Journal of Management Studies 46:3 May 2009

0022-2380

Governance, Negotiations, and Alliance Dynamics:

Explaining the Evolution of Relational Practice

Håvard Ness
Buskerud University College

abstract This paper explores relational practice in strategic alliances. Focusing on how
governance mechanisms and negotiation strategies evolve and form patterns of interaction that
are under influence of both recursive and adaptive forces, four modes of relational practice are
discussed: recursively integrative, recursively distributive, adaptive toward integrative, and
adaptive toward distributive. Three longitudinal cases are examined. I find that different
contractual conditions enable (or restrain) interaction patterns. Negotiation behaviour, in turn,
affects future contractual conditions and their combination and recombination form the
evolution of relational practice. As relationships progress, initial governance mechanisms are
complemented and negotiation behaviours change. I explain how perceptions of both process
and relational outcomes influence relational practice. In addition, I explicate why relational
practices have both recursive and adaptive characteristics, and identify conditions that cause
relational practices to move towards more integrative or distributive modes.

INTRODUCTION
Alliance dynamics is not very well understood. In 1994, Ring and Van de Ven (p. 113)
called for further exploration of ‘how equity, trust, conflict-resolution procedures, and
internal governance structures emerge, evolve, and dissolve’. Zaheer and Venkatraman
(1995, p. 388) found that ‘governance structure explained only a small amount of
variance in governance process’, and Blois (2002, p. 523) observed that ‘the behaviour
that occurs within exchanges is not determined by the forms of governance used’. What
seems to be seriously left out in empirical research on alliance dynamics is an account of
what managers actually do to govern alliances over time (Spekman et al., 1998; Wright
and Lockett, 2003), and how governance mechanisms enable and restrain action, yet
at the same time are being formed by these actions (Boddy et al., 2000; Sydow and
Windeler, 1998).
This paper aims to alleviate this shortcoming, and integrates governance with action
to understand management practice in alliances. In particular, I argue for including a
Address for reprints: Håvard Ness, Buskerud University College, Postboks 164 Sentrum, N-3502 Hønefoss,
Norway ([email protected]).

© Blackwell Publishing Ltd 2009. Published by Blackwell Publishing, 9600 Garsington Road, Oxford, OX4 2DQ, UK
and 350 Main Street, Malden, MA 02148, USA.
452 H. Ness
negotiation perspective to understand alliance dynamics and management, as negotia-
tions are at the core of decision making in alliances (Ring and Van de Ven, 1994). It is
important to include this perspective because alliances are typically founded on incom-
plete contracts with a substantial planning gap where renegotiation of initial terms and
ongoing negotiation takes place to settle issues and guide further progress (Ariño and de
la Torre, 1998). Moreover, alliances involve both joint value creation and value appro-
priation, and thus represent mixed motive settings involving a tension between shared
and private goals (Madhok, 1995). Although research on negotiations in alliances, and in
particular the link between governance and negotiations, has been called for (Ariño and
de la Torre, 1998; Kumar and Nti, 1998), this area has remained largely unexplored
empirically.
In the endeavour to bridge governance processes and ongoing negotiations in alli-
ances, I argue that the concept of relational practice, defined as the interaction pattern
‘agents typically use for coordinating activities and relations’ (Windeler and Sydow,
2001, p. 1041), is particularly suitable.
First, the chosen governance structure an alliance builds on represents a strong context
for interaction, as it gives direction and both enables and constrains developmental
paths. Second, due to the planning gap and the mixed motives of the two partners, these
interaction patterns are in fact negotiation processes. Hence, this paper explores how
governance mechanisms and negotiation strategies evolve, how they form the evolution
of relational practice, and explains why relational practice changes. The literature
distinguishes between equity and non-equity alliances (Wright and Lockett, 2003).
Equity forms (e.g. joint ventures) involve common ownership, and non-equity alliances
include a wide range of contract-based forms. This paper focuses on vertical non-equity
alliances, defined as ‘purposive strategic relationships between independent firms who
share compatible goals, strive for mutual benefit, and acknowledge a high level of mutual
interdependence’ (Mohr and Spekman, 1994, p. 135).
Alliances evolve through different phases. Some argue that different phases involve
different challenges (Larson, 1992; Spekman et al., 1998), while others maintain that
although relationships evolve from initiation towards dissolution or reconfiguration, the
management of alliances involves cycles of managerial actions (Das and Teng, 2002;
Heide, 1994; Ring and Van de Ven, 1994). Either way, of particular interest have been:
(1) the importance of contracts (in the broad sense) and governance mechanisms, and
the adjustment of these as relationships evolve (Mayer and Argyres, 2004; Ness and
Haugland, 2005); (2) the ongoing negotiation that is necessary to deal with the planning
gap that characterizes alliances (Ness and Haugland, 2005; Ring and Van de Ven, 1994);
and (3) learning efforts that enhance coordination and interaction as relationships evolve
(Ariño and de la Torre, 1998; Holmqvist, 2004). In order to understand alliance man-
agement by focusing on the emergence and evolution of relational practice, these aspects
seem to be of crucial importance. Bell et al. (2006) argued that the practical relevance
of much alliance process research is limited, and that little cross-fertilization between
perspectives has occurred. Hennart (2006) even argued that structural, as opposed to
processual, explanations contribute to both stronger theory and better managerial
advice. This paper attempts to meet these challenges by focusing on relational practice.
This approach adds value to the current research as relational practice takes centre stage

