Running head: MULTIPARTNER R&D ALLIANCE FORMATION

A Systematic Literature Review of the Key Formation Antecedents, and Processes

of Multipartner Research and Development Alliances

Janice A.C. Lalu

Faculty of Economics and Business

University of Groningen, The Netherlands

Supervisor: Dr. I. Estrada Vaquero

Co-assessor: Dr. Q. J. Dong

Groningen, Monday 22 June 2015

Abstract

Despite an increasing popularity of multipartner research and development (MR&D)

alliances, research in the field remains limited, and is fragmented. This paper focuses on the aspect

of alliance formation. A systematic literature review is conducted with the purpose to consolidate

the fragmented field by constructing a theoretical framework, reveal patterns in MR&D alliance

formation. Outcomes reveal the following influential factors in alliance formation decisions:

alliance formation motives, enablers, barriers, processes, and mechanisms. Findings show that

relevance of formation motives, enablers, barriers vary across context, namely R&D and non-R&D.

This paper also provides suggestions for further research based on the study outcomes combined

with its limitations.

Keywords: multipartner research and development alliance, alliance formation, R&D

collaboration, formation process

Table of Contents

Abstract

2

Table of Contents

3

Tables and Figures

5

Introduction

6

Methodology

10

Planning

10

Execution

11

Eligibility criteria.

11

Keyword exploration and snowball sampling.

12

Reporting

13

Results, Discussion

13

Motives for Multipartner Alliance Formation

13

Firm level motives.

15

Skills, knowledge, capabilities (SKCs), resources, and assets development.

15

Cost savings, efficiency, scale economies.

17

Risk sharing.

17

Market access.

18

Technology development, innovation boost.

18

Alliance level motives.

19

Industry level motives.

19

Institution level motives.

20

Customer level motives.

20

Enablers and Barriers for Multipartner Alliance Formation

21

Formation enablers.

22

Firm level enablers.

22

Alliance level enablers.

25

Formation barriers.

26

Firm level barriers.

26

Alliance level barriers.

29

Institution level barriers.

30

Multipartner Alliance Formation Processes

31

Stage-game.

32

Emergent, engineered, embedded.

32

Multipartner Alliance Formation Mechanisms

33

Coordination mechanisms.

34

Incentives.

35

Partner selection.

36

Cross Comparison: R&D Collaboration Versus Non-R&D Collaboration

36

Conclusion, Future Research

38

References

40

Appendices

51

Appendix A: Multipartner Alliance Formation Motives

51

Appendix B: Enablers and Barriers for Multipartner Alliance Formation

52

Appendix C: Multipartner Alliance Formation Processes

54

Appendix D: Multipartner Alliance Formation Mechanisms

56

Appendix E: Codification of Empirical Evidence

58

Tables and Figures

Table 1: Data sample report

12

Table 2: Ranking framework of multipartner alliance formation motives

14

Figure 1: Contextual motives for multipartner alliance formation

15

Figure 2: Enablers and barriers for multipartner alliance formation

22

Table 3: Ranking framework of multipartner alliance formation enablers

23

Table 4: Ranking framework of multipartner alliance formation barriers

27

Table 5: Ranking framework on formation processes

31

Table 6: Ranking framework on formation mechanisms

34

Table A1: Multipartner alliance formation motives

51

Table B1: Multipartner alliance formation enablers

52

Table B2: Multipartner alliance formation barriers

53

Table C1: Multipartner alliance formation processes

54

Table C2: Empirical evidence on multipartner formation processes

55

Table D1: Multipartner alliance formation mechanisms

56

Table D2: Empirical evidence on multipartner formation mechanisms

57

Table E1: Codification empirical evidence on formation processes and mechanisms

58

Introduction

As a result of globalisation, fierce competition, and rapid technological changes, firms are to

innovate and invest in research and development (R&D) in order to gain a sustained competitive

advantage (Hsu & Lin, 2014). In view of building and sustaining competitive advantage, firms are

increasingly considering multipartner research and development (MR&D) alliances as a business

strategy, for they represent valuable sources of knowledge and innovation, and considerable market

penetration opportunities (Sakakibara, 2002). By entering alliances, firms increase their knowledge

base by accessing that of partner firms. From the knowledge-based view perspective, firms achieve

competitive advantage by enhancing their existing knowledge by increasing it through access to

partner firms' knowledge-base, and by acquiring diverse knowledge (Peteraf, 1993). Besides from a

knowledge perspective, MR&D alliances are also source of competitive advantage from a relational

perspective. Indeed, firms that enhance their network can achieve competitive advantage with

critical resources located outside the boundaries of the firm, by exploiting resources that are

embedded in the interfirm interaction, such as knowledge sharing routines, relation-specific assets,

or complementary resources (Dyer & Singh, 1998).

alliance performance (Lin, Fang, Fang & Tsai, 2009; Sampson, 2007; Hagedoorn & Schakenraad,

1994). Though the body of research on the topic of MR&D alliances is currently still limited, a

growing interest in partnering in MR&D alliances resulted in a variety of literature streams that

investigated the various facets of MR&D alliances such as, alliance formation, alliance governance,

and alliance performance. Moreover, the duality of opportunities to accessing relevant knowledge,

and risks of knowledge misappropriation has captured the attention of many researchers (Li, 2013).

Despite a growing popularity, there appears to be a high failure rate of strategic alliances (Parkhe,

1993) of which interfirm rivalry and managerial complexity are at the root (Park & Ungson, 2001),

especially when partner firms are based in different countries (Hennart & Zeng, 1997), or have

diverging strategic paths (Dacin, Hitt & Levitas, 1997).

