and
operation of ajoint venture is not independent of the parent ﬁrms, and the termination
of a joint venture is not always associated to low performance, but also possibly
related to changed motivations of parent ﬁrms. From the parent ﬁrm’s perspective,
terminating a joint venture may not be a bad thing, especially consideringjoint
ventures are oﬁen used an alternative form of investment that provides more
ﬂexibility than fully-owned subsidiaries. Therefore, to understand when JV

termination results from failure, the reasons for termination need to be considered.

1.1.2. The reasons for JV termination: JV process and JV outcome

JV termination is an outcome of a joint venture. The life span of a joint venture
starts with its formation and ends with its termination. During its life span, a joint
venture experiences frequent changes and evolves over time (Day 1995; D02 1996;
Ring and Van de Ven 1994). JV process is the evolutionary process of a joint venture
that is characterized by changes and adaptations (Ring and Van de Ven 1994).

Though the importance of studying the process of JV partnerships has been
long recognized, very few studies have been done from this perspective. Among the
limited process studies, most are conceptual or case studies (D02 1996; Hamel 1991;
Ring and Van de Ven 1994). Case studies have the advantage of developing new
theories and investigating marketing issues in detail, but case studies cannot substitute
for quantitative empirical studies in terms of the generalizability of theory testing.

More importantly, even fewer studies have related the partnering process to the
outcome of a joint venture, investigating how the JV process inﬂuences the ﬁnal
outcome of a joint venture such as termination. Most previous studies on JV
termination associate the reasons for termination with the initial formation conditions
of the partnership, identifying factors such as ownership distribution among partners
(Blodgett 1992), partner asymmetry (Harrigan 1988), joint venture type (Park and
Russo 1996), and so on. Accordingly, the independent variables used in empirical
studies are predominantly static, i.e., not varying over time. The initial conditions of
ajoint venture have important inﬂuences on the JV outcome. However, joint ventures
experience frequent changes over the life span (Day 1995), and these changes that
unfold after JV formation directly affect the JV outcome. Neglecting the JV process

fails to consider a very important aspect of JV termination.

1.1.3. The reasons for JV termination: exogenous factors versus factors under JV
management control

Further, JV studies have mainly focused on factors under JV management
control, such as equity control (Blodgett 1992; Dhanaraj and Beamish 2004; Lu and
Hebert 2005), managerial support (Steensma and Lyles 2000), cultural distance
(Barkema et al. 1997; Park and Ungson 1997), etc. The inﬂuences of factors
exogenous to a joint venture, such as competitive nature of the industry (Kogut 1988)
and the strategies of parent ﬁrms, are largely neglected. While JV management factors
reveal the managerial complexity of joint ventures and provide important explanation
for JV termination, parent ﬁrm overall strategy and the market environment also have
important inﬂuences on JV termination (Franko 1971; Kogut 1989). Recognizing the
important inﬂuences of exogenous factors on JV termination can help clarify the
relationship between JV termination and failure, because termination due to
exogenous factors that are out of JV management control is not necessarily related to

failure.

1.1.4. Types of JV termination

Lastly, most previous studies employ duration, longevity or survival as the
dependent variable, not considering how a joint venture is terminated, i.e., dissolved
(liquidation), or acquired by one of the parent ﬁrms (Barkema et al. 1997; Lu and
Hebert 2005). The few exceptions are Kogut (1991), who studied only acquisition,
including acquisition by a partner ﬁrm and by a third party, and Park and Russo
(1996), who excluded acquisition by the parents from their analysis. But these studies
did not simultaneously consider dissolution and acquisition or compare them. How a

joint venture is terminated is related to the reason why it is terminated, and different

types of termination may be underlined by different considerations of the parent
ﬁrms. The importance of distinguishing JV termination types not only falls out of the
perspective of viewing joint ventures within the context of parent ﬁrms’ overall
strategy (Reuer and Koza 1998), but also lies in understanding the reasons of
termination and how they are related to joint venture failure (Hennart et al. 1998).

In sum, previous research has limited our understanding of JV termination in
three ways. First, initial JV formation conditions have been given the most attention
in neglect of factors that necessarily change over time. Secondly, most studies
emphasize factors under JV management control, omitting exogenous forces such as
parent ﬁrm overall strategy and the market environment. These focuses
underestimate the inﬂuence of external changes on JV termination, and consequently
fail to consider strategic adaptation as a possible motivation for IV termination.
Without considering external change as a possible reason for JV termination, the
previously found high correlation between low JV performance and termination may
be spurious. Lastly, not distinguishing between different types of JV termination

inhibits our understanding of the reasons for IV termination.

1.2. The co-evolutionary theory view

Co-evolutionary theory of alliances is an extension of co-evolutionary theory
in the context of alliances. It offers a new perspective of the alliance evolution in
relation to the market environment and parent ﬁrm strategy (Koza and Lewin 1998;
Lewin et al. 1999; VoIberda and Lewin 2003). This theory views alliances in the
context of the adaptation choices of a ﬁrm (Koza and Lewin 1998). In this view,
alliances are embedded in parent ﬁrm overall strategy, and co-evolve with parent ﬁrm

overall strategy and the market environment (Koza and Lewin 1998). It emphasizes

change and adaptation over time, and views alliance evolution through series of
reevaluation and readjustment cycles. During the process of alliance management,
parent ﬁrms learn about the alliance and its environment and periodically reevaluate
the alliance, which in turn leads parent ﬁrms to make adjustment to the alliance by
moving away from the initial conditions (D02 1996). It also argues that alliance
evolution takes place not only within an alliance itself, but also together with the
market environment and the parent ﬁrm’s overall strategy, therefore co-evolution
(Koza and Lewin 1998).

The co-evolutionary theory perspective provides a very good foundation to
incorporate both change over time and exogenous factors to the study of JV
termination. It brings a new perspective of viewing the outcome of a joint venture as a
result of the co—evolutionary process (Arino and de la Torre 1998; D02 1996; Ring
and Van de Ven 1994). With this perspective, JV termination itself can be viewed as
an adjustment action of the parent ﬁrm based upon its changed evaluation of the joint
venture. Studying JV termination within the co-evolutionary framework will also
improve our understanding of JV termination by showing how JV termination may

result from ﬁrm strategic adaptation rather than failure.

1.3. This dissertation study
1.3.1. Deﬁnitions

1. Strategic alliances and joint ventures. A strategic alliance is a contractual
arrangement between two or more independent companies that choose to carry out a
project or operate in a speciﬁc business area by coordinating the necessary skills and
resources jointly rather than operating on their own or merging their operations

(Dussauge et al. 2000). A joint venture is a contractual arrangement that creates a

separate legal entity in which the parent ﬁrms hold ownership interests. According to
these deﬁnitions, a joint venture is one type of strategic alliance that is equity-based.
This study is in the context of joint ventures.

2. JV Termination. JV Termination is deﬁned as the end of a joint venture,
including dissolution (liquidation) and acquisition (i.e., when a joint venture is
acquired by one of the parent ﬁrms).

3. JV Failure. JV Failure is deﬁned as unplanned termination without
achieving the goal of a joint venture. This deﬁnition is consistent with the work of
Park and Ungson (2001), who pointed out that “unexpected” termination is more
directly related to failure, and Khanna (1998), who argued that not achieving the
beneﬁts is an indication of failure. Failure is looked at from the parent ﬁrm’s
perspective, rather than the success or failure of a joint venture in isolation.

4. Open-ended joint ventures. Open-ended joint ventures are joint ventures
that are expected to last for an unlimited time, without a prespeciﬁed duration. To
look at JV co-evolution and termination, this study only considers open-ended joint

VCI'IIUI‘CS.

1.3.2. Purpose of the study

This study develops a model of JV termination. Employing a co-evolutionary
theory view of alliances, this study views JV termination as an outcome of the JV
process characterized by changes in parent ﬁrm strategy and the market environment.
Considering both exogenous factors and over time change, it incorporates external
change as a determinant of JV termination. In addition, two types of JV termination,

i.e. dissolution and acquisition, are separately analyzed and compared.

 

The research addresses the call for process studies in the joint venture
literature. Relating the evolutionary process to the outcome of ajoint venture, this
study highlights the difference between change-induced terminations and terminations
resultant from low performance and argues that change-induced terminations are ﬁrm
adaptations, not failures.

Distinguishing between JV termination and failure can provide a better
understanding of JV success and failure as well as new insights for JV management.
Realizing that joint ventures are not supposed to last forever, the management
philosophy of joint ventures may change from simply trying to avoid JV termination
to actively preparing, predicting and even planning for terminations. Also, oﬁen
calculated from terminations, the high JV failure rate known in the literature may not
reﬂect the actual percentage of failed joint ventures. The distinction can help to

empirically clarify the actual JV failure rate.

1.3.3. Research questions

This study intends to answer the following research questions.

1. What factors inﬂuence the propensity of JV termination?

2. Given the impact of JV performance, what is the inﬂuence of external
change, in both the market environment and parent ﬁrm overall strategy, on
the propensity of JV termination?

3. Are the inﬂuences of the above factors different for JV dissolution and

acquisition?

1.3.4. Research methodology

A conceptual framework is developed based on co-evolutionary theory of
alliances, and tested with event history analysis (EHA) and time series data. EHA is
suitable for studying JV termination because it models the possibility of the
occurrence of an event (JV termination is the event in this case) and captures the

inﬂuences of explanatory variables over time.

1.4. Overview of the dissertation

This chapter provides an overview of this dissertation research. The confusion
between JV termination and failure is discussed and the importance of distinguishing
these two concepts is emphasized. Looking at reasons for JV termination is necessary
to distinguish between JV termination and failure. The shortcomings of current
research on JV termination are identiﬁed and it is pointed out that the neglect of JV
process and exogenous factors such as parent ﬁrm overall strategy and market
environment prevents current research from recognizing external change as a possible
source of JV termination. The research questions are then presented.

Chapter 2 reviews the previous literature on JV termination within the co-
evolutionary theory view of alliances. The arguments of co-evolutionary theory of
alliances are discussed and applied to the context of JV termination. Previous
literature on JV termination is then reviewed, and the need for adopting a co-
evolutionary theory view in JV termination research is pointed out.

Chapter 3 develops the conceptual model of JV termination. Hypotheses are

developed for empirical testing.

Chapter 4 presents the research methodology. Detailed information is
provided for sampling frame, data collection method, measures, data sources, and
event history analysis method.

Chapter 5 presents and discusses the analysis results. Explanations for the
testing results are provided.

Chapter 6 evaluates the ﬁndings and contributions of this study and presents
the conclusions. Implications for academics and managers, limitations and

opportunities for future researched are discussed.

CHAPTER 2
REVIEW OF JV TERMINATION RESEARCH

WITHIN THE CO-EVOLUTIONARY THEORY OF ALLIANCES

2.1. Co-evolutionary theory of alliances

The co-evolutionary theory of alliances is an extension of the co-evolutionary
theory in the context of alliances. Co-evolutionary theory has recently been more
prominent in the management and organization theory literature (Baum 1999; Lewin
and VoIberda 1999). It provides a fresh view of organizational adaptation and
selection that integrates the interplay between the adaptation of organizations, their
competitive dynamics and the environment organizations are embedded in. It argues
that organizations and their environment are interdependent and evolve together. It
introduces the notion of “multi-levelness” and “hierarchical nestedness”, which
indicates that evolution takes place at multiple levels (i.e., within organizations,
between organizations, and between organizations and the market environment), and
the units of evolution are nested within one another (Baum and Singh 1994; Lewin et
al. 1999; March 1994).

Built on the co-evolutionary theory view, the co-evolutionary theory of
alliances views alliances in the context of the adaptation choices of a ﬁrm (Koza and
Lewin 1998) and emphasizes the embeddedness of an alliance in parent ﬁrm overall
strategy. An alliance is part of the parent ﬁrrn’s strategic portfolio, and alliance
decisions are functions of parent ﬁrm overall strategy. During the alliance process, an
alliance co-evolves with parent ﬁrm overall strategy and the market environment
(Koza and Lewin 1998). Because an alliance is a special organizational form, the co-

evolution of an alliance with the parent ﬁrm is a form of evolution between

organizations, i.e., evolution takes place not only within an alliance but also between
the alliance and its parent ﬁrms. At the same time, an alliance participates in
competition in its market and co-evolves with the market environment. The co-
evolution of alliances with their parent ﬁrms and the market environment is an
example of multi-level and hierarchically nested evolution. Speciﬁcally, the key
arguments of co—evolutionary theory of alliances include the following.

First, the alliance process is characterized by change and adaptation over time,
and alliances evolve through series of reevaluation and readjustment cycles. As a
special form of organizations, alliances are dynamic systems of adaptation and
evolution (Arino and de la Torre 1998; D02 1996; Ring and Van de Ven 1994). After
an alliance is formed, parent ﬁrms are involved in an on-going process of reevaluation
and readjustment of the alliance, which starts with the initial conditions but go beyond
the initial conditions (D02 1996). During the process of alliance management, parent
ﬁrms learn about the alliance and its environment, and periodically reevaluate the
alliance, which in turn leads parent ﬁrms to make adjustment to the alliance by
moving away from the initial conditions (D02 1996). An alliance evolves through a
sequence of learning-reevaluation-readjustment cycles over time (D02 1996; Ring and
Van de Ven 1994).

Secondly, alliance evolution takes place not only within an alliance itself, but
also together with the market environment and the parent ﬁrm’s overall strategy,
therefore it is referred to as co-evolution (Koza and Lewin 1998). The parent ﬁrm’s
reevaluation and readjustment go beyond the alliance itself to include the alignment of
the alliance with the overall strategy of the parent ﬁrm and the market environment

(D02 1996; Franko 1971).

ll

Alliances compete with other ﬁrms in its market. The level of competition
inﬂuences the competitive position of an alliance in the market (Porter 1985; Porter
1980). Changes in market competition alter the competitive position of an alliance,
and an alliance constantly adjusts to these changes to achieve an advantageous
position (Kumar and Nti 1998). For example, increased level of competition and
increased market power of other ﬁrms in the same market weakens an alliance’s
competitive position. When these changes occur, the parent ﬁrm reevaluates the
alliance and adjustment actions are taken either to strengthen its position or exit the
market when the cost of strengthening it is higher then the beneﬁt (D02 1996; Franko
1971). Through this process of reevaluation and readjustment, alliances evolve with
changes in the market environment.

At the same time, an alliance is part of the strategic portfolio of the parent
ﬁrm. It is set up to carry out certain activities, such as production, marketing or R& D,
which contribute to the overall strategy of the parent ﬁrm. Changes in the parent
ﬁrm’s overall strategy alter the relative contribution and importance of the alliance to
the parent ﬁrm, which induces the parent ﬁrm to reevaluate the alliance and take
adjustment actions regarding the activities or structure of the alliance (Arino and de la
Torre 1998). Through this process of reevaluation and readjustment, alliances evolve
with parent ﬁrm overall strategy.

Lastly, the outcome of an alliance is a result of the co-evolutionary process
(Arino and de la Torre 1998; D02 1996; Ring and Van de Ven 1994). During the co-
evolutionary process, changes induce the parent ﬁrms to reevaluate the alliance, and
changed evaluation motivates the parent ﬁrms to take adjustment actions regarding

the alliance (Arino and de la Torre 1998; D02 1996; Ring and Van de Ven 1994).

 

Terminating an alliance itself is an adjustment action of the parent ﬁrm based upon its
changed evaluation of the alliance. Incremental readjustment actions are often taken
to improve the situation before terminating an alliance, but unsuccessful incremental
adjustments will reinforce previous evaluation, which reaches a point where the
evaluation of the alliance falls below alternative arrangements accomplishing the
same purpose, and consequently leads to termination of the alliance (Arino and de la
Torre 1998; D02 1996). Though small changes may not immediately lead to
termination, the occurrence of changes increases the need for readjustment and
instability in an alliance (Yan 1998), therefore it increases the propensity of
termination of an alliance.

The application of co-evolutionary theory in the context of alliances has been
limited to some case studies (Arino and de la Torre 1998; Koza and Lewin 1999).
Though some studies have included a time dimension (D02 1996; Gulati 1995), the
majority of alliance studies are cross-sectional, and research that considers the time
dimension tends to remain at the level of conceptual development and fails to
empirically capture the process of alliance evolution (D02 1996). Little systematic
attention has been paid to the inﬂuence of the evolutionary process on the alliance
outcome. Rarely has research explicitly considered alliances as embedded within the
strategy portfolio of parent ﬁrms and/or the market environment they are in (Koza and
Lewin 1998). The co-evolution of alliances with parent ﬁrm strategy and the market

environment remains an unexplored area of research (Koza and Lewin 1998).

2.2. JV termination within a co-evolutionary perspective
Joint ventures are a special form of alliances. Joint ventures also co-evolve

with their market environment and parent ﬁrm overall strategies. JV termination, the

13

outcome of a joint venture, results from the co-evolutionary process of the joint
venture with its market environment and parent ﬁrm overall strategy. JV termination
is the parent ﬁrm’s adjustment actions based on its changed evaluation of the joint
venture. Viewing JV termination within the perspective of JV co-evolution provides
new understandings of JV termination and reveals several needs in JV termination

research.

2.2.1. JV Termination is a result of the co-evolutionary process.

During the co-evolutionary process, parent ﬁrms are involved in the
reevaluation of ajoint venture and take adjustment actions accordingly. Changes in
the market environment and parent ﬁrm overall strategy alter the parent ﬁrm’s
evaluation of the joint venture and motivate the parent ﬁrms to take readjustment
actions. Termination of a joint venture, encompassing both dissolution and
acquisition, is an adjustment action of the parent ﬁrms. Dissolution is a decision to
withdraw investment and is associated with lowered evaluation of a joint venture.
Acquisition of a joint venture can be associated with increased evaluation of the joint
venture because it enables the parent ﬁrm to fully exploit the beneﬁts of the joint
venture. Acquisition can also be associated with lowered evaluation because it
provides the parent ﬁrm more control over the joint venture so as to improve the JV
situation. Therefore, JV termination is the parent ﬁrm’s adjustment actions based on
its changed evaluation of the joint venture. It is resultant from changes in the JV co-

evolutionary process.

14

2.2.2. Change during the co-evolutionary process drives JV termination

Joint ventures experience frequent changes over the life span (Day 1995). The
changes that unfold aﬁer JV formation induce the parent ﬁrms to reevaluate the joint
venture (Kumar and Nti 1998; Ring and Van de Ven 1994) and take readjustment
actions (Kumar and Nti 1998). Though the co-evolutionary process starts with the
initial conditions, the cycle of reevaluation and readjustment quickly brings the joint
venture away from the initial conditions (D02 1996). Changes after JV formation
invoke parent ﬁrm reevaluation and readjustment of a joint venture and are sources of
instability. Changes, rather than initial formation conditions, cause changed

evaluation and consequently increase the propensity of termination.

2.2.3. Parent ﬁrm overall strategy and the market environment have important
inﬂuences on JV termination

Joint ventures are subjected to environmental change and shift in parent ﬁrm
overall strategies (Kumar and Nti 1998). The parent ﬁrm’s reevaluation of the joint
venture also takes into consideration the market environment and the parent ﬁrm’s
over all strategy (Kumar and Nti 1998). Changes in the market environment and in
parent ﬁrm overall strategy are exogenous to a joint venture, but they alter the parent
ﬁrm’s assessment of the joint venture, and lead to adjustment actions. Thus, besides
the characteristics of the joint venture itself, factors that are exogenous to a joint
venture, such as parent ﬁrm overall strategy and the market environment, also have

important inﬂuences on JV termination.

15

2.2.4. JV Termination can be resultant from ﬁrm adaptation rather than failure

Same as the decision of forming a joint venture, the decision of terminating a
joint venture is a function of the parent ﬁrm’s overall strategy portfolio and market
competition. During the co-evolutionary process, the parent ﬁrm’s reevaluation of the
joint venture is based upon not only the performance of the joint venture but also the
alignment of the joint venture with parent ﬁrm strategy and the market environment.
Undesirable performance induces lowered evaluation of a joint venture. Changed
market environment alters the competitive position of a joint venture in the market,
and changed parent ﬁrm overall strategy alters the position of a joint venture in the
parent ﬁrm overall strategic portfolio, which both change the parent ﬁrm’s evaluation
of the joint venture and consequently increase the propensity of JV termination.
Therefore besides low performance, change in parent ﬁrm overall strategy and the JV
market environment is also a cause of JV instability and termination (Yan 1998).