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 453
and evolution is explained by combining perspectives that are core to alliance
management.

RELATIONAL PRACTICE: THE DYNAMICS OF ACTING OUT

AGREEMENTS
Characteristics of Relational Practice
As alliance partners share a common governance structure and sense of direction, it is
reasonable to assume that they will establish a form of collaborative practice which to
some extent persists temporally. Such a recursive practice often exists as managerial habits
or routines ( Jarzabkowski, 2004). Initial relational practice is likely to have recursive
characteristics because short-term planning is more specific than long-term planning and
the initial governance mechanisms (or contract in a broad sense) the relationship builds
on represents a strong context for interaction that both enables and constrains develop-
mental paths. Recursive relational practice might also be desirable as it enhances pre-
dictability of partner behaviour. However, over time recursiveness might lead to a
relational practice that is ‘prone to inertia’ ( Jarzabkowski, 2004, p. 531; Walter et al.,
2008).
Although alliances can benefit from a stable and predictable managerial practice, they
often seek (strategic) change that requires adaptation (Boddy et al., 2000), where ‘interac-
tion between agent and structure does not sustain sedimented behaviours – it is “becom-
ing”, not became’ ( Jarzabkowski, 2004, p. 535). Longitudinal studies of alliances show
that agreements and governance mechanisms are dynamic (Ariño and de la Torre, 1998;
Mayer and Argyres, 2004; Ness and Haugland, 2005). Thus, the ongoing structuring
(Ring and Van de Ven, 1992) of relationships is important for exploring adaptive
characteristics of relational practice.
While alliances are typically based on common goals, it is recognized that partners also
have private goals. As a result, partners often face a series of joint decision situations over
the course of the relationship in which the private goals of each partner are at odds with
the shared goals of the alliance. Hence, to understand relational practice, it is useful
to explore interaction as sequences of negotiation incidents where the partners may
choose to deal with each decision in either an integrative (cooperative) or distributive
(non-cooperative) manner. Furthermore, as alliance processes involve both joint value
creation and appropriation (Ness and Haugland, 2005; Zajac and Olsen, 1993), and both
stable and contested states (Kumar and Nti, 1998), relational practice is likely to move
between recursiveness and adaptation, as well as between integrative and distributive
interactions.