When considering collaborating in MR&D alliances, firms must first evaluate and assess the

potential benefits, and even more thoroughly the risks of such collaboration, prior to actually

engaging in MR&D alliances. Therefore, to assess the potential of forming an MR&D alliance, it

seems appropriate to first focus on ex ante characteristics of an MR&D alliance, such as the

antecedents that determine MR&D alliance formation, before looking at ex post characteristics,

such as alliance management, or alliance performance. Although alliance management, and alliance

performance assessment are crucial to ensure effective functioning of alliances, it seems logical to

initially focus on the formation stage of an alliance prior to thinking of managing an alliance and

measuring its performance.

throughout the coopetitive relationship (Santos & Eisenhardt, 2009), such tensions are even more

prominent during the formation stages (Das & Teng, 2000). The alliance formation stage affects the

creation, and also the success of cooperation (Walker et al., 1997), since it is at this stage that

managers gain understanding of their roles within the alliance, prepare for the challenges of forming

an alliance, and are able to prevent them from missed opportunities. Therefore, focusing on alliance

formation is relevant for the success of the MR&D alliance. Yet, it is found that managers tend to

disregard formation processes, and prefer to draw more attention to alliance performance (Ring,

Doz & Olk, 2005). Assigning a higher priority of concern on activities relating to the alliance

management and performance, while paying superficial attention to the initial formation phase,

could endanger the viability of the alliance. Indeed, it is during the formation stage that partner

seeking and selection take place which proves to be a challenging and complex process (Bierly &

Gallagher 2007). Moreover, partner firms should negotiate the terms of their collaboration and

define their level of interdependence in order to ensure a strategic fit (Das & Kumar, 2011).

Given the high dissolution rate, managers must make crucial decisions when determining to

form an MR&D alliance. However, research field on the topic of MR&D alliance formation is

currently still fragmented and incoherent. Therefore, there is a need of a systematic literature review

in order to synthesise existing research on the topic. Given the fragmented field, research only gives

a descriptive view, and lacks in providing a deep understanding of the importance of MR&D

alliance formation and what the key antecedents and determinants are. As a result, the purpose of

this study is to identify, and investigate the key influential factors on alliance formation decision, of

which can be derived the following research questions:

(1) What are the main MR&D alliance formation processes?

(2) What are the underlying factors that explain why MR&D alliances follow particular

formation processes?

This is done by constructing a theoretical framework that helps consolidate the fragmented

field, and reveal patterns in MR&D alliance formation. This review integrates the diverging aspects

and perspectives, in an attempt to contribute towards a better understanding of the processes, and

mechanisms of MR&D alliance formation. In addition, the objective of integrating the different

mechanisms is twofold; to provide managers with the necessary knowledge critical for MR&D

alliance formation decisions; and to support future researchers with a structure in an attempt to

reveal potential research gaps, and to extend research in the field of MR&D alliance, and more

specifically MR&D alliance formation.

theoretical framework. Lastly, a conclusion reiterates main findings and limitations of the research,

based on which future research suggestions are presented.

Methodology

As discussed in the previous section, the focus on MR&D alliance formation is motivated by

an objective of providing a clear overview and practical suggestions, which managers could use to

formulate strategies, and develop decision making processes in the context of MR&D alliance

formation. Simultaneously, it is intended to reveal research gaps, and derive propositions for future

researchers to extend the understanding of the topic. To do so, a systematic literature review was

conducted on the topic of MR&D alliance formation. As opposed to a narrative literature review, "a

systematic review uses an explicit algorithm (…) to perform a search and critical appraisal of the

literature" (Crossan & Apaydin, 2010, p. 1156). This paper follows the systematic review principles

by Tranfield, Denyer, and Smart (2003), which consists of a three-stage procedure that comprise 1)

planning, 2) execution, and 3) reporting.

Planning

Execution

Eligibility criteria.

The following eligibility criteria were set up when selecting and assessing potential

researches. To be eligible for this systematic literature review, the study must:

-

Cover at least one aspect of MR&D alliance formation, such as formation motives,

enablers, barriers, processes, or mechanisms. Articles dedicated to post-formation stages of

MR&D alliance, such as alliance management, or performance measurement, were not

included;

-

Deal with collaborative relationships between more than two partners. Thus, in the event

that the study was about non-R&D related collaboration, it was retained if and only if it

concerned a multipartner cooperation. As a result, studies dedicated to dyadic alliances

were not considered;

-

Be a scientific article published in an academic peer review journal. Single case studies and

Keyword exploration and snowball sampling.

Data was collected from the Business Source Premier (EBSCO Host) database for the period

between January 1990 and April 2015, using the keywords listed in Table 1:

1

A trial search using the mentioned keywords yielded an initial sample of 321 articles. As this

systematic review is about MR&D alliance formation, a prior search was also made with the above

mentioned keywords including the terms R&D, and research and development, however this

resulted in a total of merely 15 articles. Therefore, to yield a larger pool of articles, it was chosen to

exclude the terms R&D, and research and development. In the event of articles not being retrievable

from Business Source Premier (EBSCO Host) database, these were searched for via ScienceDirect

Table 1: Data sample report

Keywords

Initial sample Off topic or duplicates Final sample

multi* alliance* form*

28

23

5

multi* alliance* creat*

3

3

0

multi* alliance* organi*

16

14

2

consort* form*

159

130

29

consort* creat*

71

62

9

multi* joint venture* form*

14

13

1

multi* joint venture* creat*

1

1

0

alliance constellation* form*

2

2

0

alliance constellation* creat*

0

0

0

yielded from snowballing

27

14

13

TOTAL

321

262

59

(Numbers in columns indicate the number of articles)

To ensure an exhaustive research, an asterisk (*) was placed at the end of a term so that all possible terms

1

or JSTOR. The chosen period concentrates this systematic literature review on most recent research

in the field of MR&D alliance formation.

To avoid overlooking fundamental studies and theory, the snowball method was

implemented, meaning that references within the selected papers retrieved from Business Source

Premier (EBSCO Host), ScienceDirect, and JSTOR, were also scrutinised and examined. By doing

so, it was the intention to broaden and deepen the consideration set of the theory field, and resulted

in 13 additional articles. Once the data selected, the author proceeded to analysing the data, by first

reading the abstracts in order to filter, and verify relevance of the topic in each article. The final

sample size, including articles derived from the snowball method, was of 59 articles (see Table 1).

Reporting

The remaining articles were thoroughly examined to find similarities or dissimilarities, and

analysed for patterns. For a better overview, articles used were categorised in Table D1 of Appendix

D. Findings of studies were investigated in an attempt to find heterogeneity or convergence, and

identify factors that could explain this. The objective of this taxonomy is threefold: 1) to synthesise

fragmented and extended research field in order to make it more understandable for managers so

they can use findings as guidelines to make decisions (Tranfield et al., 2003); 2) to find influential

determinants and antecedents of MR&D alliance formation; and 3) to discover issues, and gaps that

suggest for further research.