If a joint venture is terminated due to changes in parent ﬁrm overall strategy or
in the market environment, it is a result of the parent ﬁrm’s adaptive activities, rather
than a failure. From the parent ﬁrm’s perspective, terminating ajoint venture may not
be a bad thing; on the contrary, it can be beneﬁcial to the parent ﬁrm’s overall
strategic development. Viewed within the context of the co-evolutionary process, JV
termination is not necessarily failure, and change-induced terminations are ﬁrm

adaptation activities rather than failure.

2.2.5. Commonalities and differences between JV dissolution and acquisition
A joint venture can be terminated through dissolution or acquisition.
Dissolution is the liquidation of a joint venture. Acquisition in this study is deﬁned as

when a joint venture is acquired by one of the parent ﬁrms. Both dissolution and

16

acquisition are resultant from changed reevaluation during the co-evolutionary
process. The mechanisms underlining these two types of JV termination have both
commonalities and differences.

For commonalities, both dissolution and acquisition terminates the
cooperation between parent ﬁrms, which eliminates the beneﬁts and costs of
cooperation. Beneﬁts of cooperation involve resource sharing, while costs of
cooperation include the exposure of ﬁrm capabilities to the partner and the risk of
having these valuable capabilities appropriated by the partner (D02 and Hamel 1998).
When a joint venture is dissolved or acquired, the parent ﬁrms are no longer involved
in COOperation; therefore they are not be able to enjoy the beneﬁts of resource sharing,
and at the same time no longer face the risk of appropriation.

Differences between the mechanisms of JV dissolution and acquisition also
exist. Both dissolution and acquisition can be adaptive actions of the parent ﬁrm, but
they are driven by different motivations and have different implications. From the
investment strategy perspective, dissolution is a divestment decision, while
acquisition is related to the parent ﬁrm’s expansion and increased resource
commitment in the JV business (Kogut 1991). From the perspective of governance
strategy, dissolution is the change of governance form from joint venture to market
transaction, while acquisition is the change of governance form from joint venture to
internal operation. Acquisition is the parent ﬁrm’s internalization decision to achieve
more control over the joint venture (Buckley and Casson 1996). In addition,
dissolution terminates a joint venture through liquidation, which usually involves
direct loss. In the case of acquisition, the joint venture stays in operation, which

enables the parent ﬁrms to avoid the loss of liquidating assets. Thus, to fully

17

understand the reasons and implications for JV terminations, it is necessary to

separately study JV dissolution and acquisition.

In sum, JV termination research within the co-evolutionary perspective

requires:

1.

Distinguishing between JV termination and failure. JV termination can result
from ﬁrm adaptation instead of failure.

Considering JV termination as a result of the co-evolutionary process and
study change and adaptation during the co-evolutionary process of a joint
venture with its market environment and parent ﬁrm overall strategy. This
involves studying joint ventures over time, by using longitudinal time series of
adaptation events and measures of rates of change or pace of change (Koza
and Lewin 1998; Lewin et al. 1999; Lewin and VoIberda 1999), and
incorporating a historical perspective by considering a long period of time
(McKelvey 1997).

Considering the embeddedness and co—evolution of a joint venture with parent
ﬁrm overall strategy and the market environment. This requires researchers to
incorporate exogenous factors related to parent ﬁrm overall strategy and the
market environment, not only a joint venture itself, in studying the reasons for
JV termination (Reuer and Koza 1998).

Distinguishing between and separately model JV dissolution and acquisition.
JV dissolution and acquisition both terminate a JV partnership, but different
mechanisms underline these two types of termination, and they have different

implications for the parent ﬁrm (Hennart et al. 1998; Reuer and Koza 1998).

2.3. Previous research on JV termination

The COO-evolutionary perspective has not yet been applied in JV termination
research. Many previous studies on JV termination have been empirically driven,
without explicitly employing a theoretical framework to develop a testable model
(e.g. Blodgett 1992; Hennart et al. 1998; Kogut 1989; Park and Ungson 1997). Other
studies have adopted theoretical perspectives including transaction cost theory
(Dhanaraj and Beamish 2004; Lu and Hebert 2005; Park and Russo 1996), interﬁrm
learning (Barkema et al. 1997; Dussauge et al. 2000; Parkhe 1991), real option theory
(Chi 2000; Kogut 1991), and social exchange (Steensma and Lyles 2000). However,
these theoretical perspectives do not explicitly incorporate the evolutionary view. For
example, transaction cost theory identiﬁes equity control as an element of governance
structure, but does not explicitly address how governance structure may change and
therefore inﬂuence the outcome of a joint venture. The lack of a co-evolutionary

perspective has limited previous JV termination research to:

#

. Viewing JV termination as analogs to failure.

I"

Focusing on JV initial formation conditions and neglecting over time

change.

3. Neglecting factors exogenous to a joint venture, such as parent ﬁrm
overalls strategy and the market environment.

4. Not separately modeling JV dissolution and JV acquisition.

5. Empirical tests being mainly cross-sectional with static variables,

rarely considering over time effects.

2.3.1. JV termination and failure

JV duration and survival are widely used as performance measures (Reuer and
Koza 1998), assuming that long-lasting ventures are successful and short-lived ones
are failures. Though some researchers recognized the difference between JV
termination and failure (Yan 1998; Yan and Zeng 1999), termination’s relation to
failure is not well studied. Park and Russo (1996) deﬁned failure as JV dissolution or
JV spun-off to third parties, but do not include acquisition by one partner. While this
is a practical deﬁnition, it does not provide theoretical insight into the relationship

between termination and failure.

2.3.2. Focus on initial formation conditions

Due to the lack of a perspective of change in theorization and the limitation of
data availability, previous research has focused on relating the JV outcomes to initial
characteristics of the joint venture or its parents (D02 1996), such as JV type, whether
parents are direct competitors, etc. Table 1 presents a summary of key JV termination
studies. Even some of these characteristics may change over time, usually only the
initial conditions are included in the model, without considering changes over time
after formation. For example, parent experience, culture difference, differences in
parent age and size, etc, may change aﬁer a joint venture is formed, but in most
studies, only the conditions at JV formation are used.

Focusing on initial conditions assumes that initial conditions drive the
evolution of a joint venture, and if a joint venture is wrongly conﬁgured, the
misspeciﬁcation can not be compensated in the following process of evolution (D02
1996). However, changes frequently occur during the cooperation process, and ﬁrms

are involved in reevaluation and readjustment of the joint venture, which brings the

20

joint venture away from initial conditions (Doz I996). The changes in the adjustment

process have important inﬂuences on the outcome of a partnership (Arino and de la

Torre 1998; Doz 1996). Focusing only on initial conditions underestimates the

inﬂuence of changes and creates a deterministic bias (Arino and de la Torre 1998;

D02 1996).

Table 1. Previous studies on joint venture termination

 

 

 

 

 

 

Time-
dependent
Article Dependent variable Independent variables variables
Parent age, parent size. and JV
Lu and Xu 2006 Survival industry relatedness. No
Lu and Hebert 2005 Survival Equity control No
Dhanaraj and
Beamish 2004 Survival Equity ownership No
Hennart and Zeng 2002 Longevity National culture difference No
lnterﬁrrn rivalry,
Park and Ungson 2001 Dissolution managerial complexity. Conceptual

 

Chi 2000

Dussauge. Garrette

Acquisition and divesture

Reorganization,
takeover, dissolution.

Value ofthe JV option

Conceptual

 

and Mitchell 2000 and continuity Scale versus linkage alliances No
Management control
imbalance, ownership control

Steensma and Lyles imbalance, managerial support.

2000 Survival technical support. No

 

Tas and Teng 2000

Hennart, kim and zeng
I998

Instability = unplanned
major changes or
dissolution

Longevity

Differences between
cooperation level and
competition level.
Differences between rigidity
level and ﬂexibility level.
differences between
short-term orientation and
long-term orientation.

JV related product, JV formed
through acquisition. parent
experience, market growth in
JV sector. parent size.

Conceptual

NO

 

Barkema, Shenkar,
Vermeulen and Bell 1997 Longevity Experience, culture distance. No

Culture distance, differences in
strategic scope, size and age.
direct competitors, overlap in
product market scope between
JV and the partners,
involvement of technology
Park and Ungson 1997 Dissolution transfer. No

Direct competitors; pattern of
interdependence (integrative
Dissolution and spun-off vs. sequential); past

 

to third parties, not experience; multiple linkage

including acquisition by between partners; number of
Park and Russo 1996 one of its partners partners. No
Barkema and
Vermeulen 1997 Longevity Culture difference No

 

Industry concentration, annual
industrial growth, annual
Acquisition, including growth residual, JV type
selling to a third party (Manufacturing, marketing and
Kogut 1991 and to the partner. R&D ). Yes

 

Parkhe 1991 Longevity Interﬁrm diversity No

Number of ties between
Instability = termination, parents, R&D intensive,

 

including dissolution, marketing intensive, scale
sold out to the partner intensive, market growth,
Kogut 1989 or a third party. market concentration change. Yes
Partner asymmetry,
Harrigan 1988 Survival and duration diversiﬁcation relatedness. No

 

Ownership control, parent size.
market growth, market
concentration, marketing
Kogut 1988 Termination activity. No

Parent corporate
Instability = holdings of characteristics, including
the MNE cross the 50% organizational stage, foreign
or 95% ownership lines, experience, percentage of
or the interests of MNE foreign sales, total sales,
are sold, or the venture is advertising intensity, product
F ranko 1971 liquidated. tradability, etc. No

 

2.3.3. Neglecting of exogenous factors
Co-evolutionary theory indicates JV evolution takes place not only within a

joint venture but also together with its market environment and parent ﬁrm overall

22

strategy. Factors under JV management control, such as scope of JV activities, are
directly related to the evolution and termination of a joint venture. Factors concerning
the market environment and parent ﬁrm overall strategy are exogenous to a joint
venture, but they also have inﬂuences on JV evolution and consequently the JV
outcome. Parent ﬁrm overall strategy is the overall strategic plan of the parent ﬁrm,
for instance, overall investment plan or product portfolio, and it is not concerned with
the joint venture itself. Thus, parent ﬁrm overall strategy is exogenous to a joint
venture. This study categorizes the factors inﬂuencing JV termination into two
categories: factors under control of JV management and exogenous factors, i.e.,
factors exogenous to a joint venture and out of the control of JV management.
Previous research has investigated factors on the characteristics of a joint
venture itself, such as JV type, JV activities and number of partners; the differences or
relationships between parents, such as culture difference, whether parents are direct
competitors, parent differences in age, size and scope; the relationship between
parents and a joint venture, such as product relatedness and equity control; and the
market environment, such as JV industry grth and market concentration.
According to this classiﬁcation, factors on the characteristics of a joint
venture, the differences or relationships between parents, and relationship between
parent and a joint venture are all elements of JV management and under control of JV
management. Understandably, factors under control of JV management have been the
focus of investigation. The only studies that incorporated market environment
variables are Kogut (1991; 1989). The inﬂuence of parent ﬁrm overall strategy is
rarely considered. In an early study of multinational corporations and international
joint ventures by Franko (1971), parent ﬁrm characteristics such as percentage of

foreign sales, advertising intensity were found to inﬂuence the JV outcome. Franko

23

(1971) pointed out that multinational ﬁrms’ decisions on international JV terminations
depend on the multinational ﬁrms’ international business strategies. However, this
line of research that relates JV outcomes to parent ﬁrm overall strategy was not well
extended by later studies.

Looking at JV management factors reveals the managerial complexity of joint
ventures and provides important explanations for JV termination. However, it
neglects the inﬂuences of factors exogenous to JV management, such as parent ﬁrm
overall strategy and the market environment, which also have important inﬂuences on
JV termination. Joint ventures are embedded in and evolve with partner ﬁrm overall
strategies and the market environment (Koza and Lewin 1998). From the parent
ﬁrms’ strategic viewpoint, a joint venture may be terminated because the conditions
supporting the existence of the joint venture have changed. Recognizing the important
inﬂuence of exogenous factors on JV termination can help clarify the relationship
between JV termination and failure, because termination due to exogenous factors out
of JV management control is not related to failure.

Table 2 summaries the JV termination factors studied in previous research
within the classiﬁcation of change versus initial formation conditions, and factors

under JV management control versus factors exogenous to a JV.

24

Table 2. A classiﬁcation of explanatory variables in previous JV termination
studies

 

 

 

 

 

Initial formation conditions Changes after
formation

JV type;

Number of partners;

JV experience;

Ownership distribution among

partners; Case studies on

Involvement of director internal cooperative
Factors under control of . . .

competitors; and competitive
JV managers .

Involvement of technology dynamics between

transfer; partnersz.

Partner asymmetry in terms of

age, size and market overlap;

Cultural distance;

Multiple linkage between

partners].

Industry growth; Industry growth;
Factors exogenous to a JV Industry concentration; Industry

Parent ﬁrm age and size3. concentration“.

 

 

 

1. Dussauge, Garette and Mitchell 2000; Barkema, Shenkar, Vermeulen and Bell
1997; Park and Ungson 1997; Park and Russo 1996; Blodgett 1992; Kogut 1989;
Harrigan 1998.

2. Doz 1996; Ring and Van de Ven 1994; Hamel 1991.

3. Hennart, Kim and Zeng 1998; Kogut 1991; Kogut 1989.

4. Kogut 1991; Kogut 1989.

In sum, the focus on initial JV formation conditions and the neglecting of
exogenous factors has limited our understanding of JV termination. One consequence
is that external change is largely neglected as a possible reason for JV termination.
The literature relates JV termination to low performance and considers termination as
analogous to failure. Without considering external change, the previously found
relationship between low JV performance and JV termination may be spurious.
Investigating changes of JV conditions over time can provide new understanding of

the relationship between termination and failure, as termination may not be a problem

if it represents adaptations to changes (Gomes-Casseres 1987).

25

2.3.4. Different types of JV termination

Most previous studies do not separate the analyses across different types of
terminations. These studies employ duration, longevity or survival as the dependent
variable, not considering how a joint venture is terminated (i.e., dissolved, or acquired
by a parent ﬁrm) (Barkema et al. 1997; Lu and Hebert 2005). The few exceptions that
took termination type into consideration are Kogut (1991), who studied only
acquisition, including acquisition by a partner ﬁrm and by a third party, and Park and
Russo (l 996), who excluded acquisition by the parents from their analysis. The only
study that simultaneously investigated both dissolution and acquisition is by Hennart
et al (1998), who studied exit through liquidation and sales for Japanese stakes in US.

How a joint venture is terminated is related to the reason why it is terminated.
and different types of termination reﬂect different considerations of the parent ﬁrms.
Focusing only on general behaviors without investigating speciﬁc behavior is not
sufﬁcient to understand the reasons for these behaviors and may cause bias (Warshaw
l980b; Warshaw 19803). The importance of distinguishing and simultaneously
considering different types of JV termination not only falls out of the perspective of
viewing joint ventures within the context of parent ﬁrms’ overall strategy (Reuer and
Koza 1998), but also lies in understanding the reasons of termination and how they

are related to joint venture failure (Hennart et al. 1998).

2.3.5. Methods for empirical testing
The majority of JV termination studies have been cross-sectional. With the
exception of Kogut (1991; 1989), who investigated time-dependent variables such as

market grth and market concentration, other empirical studies have included only

26

static variables. The dominance of static variables prevents empirical studies from
examining the inﬂuence of the co-evolutionary process on JV outcome.

Logistic regression and event history analysis have been used in empirical
testing. While logistic regression models the occurrence of an event, it does not
consider the timing of the event, i.e., a joint venture that is terminated one year after
formation is treated same as a joint venture terminated ﬁve years after formation,
which omits important information of termination. Event history analysis not only has
the advantage of considering both the occurrence of termination and the timing of
termination at the same time, but also allows for the modeling of event probability
over all observation periods, and can readily incorporate time-dependent covariates to
capture the effects of non-static explanatory variables over time. However, only a
few recent JV termination studies have applied this method, and due to the lack of
time-dependent covariates, the advantage of event history models in incorporating
over time change was not fully utilized.

This study tests co-evolutionary theory of alliances in the context of JV
termination by developing and empirically testing an event history model with a
longitudinal research design. It addresses the needs for a co-evolutionary perspective
in JV termination research by:

1. Explicitly considering changes during the co—evolutionary process as
determinants of JV termination. It also incorporates an historical perspective
by covering the JV life span, starting from its formation and ending at its
termination.

2. Incorporating antecedent factors exogenous to a joint venture, including parent
ﬁrm overall strategy and the market environment.

3. Separately modeling and comparing JV dissolution and acquisition.

27

Distinguishing between JV termination and failure by showing JV termination
due to external change is ﬁrm adaptation rather than failure.
Utilizing event history analysis and a longitudinal research design with time-

dependent covariates to examine the inﬂuences of change over time.

28

CHAPTER 3
THE CONCEPTUAL MODEL AND HYPOTHESES
3.1. The conceptual model

This dissertation study applies co-evolutionary theory of alliances in the
context of joint ventures and develops a model of JV termination. JV termination is
viewed as resultant from the co-evolutionary process of a joint venture with its market
environment and parent ﬁrm overall strategy. Terminating ajoint venture is an
adjustment action of the parent ﬁrm based upon its changed evaluation of the joint
venture, which is induced by changes in the co-evolutionary process.

Change in market competition and change in parent ﬁrm overall strategy are
incorporated as determinants of the propensity of JV termination. The model not only
applies co-evolutionary theory of alliances by incorporating JV embeddedness and
change variables, but also connects the theory to previous theoretical development in
alliance research by explicitly considering two types of parent ﬁrm strategies, i.e.,
investment strategy and governance strategy. Figure 1 presents the conceptual

framework.

29

 

ooanotom 8.8 a .

 

vengebukx:

 

 

maﬁﬁogn wESanu mo aosuncom .
Eon—tun grep—83 E 03885 .
gnaw—moans we ﬁt 5 032qu .

 

 

 

 

 

 

 

 

R955. uuauEuEa
. zasuae «Embauuek 5 mam-3.5
V
5:65ch . Ego 02:35.5 . 0
eons—omega . 32:82 Evan E 03205 . 3
mi Ema on: 838m .
acacia—huh. >H
we Emaonehm Regan “5529—5

 

 

_ -E§o ELKEESN S cute—B

 

 

nomwubcooeoo 8x58 5 3.355 .
53on Hod—32 .

 

 

 

 

a Loumaoaaooéoohwéoz
.m> Bataan—8-80:9 .
>H coca—2:: .m> none—om .

 

EOEOEES “9.25: 5 OMSED

 

 

 

 

 

25:2.

 

 

53:35.3 25:?» E33: .38.: < A v.5»:—

The dependent variable in the model is the propensity of JV termination. Two
types of JV termination are considered: dissolution and acquisition. Dissolution is
when a joint venture is liquidated, and acquisition is when a joint venture is acquired
by one of the parent ﬁrms.

According to co-evolutionary theory of alliances, ajoint venture, as a special
form of alliances, evolves together with its market environment and the parent ﬁrm’s
overall strategy. Therefore, both change in market competition and change in parent
ﬁrm overall strategy affect the propensity of JV termination.

Speciﬁcally, change in market competition alters the competitive position of a
joint venture in its market and makes the joint venture more or less effective for the
parents to achieve their strategic goals (Kumar and Nti 1998). This leads to the parent
ﬁrm’s changed evaluation of the joint venture, and consequently increase the
prOpensity of dissolving or acquiring the joint venture (Doz 1996; Franko 1971;
Kumar and Nti 1998). Thus, change in competition in the JV market has an impact on
the propensity of JV termination.