Governance Mechanisms as Structuring Means

Governance is ‘the means by which order is accomplished in a relation in which potential
conflict threatens to undo or upset opportunities to realize mutual gains’ (Williamson, 1996,
p. 12). Such order is normally attained through the implementation of various mecha-
nisms often referred to as formal and social controls (Fryxell et al., 2002; Gulati and

© Blackwell Publishing Ltd 2009

454 H. Ness
Singh, 1998; Inkpen and Currall, 2004). Research has in particular identified three
different kinds of mechanisms that ‘operate in conjunction’ (McEvily et al., 2003, p. 92):
price, authority, and trust (Adler and Kwon, 2002; Bradach and Eccles, 1989).
A basic form of governance is accomplished through use of incentives, and price (a
shorthand expression) refers to the economic structure of relationships (Bradach and
Eccles, 1989; Ghosh and John, 1999; Ring and Van de Ven, 1992). Under market
governance, incentives ‘are inherently short term in nature’, while in a relational context
‘incentive systems become inherently long term in nature, and parties could forego
present rewards on the basis of expectations of long-run equity’ (Heide, 1994, p. 77). The
incentive structure might also include relation-specific investments (Dyer and Singh,
1998), bonus structures, sharing of costs, and principles for future compensation (Ness
and Haugland, 2005).
Hierarchical features, often referred to as authority, are ‘written into contracts and also
exist implicitly by virtue of industry practice’ (Bradach and Eccles, 1989, p. 103).
Stinchcombe (1985, p. 126) observed that a ‘structure with legitimate authority, with a
manipulable incentive system, with a method of adjusting costs, quantities, and prices,
with a structure for dispute resolution, and with a set of standard operating procedures,
looks very much like a hierarchy, very little like a competitive market. Yet all these
features of hierarchy are routinely obtained by contracts between firms’. Elements of
authority include oversight committees and joint teams (Reuer et al., 2002), joint admin-
istrative rules and procedures, and assignment of decision rights (Heide, 1994; Larson,
1992).
Trust and norms are important to understand governance in alliances (Blois, 2002;
Hagedoorn and Hesen, 2007), and trust has in particular been defined as goodwill trust,
i.e. ‘one party’s confidence that the other party in the exchange relationship will not
exploit its vulnerabilities’ (Dyer and Chu, 2003, p. 58). In addition, alliances involve
competence trust, which is a belief that the partner is technically competent to perform
their role in the relationship (Das and Teng, 2001). Furthermore, norms, defined as ‘a
principle of right action binding upon the members of a group and serving to guide,
control, or regulate proper and acceptable behaviour’ (Macneil, 1980, p. 38) has been
considered a related mechanism (Cannon et al., 2000; Ring and Van de Ven, 1992).
Although cross-sectional research has shown that ex-post governance changes are
common (Reuer and Ariño, 2002; Reuer et al., 2002), contractual dynamics and their
implications for relational practice remain unexplored. Two dynamic aspects are impor-
tant: (1) governance mechanisms may change per se (e.g. trust may emerge or erode);
and/or (2) their relative importance may change. The long-term nature of alliances makes
both kinds of dynamics likely, and both are important in order to explain the evolution
of relational practice.

Negotiations: Joint Decision Making to Close the Planning Gap

Negotiation is defined as ‘a process of potentially opportunistic interaction by which two
or more parties, with some apparent conflict, seek to do better through jointly decided
action than they could otherwise’ (Lax and Sebenius, 1986, p. 11). Research on strategic
alliances has differentiated between integrative and distributive negotiation strategies