Results, Discussion

Motives for Multipartner Alliance Formation

(Allarkhia & Walsh, 2012; Li, 2013; Reid et al., 2001), scale and scope economies (De Ridder &

Rusinowska, 2008; Evan & Olk, 1990; Lee et al., 2013), industry standard setting (Barnett et al.,

2000; Doz et al., 2000; Glassey, 2004; Hart, 1993), customer service improvement (Inkpen, 1999;

Johnson, 1999; Koza & Lewin, 1999), to responding to environmental and institutional changes

(Clout et al., 2006; Ouch & Bolton, 1988; ul-Haq & Howcroft, 2007). The fragmented empirical

evidence shows that formation motives are mainly located in five key contexts, namely 1)

firm-level, 2) alliance-firm-level, 3) industry-firm-level, 4) institution-firm-level, and 5) customer-level (see Figure 1).

The following sub-sections address each context by analysing and synthesising the fundamental

motives for MR&D alliance formation within the key findings of extant research.

Table 2: Ranking framework of empirical evidence on multipartner alliance formation motives

Motive context

Formation motives

R&D*

Non-R&D**

Total studies

Firm level

SKCs, resources and assets development

12

13

25

Firm level

Efficiency

8

12

20

Industry level

Set industry standards

7

9

16

Industry level

Counter competition, gain market power

3

10

13

Firm level

Risk sharing

6

4

10

Alliance level

Synergy creation

6

3

9

Firm level

Innovativeness boost

5

2

7

Firm level

New markets access

4

3

7

Firm level

Technology, product development

4

3

7

Customer level

Customer service improvement

-

6

6

Institution level

Respond to environmental changes

3

2

5

Firm level

Network position improvement

1

3

4

Firm level

Technology, SKC protection

3

1

4

Firm level

Core competence development

1

2

3

Firm level

Legitimacy, reputation enhancement

1

2

3

Alliance level

Avoidance of work duplication

1

1

2

Institution level

Boost employment

-

1

1

Institution level

Standardisation of governmental information

-

1

1

TOTAL NUMBER OF ARTICLES

65

78

143

Numbers in columns indicate the number of articles

*R&D: collaboration in R&D setting

!

Firm level motives.

Skills, knowledge, capabilities (SKCs), resources, and assets development.

Rapid changing technological landscape pushes firms to respond quickly by developing their

SKCs, in order to achieve, or maintain a competitive market position. Firms in fast-paced industries

such as nanotechnology, and semiconductors, are often required to acquire, or generate

technological knowledge, in order to improve their technological capabilities (Chang & Tsai, 2000),

which in turn enable firms to face the challenges of new product development (Allarakhia & Walsh,

2012). Such challenges are incentives for organisations and researchers to cooperate, and create

MR&D alliances to access, pool, and exchange their respective knowledge-bases in view of

executing research projects effectively (Schall, 2014). Having access to partner firms' SKCs-bases

broadens the focal firm's range of competences, facilitating its ability to respond to changing

Figure 1: Contextual motives for multipartner alliance formation

(numbers in brackets indicate the number of articles)

Multipartner

Alliance

Formation

Firm:

-

SKC, resources, and assets development [25]

-

Efficiency [20]

-

Risk sharing [10]

-

Innovativeness boost [7]

-

Technology, product development [7]

-

New markets access [7]

-

Network position improvement [4]

-

Technology, SKC protection [4]

-

Core competence development [3]

-

Legitimacy, reputation enhancement [3]

Industry:

-

Counter competition, gain market power [13]

-

Set industry standards [16]

Alliance:

-

Synergy creation [9]

-

Avoidance of work duplication [2]

Institution:

-

Respond to environmental changes [5]

-

Standardisation governmental information [1]

-

Boost employment [1]

Customer:

environmental demands (Koza & Lewin, 1999). The competitive climate prevailing high-tech

industries compels incumbents to accelerate their innovation rate to maintain a competitive position.

However, maintaining a high-pace innovation rate is costly, especially when done independently. As

a consequence, alliances represent a valuable option for organisations to join forces to facilitate

access to, and exchange of SKCs, resources, and assets (Cloodt et al., 2006). Moreover, new

knowledge generated by an MR&D alliance is often idiosyncratic, as it emanates from a unique and

path-dependent collaborative relationship (Gulati et al., 2000). Resultantly, such knowledge

contributes towards each partner firm's competitive advantage (Reid et al., 2001).

Cost savings, efficiency, scale economies.

Firms facing intense competitive pressures should seek ways to decrease costs in order to be

profitable. In this respect, forming cost-reducing alliances enables partner firms to generate higher

profits relative to non-partner firms. That is to say, cost-reducing alliances tend to exert negative

network externalities on outside firms. Thus, outsiders have a motive to enter the alliance, whereas

incumbents have no motive to exit it (Catilina & Feinberg, 2006). In the context of gamblers

consortia -of which main formation motivation is based on sharing risks by spreading costs-

participants are able to join efforts to innovate efficiently. Community builders, on the other hand,

are essentially motivated to ally in view of benefiting from network externalities and knowledge

spillovers. Nevertheless, in both cases, member firms can improve efficiency by sharing costs, and

achieve scale economies once the critical mass is reached resulting from positive network

externalities (Eisner et al., 2009). When cross comparing findings of extant studies about

collaboration in R&D setting with that of collaboration in non-R&D setting, non-R&D related

collaborations are more efficiency oriented, relative to R&D collaborations. This finding could be

supported by the importance given to the detrimental effect of higher coordination costs involved

with alliance formation (see Sub-Section Firm Level Barriers).

Risk sharing.

and are continuously pushed to make crucial investment decisions within highly volatile contexts,

and thus are confronted with considerable opportunity costs.

Market access.

Globalisation incites firms to develop their business environment by expanding across

borders. However, such strategy is costly, especially when carried out independently. Therefore,

forming international interfirm partnerships creates opportunities to access or even create new

markets (Cloodt et al., 2006). Firms that project to expand into new markets need to acquire local

market knowledge. Therefore, firms may want to explore interorganisational relationships by means

of cross-border alliance with firms located in the targeted market (Koza & Lewin, 1999).