Two types of parent ﬁrm overall strategies are considered, i.e., investment
strategy and governance strategy. For the parent ﬁrms, 3 joint venture is both an
investment decision and a governance decision (Buckley and Casson l998a; Buckley
and Casson l998b; Cristina Lopez and Esteban 2004). A joint venture is formed when
the parent ﬁrm invests in a new production, a new marketing program or a new
research project etc. The new investment is part of the parent ﬁrm’s overall
investment strategy. At the same time, an investment can be carried out through forms
other than a joint venture, such as internal operation through wholly-owned
subsidiaries (Anderson and Gatignon 1986; Buckley and Casson l998a). The decision

of internal operation or partnering with other ﬁrms is a governance choice, and it is

31

part of the parent ﬁrm’s governance strategy (Buckley and Casson l998a; Buckley
and Casson l998b). Similarly, termination of a joint venture may be due to the parent
ﬁrm’s motivation of withdrawing the investment or changing the governance form,
i.e., internalizing the JV activity or switching to a different JV partner. Therefore two
fundamental parent ﬁrm strategies, investment strategy and governance strategy, need
to be considered.

JV performance is included as an antecedent not only because JV performance
directly inﬂuences a ﬁrm’s decision of continuing or terminating a joint venture, but
also because JV performance is representative of the factors under JV management
control. While changes in the market environment and parent ﬁrm overall strategy are
exogenous to a joint venture, JV performance captures the ultimate result of JV
management and therefore the inﬂuences of factors under control of JV management.
JV characteristics such as number of partners, differences or relationships between
parents such as culture difference and parent asymmetry, and the relationship between
parents and a joint venture such as equity control, all inﬂuence JV performance, and
their inﬂuences on JV outcome can be represented by JV performance.

Further, by including JV performance in the model, this study is able to look at
the inﬂuence of exogenous factors given JV performance. If with the effect of JV
performance controlled for in the model, exogenous change factors are found to
signiﬁcantly inﬂuence the propensity of JV termination, then it can be concluded
external change is an important factor and should not be neglected.

In addition, JV performance represents the extent to which a joint venture has
fulﬁlled its intended beneﬁts. Unfulﬁlled beneﬁts are directly related to the deﬁnition
of failure (Park and Ungson 2001), as opposed to change in market competition and

parent ﬁrm overall strategy, which is a result of parent ﬁrm adaptive actions.

32

Therefore the inclusion of JV performance also enables us to distinguish between JV
terminations and failure by showing different reasons for termination.

Thus, four sets of factors are included as determinants of the propensity of JV
termination: change in market competition, change in parent ﬁrm overall investment
strategy, change in parent ﬁrm overall governance strategy, and JV performance.
These four set of factors inﬂuence the parent ﬁrm’s reevaluation and readjustment of
ajoint venture and consequently the propensity of JV termination.

Based upon the commonalities and differences between JV dissolution and
acquisition, different hypotheses for the propensity of dissolution and acquisition are
developed when necessary. For commonality, both dissolution and acquisition
terminates the cooperation between parent ﬁrms, which eliminates both the beneﬁts of
cooperation, such as resource sharing, and costs of cooperation, such as risk of having
capabilities appropriated by the partner (Doz and Hamel 1998). For differences, from
the investment strategy perspective, dissolution is a divestment decision, while
acquisition indicates the parent ﬁrm’s expansion actions (Kogut 1991). From the
governance strategy perspective, dissolution is the change of governance form from
joint venture to market transaction, while acquisition is the change of governance
form from joint venture to internal operation, and it is the parent ﬁrm’s internalization
decision to achieve more control over the joint venture. These rationales are applied in
hypothesis development.

Further, the model considers JV type as a moderator, and proposes that the
effects of antecedent factors on the propensity of JV termination are different for
different types of joint ventures. Two classiﬁcations of joint ventures are considered:
related JV versus unrelated JV; direct-competitor JV, when the parents are direct

competitors, versus non-direct-competitor JV, when the parents are not direct

33

competitors. The following section ﬁrst develops hypotheses for the main effects of

the antecedent factors, and then proceeds to the moderating effects of JV type.

3.2. Hypotheses

Hypotheses on the effects of antecedent factors on the propensity of JV
termination are developed in this section. Hypotheses are developed regarding the
change variables. The effects of level variables are controlled for when it is necessary.
Different hypotheses are developed for JV dissolution and acquisition when

necessary.

3.2.1. Change in market competition

Joint ventures compete with other ﬁrms in the market. Market competition in
this study refers to competition in the industry of the joint venture, not that of the
parent ﬁrms. To examine change in market competition, this study combines the co-
evolutionary theory view with industry organization literature, which has a long
history of studying market competition in an industry (Porter 1985; Porter 1980;
Scherer 1980). In the industry organization literature, two fundamental characteristics
used to describe market competition are market growth and market concentration

(Caves 1980; Caves and Porter 1980; Horowitz 1984).

Market growth

Market growth indicates increasing market demand in an industry, and it is an
important consideration for ﬁrms’ market entry and exit decisions (Horowitz 1984).
Joint ventures are formed by parent ﬁrms to explore new opportunities in the JV
industry (Park et al. 2002). Joint ventures in a fast growing market enjoy greater

market potential and more future opportunities.

34

 

During the JV co-evolutionary process, the level of market growth in ajoint
venture’s market may vary, and it inﬂuences the parent ﬁrm’s reevaluation and
readjustment of the joint venture. Through the reevaluation and readjustment process,
a joint venture co-evolves with its market. When market demand is quickly growing,
the parent ﬁrm develops positive evaluation of the joint venture, because increasing
market demand brings more potential customers and future opportunities for
expansion, and a joint venture in a fast-growing market is less likely to go bankrupt
(Hennart et al. 1998). The increased positive evaluation of a joint venture reduces the
propensity of dissolving the joint venture. Therefore market growth has a negative
impact on the propensity of JV dissolution. Formally stated:

H I a. Market growth is negatively associated with the propensity of J V

dissolution.

Joint ventures are used as a way to expand to new markets, and ﬁrms are ready
to increase resource commitment when market condition is favorable (Kogut 1991 ).
High market grth signals favorable market conditions. In presence of high market
growth, the parent ﬁrm develops positive evaluation of the joint venture investment,
and internal operation becomes a more attractive alternative because internalizing the
JV activity enables the parent ﬁrm to solely exploit the beneﬁts from the joint venture
(Kogut 1991). Acquiring the joint venture is an adjustment action taken by the parent
ﬁrm to fully exploit the beneﬁts of the fast growth in the JV market. Therefore parent
ﬁrms are more likely to acquire a joint venture in a fast growing market. Formally
stated:

H] b. Market growth is positively associated with the propensity of J V

acquisition.

35

Increase in market concentration

Market concentration is the level of seller concentration in a market. It is
associated with the level of rivalry between existing competitors, which is a key
element of the competitive forces in a market (Porter 1985; Porter 1980). It also
reﬂects the size distribution of ﬁrms (Caves 1980), with higher level of market
concentration indicating the concentration of market power to a few large ﬁrms. In a
highly concentrated market, ﬁrms are faced with large and powerful competitors, and
the minimum scale needed for a ﬁrm to compete in the market is larger (Caves 1980).

During the JV co-evolutionary process, concentration level of the JV market
may change. When there is an increase in market concentration and market power is
more concentrated to a few ﬁrms, the joint venture’s competitive position in the
market is changed, which induces the parent ﬁrm’s reevaluation and readjustment
actions of the joint venture. When market power is more concentrated to a few ﬁrms,
the joint venture is faced with more difﬁcult competitive environment and its
competitive position in the market is weakened, unless increase in market
concentration ratio is directly caused by growth of the focal joint venture, which is not
a concern in this study because of the simultaneous consideration of JV sales
performance in a longitudinal study design. The weakened competitive position
induces the parent ﬁrm to lower its evaluation of the joint venture. Given the
existence of other alternative investment opportunities and the cost of supporting the
joint venture, the possibility of dissolving ajoint venture is increased. Therefore, with
other conditions the same, the propensity of JV dissolution is increased when there is
an increase in market concentration in the JV market. Formally stated:

HZa. Increase in market concentration is positively associated with the

propensity of J V dissolution.

36

When a market is more concentrated to a few ﬁrms, the average ﬁrm size is
increased and the minimum scale needed to participate in competition is higher
(Caves 1980). The parent ﬁrm’s evaluation of a joint venture with respect to
internalization is changed, and internalization becomes a more attractive alternative,
because integrating the JV activity and directly participating in competition in the JV
market enables the parent ﬁrm to more effectively utilize the parent ﬁrm’s market
power and strengthen the joint venture’s competitive position in the market. As a
readjustment action, the parent ﬁrm is more likely to acquire and achieve full control
of the joint venture. Therefore:

H2b. Increase in market concentration is positively associated with the

propensity of J V acquisition.

3.2.2. Change in parent ﬁrm overall investment strategy

F irms’ investment strategies have been a central topic of research for strategic
management. Products and resources are viewed as two sides of the ﬁrm (Wemerfelt
1984). Product line management and resource allocation within a ﬁrm are both among
the most important investment decisions facing top management (Bergh 2001;
Wemerfelt 1984). They are both directly related to joint ventures because ajoint
venture is part of the parent ﬁrm’s overall investment portfolio. Changes in the parent
ﬁrm’s product line or resource availability alter the position of a joint venture in the
parent ﬁrm’s overall investment portfolio, and induce the parent ﬁrm to reevaluate the
joint venture and consequently take readjustment actions regarding the existence of
the joint venture. Therefore, changes in parent ﬁrm product lines and resources are

both included as antecedent factors inﬂuencing the propensity of JV termination.

37

In addition, disruptive events to the parent ﬁrm, such as acquisition of the
parent ﬁrm by another company, also lead to changes in the parent ﬁrm’s overall
investment strategy and induce adjustment decisions regarding the existence of a joint
venture. Therefore disruptive events are also considered as a determinant of the

propensity of JV termination.

Product line shift

Product diversiﬁcation has been a popular strategy pursued by many ﬁrms
(Ramanujam 1989). Firms engage in active product line management, developing
and implementing plans that concern which business to be in and which to avoid (Day
1977). Product line management activities include supporting newly introduced
products, consolidating the product range to focus on one segment, and divesting
products, etc. (Day 197 7). Product line shift in this study is deﬁned as a shift in the
primary business a company is in. When a company’s primary business changes, its
overall product portfolio is affected, with some products receiving more support while
others less, and some products being withdrawn from the market (Burton 1994;
Devinney and Stewart 1988).

Joint ventures are used as strategic vehicles to more efﬁciently explore new
market opportunities (Contractor and Lorange 1988; Gomes-Casseres 1987). The JV
product is part of the parent ﬁrm’s product portfolio (Kogut 1988), and JV
termination is related to the parent ﬁrms’ investment or divestment decisions
regarding the JV product. During the JV co-evolutionary process, the parent ﬁrm
experiences product line shift, which alters the JV product’s relative position in the
parent ﬁrm’s overall product portfolio and changes the parent ﬁrm’s evaluation of the

joint venture.

38

 

 

Parent ﬁrm product line may shift away or closer to the JV business. The shift
of parent ﬁrm primary business away from the JV business reduces the relative
importance of the joint venture in the overall product portfolio. Given that different
products share limited resources and even same customer base under some
circumstances (Bergh 2001; Burton 1994), lowered evaluation of the joint venture
increases the propensity of the parent ﬁrm dissolving the joint venture to focus on
core business. The shift of the parent ﬁrm’s primary business closer to the joint
venture business increases the relative importance of the joint venture to the parent
ﬁrm, and increased evaluation of the joint venture increases the propensity that the
parent ﬁrm acquires the joint ventures to achieve full control over its operation
(Buckley and Casson 1996). Therefore, when the parent ﬁrm is involved in product
line shift, the likelihood of terminating a joint venture is higher, either dissolution or
acquisition, as compared to when there is no such change. Formally stated:

H3a. Parent ﬁrm product line shift is positively associated with the propensinr

of JV dissolution.

H3b. Parent ﬁrm product line shift is positively associated with the propensity

of JV acquisition.

Increase in parent ﬁrm resources

Looking at ﬁrms from the perspective of their resource endorsement has a
long tradition in economics, and the traditional concept of corporate strategy is also
phrased in terms of the resource position of a ﬁrm (Andrews 1974). The more recent
resource based view extends the concept of resources from the traditional physical
resources to intangible resources and emphasizes the importance of rare resources for

ﬁrm performance (Peteraf 1993; Wemerfelt 1984).

39

From the resource perspective, joint ventures are formed to access and
combine resources of different ﬁrms (Doz and Hamel 1998; Dyer and Singh 1998).
Parent ﬁrm resources, though not directly invested in the joint venture, represent the
potential of resource sharing in a joint venture (Doz and Hamel 1998) because the
parent ﬁrm may decide to bring more resources into the joint venture at later stage of
the cooperative process. Further, even if parent ﬁrm resources are not directly
invested in a joint venture, parent ﬁrms can beneﬁt from the transfer of intangible
resources through their interaction in a joint venture (Chung 2001). Joint ventures
often utilize the same manufacturing facilities and technology as the parents, share
same brand name or marketing program with the parents, or carry out related R&D
programs. Through the coordination of such activities, parent ﬁrms transfer their
intangible resources to the joint venture and learn from each other (Chung 2001).

During the JV co-evolutionary process, both the joint venture and the parent
ﬁrms experience changes. Besides managing a joint venture, the parent ﬁrm also
proceeds with its overall strategic development and its resources may increase or
decrease during the process. Increase in parent ﬁrm resources increases the potential
of resource sharing and learning between JV partners, and consequently raises the
parent ﬁrm’s evaluation of the joint venture. Increased evaluation enhances the
partner’s willingness to continue the partnership and lowers the propensity of
termination, including dissolution and acquisition, because both dissolution and
acquisition of a joint venture terminates the cooperation between partners and
deprives the opportunity of resource sharing and learning.

Speciﬁcally, this study considers three types of resources that are shared
during the JV process, i.e., manufacturing resource, marketing resources and R&D

resources. Manufacturing resources are resources utilized during the production

40

process, such as buildings, machines, equipments, and related technologies and
procedures, etc. The parent ﬁrm’s manufacturing resources are not directly invested in
a joint venture, but they can be shared with the joint venture, and manufacturing
knowledge, such as operation management procedures, can be transferred from the
parent ﬁrm to the joint venture during the process of cooperation. Increase in parent
ﬁrm manufacturing resources raises the potential for sharing these resources and
learning manufacturing knowledge in the partnership. Thus, during the co-
evolutionary process, increase in parent ﬁrm manufacturing resources increases the
parent ﬁrm’s evaluation of the joint venture and consequently reduces the propensity
of dissolving or acquiring the joint venture, as both dissolution and acquisition
terminates the opportunity of resource sharing and learning between JV parents.
Therefore:

H4a. Increase in parent ﬁrm manufacturing resources is negatively

associated with the propensity of JV dissolution.

H4b. Increase in parent ﬁrm manufacturing resources is negatively

associated with the propensity of J V acquisition.

Marketing resources are resources utilized to carry out marketing activities,
such as market research, promotion programs, advertising, etc. Marketing resources of
the parent ﬁrms are not directly invested in a joint venture, but because a joint venture
often shares the same brand name or marketing program with its parent, marketing
resources invested in the parent ﬁrms reinforce the marketing image of the joint
venture product therefore provide support to the joint venture. More importantly, the
parent ﬁrm’s knowledge on marketing activities or the market itself can be transferred
to the joint venture through management interaction during the process of

cooperation. During the co-evolutionary process, increase in parent ﬁrm marketing

41

resources increases the potential for sharing these resources and learning marketing
knowledge in the partnership, and raises the parent ﬁrm’s evaluation of the joint
venture, which consequently reduces the propensity of dissolving or acquiring the
joint venture, as both dissolution and acquisition terminates the opportunity of
resource sharing and learning between JV parents. Therefore:

H5a. Increase in parent ﬁrm marketing resources is negatively associated

with the propensity of J V dissolution.

H5b. Increase in parent ﬁrm marketing resources is negatively associated

with the propensity of acquisition.

Parent ﬁrm R&D resources can also be shared with a joint venture because the
joint venture often carries out similar research and development activities or share
similar technological bases. The intangible R&D knowledge can be transferred to a
joint venture through management interaction during the cooperative process.
Increase in parent ﬁrm R&D resources increases the potential for knowledge transfer
between parents in the JV partnership, and consequently raises the parent ﬁrm’s
evaluation of the joint venture. Therefore, during the co-evolutionary process, an
increase in parent R&D resources reduces the propensity of dissolving or acquiring
the joint venture, as both dissolution and acquisition terminates the opportunity of
resource sharing and learning between JV parents. Therefore:

H6a. Increase in parent ﬁrm R&D resources is negatively associated with the

propensity of J V dissolution.

H61). Increase in parent ﬁrm R&D resources is negatively associated with the

propensity of J V acquisition.

Further, among the three types of resources, R&D resources are of a higher

degree of ambiguity because R&D activities involve complex technologies and are

42

exploratory in nature (Cummings and Teng 2003). Complexity and ambiguity of
knowledge reduces the effectiveness of knowledge transfer (Nonaka 1994; Simonin
1999; Spender 1996), therefore transfer of parent ﬁrm R&D knowledge is relatively
more difﬁcult than manufacturing and marketing resources. Because of the difﬁculty
in transferring R&D resources, though increase in parent ﬁrm R&D resources
increases the potential for resource sharing and learning between JV parents, the
potential increase is not as much as for manufacturing or marketing resources. Thus
for the same resource increase, the parent ﬁrm’s evaluation of the joint venture does
not rise as much for R&D resources as for manufacturing and marketing resources.
Therefore, it is hypothesized that the effects of increase in parent ﬁrm R&D resources
on the propensity of JV termination is not as strong as the effects of increase in
manufacturing and marketing resources. The same argument applies to both JV
dissolution and acquisition. Formally stated:
H 7a. The eﬂect of increase in parent ﬁrm R&D resources on the propensity of
JV dissolution is not as strong as the effects of increases in parent ﬁrm
manufacturing and marketing resources.
H 7b. The effect of increase in parent firm R&D resources on the propensity of
JV acquisition is not as strong as the eﬂects of increases in parent/inn

manufacturing and marketing resources.

Disruptive event

During the JV co-evolutionary process, the parent ﬁrm may experience some
events that are disruptive to its overall strategy, such as being acquired by or merged
with other ﬁrms. When a parent ﬁrm experiences strategically disruptive events, the

overall strategy of the parent ﬁrm is subjected to change. The resultant change in

43

parent ﬁrm overall strategy induces the parent ﬁrm to reevaluate the joint venture’s
relative position in its overall strategic portfolio and take adjustment actions, which
inﬂuences the existence of a joint venture. Through this reevaluation and readjustment
process, the joint venture co-evolves with its parent ﬁrm’s overall strategy.

This study considers one type of disruptive event: when the parent ﬁrm is
acquired by or merged with another ﬁrm. When merger and acquisition (M&A)
happen to the parent ﬁrm, adjustment is required for the new ﬁrm to integrate
previous business. Post merger integration involves changes from operation to
management and strategy, and often with replacement of the top management team
(Goh 2001; 20110 and Singh 2004). During the integration process parent ﬁrm overall
strategy experiences signiﬁcant changes, which alter the necessity or importance of a
joint venture in the parent ﬁrm overall strategic plan. The changed evaluation of a
joint venture puts the existence of a joint venture into question. The propensity of
dissolving or acquiring the joint venture is increased compared to before the M&A
event. Formally stated:

H8a. The propensity of J V dissolution is increased when the parent company

is acquired by or merged with another company.

H8b. The propensity of JV acquisition is increased when the parent company

is acquired by or merged with another company.

3.2.3. Change in parent ﬁrm overall governance strategy

A joint venture is not only an investment for the parent ﬁrm, but also a form of
governance for the investment activities (Buckley and Casson 1996). When a ﬁrm
invests in a certain product or activity, it can choose from different governance forms

including internalization, market contracts, and joint ventures (Buckley and Casson

44

1996). Firms engage in adjusting their governance strategies to achieve the optimal
governance form. The termination decision of a joint venture is part of the parent
ﬁrm’s governance adjustment, as both JV dissolution and acquisition are changes of
governance forms. Dissolution is a change from joint venture to market transaction,
and acquisition is a change from joint venture to internalization. Both dissolution and
acquisition of a joint venture terminates the cooperation between JV parents.