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 455
(Kumar and Nti, 1998; Larsson et al., 1998). Integrative strategies contribute to value
creation and goal realization in alliances, while distributive strategies are linked to
realization of private goals and value claiming. In negotiation research (see, e.g. Pruitt
and Carnevale, 1993), two different strategies are considered integrative: problem
solving and inaction. Problem solving tactics (trading on issues by identifying different
priorities, contributing resources, offering (future) compensation, solving underlying
concerns, and extensive information sharing) aim at securing both parties’ primary
interests in the decision that is made. Inaction (avoidance, withdrawal) is a passive
strategy, which is often used to deal with unimportant issues or when concern for
immediate outcomes of the incident is low. While both problem solving and inaction
reflect an integrative dimension, they have different implications. Problem solving
implies active value creation. In contrast, inaction implies a passive response, which
creates little value.
Distributive negotiations also involve two primary strategies: contending and conces-
sion making. Contending is aimed at securing one’s own interests without concern for the
other party. Key tactics include threats, positional commitments, and persuasive argu-
mentation. Concession making reduces demands and goals. Consequently, these strat-
egies represent actions of claiming and abstaining from value, respectively. What
basically sets integrative negotiations apart from distributive is the value creating poten-
tial that follows from addressing several issues simultaneously; integrative negotiations
are variable-sum situations, while distributive are fixed-sum situations that deal with one
issue at a time. However, as integrative negotiations deal with several issues simulta-
neously, value sharing aspects might be one of the issues. Thus, integrative negotiations
might also involve a distributive aspect if they also assign sharing solutions. Cross-
sectional research on alliances suggests that both integrative and distributive orientations
are commonly used (Ganesan, 1993; Lui and Ngo, 2005; Mohr and Spekman, 1994).
Furthermore, Madhok (1995), exploring the relation between contracts and conflict
resolution in alliances, argued that when trust is present more information is shared and
enhance integrative decisions. Thus, governance structures matter, and alliance pro-
cesses are likely to include a range of integrative and distributive behaviours when viewed
over time.

The Evolution of Relational Practice

The conceptual idea that is explored in this paper is summarized in Figure 1. The
process of reaching an initial agreement is considered important because early events
foreshadow how subsequent interaction will be conducted (Doz, 1996; Inkpen and
Currall, 2004). Studies have emphasized the importance of identifying mutual economic
advantages (Larson, 1992), including the distribution of costs and revenues (Ariño and de
la Torre, 1998). This reflects a focus on economic incentives (‘price’) as a governance
mechanism, and the distribution of money tends to be a contentious issue involving
opposing interests. However, if aspects of value claiming are left for future settlement, it
may be easier to adopt a more integrative interaction mode that focuses on joint value
creation (Ness and Haugland, 2005). Furthermore, if several economic issues are
addressed simultaneously, different priorities between the partners enable trading on

© Blackwell Publishing Ltd 2009

456 H. Ness
Initial structuring Evolving structure
of governance of governance
mechanisms mechanisms

Proposition 1a–b Proposition 4a Proposition 5a

Socially embedded / Proposition 3 Proposition 4c

market context
Initial relational Alliance
Evolution of
Alliance initiation practice reconfiguration /
relational practice
termination
Initial issue structure
Value creation / appropriation Proposition 5b
Developmental / maintenance
Proposition 2a–b Proposition 4b

Initial Evolving
negotiation negotiation behaviour
behaviour

Figure 1. The evolution of relational practice

issues and increase the value of the agreement. Hence, depending on the issue structure
and expectations for future interaction, initial interactions can initiate either an integra-
tive or distributive relational practice. This suggests that the establishment of an initial
governance structure is strongly linked to negotiation behaviour.
Although alliances can be initiated from a predominantly market context, which
emphasizes transaction value in the narrow sense, alliances may also be initiated to gain
access to knowledge, combine capabilities, and stimulate learning. As closer integration
and coordination is required, some hierarchical features (‘authority’) such as joint work
groups and routines for information sharing may be perceived as useful (Vlaar et al.,
2007). Assuming that neither party can unilaterally force such arrangements into the
relationship, the introduction of authority elements is likely to reflect problem solving (an
integrative practice). Furthermore, this can enable further integrative interaction or,
given that shared decision rules often restrain contentious behaviours and secure proce-
dural justice, move a predominantly distributive practice towards a more integrative
mode.
While alliances may initially involve competence trust, goodwill trust and norms often
emerge from problem solving interaction. If early interaction patterns focus on economic
structures and involve a ‘substantial’ use of contentious tactics, suspicions and perceived
risk of opportunism may arise, which can constrain the development of goodwill trust. In
such a situation, relational practice may become recursively distributive and constrain
joint value creation if successful implementation of authority mechanisms does not
foreclose such behaviours. On the contrary, if relational practice emerges as predomi-
nantly problem solving, formation of goodwill trust is enabled, and can in turn serve as
a basis for further integrative relational practice.
As alliances evolve, they are frequently evaluated with regard to efficiency and equity
(Ring and Van de Ven, 1994). A key challenge is to balance value creation and appro-
priation concerns in order to sustain relational rents. Hence, the partners need to keep
the alliance in a state of relative equilibrium, and avoid negative outcomes and process