Furthermore, the research conducted by Hitt and colleagues (2000), in the field of international

partner selection, supports that firms pursuing an international expansion strategy are motivated to

enter alliances with local firms of the targeted market to access local market knowledge.

Technology development, innovation boost.

Alliance level motives.

The constant pressures of fierce competition, high-velocity changes, and globalisation, drive

organisations to join forces and collaborate. First, interfirm collaborative relationships enable

participants to create synergy among partner firms, so that all member firms are involved in the

process of enhancing technical capabilities (Chang & Tsai, 2000). Second, with emphasis on the

aspect of high-velocity of certain industries, R&D consortia enable firms to avoid costly work

duplication (Sakakibara, 2002; Winter & Wagenknecht, 2003).

Industry level motives.

At industry level, the two key underlying rationales found in this review for multipartner

alliance formation are industry standard- and market power-related (see Table 2, p.12). First, allying

enables smaller players to join forces and enhance their respective competitive position so that they

can compete with dominant players in the industry (Albers & Klaas-Wissing, 2012; De Ridder &

Rusinowska, 2008; Griffith et al., 1998). A striking yet logical finding is that, organisations

motivated by the need to collaborate on the basis of countering competition, or gaining market

power, are predominantly non-R&D related collaborations. This reflects the differential objectives

of research entities such as laboratories or universities, and for-profit organisations. The former are

usually learning-driven, meaning that their main concern is to create knowledge for the sake of

advancements in the field, whereas the latter are considerably more profit and power-oriented.

exchanging knowledge, but it also shapes the structure of the industry, because they can forge entry

and mobility barriers.

Institution level motives.

Institutional factors, such as legal changes, appear to play a less significant role in MR&D

alliance formation decisions. Japan developed special laws to create a responsive legal environment

allowing emerging industries to prosper (Ouchi & Bolton, 1988). On the other hand, in Taiwan's

semiconductor industry, incumbents invest in R&D by means of alliance formation, to respond to

environmental changes (Chang & Tsai, 2000). This suggests that institutional factors vary across

industries and countries, making it complex to distinguish a dominant pattern.

Customer level motives.

Lastly, one locus of formation motive was found at customer level (see Table 2 and Figure 1).

Within the tourism industry, firms were facing quality issues, which were mainly caused by

differential perspectives of quality between firms and customers. In order to close this gap, it was

found that organisations could address these problems by forming total quality tourism consortia

(TQTC) for the purpose of unifying the quality standards of the industry and to improve the level of

customer service (Augustyn, 1998).

Enablers and Barriers for Multipartner Alliance Formation

To decide whether firms should enter into a collaboration agreement, or to determine the

governance mode of such collaboration depends on endogenous and exogenous factors. Indeed,

Augustyn (1998) explains that, in the context of European tourism industry, three crucial conditions

should be fulfilled when establishing a TQTC. These focus on quality of inputs, management

processes, and relationships. Similarly, in another type of non-R&D collaboration, Kakabadse and

Kakabadse (2000) find that the choice of collaboration arrangement depends on the value of inputs

and activities. As a result, in the context of non-R&D collaborations, it seems that value of

contributions and desired outputs are the focal point when considering alliance formation.

In contrast, when envisaging an R&D collaboration agreement, the choice of alliance form

depends more on relational factors. Rather than focusing on the value of contributions, participants

tend to concentrate on relational aspects, such as type of reciprocity (Das & Teng, 2002), and level

of interdependence in a relationship (Reid et al., 2001). Likewise, Ring and colleagues (2005)

suggest that the type of formation process depends on three critical success factors, namely the

degree of similarity in business interests, the preexistence of social relationships, and of strategic

relationships.

the intention of building longterm collaborative relationships without primarily emphasising the

value of inputs.

!

Formation enablers.

Firm level enablers.

From extant multipartner alliance formation literature, two main enablers strikingly stand out,

namely organisational learning capabilities, and prior experience (see Figure 2, Tables 3, and B1 of

Appendix B). As discussed in the section concerning formation motives, the most important motive

for organisations to ally, is to access and exchange knowledge and resources. In order to do so,

participants must have suitable capabilities to acquire such critical knowledge, and internalise it. In

this respect, Sakakibara (2002) explains that organisational learning capabilities positively affect

Figure 2: Enablers and barriers of multipartner alliance formation

(numbers in brackets indicate the number of articles)

Multipartner

Alliance

Formation

Firm:

-

Absorptive capacity/Dynamic capabilities [14]

-

Prior experience [10]

-

Reputation [3]

-

Incoming spillovers [1]

Alliance:

-

Synergy, alignment [21]

-

Mutual trust, transparency [8]

-

Alliance size [7]

-

Longterm goals [5]

-

Cost discipline [2]

-

Sensemaking [1] Industry:

-

Industry context/conditions [3] Firm:

-

Opportunism, free riding [13]

-

Coordination costs [8]

-

Increased complexity [7]

-

Loss of SKCs [7]

-

Loss of control [3]

-

Weaker innovation capacity [1]

Alliance:

-

Goal divergence; unbalanced coopetition [13]

-

Uncertainty, instability, tension [6]

-

Governance mode disagreement [2]

-

Collaborative inertia [1]

Institution:

-

Cultural differences [4]

-

Compliance to requirements changes [1]

-firm participation in alliances; that is to say, the more developed the learning capabilities, the more

likely the firm enters an alliance with the objective of creating or accessing new knowledge. Yet,

mere access is insufficient to fully benefit from knowledge. Indeed, access does not systematically

imply that firms have the ability to exploit new knowledge, and internalise it. Hence, firms are

urged to develop knowledge capturing, and retention capabilities (Tukel et al., 2011), or absorptive

capacity, which Cohen and Levinthal (1990) define as a firm's: "ability to recognize the value of

new information, assimilate it, and apply it to commercial ends." (p.128). In the context of

knowledge based enterprises, Reid and colleagues (2001) emphasise the importance of absorptive

capacity, particularly because it influences the capability of firms to internalise information

acquired from partners, whereas the lack of it impedes interorganisational knowledge creation.

Furthermore, absorptive capacity is crucial in order to recycle the significant amounts of unused

knowledge that many firms accumulate resulting from dormant or rejected projects (Tukel et al.,

2011).