As the classic theory in governance literature, transaction cost theory explains
ﬁrms’ governance decisions in interﬁrm partnerships with the match between
governance form and exchange attributes so as to reduce transaction cost (Rindﬂeisch
and Heide 1997; Williamson 1985; Williamson 1975). Due to the existence of
opportunism, in a partnership, self-interest seeking ﬁrms may involve in activities that
are detrimental to the other partner, such as appropriating the technology of the
partner. The risk of having its technology appropriated by the partners contributes to
transaction cost and inﬂuences on ﬁrm’ governance choice. Therefore risk of
appropriation is considered as a governance strategy variable in the model.

Factors that inﬂuence the level of dependence between partners are also found
to have impacts on ﬁrms’ governance decisions. High level of dependence enhances
trust, commitment and long-term orientation in partnerships (Ganesan 1994;
Goodman and Dion 2001), which serves a form of relational governance (Heide and
John 1992). Dependence contributes to the stability and continuity of a partnership
(Ganesan 1994; Goodman and Dion 2001; Kumar et al. 1995). Two factors that
inﬂuence the level of dependence, i.e., alternative partners and competing
partnerships (Heide and John 1988; Pfeffer and Salancik 1978), are included as

governance variables in the model.

45

Increase in risk of appropriation

When two ﬁrms enter a joint venture, their unique knowledge or capabilities
are exposed to the risk of appropriation by the partner (Hamel 1991). During the JV
co-evolutionary process, the level of potential risk of appropriation may change as the
parent ﬁrms’ technology structures change. Though the parent ﬁrms’ decisions of
entering the joint venture indicate their acceptance of a certain level of risk, change in
risk of appropriation after JV formation induces the parent ﬁrms to reevaluate the
situation and adjust the governance form accordingly.

An increase in risk of appropriation indicates more need for the parent ﬁrm to
protect its knowledge and higher level of transaction cost for the joint venture. When
risk of appropriation increases, the parent ﬁrm’s reevaluation of the joint venture as
compared to other governance forms, such as market transaction or internalization, is
lowered, because other governance forms prevent the exposure of ﬁrm knowledge to
the partner. In presence of an increase in risk of appropriation, the parent ﬁrm is more
motivated to change the governance form from joint venture to market transaction or
internalization, which results in dissolution and acquisition of the joint venture
respectively. Thus, the propensity of dissolving or acquiring a joint venture is
increased when the risk of appropriation in the partnership increases. Formally stated:

H9a. Increase in risk of appropriation is positively associated with the

propensity of J V dissolution.

H9b. Increase in risk of appropriation is positively associated with the

propensity of J V acquisition.

46

Decrease in alternative partners

In this study, alternative partners for the focal parent ﬁrm (P l) are deﬁned as
ﬁrms that are in the same industry as the other partner ﬁrm (P2) and have the
intention to carry out activities in the same industry as the focal joint venture. These
ﬁrms can serve as alternative partners for the focal parent ﬁrm. The more alternative
partners, the less dependent the focal parent ﬁrm is on the current partner, because the
parent ﬁrm is able to switch to other partners (Pfeffer and Salancik 1978).

During the JV co-evolutionary process, the number of alternative partners for
the parent ﬁrm may change. When there is a decrease in the number of alternative
partners, the dependence of the focal parent ﬁrm on its current partner is increased
because it is more difﬁcult to switch partners. Though dependence exists at the start
of a joint venture, increased dependence after JV formation induces the parent ﬁrms
to reevaluate the situation and to adjust its decisions accordingly. Through this
process of reevaluation and readjustment, the joint venture co-evolves with its parent
ﬁrms.

Increased dependence makes the JV partnership more critical in the parent
ﬁrm’s overall strategy and increases the parent ﬁrm’s positive evaluation of the joint
venture. The parent ﬁrm is more motivated to continue the dependent partnership, and
less likely to dissolve or acquire the joint venture, as both these actions will terminate
the cooperation. Therefore:

H 1 0a. Decrease in alternative partners is negatively associated with the

propensity of J V dissolution.

H 1 0b. Decrease in alternative partners is negatively associated with the

propensity of J V acquisition.

47

Formation of competing partnerships

It is common that one ﬁrm is involved in more than one partnerships (D02 and
Hamel 1998). Some of these partnerships may serve similar purposes for the parent
ﬁrm, for instance, they may operate in the same industry as the focal JV. Competing
partnerships in this study are deﬁned as other partnerships that are formed by the focal
parent ﬁrm and are in the same industry as the focal JV. These partnerships are
competing with the focal JV in the sense that they are to a certain degree substitutable
for the focal JV and they compete for similar resources from the parent ﬁrm.

During the JV co-evolutionary process, the parent ﬁrm may form other
competing partnerships. Formation of competing partnerships reduces the dependence
of the parent ﬁrm on the focal joint venture because it provides the possibility of
switching the JV business to the competing partnerships (Pfeffer and Salancik 1978).
Though dependence exists at the start of a joint venture, reduced dependence after JV
formation induces the parent ﬁrms to reevaluate the joint venture and adjust its
decisions accordingly. Decreased dependence makes the focal JV partnership less
critical to the parent ﬁrm and consequently lowers the parent ﬁrm’s evaluation of the
focal joint venture. The parent ﬁrm is more likely to discontinue the JV partnership in
presence of reduced dependence, and the propensity of JV dissolution and acquisition
is both increased. Therefore:

[-1 I 1 a. Formation of competing partnerships is positively associated with the

propensity of J V dissolution.

H 1 1b. Formation of competing partnerships is positively associated with the

propensity of J V acquisition.

48

3.2.4. Joint venture performance

JV performance directly inﬂuences the parent ﬁrm’s evaluation of the joint
venture. Undesirable performance lowers the joint venture’s contribution to the
overall strategy of the parent ﬁrm, and consequently the parent ﬁrm’s evaluation of
the joint venture. Firms are more likely to dissolve a low-performing partnership than
a well-performing one (Porter 1987). Therefore:

H 1 2a. JV performance is negatively associated with the propensity of J V

dissolution.

Joint ventures are used as a way to expand to new markets, and ﬁrms are ready
to increase resource commitment when market conditions are favorable (Kogut 1991).
Good JV performance indicates favorable market conditions. In presence of high JV
performance, the parent ﬁrm raises its evaluation of the joint venture, and internal
operation becomes a more attractive alternative because internalizing the JV activity
enables the parent ﬁrm to solely exploit the beneﬁts from the joint venture (Kogut
1991). Acquiring the joint venture is an adjustment action taken by the parent ﬁrm to
fully exploit the beneﬁts of the joint venture. Therefore parent ﬁrms are more likely to
acquire a well-performing joint venture. Formally stated:

H 1 2b. J V performance is positively associated with the propensity of 'J V

acquisition.

3.2.5. Joint venture type
The effects of antecedent factors on JV termination are proposed to be
different for different types of joint ventures. Two classiﬁcations of joint ventures are

considered: related JV versus unrelated JV; and direct-competitor JV, when the

49

parents are direct competitors, versus non-direct-competitor JV, when the parents are

not direct competitors.

Related JV versus unrelated JV

From the product portfolio perspective, the joint venture’s product can be
related or unrelated to the parent ﬁrm’s product (Harrigan 1988; Luo 2002b). In this
study, a related I V is a joint venture that is in the same industry as the parent ﬁrm;
whereas unrelated JV refers to the scenario when the joint venture and the parent ﬁrm
are not in the same industry (Harrigan 1988; Luo 2002b).

When the JV product is related to the parent ﬁrm’s primary product, the parent
ﬁrm has more knowledge or capability in managing the JV activity (Luo 2002b). The
diversiﬁcation literature has found related diversiﬁcation is more likely to be
persistent (Harrigan 1988; Luo 2002a; Pennings et al. 1994), while unrelated
diversiﬁcation suffers from higher exit rate (Harrigan 1988; Hennart et al. 1998; Luo
2002a). With regard to acquisition, when the JV product is closely related to the
parent ﬁrm’s core product, the parent ﬁrm is more likely to have the capability needed
to internalize (acquire) the JV activity. Therefore the propensity of dissolution is
lower for related JVs than for unrelated JV 8, while the propensity of acquisition is
higher for related JV than unrelated JV.

Besides the direct effect, which is controlled for, this study hypothesizes that
related versus unrelated J Vs plays a moderating role in the relationship between
change in parent ﬁrm overall investment strategy and the propensity of JV
termination. From the product aspect, during strategic adjustment, parent ﬁrms are
more likely to change lines of business that are with different proﬁles, and products

that are closely related to the ﬁrm’s primary business are given priority (Chang 1996).

50

While changes in parent ﬁrm overall investment strategy alter the evaluation of a
related JV, parent ﬁrms are less willing to take adjustment actions toward a related
JV, as compared to an unrelated JV. From the resource aspect, parent ﬁrm resources
are more transferable to a related JV than to an unrelated JV (Luo 2002b), and
accessing unique resources plays a more important role in a related JV. Therefore it is
expected the effects of covariates regarding investment strategy change to be different
for related and unrelated JVs.

Speciﬁcally, when there is a product line shift in the parent ﬁrm’s strategy, the
relative importance of a joint venture in the overall strategic portfolio is reevaluated
and necessary adjustment actions are taken. Because a related JV is more closely
related to the parent ﬁrm’s primary business and plays a more important role in the
parent ﬁrm’s strategic portfolio, more efforts are made to keep this investment stable
(Chang 1996), therefore the continuity of a related JV is less likely to be inﬂuenced
by changes in the parent ﬁrm’s product line. Therefore, the effects of product line
shift on JV dissolution and acquisition are weaker for related J Vs than for unrelated
J Vs. Considering the positive association between product line shift and the
propensity of JV dissolution and acquisition, the interaction effect of product line shift
and related JV is hypothesized to be negative. Formally stated:

H13a. The interaction between product line shift and related JV has a

negative effect on the propensity of J V dissolution.

H13b. The interaction between product line shift and related JV has a

negative effect on the propensity of J V acquisition.

Increases in parent ﬁrm manufacturing, marketing and R&D resources provide
larger potential of resource sharing for the parent ﬁrms and reduces the propensity of

JV dissolution and acquisition. For related JV 5, the parent ﬁrm’s resources are more

51

transferable to the joint venture, which is in the same industry as the parent ﬁrm (Luo
2002b). When parent ﬁrms set up a related JV, accessing unique resources plays a
more important role, as compared to an unrelated JV where the resources are less
transferable. Therefore increased potential of resource sharing is more likely to
enhance the continuity of a related JV than an unrelated JV. Thus it is hypothesized
that the inﬂuence of parent ﬁrm resource increase is stronger for related JVs than for
unrelated JVs. Considering the negative association between increase in parent ﬁrm
resources and the propensity of JV dissolution and acquisition, the interaction effects
of related JV and increase in parent manufacturing, marketing and R&D resources are
hypothesized to be negative. H13 to H15 state the hypotheses for the interaction
effects of related JV and increase in manufacturing, marketing and R&D resources
respectively:

H14a. The interaction between increase in parent manufacturing resources
and related JV has a negative effect on the propensity of JV
dissolution.

H 1 4b. The interaction between increase in parent manufacturing resources
and related JV has a negative effect on the propensity of JV
acquisition.

H15a. The interaction between increase in parent marketing resources and
related JV has a negative effect on the propensity of J V dissolution.

H15b. The interaction between increase in parent marketing resources and
related JV has a negative eﬂect on the propensity of J V acquisition.

H16a. The interaction between increase in parent R&D resources and related

JV has a negative effect on the propensity of J V dissolution.

52

H16b. The interaction between increase in parent R&D resources and related
JV has a negative effect on the propensity of J V acquisition.
During the JV co-evolutionary process, disruptive event such as parent ﬁ rm
M&A changes the parent ﬁrrn’s overall strategy and induces the parent ﬁrm to
reevaluate the role of a joint venture in the overall strategic plan. Because related JV
is more closely related to the parent ﬁrm’s primary business and a plays a more
important role in the parent ﬁrm’s strategic portfolio, parent ﬁrms try to keep this
investment stable and related JVs are given priority in the process of strategic
adjustment (Chang 1996). Therefore the continuity of a related JV is less likely to be
inﬂuenced by disruptive event as opposed to an unrelated JV. Therefore, the effects of
disruptive event on JV termination, including dissolution and acquisition, are weaker
for related JV 5 than for unrelated JV 3. Considering the positive association between
disruptive event and the propensity of JV dissolution and acquisition, the interaction
effect of disruptive event and related JV is hypothesized to be negative. Formally
stated:
H1 7a. The interaction between disruptive event and related JV has a negative
effect on the propensity of JV dissolution.
H17b. The interaction between disruptive event and related JV has a negative

effect on the propensity of JV acquisition.

Direct-competitor JV versus non-direct-competitor JV

When JV parents are direct competitors, they are involved in competition both
inside and outside of the joint venture, and the relationship between JV partners are
more competitive than otherwise (Park and Russo 1996). Due to the competitive

nature, the risk of opportunistic behavior is higher and parent ﬁrms are more

53

concerned about the protection of self-interest (Park and Russo 1996). When the risk
of appropriation increases, it inﬂuences the parent ﬁrm’s evaluation of the joint
venture more when the parents are direct competitors than when they are not.
Considering the positive main effect of increase in risk of appropriation, the
interaction effect between direct-competitor JV and increase in risk of appropriation is
hypothesized to be positive. Formally stated:

H18a. The interaction between increase in risk of appropriation and direct-

competitor JV has a positive effect on the propensity of J V dissolution.
H18b. The interaction between increase in risk of appropriation and direct-
competitor JV has a positive eﬂect on the propensity of J V acquisition.

Decrease in alternative partners reduces the propensity of JV dissolution and
acquisition because it raises the level of dependence between JV partners, which
supports the continuity of the JV partnership. When JV parents are in the same
industry, not only they compete with each other in the market, but also their resources
and capabilities are more similar and substitutable than when parents operate in
different industries. When the number of alternative partners decreases, the parent
ﬁrm has more ﬂexibility of obtaining the needed resources internally instead of
becoming more dependent on the current partner. Therefore the level of dependence
between partners is not inﬂuenced by the number of alternative partners as much in a
direct-competitor JV as it is in a non-direct-competitor JV. Consequently, the effects
of decrease in alternative partners on the propensity of JV dissolution and acquisition
are not as strong for a direct-competitor JV as for a non-direct-competitor JV.
Considering the negative main effect of decrease in alternative partners, the
interaction effect between direct-competitor JV and decrease in alternative partners is

hypothesized to be positive. Formally stated:

54

H19a. The interaction between decrease in alternative partners and direct-

competitor JV has a positive effect on the propensity of J V dissolution.

H19b. The interaction between decrease in alternative partners and direct-
competitor JV has a positive effect on the propensity of J V acquisition.

Formation of competing partnerships increases the propensity of JV
dissolution and acquisition because it provides substitutes for the focal JV
partnerships and reduces the level of dependence between current parents. When
current parents directly compete with each other in the market, their JV partnership is
subjected to higher level of competition and conﬂict (Doz and Hamel 1998), and
reduced dependence is more likely to cause instability in the partnership. Therefore,
the inﬂuence of formation of competing partnerships on the propensity of JV
dissolution and acquisition is stronger for direct-competitor JVs. Considering the
positive main effects of formation of competing partners, the interaction effect
between formation of competing partnerships and direct-competitor IV is
hypothesized to be positive. Formally stated:

HZOa. The interaction between formation of competing partnerships and
direct-competitor JV has a positive effect on the propensity of J V
dissolution.

H20b. The interaction between formation of competing partnerships and
direct-competitor JV has a positive effect on the propensity of .l V

acquisition.

3.3. Further statements

Based upon the above hypotheses, this study also proposes the following

statements. By testing the model, it seeks evidences for these statements:

55

1. Change during the JV co-evolutionary process is a cause for JV termination,
including both dissolution and acquisition. If the variables regarding changes in the
market environment and parent ﬁrm overall strategy show signiﬁcant effects on the
propensity of JV dissolution and acquisition, with the effects of corresponding level
variables controlled for, then it can be concluded that change is a cause for JV
dissolution and acquisition.

2. As a result of the co-evolution of a joint venture with parent ﬁrm overall
strategy and the market environment, factors exogenous to the joint venture, such as
changes in parent ﬁrm overall strategy and the market environment, have equally
important impact on JV termination, including both dissolution and acquisition. lf
with JV performance being controlled for, these exogenous factors show signiﬁcant
effects on the propensity of JV dissolution and acquisition, then it can be concluded
that exogenous factors are important antecedent factors of JV termination, and JV
termination can be resultant from parent ﬁrm strategic adjustment.

3. Based upon the model, this study intends to identify two primary reasons
for JV termination: low JV performance and external change. External change is the
change in factors that are exogenous to a joint venture, including change in the market
environment and parent ﬁrm overall strategy. The model simultaneously tests the
effects of external change and JV performance on the propensity of JV termination,
including both dissolution and acquisition.

4. By identifying the two primary reasons for JV termination, this model
intends to show that JV termination is not necessarily failure. If ajoint venture is
terminated due to low performance then it is associated with failure; but if it is
terminated due to external change, then it is not necessarily a failure, but rather a

result of strategic adaptation.

56

5. By distinguishing and separately modeling JV dissolution and acquisition,
this model aims to show that although dissolution and acquisition both terminate a JV
partnership, they are driven by different motivations and have different implications
for the parent ﬁrm. If different effects of the antecedent factors are found for JV
dissolution and acquisition, then it can be concluded that different mechanisms

underline JV dissolution and acquisition.

57

CHAPTER 4
RESEARCH METHODOLOGY

Event history analysis is used to empirically test the model and is suitable for
the study because it allows for the modeling of event probability over all observation
periods, and can incorporate time-dependent covariates to capture the effects of non-
static explanatory variables over time. Speciﬁcally, semiparametric Cox models are
employed to minimize bias arising from incorrect parametric speciﬁcation for JV
termination. Multiple data collection methods are used to obtain time series data,
with secondary data from various public sources complemented by survey and

interviews with parent companies.

4.1. Model formulation

Event history analysis looks at the hazard rate of an event, which is deﬁned as
the conditional probability of the event occurring at time t given that it has not
occurred until time t-l (Allison 1984; Allison 1995). In this study, the events are JV
termination. It looks at the likelihood of a joint venture being terminated at a certain
time point, given that it has survived the past.

From previous discussion, this study identiﬁed two types of JV termination,
i.e., dissolution and acquisition. These two types of terminations are modeled as
competing risks. A competing risk situation is where the occurrence of one event
removes the individual from the risk of all other types of events (Allison 1995; Box-
Steffensmeier 2004). In this context, dissolution and acquisition compete with each
other in the sense that a joint venture can only be terminated through one of the two
ways, i.e., dissolution or acquisition. A dissolved joint venture is removed from the

risk set for acquisition and vice versa.

58

The hazard rates for dissolution, acquisition, and termination in total are
deﬁned as follows. The hazard rate for termination is equal to the sum of the hazard

rates for dissolution and acquisition.

P(t<T,-St+At,D=d |T,- >1)
At

 

Dissolution: hiD(t) = limA,_,0

P(t<]}st+At,A=a |T, >t)
At

 

Acquisition: hm (t) = limA,_,0

P(t<T,- St+At,T=a [7",- >1)

Termination: hiT (t) = limA, —>O A
t

 

hi0) = hiD(t) + lit/10)

Among different types of EHA models, semiparametric Cox models are
chosen for this study. Semiparametric Cox models are more suitable than parametric
models for this study because it is difficult to make a reasonable assumption of the
baseline hazard function for JV dissolution and acquisition, and unrealistic
assumption of the baseline form introduces bias to the analysis. Semiparametric
models can avoid bias by leaving the baseline hazard function form unspeciﬁed, and
still be able to estimate the effects of covariates (Box-Steffensmeier 2004; Therneau
2000). Moreover, Cox models can readily incorporate time-dependent covariates.
which is essential in this study that emphasizes change over time (Allison 1995; Box-
Steffensmeier 2004; Therneau 2000).