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 457
discrepancies (Kumar and Nti, 1998). Research suggests that partners learn and improve
their contracting practice over time (Mayer and Argyres, 2004; Vlaar et al., 2007),
incrementally create more functional structures, and accumulate experience with regard
to how to deal with unexpected issues and adapt their actions accordingly (Ness and
Haugland, 2005). Holmqvist 〈2004) found that learning processes in alliances involve
relatively stable and rule-like behaviours that enable activities to be carried out effec-
tively, and that such recursiveness enables the formation of accurate outcome expecta-
tions and the identification of beneficial adaptation. Holmqvist also suggests that when
outcome expectations are not met, dissatisfaction triggers adaptation of learning pat-
terns. Such learning might be related to both the ongoing structuring and the use of
negotiation strategies. As governance structures increasingly combine different gover-
nance mechanisms into a common framework for interaction, flexibility is likely to
increase, and create a greater potential for different practices to emerge. As the partners
over time experience the effectiveness of different interaction patterns in different situ-
ations, they might over time develop a broad repertoire that might be put to use in a
given situation. Thus, different observations and experiences over time might constitute
a set of potential practices available for use. Such relational practices can be defined as
‘those traditions, norms, rules, and routines through which’ the alliance is managed
( Jarzabkowski, 2004, p. 545). These patterns may operate at different times in the
relationship or as simultaneous sub-processes. However, how relational practice(s)
emerge, evolve, and influence outcomes remains unclear, and I will now explore these
issues empirically.

RESEARCH METHODS
Longitudinal approaches are particularly appropriate to study processes ( Van de Ven,
1992). I chose a multiple case study approach (Yin, 2003) as it allowed for an in-depth
exploration of relational practice as it occurred naturally. Comparative analysis allowed
me to control for idiosyncratic developmental aspects, enabling theory development
(Eisenhardt, 1989). I conducted the study in the Norwegian retail sector, as I was able to
gain open and sustained access. During the late 1980s and 1990s, Norwegian retailing
went through substantial change in terms of horizontal integration (e.g. chain store
formats). Vertical integration also increased, but the degree of integration varied sub-
stantially. Thus, hybrid organizational arrangements were common, and selection of
appropriate firms could be done. Furthermore, the ‘non-legalistic’, yet often contract
based business culture provided an opportunity to explore the aspects assumed to be
of importance to relational practice. These aspects are important, as selection ‘of an
appropriate population controls extraneous variation and helps to define the limits for
generalizing the findings’ (Eisenhardt, 1989, p. 537). However, the selected cases are not
drawn for statistical reasons, and generalization is made ‘from one case to the next on the
basis of a match to the underlying theory, not to a larger universe’ (Miles and Huberman,
1994, p. 29). A 2 ¥ 2 sampling frame (see Figure 2) was developed to guide case selection
and distinguished between cases on dimensions that are conceptually meaningful (Miles
and Huberman, 1994). The degree of formalization of contracts (formal/informal) is one
dimension. This allowed for selection of cases that varied with regard to governance