Table 3: Ranking framework of empirical evidence on multipartner alliance formation enablers

Enabler level

Formation enabler

R&D*

Non-R&D**

Total studies

Alliance level

Synergy, alignment

11

10

21

Firm level

Absorptive capacity; dynamic capabilities

8

6

14

Firm level

Prior experience

5

5

10

Alliance level

Mutual trust, transparency

4

4

8

Alliance level

Alliance size

3

4

7

Alliance level

Longterm goals

3

2

5

Firm level

Reputation

2

1

3

Alliance level

Partner selection

1

2

3

Industry level

Industry context/conditions

1

2

3

Alliance level

Cost discipline

1

1

2

Firm level

Incoming spillovers

1

-

1

Alliance level

Sensemaking

1

-

1

TOTAL NUMBER OF ARTICLES

41

37

78

Numbers in columns indicate the number of articles

*R&D: collaboration in R&D setting

In fast-changing environments and in competitive industries, absorptive capacity is even more

crucial to remain competitive. Authors such as De Ridder and Rusinowska (2008), and Kakabadse

and Kakabadse (2000) have found that firms' degree of flexibility plays an important role with

regard to alliance formation, in terms of determining a firm's position within an alliance.

Accordingly, researchers (see Table B1 of Appendix B) stressed the importance of firms' ability to

adjust to changing demands (Kakabadse & Kakabadse, 2000), and to evolving environments

(Johnson, 1999). To deal with such unpredictabilities, and as an extension of the concept of

absorptive capacity, Teece and colleagues (1997) introduced the dynamic capabilities approach,

which they define as: "the firm's ability to integrate, build, and reconfigure internal and external

competences to address rapidly changing environments." (p.516). Firms that develop their dynamic

capabilities are more likely to cope successfully with environmental changes, and to form alliances.

Furthermore, potential partners with prior experience in allying or with previous partner

firms, are in a more favourable position with regard to successful alliance formation, in comparison

to those without prior experience. Studies have confirmed that prior collaboration with potential

partners increases the level of trust (Gulati, 1995; Ring & Van De Ven, 1989; Shapiro et al., 1992)

which is a disincentive to opportunistic behaviours within an alliance (Corley et al., 2006).

Moreover, Ebers (1997) explained that prior relations was, in certain cases, even a condition for

partners to form an alliance, because trust and commitment are likely to emerge from prior

experience, which play a key role in alliance success (Jost et al., 2005; Olk, 1998; Reid et al., 2001).

Besides contributing towards successful alliance formation, prior experience also affects the

alliance formation process (Ring et al., 2005), and even alliance participation (Ebers, 1997;

Sakakibara, 2002).

internalisation. In addition, prior experience with allying and with potential experience increases the

likelihood of alliance participation and successful alliance formation.

Alliance level enablers.

At alliance level, a large majority of extant findings shows that synergy, and alignment are

two influential enablers with regard to alliance formation (see Figure 2, Tables 3, and B1 of

Appendix B). Goal alignment is considered a key success factor in the context of research consortia,

especially in an aggressively competitive industry such as Taiwan's semiconductor industry (Chang

& Tsai, 2000); misaligned goals among partner firms may impede the speed of technological

development and, therefore, also the achievement of a sustainable competitive advantage. Similarly,

political parties must make agreements, have shared interests, and form coalitions, should they want

to achieve majority governments (Hendrix et al., 2013). In the same way, it was found that, in the

US healthcare industry, a common understanding and shared incentives enabled firms to forge a

learning environment in the form of learning consortia so that participants have increased

opportunities to achieve the set goals (Hirsch & Immediato, 1999).

this respect, opportunistic behaviour makes appropriate partner selection a crucial matter to ensure a

trustworthy collaboration (Msanjilla & Afsarmanesh, 2009). Additionally, mutual trust is a key

enabler to successful alliance formation, since trust leads to less formal and hierarchical structures

(Corley et al., 2006). Hierarchy can sometimes be at the root of lengthy decision-making processes,

which can be disadvantageous in fast-moving industries. Therefore, developing relationships based

on trust can improve the smooth development of collaborative interactions within an alliance

(García-Canal et al., 2003). In line with developing relationships built on trust, Jost and colleagues

(2005) found that the main enablers of productive collaboration were staff engagement, team

building, and transparency.

Industry level enablers.

In the industry context, main findings about factors that facilitate alliance formation were on

the topic of industry conditions, such as intensity of competition, and appropriability conditions.

These factors influence not only the quality of the collaboration, but also the form of collaboration

(Ebers, 1997). In non-R&D related collaboration, firms in highly competitive industries tend to be

efficiency-driven and, therefore, pursue collaborative strategies (Albers & Klaas-Wissing, 2012). In

contrast, the rate of R&D collaboration rate is more likely to occur in industries with weak

competition (Sakakibara, 2002).

Formation barriers.

Firm level barriers.

(Barnett et al., 2000). As a result, the free-riding problem has a notable influence on the

organisational structure of the alliance (Johnson, 1999); firms implement formal control

mechanisms to manage free-riding (García-Canal et al., 2003; Hwang & Burgers, 1997).

Although opportunistic behaviour is moderated when the levels of trust and shared interests

are high, opportunism plays an influential role especially in the decision of formation process, and

governance mode (Doz et al., 2000; Hwang & Burgers, 1997). When the risk of opportunistic

behaviour is significantly high, firms tend to adopt an emergent process and establish a formal

structure. Alliances engaged in emergent process are generally related to competing firms because

they are formed between players of the same industry, resulting in a higher risk of opportunistic

behaviour. In an attempt to constrain opportunism, the emergent process emphasises the importance

of interdependence (Doz et al., 2000). Furthermore, in emergent process, alliances are created in

Table 4: Ranking framework of empirical evidence on multipartner alliance formation barriers

Motive context

Formation barriers, challenges, and concerns

R&D*

Non-R&D**

Total studies

Firm level

Opportunism, free-riding

9

4

13

Alliance level

Goal divergence, unbalanced coopetition

9

4

13

Firm level

Coordination costs

2

6

8

Firm level

Increased complexity

3

4

7

Firm level

Loss of SKCs

6

1

7

Alliance level

Uncertainty, instability, tension

5

1

6

Institution level

Cultural differences

3

1

4

Firm level

Loss of control

2

1

3

Alliance level

Governance mode disagreement

2

-

2

Firm level

Weaker innovation capacity

-

1

1

Alliance level

Collaborative inertia

-

1

1

Institution level

Compliance to requirements changes

1

-

1

Institution level

Legislative differences

1

-

1

TOTAL NUMBER OF ARTICLES

43

24

67

Numbers in columns indicate the number of articles

*R&D: collaboration in R&D setting

reaction to external events, be it opportunities or challenges such as opportunism (Eisner et al.,

2009). Alternatively, firms create a separate coordinating mechanisms that monitors alliance

activities and potential opportunism (Ouchi & Bolton, 1998).