After estimating the Cox models for JV dissolution and acquisition, for the
purpose of comparison, a model for the event of termination in total is estimated. The
formulations of Cox models for dissolution, acquisition and termination in total are
listed below. Same covariates are included in the three models. In the following

equations, the covariates x,,—(t) are time dependent covariates.

59

Dissolution:
hzoﬂ) = 100(t)explﬂioxn(t) + 3203920) + + ﬂjbxg-(I) + + ﬂip-Yn- (1)}

loghiD(t) = logdooﬂ) + .BIDxnU) + .32011'20) + + 310x90) + + ﬂip-Vii (1)
Acquisition:

lit/1(1) = 10A(t)eXPi/31Axn(t) + ﬂZAxiZU) + + ﬂj/ixijU) + + ﬁlm-WA- 0)}

log hm (I) = log/10A“) + ﬂinxnU) + 32AM“) + + ﬂijg-(t) + + ﬂk/txik (1)
Termination:

her) = dorﬂkXPiﬂiTxnU) + ﬂzrxtzﬂ) + + IBijij(t) + + ﬂerzk (1)}

loghrrﬂ) = log/10TH) + ﬂITXnU) + ﬂzrxizﬂ) + + ﬂjTXg-(t) + + ﬂkrl‘rk (f)

4.2. Data
4.2.1. Sample

Joint ventures were selected from Thomson Financial SDC Platinum database.
The study selected manufacturing, marketing, and R&D joint ventures located in US
and formed during the time period from 1990 to 2001, leaving 3 years between the
formation time and the time when observation was ended (2004), considering the
average life span of joint ventures is 3—4 years (Harrigan 1988). Twelve major
manufacturing industries were included, speciﬁcally, chemical and allied products,
drugs, electronic and electrical equipment, communication equipment, food and
kindred products, computer and ofﬁce equipment, machinery, transportation
equipment, paper and allied products, soaps, cosmetics, and personal-care products.
rubber and miscellaneous plastic products, metal and metal products. To be able to
obtain ﬁnancial data of parent companies and to contact parent companies, the sample

was further restricted to joint ventures with at least one US public company as parent.

60

This resulted in 925 joint ventures. Matching thesejoint ventures with the Dun &
Bradstreet (D&B) database resulted in 465 joint ventures with JV sales information.
Each of these joint ventures was then tracked to see whether it was terminated or still
in operation at the end of 2004. Among these 465 joint ventures, information on
termination was obtained for 255 joint ventures, including whether it is terminated or
not, and the time and type of termination if terminated.

Missing values of the covariates reduced the ﬁnal sample for analysis to l50,
among which 89 joint ventures were terminated, including 13 dissolutions and 76
acquisitions. This resulted in a termination rate of 8.67% for dissolution and 50.67%
for acquisition, and a total termination rate of 59.33%.

To test for sampling bias, sales, net worth and gross proﬁt of joint ventures in
the ﬁnal sample were compared with those of joint ventures that missed covariate
values, and with those of joint ventures that missed JV termination information. T
tests were conducted for each year from 1990 to 2001. For most years, t tests did not
show signiﬁcant difference between the ﬁnal sample and excluded cases due to

missing values.

4.2.2. Data collection method

The core data needed for the study include: the time of joint venture formation
(i.e., when the contract was completed); whether the joint venture had been
terminated at the end of the data period; if it had been terminated, the time of
termination (i.e., when the contract was terminated); and how it was terminated (i.e.,
dissolution, or acquired by a parent ﬁrm). These data were gathered from Thomson
Financial SDC Platinum database, and validated through Corporate Affiliations and

Factiva. For those joint ventures that still miss this information, the parent companies

61

were contacted. The contact was done through email and telephone. Emails were ﬁrst
sent to the investor relations contact of the parent companies, asking for cooperation
and identifying the person with knowledge of the joint venture. Multiple telephone
follow-ups were made to obtain the information. Other potential sources were also
utilized as complementary means of identifying the person involved in the joint
venture.

To build time-dependent covariates, the values of the covariates are needed
from the year when a joint venture was formed until it was terminated or until the end
of observation period, which is the end of year 2004. Time series data of covariates
were collected from various secondary sources, including bureau of census,
Compustat, Corporate Afﬁliations, Delphion; and Dun & Bradstreet. The
operationalization of the covariates and their data sources are presented in the

following section.

4.3. Measures
Table 3 listed the measures of the covariates in this study and the

corresponding data sources.

Market growth

Market growth is measured with the annual growth rate of value of shipments
for different industries at 4-digit SIC level (Kogut I991; Kogut 1989). Data were
obtained from Bureau of Census Annual survey of manufacturers. To create time
series data, for the years after 1997, when NAICS system was used instead of the SIC,
NAICS codes were converted to SIC codes according the matching table provided by

NAICS.

62

Table 3. Variables, measures and data sources

 

Variables

1 Measures

I Data sources

 

Changes in market competition

 

Market growth

Annual growth rate of value of
shipments

Bureau ofcensus

 

Increase in market
concentration

Market concentration ratio increase

 

Bureau of census

 

 

Changes in partner ﬁrm overall investment stratggy

 

Change of parent ﬁrm primary SIC

 

 

 

 

 

 

 

 

 

appropriation

Product line shift Compustat
code
Manufacturin . .
g Capital expenditure Compustat
. resources
Increase in Marketin
parent ﬁrm g Marketing expenditure Compustat
resources resources
R&D resources R&D expenditure Compustat
. . Par t ﬁ m w'
Disruptive event en rm erged “h or Compustat
jacqurred by others
Changes in partner ﬁrm overall governance strategy
Increase in risk f . . .
0 Increase In patent correlation Delphion

 

Decrease in alternative
partners

Formation of partnerships
involving same activities by ﬁrms
in the partner’s industry

Thompson Financial
SDC Platinum

 

Formation of competing

Formation of similar partnerships

 

Thompson Financial

 

 

 

 

 

 

 

 

 

artnerships by the focal parent SDC Platinum
JV performance
JV sales Performance JV sales Dun & Bradstreet
JV type

Thompson Financial

Related vs. unrelated JV Parent and JV same SIC codes SDC Platinum
Direct-competitor JV vs. Thompson Financial
Non-direct-competitor JV Two parents same SIC codes SDC Platinum

 

63

 

Increase in market concentration

Market concentration is measured with the percentage of value of shipments
accounted for by the top 50 companies in an industry at 4-digit SIC code level.
The Economic Census published by Bureau of Census provides market concentration
data for manufacturers. Economic Census is published every ﬁve years. So for years
that this information is not available, the closest available year’s information is
applied. This is consistent with the study by Kogut (1991; 1989). Increase in market
concentration is calculated by subtracting the value of market concentration ratio at

time H from the value at time t.

Product line shift

Product line shift is measured with change in parent ﬁrms’ primary SIC codes
at 4-digit level. At each time point, if there is a change in the parent ﬁrm’s primary
SIC code, product line shift is coded as l, and 0 otherwise. Because substantial time is
needed for companies to adjust their lines of business, the carryover effects of product
line shift are considered by using a Koyck-lag structure (Dutta et al. 1999). Formally,

k=t t-k

product line shift (PLS) for time t is deﬁned as PLS, = k=17

x PLSk . Here 7

represents the weight attached to past values of product line shift. Primary SIC codes

of the parent companies in each year are obtained from Compustat database.

Increase in parent ﬁrm resources

Three types of resources, manufacturing resources, marketing resources and
R&D resources, are measured respectively by capital expenditure, sales, general and
administrative expenses, and R&D expenditure of the parent ﬁrms (Dutta et al. 1999;

Kogut 1991; Kogut 1989). Although sales, general and administrative expenses also

64

includes items that are in the domain of marketing, it has been used as to measure
marketing resources in the literature and found to be a good proxy for the resources
ﬁrms spend on their market research, sales efforts, and other related activities (Dutta
et al. 1999). Increase in parent ﬁrm resource is calculated by subtracting the value of
parent resources at time t-l from the value at time t. Yearly data on capital
expenditure, sales, general and administrative expenses, and R&D expenditure were

obtained from Compustat.

Disruptive event

For disruptive event, this study considers when a parent ﬁrm is merged with or
acquired by another ﬁrm. At each time point, the variable is coded as 1 if there is a
M&A event for the parent, and 0 otherwise. This information was obtained form
Compustat event database. Because substantial time is needed for the adjustment after
M&A events, the carryover effects of disruptive events are considered by using a

Koyck-lag structure (Dutta et al. 1999).

Increase in risk of appropriation

Risk of appropriation is measured by the correlation between the patent
structures of different parents. Patent correlation has been used to measure of
potential of technology spillover (Jaffe I986). The distribution of a ﬁrm’s patents
over patent classes is used to characterize the technology position of the ﬁrm. A
vector is constructed for each parent ﬁrm at each time point, with each element of the
vector being the number of patents published in a certain class. The length of the
vector therefore is the number of patent classes in the classiﬁcation system.

International patent classiﬁcation (IPC) code is used, and the 3-digit class level is

65

adopted in the analysis, which results in altogether 129 classes. The correlation
between the two vectors of the two parents represents how close their technology
structures are, and the closer they are, the higher chance of technology spillover,
which is related to a higher risk of appropriation. Increase in risk of appropriation is
calculated by subtracting the value of risk of appropriation at time t-l from the value

at time t. Patent information was obtained from Delphion.

Decrease in alternative partners

Alternative partners should be ﬁrms that can provide similar needs as the
current partner. Therefore when ﬁrms in the same industry as the current partner form
joint ventures that carry out the same type of activity in the same industry as the focal
joint venture, the number of alternative partners for the focal parent decreases. From
Thompson Financial SDC Platinum, all the joint ventures formed in the current
partner’s industry were gathered and compared with the focal joint venture. Joint
ventures that are in the same industry as the focaljoint venture and are involved in the
same type of activities, whether it is manufacturing, marketing or R&D, were
selected. Considering the possible general trend of JV formation in an industry, the
ratio of the number of selected joint venture formations to the total number of joint
venture formations was used as the measure for decrease in alternative partners.
Because of the time needed for companies to switch partners, the carryover effects of
decrease in alternative partners are considered by using a Koyck-lag structure (Dutta

et al. 1999).

66

Formation of competing partnerships

From Thompson Financial SDC Platinum, all other partnerships formed by the
focal parent were gathered, and those in the same industry as the focal joint venture
were selected as competing partnerships. Considering the possible general trend of JV
formation of a company, the ratio of the number of competing JV formations to the
total number of JV formations by the focal parent is used as the measure for formation
of competing partnerships. Same as before, because substantial time is needed for
companies to switch partners, the carryover effects of formation of competing

partnership are considered by using a Koyck-lag structure(Dutta et al. 1999).

JV sales performance
Joint venture sales are used to measure the JV performance. This data were

obtained from D&B.

Related JV
Related versus unrelated JVs are determined by comparing the 2-digit primary
SIC codes of a joint venture and its parent. It is coded as one if they are the same, and

0 otherwise (Li 1995; Lu and Xu 2006).

Direct-competitor JV

Whether ajoint venture is between director competitors is determined by
comparing the 2-digit primary SIC codes of the two parents. It is coded as one if they
are the same, and 0 otherwise (Park and Russo l996; Park and Ungson 1997).

Due to the nature of the analysis, for some covariates for parent ﬁrm strategy,
only one parent can be selected. Among different parents of ajoint venture, public

ﬁrms are chosen over private ﬁrms to ensure the availability ﬁnancial information.

67

Then parents with a higher ownership are chosen over parents with lower ownership,
as this study looks at the inﬂuence of parent strategy on joint ventures and dominant
parents are more likely to be able to inﬂuence the joint venture. When two parent
ﬁrms have equal ownership, one parent ﬁrm is randomly chosen.

Table 4 presents the descriptive statistics of the covariates.

68

.32: M 6:32 .. 85E”; quCo .2832 .1. Z
.752 5d a «58557. 5:22.50 .3
.36. mod 5 Eng-Ema coon—280 ..

 

_

omvd: bozo:

hood: moNd: wood omod: deed-.3

 

Sod-.2.. 38d: Sod-s Sod-3 oood- ovd :d 2. 55858625.?
_ «em—d1. $_.o_..._ and: wood :od omod-ga Sod... ”3d. omod-t. Sod-2. wvod: omd wed Zoos—udd—
_ wood: wood: Nood :od- ”nod-2. omod: mood: mood dood: Sud: modm v92 accustom—om .8.me .2
aﬁﬁogq wcoanou
_ momd: Sod- vood- Rod: hid: vﬂd: mmodJ oNod-E. o2d2. No.9. 3.9 dagger—cud—
$05.3
_ oood- vood- mmod-t. did... mmmd: omod- 3d-.. did: 36» oodm oﬁﬁEBEEomaucoun—d
Governance
_ mood vood- Nood mood Nood- «8.? wood om; mod do xmtﬁugocoﬁd
_ oood- ”mod-1. vood- mood: Sod- god-e cod _od E95 gasp—£06
80582
_ Sod- mNod: mood: Sod- hood-2. 3.? ohm Ody—5080856
39582
_ oEd: wood Nood- Sod 3.3 and wSBxEEEoQBoEd
30582 wagoﬂscmﬁ
_ Sod- Eod- mvod: m2: and 5030856
_ 2od- Sod-2. odd Nod «Em oéwosdoid
cougcooeoo
_ _m_d: Quo— wmd «8.38 E 08965 .N
_ e3 2.... 532» Be: ._
noradren—
S 2 c m e m N _ Review :32

 

8.3.859 he 28:23.89 one £323.56 Eat—:3... .532 .v 93:.

69

CHAPTER 5
ANALYSIS RESULTS AND DISCUSSION

To examine the effects of covariates, which are changes, the level of these
variables are included as control when possible. While changes are the differences of
values between two time points, levels are the values at a certain point of time. After
controlling for the effect of level variables, change variables are expected to have a
distinct effect on the dependent variable. The level variables controlled for are market
concentration, parent ﬁrm resource, and risk of appropriation.

To consider the moderating effects of JV type variables (interaction effects),
the main effects of the two JV type variables, i.e., related versus unrelated JV, and
direct-competitor JV and non-direct competitor JV, are also included. To investigate
the interaction of JV type variables with the other covariates, the interactions of JV
type variables and the levels of the covariates are also included. Other control
variables included in the analysis are parent size, parent diversiﬁcation level, and

number of companies in the JV industry.

5.1. Statistical results

A model without the interaction effects of JV type variables was estimated
ﬁrst, and then interaction effects were included in a hill model. SAS 9.1 was used for
the statistical analysis. The effects of covariates on the hazard rates were estimated
with partial likelihood estimation, and Breslow’s approximation method was used to
deal with event time ties. Table 5 presents the test results of the Cox models on JV
dissolution and acquisition when interaction was not considered. Table 6 presents the

test results of the Cox models on JV dissolution and acquisition when interaction was

70

included. The )6 statistics of the models are highly signiﬁcant (p<0.001), indicating

good overall ﬁt of the models to data.

Table 5. Analysis results for the model without interaction effects.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Hypothesized Dissolution Acquisition
Variables Hs relationships (D) (A)
D T A Coef. [p value Coef. lp value
Change in market competition
Market growth Hl - -0.010 0.89 -0.027 0.30
Increase in market concentration H2 + -0.001 0.97 -0.012 0.17
Change in parent ﬁrm overall investment strategy
Product line shiﬁ H3 + + *3.826 0.01 *l.328 0.09
Increase in parent ﬁrm:
Manufacturing (manu.) resourcesl H4 - - -0.006 0.90 *-0.050 0.01
Marketing (mktﬂesources H5 - - -0.007 0.59 *-0.011 0.02
R&D resources H6 - - 0.001 0.87 0.002 0.23
R&D vs. manu. resources R&D < Manu x2 =0.02. p =0.88 {=6.27, p =0.01
R&D vs. Mkt. resources H7 R&D< Mkt =0.30. p =0.58 2 =7.12, p =0.007
Disruptive event H8 + + *6.686l 0.04 *3.016l <0.01
C hang: in parergﬁrm overall governance strategy
Increase in risk of appropriation H9 + + -0.083 0.84 -0.1 17 0.51
Decrease in alternative partners H10 - - *-0.027 0.08 -0.001 0.63
Formation of competing
partnerships H1 1 + + *0.011 0.04 0.002 0.45
Joint venture performance
JV Sales performance I H 12 I - j + I -0024] 0.22] *-o.013l 0.02
C ontrol variables
Level variables
Market concentration -0.010 0.54 *0.018 0.04
Parent diversiﬁcation level -0.015 0.94 *-0.l49 0.04
Parent manufacturing resources 0.002 0.84 0.001 0.70
Parent marketing resources 0.002 0.59 0.0001 0.91
Parent R&D resources 0.005 0.69 0.002 0.75
Risk of appropriation -0.006 0.66 *-0.011 0.02
Other control variables
Direct-competitor JV -0.452 0.60 0.037 0.90
Related JV 0.464 0.54 -0.174 0.55
Parent size -3 l .855 0.61 -1.564 0.88
Number of companies in the JV
industry -0.639 0.12 0.101 0.33
72.28 74.42
Model 12 (df=21. p<0.001) (df=21,p<0.()()l )

 

 

Note: 1. Unstandardized coefﬁcients are reported.

2. Positive coefﬁcients indicate an independent variable has a positive effect on the

hazard rate.

71

 

Table 6. Analysis results for the model with interaction effects

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Hypothesized Dissolution Acquisition
Variables Hs relationships (D) (A)
D l A Coef. lp value Coef. ]p value
Change in market competition
Market growth H1 - -0.035 0.69 -0.027 0.32
Increase in market concentration H2 + 0.006 0.83 -0.01 1 0.21
C harge in parent ﬁrm overall investment strategy
Product line shift H3 + + *4.883 0.03 *2.631 <0.01
Increase in parent ﬁrm:
Manufacturing resources H4 - - -0.033 0.73 *-0.053 0.06
Marketing resources H5 - - -0.043 0.18 *-0.015 0.03
R&D resources H6 - - -0.033 0.56 0.001 0.33
R&D vs. manu. resources R&D < Manu 12 =0.0001, p =0.9912 =3.67, p =0.01
R&D vs. Mkt. resources H7 R&D< Mkt 2 =0.26. p =0.87 2 =5.46, p =0.05
Disruptive event H8 + + *7.321 0.06 *6.660l 0.01
Change in parent firm overall governance strategy
Increase in risk of appropriation H9 + + -0.062 0.85 -0.132 0.54
Decrease in alternative partners H10 - - -0.043 0.12 -0.006 0.14
Formation of competing
partnerships H l l + + *0.015 0.07 0.003 0.42
Joint venture performance
JV Sales performance H12 - + *-0.056 0.04j *-0.011 0.07
Interaction of related J V with change_in investment strategy variables
Related JV ‘ Product line shiﬂ H13 - - 0.001 0.80 -25.l22 0.53
Related JV "‘ Increase in
manufacturing resources H 14 - - -0.209 0.29 -0.032 0.64
Related JV * Increase in marketing
resources H15 - - *0.207 0.06 0.017 0.29
Related JV " Increase in R&D
resources H16 - - *0.210 0.03 -0.012 0.50
Related JV "' Disruptive event H17 - - --- --- -3.710 0.19
Interaction of direct-competitor JV with change in wovernance strategy variables
Direct-competitor JV "‘ Increase in
risk of appropriation H18 + + -0.129 0.92 0.01 l 0.98
Direct-competitor JV * Decrease in
number of alternative partners H19 + + 0.012 0.51 *0.005 0.06
Direct-competitor JV "' Formation of
competing partnerships H20 + + 0.008 0.66 -0.007 0.24
Control variables
Level variables
Market concentration -0.004 0.83 *0.019 0.04
Parent diversiﬁcation level -0.02() 0.36 *-0.285 <0.01
Parent manufacturing resources 0.004 0.50 0.002 0.68
Parent marketing resources 0.017 0.57 0.0001 0.85
Parent R&D resources -0.053 0.83 0.003 0.75
Risk of appropriation 0.0001 0.99 *-0.012 0.09

 

72

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Interaction of level variables
Related JV “ Manufacturing
resources 0.034 0. 14 -0.006 0.39
Related JV * marketingresources 0.010 0.39 *0.007 0.05
Related JV " R&D resources *-0.088' 0.05 -0.01 l 0.40
Direct-competitor JV " risk of
appropriation -0.010 0.78 -0.015 0.17
Other control variables
Direct-competitor JV -0.525 0.70 0.01 1 0.98
Related JV -1.244 0.31 -0.321 0.38
Parent size -33.940 0.66 -5.127 0.81
Number of companies in the JV
industry *-0.971 0.07 0.054 0.64
79.00 108.99
Model x2 (df=32. p<0.001) (df=33. p<0.001)

 

 

 

 

 

Note: 1. Unstandardized coefﬁcients are reported.