Beside opportunism and free-riding, alliance formation decisions are also negatively affected

by the increased costs caused by collaboration (Albers & Klaas-Wissing, 2012; Barnett et al., 2000;

García-Canal et al., 2003; Griffith et al., 1998; Johnson, 1999; Jost et al., 2005; Kakabadse &

Kkabadse, 2000; Sakakibara, 2002). In spite of the possibility of sharing costs and achieving scale

economies, alliances also incur higher costs, such as communication and coordination costs that are

mostly due to the number of partner firms involved in the alliance (Albers & Klaas-Wissing, 2012;

García-Canal et al 2003), the geographic distance between partners (Griffith et al., 1998), the

creation of a separate organisation (Johnson, 1999), and the increased need for planning (Jost et al.,

2005). Such higher costs are directly related to the increased complexity of entering an alliance

(Albers & Klaas-Wissing, 2012; Chang & Tsai, 2000; García-Canal et al., 2003; Grandersen, 2011;

Jost et al., 2005; Ring et al., 2005; Uttaro, 2003).

Furthermore, allying causes a loss of control over resources and assets (Johnson, 1999; Olk,

1998; Reid et al., 2001). Olk (1998) explains that when in an alliance, firms are to commit resources

over which they risk a loss of control. Such barrier is derived from potential information leakage

and spillovers (Elderbos et al. (2004); Kakabadse & Kakabadse (2000); Kesavayuth & Nikos

(2012); Kreiner & Schulz (1993); Li (2013); Reid et al. (2001); Winter & Wagenknecht (2003).

transferred to partner firms. As a consequence, creating an alliance can cause firms to weaken their

innovation capability.

Alliance level barriers.

One of the most important impediments to alliance formation is related to goal divergence. In

particular, MR&D alliances frequently consist of a mix of 1) nonprofit research institutions (i.e.

laboratories or universities) that are mostly knowledge-driven and 2) for-profit organisations which

are considerably more profit and power-oriented. This divergence is not only observed between

partner firms (Chung & Beamish, 2012), but also between alliance goal and partner firms goals,

which Evan and Olk (1990) designate as "supracorporate challenge" (p.40).

When forming coalitions, participants with diverging objectives must conduct lengthy

negotiations and make compromises prior to reaching an agreement (Hendrix et al., 2013).

Likewise, in consortia with partners from the public and private sectors, diverging objectives must

be reconciled in order to achieve a productive collaboration (Jost et al., 2005). As a result, the

choice of formation process also depends on the degree of divergence between partner firms'

motives or interests. According to Ring and colleagues (2005), in the event of strong diverging

interests, the emergent process is more likely to be adopted. With regard to divergence between

alliance goal and partner goals, in highly competitive and volatile industries such as the computer

industry, it sometimes occurs that divergence in objectives creates so much resistance that it makes

the collaboration impossible (Hwang & Burgers, 1997).

to be high a likelihood of motivation and objective differential. Therefore, to avoid becoming

obsolete in an alliance, firms must maintain an asymmetric bargaining power in their favour, by

pursuing a learning race in view of outlearning their partners (Li, 2013). However, such behaviour

has a negative effect on successful cooperation. As a result, the complexity of reaching an

equilibrium between competitive and cooperative interests is an impediment to successful

cooperation or even alliance formation (Jost et al., 2005; Koza & Lewin, 1999; Mothe & Quélin,

2000; Olk, 1999; Schall, 2013; Zeng & Chen, 2003). Thus, given the tensions that diverging goals

can provoke, it is vital to carry out the necessary negotiations in order to settle to an agreement to

avoid opportunistic behaviour, free-riding, or even non-cooperation. Alternatively, to counter the

negative effects of theses tensions, partner firms must opt for a suitable alliance formation process.

Institution level barriers.

respect, the larger the cognitive distance between partner firms, the higher the heterogeneity in their

knowledge fields, resulting in a reduced overlap. Too little overlap or too much diversity (such as

cultural differences) can interfere in the understanding and assimilation, thus, the exploitation of

new knowledge (Nooteboom et al., 2007). Given these challenges, it becomes obvious that

appropriate partner selection is crucial, since potential partners should share new yet understandable

knowledge, and have the capabilities to absorb it in order to benefit from the collaboration (This

issue is further discussed in the Section Alliance Formation Mechanisms).

Multipartner Alliance Formation Processes

Previously conducted research examined the differential alliance formation processes, ranging

from a wide variety of stage-game processes (Catilina & Feinberg, 2000; Corley et al., 2006; Ebers,

1997; Johnson, 1999; Kesavayuth & Zikos, 2012; Kreiner & Schulz, 1993; Msanjilla &

Afsarmanesh, 2009; Olk, 1998; Reid et al., 2001; Ring et al., 2005), to the emergent and engineered

processes (Doz et al., 2000; Eisner et al. 2009), and the embedded process (Ring et al., 2005) (See

Table 5).

Table 5: Ranking framework of empirical evidence on multipartner alliance

formation processes

Formation Process

R&D* Non-R&D** Total studies

Stage-game

11

6

17

Emergent process

3

-

3

Engineered process

3

-

3

Step-by-step approach

1

2

3

Embedded process

1

-

1

TOTAL NUMBER OF ARTICLES

19

8

27

Numbers in columns indicate the number of articles

*R&D: collaboration in R&D setting

Stage-game.