2. Positive coefﬁcients indicate that an independent variable has a positive effect on the
hazard rates.

3. The interaction between related JV and disruptive event is not tested for dissolution due
to 0 degree of freedom.

5.1.1. Change in market competition
The effects of market growth on the propensity of JV dissolution (B = -0.010,

p =0.89) and acquisition (B = -0.027, p =0.30) are both not signiﬁcant. The non-
signiﬁcance does not change when interaction effects are considered. Hla and H111
are not supported.

The effects of increase in market concentration on the propensity of JV
dissolution (B =-0.001, p =0.97) and acquisition (0 =-0.012, p =0.17) are both not
signiﬁcant. The non-signiﬁcance does not change when interaction effects are

considered. H2a and H2b are not supported.

5.1.2. Change in parent ﬁrm overall investment strategy
Parent product line is found to have signiﬁcant positive effects on both
dissolution and acquisition. In the model without interaction, the coefﬁcient estimates

are 3.826 (p=0.01) for dissolution and 1.328 (p=0.09) for acquisition. In the model

73

with interaction, the coefﬁcient estimates are 4.883 (p=0.03) for dissolution and 2.631
(p<0.01) for acquisition. Therefore H3a and H3b are both supported.

Increase in parent manufacturing resources does not show any signiﬁcant
effect for dissolution, but it shows a signiﬁcant negative effect on the propensity of
acquisition (B =-0.050, p =0.01 in the model without interaction and B =-0.053. p
=0.06 when interaction is included). H4a is not supported, but H4b is supported.
Similarly, increase in parent marketing resources does not show any signiﬁcant effect
for dissolution, but it shows a signiﬁcant negative effect the prOpensity of acquisition
(B =-0.011,p =0.02 in the model without interaction and B = -0.015, p =0.03 when
interaction is included). Therefore H5a is not supported, but H5b is supported.
Increase in parent R&D resources does not show any signiﬁcant effects on both
dissolution and acquisition. H6 are not supported. In general inﬂuences of increase in
parent ﬁrm resources are partly supported for acquisition but not for dissolution.

To test for H7, the coefﬁcient estimates for increase in parent ﬁrm R&D
resources were compared with those for increase in parent ﬁrm manufacturing and
marketing resources. Wald tests were conducted to compare the estimates. For JV
dissolution, all three types of resources did not show signiﬁcant effects, and the effect
of R&D resources was not signiﬁcantly different from those of manufacturing and
marketing resources. When the model was tested without interaction, the comparison
with manufacturing resources yielded a 12 of 0.02 (p =0.88); the comparison with
marketing resources yielded a xz of 0.30 (p =0.58). In the model with interactions
included, the comparison with manufacturing resources yielded a xz of 0.0001 (p
=0.99); comparison with marketing resources yielded a x2 of 0.026 (p =0.87). H7a is
not supported. For JV acquisition, the effect of R&D resources is found to be

signiﬁcantly lower than those of manufacturing and marketing resources. When the

74

model was tested without interaction, the comparison with manufacturing resources
yielded a 12 of 6.27 (p =0.01); the comparison with marketing resources yielded a x2
of 7.12 (p <0.01). In the model with interactions included, the comparison with
manufacturing resources yielded a 752 of 3.67 (p =0.01); comparison with marketing
resources yielded a x2 of 5.46 (p =0.05). H7b is supported.

The effects of disruptive event are found to be positive and signiﬁcant. In the
model without interaction effects, the coefficient estimates are 6.686 (p=0.04) for
dissolution and 3.016 for acquisition (p<0.01). When interaction is considered, the
coefﬁcient estimates are 7.321 (p=0.06) for dissolution and 6.660 for acquisition

(p=0.01). Therefore H8a and h8b are both supported.

5.1.3. Change in parent ﬁrm overall governance strategy

The effects of increase in risk of appropriation are not signiﬁcant for both
dissolution and acquisition. H9a and H9b are not supported.

Decrease in alternative partners is found to have a negative and signiﬁcant
effect on dissolution (B =-0.027, p =0.08) in the model without interaction. When
interaction effects are included in the model, the effect turns non-signiﬁcant with a p
value of 0.12. Considering the small sample size for dissolution, H10a is generally
supported. Decrease in alternative partners does not show signiﬁcant effect on
acquisition (B =-0.001, p =0.63 in the model without interaction and B =-0.006, p
=0.14 when interaction is included). Therefore HlOb is not supported.

A signiﬁcant and positive effect is found for formation of competing
partnerships on JV dissolution (B =0.01 1, p =0.04 in the model without interaction
and B =0.01 5, p =0.07 when interaction is included). H1 1a is supported. Formation of

competing partnerships does not show signiﬁcant effect on acquisition (B =0.002, p

75

 

=0.45 in the model without interaction and B =0.003, p =0.42 when interaction is

included). Therefore H1 1b is not supported.

5.1.4. JV performance

Even though JV sales performance did not show a signiﬁcant effect on
dissolution in the base model without interaction effects, it did show a signiﬁcant
negative effect (B =-0.056, p =0.04) on dissolution when interaction is considered.
Therefore H123 is generally supported. JV sales performance shows a signiﬁcant
negative effect on JV acquisition (B =-0.013, p =0.02 in the model without interaction
and B =-0.01 1, p =0.07 when interaction is included). Therefore H12b is generally

supported.

5.1.5. The interaction effects with JV type variables

Among the interactions with the ﬁrst JV type variable, related versus unrelated
JV, increase in marketing resource and R&D resources show signiﬁcant effects on
dissolution, with the coefﬁcient estimates being 0.207 (p=0.06) and 0.210 (p=0.03)
respectively. All other interaction effects are not signiﬁcant. The interaction effect of
disruptive events and related versus unrelated JV is not tested because there is not
enough degree of freedom due to small sample size. Therefore, H15a and H163 are
supported, while H14, H15b and H16b are not supported. In total, the interaction
effects of increase in parent ﬁrm resources and related JV are partly supported.

For interactions with the second JV type variable, JV between direct
competitors, decrease in alternative partners showed a signiﬁcant effect on JV
acquisition (B =0.005 p =0.06). Therefore H19b is supported. Other interaction effects

of direct-competitor JV and governance strategy change are not signiﬁcant. H18,

76

H19a and H20 are not supported. In total, the interaction effects of direct-competitor

JV and change in governance strategy variables are partly supported.

5.1.6. JV termination in total

For the purpose of comparison, the same Cox models for JV termination in
total was also tested, where dissolution and acquisition are not distinguished from
each other, but rather considered as the same event: JV termination. Table 7 and 8
present the comparison of test results for JV termination in total, and JV dissolution
and acquisition. Table 7 is the results when interactions are not included; Table 8 is

the results for models with interactions.

77

Table 7. Comparison of test results for JV termination in total and JV

dissolution and acquisition: without interaction effects

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Hypothesized Dissolution Acquisition Termination
Variables Hs relationships D
D A Coef. lp value Coef. Tp value Coef. p value

Change in market competition

Market growth H1 - + -0.010 0.89 -0.027 0.30 -0.024 0.31
Increase in market

concentration H2 + + -0.001 0.97 -0.012 0.17 -0.006 0.46
Change irmarentﬂrm overall investment strate

Product line shift H3 + + *3.8261 0.01 *l.328T 0.091 H.488] 0.03
Increase in parent ﬁrm:

Manufacturingresources H4 - - -0.006 0.90 *-0.050 0.01 *-0.041 0.02
Marketing resources H5 - - -0.007 0.59 *-0.011 0.02 *-0.010 0.01
R&D resources H6 - - 0.001 0.87 0.002 0.23 0.002 0.23
R&D vs. manu. resources R&D < Manu 2:2 =0.02. p =0.88 124.27, p =0.01 1’ =5.90, p =0.02
R&D vs. Mkt. resources H7 R&D< Mkt 36:030. p =0.58 z=7.12. p<0.01 1257.71, p =0.01
Disruptive event H8 + l + *6.6861 0.04 *3.016 <0.01 *3.449 <0.01
C burg; in parent ﬁrm overall governance strategy

Increase in risk of

apmomiation H9 + + -0.083 0.84 -0.1 17 0.51 -0.1 13 0.47
Decrease in

alternative partners H10 - - *-0.027 0.081 -0.001 0.63 -0.002 0.31
Formation of

competirﬁpartnersﬂps H11 + + *0.011 0.04 0.002 0.45 0.003 0.19
Joint venture performance

JV Sales performance E1121 - l + ] -0024] 0.22] *-0.013] 0.02] *-0.015] 0.01
Control variables
Level variables

Market concentration -0.010 0.54 *0.01sl 0.041 0.010 01(-
Parent diversiﬁcation

level -0.015 0.94 *-0.149 0.04 *-0.120 0.07
Parent manu. resources 0.002 0.84 0.001 0.70 0.001 0.73
Parent mkt. resources 0.002 0.59 0.0001 0.91 0.0001 0.95
Parent R&D resources 0.005 0.69 0.002 0.75 0.003 0.49
Risk of appropriation -0.006 0.66 *-0.011 0.02 *-0.011 0.01
Other control variables

Direct-competitor JV -0.452 0.60 0.037 0.90 0.042 0.88
Related JV 0.464 0.54 -0.174 0.55 -0.150 0.58
Parent size -3l.855 0.61 -1.564 0.88 ~3.934 0.73
Number of companies in

the JV industry 0639 0.12 0.101 0.33 0.014 0.89

72.28 74.42 104.09
Model 1: kdr=21, p<0.001) kdf=21, p<0.001)1(df=21 ,p<(l.lllll 1

 

Note: I. Unstandardizedcoefﬁcients are reported.
2. Posrtrve coefﬁcrents indicate that an independent variable has a positive effect on the hazard

rate.

78

 

Table 8. Comparison of test results for JV termination in total and JV

dissolution and acquisition: with interaction effects

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

lHypothesized Dissolution Acquisition Termination
Variables Hs relationships (D) (A) (T)
D l A Coef. l p value Coef. 11) value Coef. lp value

C huge in market competition

Market growth H1 - + -0.035 0.69 -0.027 0.32 -0.016 0.52
Increase in market

concentration H2 + + 0.006 0.83 -0.01 1 0.21 -0.007 0.32
C huge in parent ﬁrm overall investment strategy

Product line shift H3 + + *4.883 0.03 *2.631 <0.01 *2.562 <0.01
Increase in parent ﬁrm:

Manufacturing resources H4 - - -0.033 0.73 *-0.053 0.06 *—0.048 0.05
Marketing resources H5 - - -0.043 0.18 *—0.015 0.03 *-0.014 0.02
R&D resources H6 - — -0.033 0.56 0.001 0.33 0.001 0.31
R&D vs. manu. resources R&D < Manu {$000142 =0.99 1’ =3.67, p =0.01 1’ =3.92, p =0.01
R&D vs. Mkt. resources H7 R&D< Mk1 12 =0.26, p =0.87 12:5.46 p =0.05 11:15.07, p =0.05
Disruptive event H8 + I + *7.32 11 0.06 moon 0.01 *6.418l 0.01
Change in parent ﬁrm overall governance strategy

Increase in risk of

appropriation H9 4» + -0.062 0.85 -0.132 0.54 -0.1 16 0.51
Decrease in alternative

partners ' H10 - - -0.043 0.12 -0.006 0.14 *-0.008 0.04
Formation of competing

partnerships H1 1 + + *0.015 0.07 0.003 0.42 0.004 0.21
Joint venture performance

JV Sales performance I H12l - + *-0.056 0.041 *-0.011 0.07 *-0.013 0.02
Interaction of related JV with change in investment strategy variables

Related JV "

Product line shift H13 - - 0.001 0.80 -25.122 0.53 -0.003 0.18
Related JV "‘ Increase in

manufacturing resources H14 - - -0.209 0.29 -0.032 0.64 -0.017 0.74
Related JV " Increase in

marketing resources HIS - - *0.207 0.06 0.017 0.29 0.016 0.28
Related JV * Increase in

R&D resources H16 - - *0.210| 0.03 -0.012 0.50 -0.004 0.84
Related JV *

Disruptive event H17 - - -- --- -3.710 0.19 -2.893 0.26
Interaction of direct-competitor JV with change in governance strategy variables

Direct-competitor JV "'

Increase in risk of

appropriation H18 + + -0.129 0.92 0.01 1 0.98 -0.035 0.92
Direct-competitor JV *

Decrease in alternative

partners H19 + + 0.012 0.51 *0.005 0.06 *0.006 0.02
Direct-competitor JV "

Formation of competing

artnerships 1120 + + 0.008 0.66 -0.007 0.24 -0.008 0.13

 

79

 

 

Control variables

 

Level variables

 

*0.019I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Market concentration -0.004 0.83 0.04 *0.013 0.08
Parent diversiﬁcation

level -0.020 0.36 *-0.285 <0.01 *-0.256 <0.01
Parent manu. resources 0.004 0.50 0.002 0.68 0.002 0.58
Parent mkt. resources 0.017 0.57 0.0001 0.85 0.0001 0.94
Parent R&D resources -0.053 0.83 0.003 0.75 0.004 0.66
Risk of appropriation 0.0001 0.99 *-0.012 0.09 *-0.011 0.07
nteraction of level variables

Related JV * Parent manu.

Resources 0.034 0.14 -0.006 0.39 -0.002 0.62
Related JV * Parent mkt.

resources 0.010 0.39 *0.007 0.05 *0.006 0.04
Related JV " Parent R&D

resources *-0.088 0.05 -0.01 l 0.40 -0.012 0.3 2
Direct-competitor JV “‘

Risk of appropriation -0.010 0.78 -0.015 0.17 -0.015 0.16
Other control variables

Direct-competitor JV -0.525 0.70 0.01 1 0.98 0.013 0.97
Related JV -1.244 0.31 -0.321 0.38 -0.501 0.13
Parent size -33.940 0.66 -5. 127 0.81 -9.993 0.58
Number of companies in

the JV industry *-0.971 0.07 0.054 0.64 -0.028 0.73

79.00 108.99 137.08

 

Model x2

 

 

 

(arr-32, p<0.001)

 

(df=33.p<0.001)1(df=33.p<0.00 1)

 

Note: 1. Unstandardized coefficients are reported.

2. Positive coefﬁcients indicate that an independent variable has a positive effect on the hazard

I'ZIICS .

3. The interaction between related JV and disruptive event is not tested for dissolution due to 0

degree of freedom.

A simple comparison of the signiﬁcance of covariates in the three models (JV

dissolution, JV acquisition and JV termination in total) shows that the testing results

for JV termination in total is a lot more similar to those of JV acquisition than JV

dissolution. The reason is acquisition accounts for a dominant proportion of JV

termination. As shown in this sample, among all the joint ventures that were

terminated, 86% were acquired by one of the parents and only 14% were dissolved.

This further conﬁrms the danger of not distinguishing between JV acquisition and JV

dissolution.

80

 

5.2. Discussion of results

In general, JV performance and most external change variables were found to
have signiﬁcant inﬂuences on the propensity of JV dissolution and acquisition, and
some effects differed for JV dissolution and acquisition. Some moderating effects of

JV type on these effects were also found. The details are discussed below.

5.2.1. The inﬂuence of change in market competition

It is hypothesized that market growth decreases the propensity of JV
dissolution but increases the propensity of JV acquisition. But the test results did not
show signiﬁcant inﬂuences of market growth on JV dissolution or acquisition. While
market growth indicates increasing market potential and opportunities, it also
increases new entries and competitive rivalry between ﬁrms (Kogut 1989). Increasing
market potential raises the parent ﬁrm’s evaluation of a joint venture, but increasing
new entries and competitive rivalry have a negative inﬂuence on the parent ﬁrm’s
evaluation of the joint venture, because heated competition increases the difficulty for
the joint venture to capture the increased market demand. How much a joint venture
can contribute to the parent ﬁrm, which essentially determines the parent ﬁrm’s
evaluation of the joint venture, depends more on the actual competitive strategy and
performance of the joint venture. Though market grth can help improve the
performance of a joint venture, this effect has been captured by the JV perfomiance
covariate. Thus, market growth itself does not directly inﬂuence the propensity of JV
dissolution or acquisition.

Increase in market concentration is hypothesized to be positively associated
with the propensity of JV dissolution and acquisition, because increase in market

concentration indicates increased market competition and the minimum scale needed

81

to participate in market competition. But the test results did not show signiﬁcant
inﬂuences of market growth on JV dissolution or acquisition.

Increase in market concentration ratio reﬂects changes in the general level of
concentration in a market, but does not capture changes in the relative position of a
joint venture in the market, which is more directly associated with the parent ﬁrm’s
reevaluation of the joint venture. For example, merger of two big ﬁrms in the JV h
industry will increase the concentration ratio. If the joint venture is a big ﬁrm that

directly competes with these two ﬁrms, this has a substantial inﬂuence on its

 

competitive position and consequently the parent ﬁrm’s evaluation of the joint -' -~
venture. But if the joint venture is signiﬁcantly smaller and does not directly compete
with these ﬁrms, this merger might not inﬂuence the joint venture as much. This is
especially true for fragmented markets. Therefore without considering the relative
position of a joint venture in the structure of competition, general change in the
market concentration level does not inﬂuence the propensity of JV termination.

Considering speciﬁc events in market competition can help to capture changes
of a joint venture’s relative position in market competition. Events such as
introduction of a new product by direct competitors or value chain expansion by
direct competitors more accurately capture changes in market competition and its
inﬂuence on the relative position of a joint venture in the market. Looking at the
inﬂuences of these events would further our understanding of the co-evolutionary
process and the inﬂuences of the market competition change on JV termination.

In sum, the test results indicate that changes in the general market
environment, such as market growth and market concentration, do not have direct
impact on the propensity of JV termination. Market growth is a measure of general

market demand increase. Without considering the relative position of a joint venture

82

in the market, market concentration ratio is also a general measure of market
competition. Changes in the general market environment may inﬂuence JV
performance and the parent ﬁrm’s strategic decisions, through which these changes
have an indirect impact on JV termination. With a longitudinal study design that also
incorporates JV performance and parent ﬁrm strategy as covariates, the indirect
impact of market environmental change is captured by JV performance and parent
ﬁrm strategy change, and it can be concluded that changes in the general market
environment do not have a direct impact on JV termination. Market level factors do
not have direct effects on JV termination when organizational level factors, such as

parent ﬁrm strategy and JV performance, are considered.

5.2.2. The inﬂuence of change in parent ﬁrm overall investment strategy

Overall, change in parent ﬁrm overall investment strategy is found to have an
impact on the propensity of JV dissolution and acquisition. Product line shift and
disruptive event increase the propensity of both JV dissolution and acquisition. This
supports the argument that parent ﬁrms’ adjustment in business proﬁles changes their
strategic focus and consequently the importance of a joint venture in the investment
portfolio, which questions the existence of the joint venture. Disruptive events such as
parent M&A also cause adjustments in the parent ﬁrms’ overall strategy and
operation, and therefore affect the existent of their joint ventures. This conﬁrms that
JV dissolution and acquisition are parent ﬁrms’ adjustment actions to changes in the
JV process, and joint ventures co-evolve with parent ﬁrm overall strategies.

Increase in parent ﬁrm manufacturing and marketing resources reduces the
propensity of JV acquisition, while increase in parent R&D resources does not. The

effect of increase in R&D resources is hypothesized to be weaker than manufacturing

83

and marketing resources because the complex and ambiguous nature of R&D
resources makes it harder to transfer to the joint venture. The results indicate that the
effect of R&D resource increase is not only weaker but also insigniﬁcant. This
provides support to the complexity and ambiguity of R&D resources and its impact on
resource sharing in a joint venture. In sum, the effects of increase in parent ﬁrm
manufacturing, marketing and R&D resources conﬁrms that, parent ﬁrm resources,
though not directly invested in a joint venture, provides potentials for resource sharing
and learning between parents, and an increase in these resources enhances parent
ﬁrms’ willingness to continue the JV partnership.

While increase in parent manufacturing and marketing resources shows
signiﬁcant effects on the propensity of JV acquisition, it does not signiﬁcantly
inﬂuence the propensity of JV dissolution. One of the differences between dissolution
and acquisition is that dissolution involves substantial cost from liquidating of the
joint venture, while acquisition does not involve such cost. Besides providing
potentials for sharing, abundant parent ﬁrm resources enables the parent ﬁrm to afford
the cost of dissolving a joint venture. Increase in parent ﬁrm resources not only
reduces the propensity of JV dissolution by providing higher potential for resource
sharing, but also makes dissolution more affordable, which counters the effects of
increased resource sharing potential. Therefore, increase in parent ﬁrm resource does
not signiﬁcantly inﬂuence the propensity of JV dissolution. This indicates different
effects of parent ﬁrm resources on JV dissolution and acquisition, and provides

support to the differences between JV dissolution and acquisition.

84

5.2.3. The influence of change in parent ﬁrm overall governance strategy

Increase in risk of appropriation is hypothesized to positively inﬂuence the
propensity of JV dissolution and acquisition. However the effect did not turn out to be
signiﬁcant. Three possible reasons can explain this result. First of all, measured with
patent correlation between partners, risk of appropriation represents the potential of
technology transfer between partners. While technology transfer can be a risk of
leaking valuable technology to the partner, it also provides beneﬁts to the partners
through interﬁrm learning, which creates synergy and competitive advantage (D02
and Hamel 1998; Hamel 1991). In fact, interﬁrm learning is an important motivation
for the formation of business partnerships (Doz and Hamel 1998). Both protection of
technology and learning play important roles in the interaction between partners, and
the risk of appropriation is related to both forces. Through protection of technology,
risk of appropriation positively inﬂuence the propensity of JV termination, but the
potential for interﬁrm learning imposes a negative inﬂuence on JV termination,
because partner frrrns are more likely to continue the joint venture when there is
higher potential of interﬁrm learning (Hamel 1991). Therefore considering the
beneﬁts of technology transfer may help clarify the effects of risk of appropriation on
JV termination.

Secondly, the inﬂuence of risk of appropriation may also be related to the risk-
taking characteristics of parent ﬁrms. The positive effect of risk of appropriation is
based on the assumption of risk adverseness of ﬁrms. However, not all ﬁrms are
equally risk adverse, and some of them may not be risk adverse. For ﬁrms that are
more tolerant to risk, the effect of risk of appropriation is less or even does not exist.

Third, it is also important to note the parent ﬁrm’s ability to learn about

changes in risk of appropriation. When involved in JV partnerships. not all ﬁrms are

85

equally aware of the risk of appropriation. In the GM and Toyota partnership, while
Toyota came to the partnership with the intention to learn about the America auto
market, GM was not prepared to protect its valuable knowledge and competitive
advantage (Doz and Hamel 1998). When a parent ﬁrm is not aware of potential risk of
appropriation, its JV decisions are not inﬂuenced by this risk. When there is an
increase in the risk of appropriation, only parent ﬁrms that are aware of this change
will be able to adjust accordingly.

Decrease in the number of alternative partners reduces the propensity of JV
dissolution, but it does not inﬂuence JV acquisition. Formation of competing
partnerships by the parent ﬁrms increases the propensity of JV dissolution, but not
acquisition. Both decrease in alternative partners and formation of competing
partnerships inﬂuence the focal JV by changing the parent ﬁnn’s dependence on the
partnerships. They both provide substitutes for the focal JV partnership, and make it
easier for the parent ﬁrm to switch to other partners. The result indicates changed
level of dependence and possibility of switching partners affect the propensity of JV
dissolution but not acquisition. While dissolution can be motivated by switching
partners, acquisition is more related to parent ﬁrm’s internalization decision of the JV
activity. This provides support to the different mechanisms underlying JV dissolution

and acquisition.

5.2.4. The inﬂuence of JV sales performance

As hypothesized, JV sales performance negatively inﬂuences the propensity of
JV dissolution, which conﬁrms parent ﬁrms are more likely to dissolve a low-
performing joint venture. Different from the hypothesis, JV performance is found to

be negatively related to the propensity of acquisition, indicating that a joint venture

86

that is not performing well is more likely to be acquired one of the parent ﬁrms. This
shows the motivation of JV acquisition is not only to fully exploit beneﬁts, but very
often is to internalize and achieve more control (Buckley and Casson 1996). Poor
performance lowers the parent ﬁrm’s evaluation of the joint venture and motivates the
parent ﬁrm to achieve more control over the joint venture to improve performance.
During the JV process, low performance gives the parent ﬁrm a legitimized excuse to
propose a more active role in JV governance (Yan and Gray 1994), and undesirable
JV performance has been found to be a reason for JV instability and a stimulus to

changes in JV governance form (Yan 1998; Yan and Gray 1994).

5.2.5. The moderating effects of JV type
Related JV

Among the hypothesized interaction effects of related JV, only the interaction
between related JV and increases in marketing resources and R&D resources show
signiﬁcant positive effects on JV dissolution, though the main effects are not
signiﬁcant. This indicates that increase in these resources increases the propensity of
dissolution for related JVs, but not for unrelated JV 5. A possible explanation is,
because a related JV operates in the same industry as the parent ﬁrm, it shares very
similar resources as the parent ﬁrm, and this in a sense puts the joint venture into
competition with the parent ﬁrm’s own business for resources in the overall portfolio
of investment. When the parent ﬁrm increases its own resources, these resources are
utilized in the parent ﬁrm, not in the joint venture. Because the joint venture and the
parent ﬁrms are in the same industry, given the same amount of available resources
for this industry in the overall portfolio, increase in parent ﬁrm resources can indicate

a shift of focus from the joint venture to the parent ﬁrm’s internal business, or the

87

expansion of the parent ﬁrm’s internal business that reduces the importance of the
joint venture in the overall investment portfolio. Both of these changes will increase
the possibility of the joint venture being dissolved.

The effects of other covariate with regard to change in investment strategy,
including product line shift, increase in manufacturing resources and disruptive
events, stay same for related and unrelated JVs. In general, the moderating effects of
related JV 5 are found to be limited. Related diversiﬁcations are not always preferred
by ﬁrms to unrelated diversiﬁcation, and related JV 5 are not always given priority in
parent ﬁrm’s adjustment of investment strategy. When the parent ﬁrm is engaged in a
major strategic adjustment, related business may be ﬁrst to divest. This reminds us to
rethink the role of product relatedness in JV termination and ﬁrms’ investment
strategies.

First, the role of product relatedness can be different depending on the speciﬁc
content of strategic change. Compared to unrelated JVs, related JVs are more closely
connected to the parent ﬁrm’s primary business. During the adjustment of parent ﬁrm
overall investment strategy, related JVs can be given the priority to continue because
they are closer to the core business, but on the other hand, they are under more direct
inﬂuence of the parent ﬁrm’s strategy and therefore more likely to be terminated
during strategic change. Which inﬂuences are stronger depends on the speciﬁc
content of strategic adjustment. For example, during product consolidation to focus on
the core business, related JVs are less likely to be terminated because of its
imponance to the core business. However, when the parent ﬁrm is involved in a quick
expansion to a completely different market, a related JV may be more likely to
terminate than an unrelated JV, because resources are prioritized toward the

expansion of the unrelated business. Investigation of the inﬂuence of speciﬁc events

88

of change, such as product consolidation, diversiﬁed expansion, etc. would improve
our understanding of the role relatedness in JV termination.

Secondly, relatedness is only one dimension that how a joint venture is
connected to the parent’s primary business. Besides relatedness, the relative position
and importance of the JV business in the parent ﬁrm’s overall investment portfolio are
associated with other factors such as the size of the JV or customer base for the JV
product. F irm’s investment strategies are also becoming more complicated than just
diversiﬁcation and consolidation (Bergh 2001). Other factors such as the interaction
between market demands of different products are also important consideration of
product portfolio management, and also inﬂuence the parent ﬁrm’s decision on the JV

business.

Direct-competitor JV

No interaction effect between direct-competitor JV and parent governance
strategy change is found except for decrease in alternative partners. Decrease in
alternative partners increases the propensity of acquisition for direct-competitor J Vs
but not for non-direct-competitor J Vs. This conﬁrms the hypothesized effect that,
because JV parents in the same industry have more substitutable resources, when the
availability of alternative partners decreases, the parent ﬁrm has the ﬂexibility of
obtaining needed resources internally instead of becoming more dependent on the
current partner. Therefore for a direct-competitor JV, when less alternative partners
are available, the propensity of acquiring the joint venture is increased. This effect
does not exist for no-direct competitor JV because when parent ﬁrms are not from the
same industry and the resources they seek from the JV partner is less likely to be

internally available.

89

In general, the moderating effects of direct-competitor JV are found to be
limited. Governance strategy is an important consideration for both joint ventures
between direct competitors and joint ventures that do not direct complete.
Cooperation between competitors has become more and more common, and firms are
ready to take the risk of working with competitors and are learning about cooperation
in the presence of competition (Bleeke and Ernst 1993; Doz and Hamel 1998). This
also indicates that looking at factors that are directly related to the joint venture and
its parent ﬁrms’ strategies can reveal more information joint ventures than a simple

classiﬁcation of direct-competitor JV versus non-direct-competitor JV.

5.2.6. A comparison of JV dissolution and acquisition

It is generally conﬁrmed that antecedent factors have differentiated effects on
JV dissolution and acquisition. One difference is that increase in parent ﬁrm resources
inﬂuences the propensity of JV acquisition, but not dissolution. This can be explained
with the high cost of dissolving a joint venture, which involves liquidating the JV
asset. Firms lacking resources try to avoid dissolution by keeping theirjoint ventures
in operation or terminating them through acquisitions, which does not involve high
liquidation cost. Parent ﬁrm resources provide potential for resource sharing, which
supports the continuity of a joint venture; on the other hand, they enable the parent
ﬁrm to afford the cost of liquidating a joint venture. So increase in parent ﬁrm
resources reduces the propensity of JV acquisition, but does not have an inﬂuence on
the propensity of dissolution.

Another difference is, decrease in alternative partners and formation of
competing partnerships inﬂuence the propensity of JV dissolution but not acquisition.

This is due to the different governance strategy considerations for dissolution and

90

acquisition. Dissolution can be motivated by the intention of switching partners. Both
decrease in the number of alternative partners and formation of competing
partnerships change the possibility of partner switching and the dependence of the
parent ﬁrm on the focal joint venture. Acquisition, however, is mainly an
internalization decision (Buckley and Casson 1996), not motivated by switching
partners, and therefore not inﬂuenced by decrease in the number alternative partners
or formation of competing partnerships.

In sum, the test results show different underlying mechanisms for JV
dissolution and acquisition. The comparison of analysis results for JV dissolution,
acquisition and termination in total also reveals that due to the dominant proportion of
acquisition in JV terminations, the results for JV termination in total is very similar to
that of JV acquisition, but quite different from that of JV dissolution. Considering the
differences between JV dissolution and acquisition, only looking at JV termination in
total would very likely produce misleading results and inhibits our understanding of
the reasons and implications for JV termination. This further highlights the danger not

distinguishing and separately modeling JV dissolution and acquisition.

91

CHAPTER 6
CONCLUSION
6.1. Synthesis of signiﬁcant ﬁndings
1. External change is an important reason for JV termination.

External change is found to be an important reason for JV termination.
Although changes in the market environment did not show signiﬁcant inﬂuences,
changes in the parent ﬁrm’s overall investment strategy and governance strategy
signiﬁcantly inﬂuences the propensity of JV termination. During the JV co-
evolutionary process, changes in the parent ﬁrm’s overall investment strategy and
governance strategy induce the parent ﬁrm to reevaluate ajoint venture and readjust
its JV related decisions. With the effect of JV performance being considered, external
change still showed signiﬁcant inﬂuences on the propensity of JV termination,
including dissolution and acquisition. This indicates that external change is an
important reason for JV termination and requires more attention in JV research.

2. JV termination due to external change is not necessarily failure.

Based upon the ﬁrst ﬁnding, this study conﬁrms that external change and JV
performance are two primary reasons for JV termination. While JV terminations due
to low performance are related to failure, terminations caused by external change are
not necessarily failures, but rather strategic adaptations.

3. Change rather than initial formation condition causes JV termination.

By including both level variables and change variables in a longitudinal
model, and controlling for some initial formation condition variables such as parent
JV type, parent diversiﬁcation level, etc., this study shows that, while initial formation
condition may inﬂuence the propensity of JV termination, change in the JV co-

evolutionary process is a direct cause of JV terminations. JV termination is resultant

92

from changes in the process of JV co-evolution with the market environment and
parent ﬁrm overall strategy.
4. A joint venture is embedded in the parent ﬁrm’s overall strategy, and JV
termination can result from parent ﬁrm strategic adjustment.

This study incorporated factors exogenous to ajoint venture and showed the
important inﬂuence of parent ﬁrm overall strategy on the outcome of a joint venture.
Changes in the parent ﬁrm’s overall investment strategy such as product line shift, '
resource increase and disruptive events, and changes in the parent ﬁrm’s overall
governance strategy such as decrease in alternative partners and formation of
competing partnerships are found to have signiﬁcant impacts on the propensity ofJV
termination, including both dissolution and acquisition. These ﬁndings demonstrated
the embeddedness of joint ventures in their parent ﬁrms’ overall strategies, and how
joint ventures evolve together with parent ﬁrm overall strategies. It also supported the
argument that exogenous factors play an important role in determining the outcome of
a joint venture.

5. General market environment changes such as market growth and market
concentration increase do not have direct influences on the propensity of

JV termination.

This study shows that, while changes in the market environment inﬂuence the
performance of joint venture and may induce parent ﬁrm to adjust their decisions
regarding the joint venture, general market environment changes do not have direct
inﬂuences on the propensity of JV termination. When organizational level factors,
including joint venture performance and parent ﬁrm strategy, are considered, general
market environment changes such as market growth and market concentration

increase do not show signiﬁcant inﬂuence on JV termination. This indicates that

93

organizational level factors play a more direct role than general market environmental
factors. Faced with the same environmental changes, joint ventures that are able to
implement effective competitive strategy and produce high performance survive.

6. Different mechanisms underline JV dissolution and acquisition.

This study found some different effects of the antecedent factors for JV
dissolution and acquisition. This implies that dissolution and acquisition are grounded
in different motivations of the parent ﬁrms and have different implications for the
parent ﬁrm. Dissolution involves high cost from liquidating assets, while acquisition
can avoid such cost. From the perspective of investment strategy, acquisition can be
related to the parent ﬁrm’s expansion, indicating increased resource commitment in
the JV business, while dissolution is a divestment decision. From the perspective of
governance strategy, dissolution can be motivated by the intention of switching
partners; acquisition is an internalization decision to achieve more control over the
venture. These differences further demonstrate the importance of separately

examining and comparing these two types of terminations.

6.2. Contributions
6.2.1. Theoretical contribution

First of all, this study provides an empirical test of co-evolutionary theory of
alliances and further co-evolutionary theory. The empirical tests of this theory have
been limited to case studies. This study conducts a quantitative test of this theory in
the context of JV termination with a longitudinal study design. It provides empirical
support to the co-evolutionary process of joint ventures with their parent ﬁrms and the

market environment, and demonstrates the inﬂuences of the co-evolutionary process

94

on the JV outcome. It contributes to both the co-evolutionary theory and the area of
alliances and joint ventures.

Secondly, adopting a co-evolutionary theory view, this study brings a new
perspective of external change to JV termination studies. While previous studies
focused on JV initial formation conditions and factors under JV management control.
this study demonstrates that external change is also an important reason for JV
termination. Change, rather than initial formation conditions, is identiﬁed as a direct
cause of JV termination. This not only addresses the need to study the JV
management process, but also reveals the importance of taking an evolutionary view
in looking JV termination.

The signiﬁcant inﬂuence of external change on JV termination provides
support to and extends the evolutionary cycle of reevaluation and readjustment in the
JV process (Doz 1996). During the partnering process, parent ﬁrms reevaluate the
operation and performance of a joint venture as well as the joint venture’s
contribution to the overall parent ﬁrm strategy, and make adjustments on the joint
venture decisions, which may lead to termination of the joint venture. This study
shows that the cycle of reevaluation and readjustment (Doz 1996) not only involves
the speciﬁcations of the joint venture itself, but also concerns the relative position of
the joint venture in the overall strategic portfolio of the parent ﬁrm. By including
external changes in parent ﬁrm overall strategy and the market environment, this
study extends our understanding of the reevaluation and readjustment cycle during the
JV partnering process.

Thirdly, this study demonstrates the embeddedness of joint ventures in parent
ﬁrm overall strategy and supports the co-evolution of joint ventures with parent ﬁrm

strategies. The inﬂuence of parent ﬁrm overall strategy change on the propensity of

95

JV termination shows the importance of viewing joint ventures in their parent ﬁrms’
overall strategy portfolio, which has been largely neglected in previous literature.

Further, this study develops a comprehensive model of JV termination that
incorporates both JV performance and factors exogenous to a joint venture, including
the market environment and parent ﬁrm overall strategy. The model integrates
industry organization, diversiﬁcation and governance literature, and has substantial
explanatory power of JV termination.

In addition, based on the model, this study identiﬁes two primary reasons for
JV termination: external change and JV performance. While relating low JV
performance to failure, it argues that terminations that are caused by external change
is not necessarily failure, but rather strategic adaptations of the parent ﬁrm, which
may not be a bad thing for the parent ﬁrm. This helps clarify the misconception of
relating JV termination solely to failure.

Lastly, this study distinguishes between two types of JV termination:
dissolution and acquisition. It simultaneously investigates JV dissolution and
acquisition and identiﬁes different effects of the antecedent factors, which has rarely
been done in the literature. This reminds researchers to take into consideration of the

different nature of dissolution and acquisition in JV termination studies.

6.2.2. Empirical contribution

This study empirically tests a longitudinal model of joint venture termination
with time-dependent covariates, and ﬁnds various factors that signiﬁcantly contribute
to JV termination, such as parent product line shift, parent resources, parent disruptive
events, decrease in alternative partners, formation of competing partnerships, etc. It is

also veriﬁed the inﬂuence of JV performance on termination, which has been

96

assumed in the literature but rarely tested in a comprehensive model with the
longitudinal study design. These ﬁndings contribute to our understanding of the
reasons for JV termination.

In addition, this study ﬁnds that acquisition accounts for a large proportion
(86% in this study) of JV termination, while dissolution is actually a very small
proportion (14% in this study) of total terminations. Total JV termination rate found
in this study is 59.33%, while acquisition rate is 50.67% and dissolution rate is only
8.67%. The previously found high termination rate (referred as failure rate in some
studies) of joint ventures needs to be reevaluated. Because acquisition and dissolution
have different implications for the parent ﬁrms, separately evaluating dissolution rate
and acquisition rate is necessary. This will also contributes to our understanding of the

actual failure rate of joint ventures.

6.2.3. Managerial implications

Failure studies provide different insights from success studies. Understanding
the reasons for JV termination can help managers to anticipate and be prepared for
terminations. Despite of the frequent occurrence of JV terminations, managers are
often caught off guard by the termination and are ill prepared to cope with the
tensions and uncertainties associated with the breakdown of the relationship (Pen g
and Shenkar 2002). Being able to anticipate possible terminations helps managers to
be prepared and better manage terminations.