In the various stage-game processes, formation occurs in series of stages ranging from two to

eight separate stages (See Table C1 of Appendix C for an overview of the various formation

processes). In general, the initial stages are the preformation steps during which decisions are made

concerning whether or not to form an alliance. During these introductory steps, conditions of

collaboration are outlined, and the purpose and structure of collaboration are discussed. Following

stages are, in general, about establishing the direction of collaboration. Lastly, further stages deal

with operating structure and procedures.

From extant literature of this review, there appears to be a majority of researches about

alliance formation that study stage-game processes and that are R&D related; out of the 17 studies

that involve stage-game processes, 11 are R&D related against six non-R&D related (see Table 5).

The main barriers to alliance formation for R&D related collaboration involved loss of control over

resources or assets (Olk, 1998; Reid et al., 2001), loss of SKCs and information spillover

(Kesavayuth & Zikos, 2012; Kreiner & Schulz, 1993; Reid et al., 2001); and increased complexity

resulting from intercultural tensions and instability (Catilina & Feinberg, 2000; Mothe & Quélin,

2000; Ring et al., 2005). On the other hand, the main motives for alliance formation are related to

SKC access and protection (Kesavayuth & Zikos, 2012; Reid et al., 2001).

Emergent, engineered, embedded.

(Doz et al., 2000; Ring et al., 2005). In certain cases, alliances are formed based on the awareness of

one dominant firm becomes of a particular threat, or opportunity (Eisner et al., 2009).

In engineered process, consortia are formed by means of a triggering entity -which could be a

person as well as a firm- that is able to demonstrate, to potential partner firms, the need and benefits

of allying (Doz et al., 2000; Ring et al., 2005). When adopting the engineered approach, the

triggering entity becomes the dominant coordinating firm (Eisner et al., 2009).

The embedded approach results from a growing interdependence awareness, and continuity

expectation among potential partner firms. In such process, participants already enjoy a strong

relationship. Thus, potential member firms are embedded in a common social network, resulting in

less managerial activities compared to the emergent, and engineered processes (Ring et al., 2005).

The three studies concerning emergent, engineered, and embedded processes were all set in an

R&D-related context. Furthermore, the common factor is that opportunism is considered an

influential barrier to successful cooperation, thus affects the choice of multipartner alliance

formation process. Indeed, the higher the risk of opportunistic behaviour, the more likely partner

firms adopt the emergent approach and create a more formal structure and engage protection

mechanisms (Doz et al., 2000; Eisner et al., 2009; Ring et al., 2005).

Multipartner Alliance Formation Mechanisms

1988). Furthermore, certain researches explore the different partner selection approaches, i.e.

simultaneous versus step-by step (see Tables 6, and D1, D2 of Appendix D).

Coordination mechanisms.

Certain organisations decide to form MR&D alliances by designating a dominant partner firm

as a central hub, which acts as a coordinating mechanism to lead the collaboration, and manage the

various projects (Albers & Klaas-Wissing, 2012; Eisner et al., 2009; Gupta & Zhdanov, 2012; Koza

& Lewin, 1999; Ouch & Bolton, 1988; Safford, 2010), or even provide financial support (Das &

Teng, 2002). The main benefit of appointing a core firm as a coordinator within the alliance is that it

facilitates collaborative interactions among participants so that research projects can be carried out

and developed effectively. Indeed, the champion firm is able to oversee and manage the overall

progress of the collaboration, while participants can focus on their research projects (Koza &

Lewin, 1999). Furthermore, depending on the environment and context, the dominant firm plays a

distinctive role. For instance, in a monopolistic setting, besides fulfilling a coordinating role, the

dominant firm also has decision-making power over choices concerning price setting or

membership (Gupta & Zhdanov, 2012).

Table 6: Ranking framework of empirical evidence on multipartner alliance

formation mechanisms

Formation Mechanism

R&D* Non-R&D** Total studies

Champion firm, central hub

4

4

8

Simultaneous approach

3

3

6

Membership fee

3

1

4

Independent office or laboratory

3

-

3

Outsiders

-

2

2

TOTAL NUMBER OF ARTICLES

13

10

23

Numbers in columns indicate the number of articles

*R&D: collaboration in R&D setting

Alternatively, organisations may choose for outsiders, such as external management firms, to

lead the partnership (Inkpen, 1999; Ouch & Bolton, 1988). Both studies in this review are in the

context of non-R&D collaboration, and take place in the telecommunication industry, with as

formation motive located in the industry context, namely enhancing the quality of the industry

(Inkpen, 1999), and setting industry standards (Ouch & Bolton, 1988). Besides the fact that both

alliances are non-R&D related, their structures are both complex, composed of a board of directors

where each partner firm is represented, an executive committee, and an external management firm.

Similar to internal coordinating mechanisms, this option enables participants to solely focus

attention on the progress of the research projects, while leaving the external participants manage

and coordinate the progress of the alliance. Additionally, this particular structure choice allows

partner firms to exploit and develop their core business; and simultaneously, external firms can

apply their core business competences in managing and coordinating the collaboration. This

approach could lead the alliance towards developing, and maintaining a competitive advantage.

Incentives.

Partner selection.

The differing partner selection criteria, and approaches are found to vary according to the type

of partners (Koza & Lewin, 1999), and the form of collaboration (De Ridder & Rusinowska, 2008;

Hendrix et al., 2013). For instance, in the case of Nexia International, a non-equity multiparty

alliance composed of independent public accounting firms, different types of memberships are

offered with distinctive privileges and obligations (Koza & Lewin, 1999). Extant research found

that horizontal alliances, or alliances with a non-hierarchical structure tend to adopt the

simultaneous approach to select partner, where all potential partner firms gather and negotiate

simultaneously their collaborative links (Catilina & Feinberg, 2000; De Ridder & Rusinowska,

2008; Hendrix et al., 2013; Kesavayuth & Nikos, 2012). In contrast, alliances with a hierarchical

structure are inclined to follow a step-by-step approach, meaning that founding core members

gather to seek further partners until the formation criteria are fulfilled (De Ridder & Rusinowska,

2008; Hendrix et al., 2013; Mothe & Quélin, 2000).