More importantly, this study emphasizes that termination is not necessarily a
failure if it is caused by external changes. Terminating a joint venture that no longer
ﬁt into the overall strategy is a necessary adaptation action rather failure. Even

further, being able to do so in time contributes to the overall welfare of the firm.

97

Therefore knowing termination is not necessary failure enables the transformation of
JV management philosophy from avoiding termination to anticipating, preparing and
even actively planning termination.

Such a proactive view in JV management reminds mangers to specify terms of
termination upfront when a joint venture is formed. Most joint ventures are formed
without detailed terms regarding possible terminations. Managers need to be
comfortable to discuss termination at the formation stage, as these prespeciﬁed terms

will greatly reduce the uncertainty and tension coming with unexpected termination.

6.3. Limitations and directions for future research

One limitation of this study is that only one parent of a joint venture is
considered. This is largely due to the limited data availability. If future research could
include both parents and investigate the interaction between parents during the JV
process characterized by change, it would enrich our understanding of the co-
evolutionary process of joint ventures with the parent ﬁrms, especially the mechanism
through which parent ﬁrms’ adaptation strategies interact with each other and
therefore inﬂuence the outcome of a joint venture.

To fully explore the co-evolution of joint ventures with the parent ﬁrms,
further studies can investigate the inﬂuences of joint ventures on parent ﬁrm strategy.
for example, how unexpected termination of a joint venture inﬂuences parent ﬁrm
strategy and performance. Some studies have been done on the value creation effects
of JV acquisition (e.g. Kumar 2005) in ﬁnance and management literature, but
comprehensive studies are still needed on the inﬂuence of JV termination on parent
ﬁrm strategic adjustment, especially the shock and adjustment needed after an

unexpected JV termination.

98

Further, under this framework, other factors on changes in parent ﬁrms overall
strategy and the market competition can be investigated. Especially valuable would be
variables that capture the speciﬁc position of a joint venture in the market
competition, and the relative position of a joint venture in the parent ﬁrm’s overall
strategic portfolio. For example, market concentration increase may have different
inﬂuences depending on the joint venture’s competitive position in the market, but
value chain expansion by direct competitors will directly increase competition
intensity for the focal joint venture. Parent product line shift is the general adjustment
of the whole product portfolio, but the event of introducing a new product in the same
category as the JV product considers the relative position of the JV product in the
parent ﬁrm’s overall product portfolio, and captures the speciﬁc inﬂuence of parent
ﬁrm strategy change on the focal joint venture. Compared to variables on general
changes of parent ﬁrm strategy and market competition, these variables would be able
to better capture the connection between the joint venture and its parent ﬁrm strategy
or the market environment, and be able to provide valuable contribution to the
inﬂuence of external change on JV termination.

Lastly, this study reveals the different motivations and implications for JV
dissolution and acquisition. Further investigations on the different mechanisms
underlying dissolution and acquisition, and ﬁrms’ choices between dissolution and
acquisition would be very interesting and helpful for the understanding of JV
termination and failure. In addition, from a parent ﬁrm’s perspective, selling off a
joint venture and acquiring a joint venture may have different reasons and
implications. Both these two situations are included in acquisition in this study.
Separately investigating parent sell-off and parent acquisition would be interesting

future research.

99

6.4. Conclusion

This study adopts a co-evolutionary theory view of joint ventures and
investigates the inﬂuences of external change, including changes in parent ﬁrm
overall strategy and in the market competition, on the propensity of JV termination. It
develops and tests a model of JV termination that incorporates external change as well
as JV performance as determinants of the propensity of JV termination.

This study emphasizes change, rather than initial JV formation conditions, as
an important cause of JV termination, and employs a longitudinal study design
examine the over time effects of the antecedent factors. Viewing a joint venture as
evolving in accompaniment with parent ﬁrm overall strategy and the market
environment, it demonstrates the embeddedness of joint ventures in parent ﬁrm
overall strategy, and how JV termination can result from parent ﬁrm strategic
adjustment. It also contributes to the literature by simultaneously examining and
comparing the differences between two types of JV termination: dissolution and
acquisition.

Based on the model, this dissertation study identiﬁes two causes of JV
termination: low JV performance and changes external to the joint venture.
Termination due to low performance is directly associated with failure, whereas

change-induced termination is a result of ﬁrm strategic adaptation.

100

REFERENCES

Allison, Paul D. (1984), Event history analysis : regression for longitudinal event
data. Newbury Park, Calif: Sage Publications.

---- (1995), Survival analysis using SAS: a practical guide. Cary,NC: SAS Institute
Inc.

Anderson, Erin and Hubert Gatignon (I986), "Modes of Foreign Entry: A Transaction
Cost Analysis and Propositions," Journal of International Business Studies, 17 (3 ), 1.

Andrews, K (1974), The concept of corporate strategy. Homewood, Illinois: Dow
Joines-Irwin.

Arino, Africa and Jose de la Torre (1998), "Learning from failure: Towards an
evolutionary model of collaborative ventures," Organization Science, 9 (3), 306.

Barkema, Harry G, Oded Shenkar, Freek Vermeulen, and John H J Bell (1997),
"Working abroad, working with others: How ﬁrms learn to operate international joint
ventures," Academy of Management Journal, 40 (2), 426.

Baum, J. A. C. (1999), "Whole-part coevolutionary competition in organizations," in
Variations in organization science: In honor of Danald T. Campbell, J. A. C. Baum
and B. McKelvey, Eds. London: Sage.

Baum, J. A. C. and J. V. Singh (1994), "Organizational hierarchies and evolutionary
processes: some reﬂections on a theory of organizational evolution," in Evolutionary
dynamics of organizations, J. A. C. Baum and J. V. Singh, Eds. New York: Oxford
University Press.

Bergh, Eonald D. Ed. (2001), Diversiﬁcation strategy at crossroad: established,
emerging and anticipated paths. Oxford: Blackwell Publishers.

Bleeke, Joel and David Ernst (1993), Collaborating to compete: using strategic
alliances and acquisitions in the global market place: John Wiley & Sons, Inc.

Blodgett, Linda Longfellow (1992), "Research Notes and Communications Factors in
the Instability of International Joint Ventures: An Event History Analysis," Strategic
Management Journal, 13 (6), 475.

Box-Steffensmeier, Janet M (2004), Event history modeling : a guide for social
scientists. Cambridge ; New York: Cambridge University Press.

Buckley, Peter J and Mark Casson (1996), "An economic model of international joint
venture strategy," Journal of lntemational Business Studies, 27 (5), 849.

Buckley, Peter, J. and Marc Casson, C. (l998a), "Analyzing foreign market entry

strategies: Extending the internalization approach," Journal of lntemational Business
Studies, 29(3), 539.

101

Buckley, Peter, J. and Mark Casson, C. (l998b), "Models of the multinational
enterprise," Journal of lntemational Business Studies, 29(1), 21.

Burton, Peter S (1994), "Product portfolios and the introduction of new products: A11,"
The Rand Journal of Economics, 25 (l), 128.

Caves, Richard E. (1980), "Industrial Organization, Corporate Strategy and
Structure," Journal of Economic Literature, 18(1), 64.

Caves, Richard E. and Michael E. Porter (1980), "The Dynamics of Changing Seller
Concentration," The Journal of Industrial Economics, 29 (l ), 1.

Chang, S. J. (1996), "An evolutionary perspective on diversiﬁcation and corporate
restructuring: Entry, exit, and economic performance during 1981-89," Strategic
Management Journal, 17 (8), 587-611.

Chi, T. L. (2000), "Option to acquire or divest a joint venture," Strategic Management
Journal, 21 (6), 665-87.

Chowdhury, Jafor (1992), "Performance of international joint ventures and wholly
owned foreign subsidiaries: A comparative perspective," Management lntemational
Review, 32 (2), 115.

Chung, Wilbur (2001), "Identifying technology transfer in foreign direct investment:
Inﬂuence of industry conditions and investing ﬁrm motives," Journal of International
Business Studies, 32(2), 211.

Contractor, F. J. and P. Lorange (1988), "why should ﬁrms cooperate? the strategic
and economic basis for cooperative ventures," in Cooperative strategies in
international business, Contractor, F. J. and Lorange, P. eds. (Lexington books:
Lexington, MA.).

Cristina Lopez, Duarte and Garcia-Canal Esteban (2004), "The Choice Between Joint
Ventures and Acquisitions in Foreign Direct Investments: The Role of Partial

Acquisitions and Accrued Experience," Thunderbird lntemational Business Review,
46(1), 39.

Cummings, Jeffrey L. and Bing-Sheng Teng (2003), "Transferring R&D knowledge:
The key factors affecting knowledge transfer success," Journal of Engineering and
Technology Management, 20 (1,2), 39.

Day, George S (1995), "Advantageous alliances," Academy of Marketing Science.
Journal, 23 (4), 297.

Day, George S. (1977), "Diagnosing the product portfolio," Journal of Marketing, 41
(2), 29.

Devinney, Timothy M. and David W. Stewart (1988), "Rethinking The Product
Portfolio: A Generalized Investment," Management Science, 34 (9), 1080.

102

Dhanaraj, Charles and Paul W Beamish (2004), "Effect of equity ownership on the
survival of international joint ventures," Strategic Management Journal, 25 (3), 295.

Doz, Yves L (1996), "The evolution of cooperation in strategic alliances: initial
conditions or learning processes?," Strategic Management Journal (1986-1998), 17
(Special issue), 55.

Doz, Yves L and Gary Hamel (1998), Alliance advantage: the art of creating value
through partnering. Boston: Harvard Business School Press.

Dussauge, Pierre, Bernard Garrette, and Will Mitchell (2000), "Learning from
competing partners: Outcomes and durations of scale and link alliances in Europe,
North America and Asia," Strategic Management Journal, 21 (2), 99.

Dutta, Shantanu, Om Narasimhan, and Surendra Rajiv (1999), "Success in high-
technology markets: ls marketing capability critical?," Marketing Science, 18 (4),
547.

Dyer, Jeffrey H and Harbir Singh (1998), "The relational view: Cooperative strategy
and sources of interorganizational competitive advantage," Academy of Management.
The Academy of Management Review, 23 (4), 660.

Franko, Lawrence G. (1971), Joint venture survival in multinational corporations.
New York: Praeger Publishers.

Ganesan, Shankar (1994), "Determinants of long-term orientation in buyer-seller
relationships," Journal of Marketing, 58 (2), l.

Goh, Swee C (2001), "Management strategies for successful post-acquisition
integration," The Academy of Management Executive, 15 (l), 152.

Gomes-Casseres, Benjamin (1987), "Joint Venture Instability: Is It a Problem?,"
Columbia Journal of World Business, 22 (2), 97.

Goodman, Lester E and Paul A Dion (2001), "The determinants of commitment in the
distributor-manufacturer relationship," Industrial Marketing Management, 30 (3 ), 287.

Gulati, Ranjay (1995), "Social structure and alliance formation patterns: A
longitudinal analysis," Administrative Science Quarterly, 40(4), 619.

Hamel, Gary (1991), "Competition for Competence and Inter-Partner Learning Within
lntemational Strategic Alliances," Strategic Management Journal, 12, 83.

Harrigan, Kathryn Rudie (1988), "Strategic Alliances and Partner Asymmetrics,"
Management lntemational Review, 28, 53.

Heide, Jan B. and George John (1992), "Do Norms Matter in Marketing
Relationships?," Journal of Marketing, 56 (2), 32.

103

---- (1988), "The role of dependence balancing in safeguarding trasaction-speciﬁc
assets in conventional channels," Journal of Marketing, 52 (l), 20.

Hennart, Jean-Francois, Dong-Jae Kim, and Ming Zeng (1998), "The impact of joint
venture status on the longevity of Japanese stakes in US. manufacturing afﬁliates,"
Organization Science, 9(3), 382.

Horowitz, Ira (1984), "Market Growth, Demand Uncertainty, and Potential Entry."
The Quarterly review of economics and business, 24 (2), 33.

Jaffe, Adam B. (1986), "Technological Opportunity and Spillovers of R&D: Evidence
from Firms' Patents, Proﬁts, and Market Value," The American Economic Review, 76
(5), 984.

Khanna, Tarun (1998), "The scope of alliances," Organization Science, 9(3), 340.

Kogut, Bruce (1991), "Joint Ventures and the Option to Expand and Acquire,"
Management Science, 37 (l), 19.

---- (1989), "The Stability Of Joint Ventures Reciprocity And Competitive," The
Journal of Industrial Economics, 38 (2), 183.

---- (1988), "A Study of the Life Cycle of Joint Ventures," Management International
Review, 28, 39.

Koza, Mitchell P and Arie Y Lewin (1998), "The co-evolution of strategic alliances."
Organization Science, 9(3), 255.

---- (1999), "The coevolution of network alliances: A longitudinal analysis of an
international professional service network," Organization Science, 10 (5), 638.

Kumar, M. V. Shyam (2005), "The value from acquiring and divesting a joint venture:
a real options approach," Strategic Management Journal, 26(4), 321.

Kumar, Nirmalya, Lisa K Scheer, and Jan-Benedict E M Steenkamp (1995), "The
effects of supplier fairness on vulnerable resellers," J MR, Journal of Marketing
Research, 32 (l), 54.

Kumar, Rajesh and Koﬁ O. Nti (1998), "Differential learning and interaction in
alliance dynamics: A process and outcome discrepancy model," Organization Science.
9(3), 356.

Lewin, Arie Y, Chris P Long, and Timothy N Carroll (1999), "The coevolution of
new organizational forms," Organization Science, 10(5), 535.

Lewin, Arie Y and Henk W. VoIberda (1999), "Prolegomena on coevolution: a

framework for research on strategy and new organizational forms," Organization
Science, 10(5), 519-34.

104

Li, J iatao (1995), "Foreign entry and survival: effects of strategic choices on
performance in international markets," Strategic Management Journal (1986-1998),
16 (5), 333.

Lu, J. W. and D. Xu (2006), "Growth and survival of international joint ventures: An
extemal-intemal legitimacy perspective," Journal of Management, 32 (3), 426-48.

Lu, Jane W and Louis Hebert (2005), "Equity control and the survival of intemational
joint ventures: a contingency approach," Journal of Business Research, 58 (6), 736.

Luo, Y. D. (2002a), "Product diversiﬁcation in international joint ventures:
Performance implications in an emerging market," Strategic Management Journal, 23
(1), 1-20.

Luo, Yadong (2002b), "Product diversiﬁcation in international joint ventures:
Performance implications in an emerging market," Strategic Management Journal, 23

(1), 1-

March, J. G. (1994), "The evolution of evolution," in Evolutionary dynamics of
organizations, J. A. C. Baum and J. V. Singh, Eds. New York: Oxford University
Press.

McKelvey, Bill (1997), "Quasi-natural organization science," Organization Science, 8
(4), 352.

Nonaka, Ikujiro (1994), "A dynamic theory of organizational knowledge creation,"
Organization Science, 5 (l), 14.

Park, Seung Ho, Roger Chen, and Scott Gallagher (2002), "Firm resources as
moderators of the relationship between market growth and strategic alliances in
semiconductor start-ups," Academy of Management Journal, 45 (3), 527.

Park, Seung Ho and Michael V Russo (1996), "When competition eclipses
cooperation: An event history analysis of joint venture failure," Management Science.
42 (6), 875.

Park, Seung Ho and Gerardo Ungson (2001), "lnterﬁrrn rivalry and managerial
complexity: A conceptual framework of alliance failure," Organization Science, 12
(1), 37.

Park, Seung Ho and Gerardo R Ungson (1997), "The effect of national culture,
organizational complementarity, and economic motivation on joint venture

dissolution," Academy of Management Journal, 40(2), 279.

Parkhe, Arvind (1991), "Interﬁrm Diversity, Organizational Learning. and
Longevity," Journal of lntemational Business Studies, 22 (4), 579.

---- (1993), "Strategic alliance structuring: A game theoretic and transaction cost
examination of interﬁrm cooperation," Academy of Management Journal, 36 (4), 794.

105

Peng, M. W. and O. Shenkar (2002), "Joint venture dissolution as corporate divorce,"
Academy of Management Executive, 16(2), 92-105.

Pennings, Johannes M, Harry Barkema, and Sytse Douma (1994), "Organizational
learning and diversiﬁcation," Academy of Management Journal, 37 (3), 608.

Peteraf, Margaret A (1993), "The cornerstones of competitive advantage: A resource-
based," Strategic Management Journal, 14 (3), 179.

Pfeffer, Jeffrey and Berald R. Salancik (1978), The external control of organizations.
New York, Hagerstown, San Francisco, London: Harper & Row publishers.

Porter, M. E. (1985), Competitive advantage: creating and sustaining superior
performance. New York: Free Press.

---- (1980), competitive strategy: techniques for analyizing industries and competitors.
New York: Free Press.

Porter, Michael E. (1987), "From Competitive Advantage to Corporate Strategy,"
Harvard Business Review, 65 (3), 43.

Ramanujam, Vasudevan and Varadarajan, P. (1989), "Research on corporate
deversiﬁcation: a synthesis," Strategic Management Journal (1986-1998), 10(6). 523.

Reuer, J. J. and Mitchell P Koza (I998), "lntemational joint venture instability and
corporate strategy," in The strategy and organization of international business, Peter J.
Buckley and Fred Burton and Haﬁz Mirza, Eds. New York: St. Martin's Press.

Rindﬂeisch, Aric and Jan B Heide (1997), "Transaction cost analysis: Past, present,
and future applications," Journal of Marketing, 61 (4), 30.

Ring, Peter Smith and Andrew H Van de Ven (1994), "Developmental processes of
cooperative interorganizational relationships," Academy of Management. The
Academy of Management Review, 19(1), 90.

Scherer, F. M. (1980), "The determinants of market structure," in Industrial market
structure and economic performance. Chicago: Rand McNally.

Simonin, Bernard L (1999), "Ambiguity and the process of knowledge transfer in
strategic alliances," Strategic Management Journal, 20 (7), 595.

Spender, J. C. (1996), "Making knowledge the basis of a dynamic theory of the ﬁrm,"
Strategic Management Journal, 17, 45.

Steensma, H. K. and M. A. Lyles (2000), "Explaining IJV survival in a transitional
economy through social exchange and knowledge-based perspectives," Strategic
Management Journal, 21 (8), 831-51.

Therneau, Terry M. (2000), Modeling survival data : extending the Cox model. New
York: Springer.

106

VoIberda, Henk W and Arie Y Lewin (2003), "Guest Editor's Introduction: Co-
evolutionary Dynamics Within and Between Firms: From Evolution to Co-evolution,"
The Journal of Management Studies, 40(8), 2111.

Warshaw, Paul R (l980a), "Predicting purchase and other behaviors from general and
contextually speciﬁc intentions," JMR, Journal of Marketing Research (pre-1986), l7
(1), 26.

Warshaw, Paul R. (1980b), "A New Model for Predicting Behavioral Intentions: An
Alternative to Fishbein," JMR, Journal of Marketing Research, 17 (2), 153.

Wemerfelt, Birger (1984), "A Resource-based View of the Firm," Strategic
Management Journal (pre-l986), 5 (2), 171.

Williamson, 0. (1985), The Economic Institutions of Capitalism. New York: The
Free Press.

---- (1975), Market and Hierarchies. New York: The Free Press.

Yan, Aimin (1998), "Structural stability and reconﬁguration of international joint
ventures," Journal of lntemational Business Studies, 29(4), 773.

Yan, Aimin and Barbara Gray (1994), "Bargaining power, management control, and
performance in Uni," Academy of Management Journal, 37 (6), 1478.

Yan, Aimin and Ming Zeng (1999), "International joint venture instability: A critique
of previous research, a reconceptualization, and directions for future research,"
Journal of lntemational Business Studies, 30(2), 397.

20110, Maurizio and Harbir Singh (2004), "Deliberate learning in corporate

acquisitions: post-acquisition strategies and integration capability in US. bank
mergers," Strategic Management Journal, 25 (13), 1233.

‘107

 

 

   

11111litigtijtttltgit1