Cross Comparison: R&D Collaboration Versus Non-R&D Collaboration

Conclusion, Future Research

Altogether, despite a scarcely developed research topic, multipartner alliance formation

research was still found to be fragmented and incoherent. Such fragmentation is explained by the

specificity of the studies, each investigating one particular feature of alliance formation and set in a

specific context. Moreover, studies unveiled a wide array of motives, opportunities, challenges,

processes, and mechanisms related to MR&D alliance formation. Additionally, the extant body of

research is rendered incoherent as researchers sometimes use terminology inconsistently or even

incorrectly; it has occurred that authors amalgamate, and confuse alliance portfolio (aggregate of

alliances of a firm), or alliance network with multipartner alliances (Wassmer, 2010). Therefore, this

study was conducted to systematically review empirical evidence in the field of MR&D alliance

formation in an attempt to organise and synthesise key findings. This approach was intended to

contribute towards an in-depth understanding by designing a framework that revealed diverging or

converging areas, and by providing a comprehensive analysis of the research field. The systematic

review of MR&D alliance formation literature revealed that the body of research revolves around

four major areas, which are formation motives, enablers and barriers for formation, formation

processes, and formation mechanisms. Furthermore, an in-depth analysis was carried out in an

attempt to identify the main influential factors (motives, enablers, and barriers) in successful

multipartner alliance formation; discuss the prevailing formation processes; and assess the

mechanisms underlying multipartner alliance formation. More specifically, this study examined

how motives, enablers, barriers, and mechanisms are interrelated, and how they affect the choice of

formation process and governance mode. Striking differences were observed between R&D-related

multipartner alliances and non-R&D related multipartner alliances.

attention. Furthermore, studies in the field are characterised by addressing a narrow aspect of the

field. Therefore, this paper was intended to consolidate the fragmented field by constructing a

theoretical framework, reveal patterns in MR&D alliance formation. Outcomes reveal that

relevance of formation motives, enablers, barriers vary across context, namely R&D and non-R&D.

References

*Albers, S., & Klaas-Wissing, T. (2012). Organisation of multilateral LTL alliances. International

Journal of Logistics Research and Applications, 15(3), 181-198.

*Allarakhia, M., & Walsh, S. (2012). Analyzing and organizing nanotechnology development:

Application of the institutional analysis development framework to nanotechnology consortia.

Technovation, 32(3), 216-226.

*Augustyn, M. M. (1998). The road to quality enhancement in tourism. International Journal of

Contemporary Hospitality Management, 10(4), 145-158.

*Barnett, W. P., Mischke, G. A., & Ocasio, W. (2000). The evolution of collective strategies among

organizations. Organization Studies, 21(2), 325-354.

*Belderbos, R., Carree, M., Diederen, B., Lokshin, B., & Veugelers, R. (2004). Heterogeneity in

R&D cooperation strategies. International Journal of Industrial Organization, 22(8),

1237-1263.

*Benjaafar, S., Heragu, S. S., & Irani, S. A. (2002). Next generation factory layouts: research

challenges and recent progress. Interfaces, 32(6), 58-76.

Bierly, P. E., & Gallagher, S. (2007). Explaining alliance partner selection: fit, trust and strategic

expediency. Long Range Planning, 40(2), 134-153.

*Boone, A. L., & Mulherin, J. H. (2011). Do private equity consortiums facilitate collusion in

takeover bidding? Journal of Corporate Finance, 17(5), 1475-1495.

*Catilina, E. P., & Feinberg, R. M. (2006). Market power and incentives to form research consortia.

Review of Industrial Organization, 28(2), 129-144.

*Chang, P. L., & Tsai, C. T. (2000). Evolution of technology development strategies for Taiwan's

semiconductor industry: formation of research consortia. Industry and Innovation, 7(2),

185-197.

Chiles, T. H., & McMackin, J. F. (1996). Integrating variable risk preferences, trust, and transaction

cost economics. Academy of Management Review, 21(1), 73-99.

*Chung, C. C., & Beamish, P. W. (2012). Multi-party international joint ventures: Multiple

post-formation change processes. Journal of World Business, 47(4), 648-663.

*Cloodt, M., Hagedoorn, J., & Roijakkers, N. (2006). Trends and patterns in interfirm R&D

networks in the global computer industry: An analysis of major developments, 1970-1999.

Business History Review, 80(4), 725.

*Cohen, W. M., & Levinthal, D. A. (1990). Absorptive capacity: a new perspective on learning and

innovation. Administrative Science Quarterly, 35(1), 128-152.

*Corley, E. A., Boardman, P. C., & Bozeman, B. (2006). Design and the management of

multi-institutional research collaborations: Theoretical implications from two case studies. Research

policy, 35(7), 975-993.

Dacin, M. T., Hitt, M. A., & Levitas, E. (1997). Selecting partners for successful international

alliances: examination of U.S. and Korean firms. Journal of World Business. 32(spring), 3-16.

Das, T.K., & Kumar, R. (2011). Inter partner negotiations in alliances: a strategic framework.

Das, T. K., & Teng, B. S. (2000). Instabilities of strategic alliances: an internal tensions perspective.

Organization Science, 11(1), 77-101.

*Das, T. K., & Teng, B. S. (2002). Alliance constellations: a social exchange perspective. Academy

of Management Review, 27(3), 445-456.

*De Ridder, A., & Rusinowska, A. (2008). On some procedures of forming a multipartner alliance.

Journal of Economics & Management Strategy, 17(2), 443-487.

*Doz, Y. L., Olk, P. M., & Ring, P. S. (2000). Formation processes of R&D consortia: which path to

take? Where does it lead? Strategic Management Journal. 21(239-266).

Dyer, J. H., & Singh, H. (1998). The relational view: Cooperative strategy and sources of

interorganizational competitive advantage. Academy of Management Review, 23(4), 660-679.

*Ebers, M., 1997. Explaining inter-organizational network formation. In- M. Ebers, ed. The

formation of inter-organizational networks. Oxford- Oxford University Press, 3–4.

*Eisner, A. B., Rahman, N., & Korn, H. J. (2009). Formation conditions, innovation, and learning in

R&D consortia. Management Decision, 47(6), 851-871.

*Evan, W. M., & Olk, P. (1990). R&D consortia: A new US organizational form. MIT Sloan

Management Review, 31(3), 37.

Fitjar, R. D., Gjelsvik, M., & Rodríguez-Pose, A. (2013). The combined impact of managerial and

relational capabilities on innovation in firms. Entrepreneurship & Regional Development,

25(5-6), 500-520.