© Blackwell Publishing Ltd 2009

458 H. Ness
COOPERATION TO

Maintain Develop

Esthetique – Scandinavia
Formal
L´Oréal Online –
Europay
CONTRACT

(Not COOP –
Informal
selected) Lilleborg

Figure 2. The selection frame

mechanisms. As negotiation behaviour is difficult to observe prior to interaction with the

research context, ‘kind of cooperation’ was chosen as the other dimension. ‘Kind’
referred to the basic intent of the cooperation, and I assumed that relationships that
required development of new solutions or processes would differ with regard to nego-
tiation behaviour compared to maintenance-focused relationships. As discussed earlier,
relationships can emerge from a predominantly market based interaction, or as a result
of identified learning and innovation opportunities that follow from resource comple-
mentarities. I initially contacted 14 firms. Some of these responded negatively, but when
the first access agreement was made, I changed my dialogue towards the remainder in
order to gain access to secure sufficient variation.
Three dyads were selected for the study: Esthetique and L’Oréal (hereafter, E–L),
COOP and Lilleborg (hereafter, C–LB), and Scandinavia Online and Europay (hereaf-
ter, SOL–EP). While no case was exclusively selected to explore an informally founded
and maintenance oriented relationship, both the E–L case and the C–LB case also relate
to this combination. The E–L dyad was initially based on extensive formal contracting,
but over time, governance became increasingly relational. Similarly, the cooperation
between C–LB also intended to maintain the corporate relationship.
Data were collected from both sides of the dyads. Informants had intimate knowledge
of the relationships and were key participants in the joint decision-making processes. A
total of 37 semi-structured interviews were conducted, recorded, and transcribed verbatim
(Table I). The cases were followed in real-time from February 1997 to autumn 1998, and
archival data were used to extend the case histories. Extensive archival data were
collected on all cases, and were actively used to shape the interviews and to establish
detailed and exact case chronologies. Archival data included contracts, project hand-
books and reports, internal and joint memos, agendas, status reports and minutes from
meetings, letters, and annual reports. Archival data enabled me to gain credible inter-
view data on former interactions. It was possible to use, for example, agendas, status
reports, and minutes from meetings to enhance informants’ memories and also to
examine accounts of incidents, as they were experienced, for corroboration. Finally,
press clippings and statistical information on the businesses and their respective markets
were used to increase understanding. Although the data were collected in 1997–98, they
have been preserved in a systematic and complete form and serve to illuminate an

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 459
Table I. Interviews

Date Company Informant Duration

(minutes)

SOL–EUROPAY
970421 SOL Espen Grimmert 75
Helge Zimmer
971103 SOL Sonja Fiskum 60
971125 SOL Sonja Fiskum 105
980603 SOL Sonja Fiskum 60
980611 SOL Sonja Fiskum 90
980729 Europay Jørgen Myhrer 120
980814 SOL Sonja Fiskum 90
980827 Europay Jørgen Myhrer 60
COOP–LILLEBORG
970403 COOP Dag Schøyen 70
970404 Lilleborg Jon Jacobsen 75
970630 COOP Dag Schøyen 60
970630 Lilleborg Jon Jacobsen 60
971003 COOP Dag Schøyen 75
971103 COOP Dag Schøyen 60
971125 Lilleborg Jon Jacobsen 120
980216 COOP Dag Schøyen 80
980216 Lilleborg Jon Jacobsen 150
980309 Lilleborg Jon Jacobsen 75
980309 COOP Beate Storsul 70
980320 COOP Dag Schøyen 75
980323 Lilleborg Jon Jacobsen 70
980403 Lilleborg Jon Jacobsen 60
980403 COOP Dag Schøyen 60
ESTHETIQUE–L’ORÉAL
970228 Esthetique Thor Hauge 60
970404 Esthetique Thor Hauge 45
970422 L’Oréal Sigurd A. Hauge 75
970606 Esthetique Thor Hauge 70
970701 L’Oréal Sigurd A. Hauge 80
971008 Esthetique Thor Hauge 75
971126 L’Oréal Sigurd A. Hauge 60
971126 Esthetique Thor Hauge 30
980223 L’Oréal Anne R. Larsen 80
980302 Esthetique Thor Hauge 75
980313 L’Oréal Sigurd A. Hauge 60
980330 L’Oréal Sigurd A. Hauge 80
980806 L’Oréal Bernhard Claudel 75
980821 Esthetique Thor Hauge 60

© Blackwell Publishing Ltd 2009

460 H. Ness
important aspect of alliance dynamics. In addition, the recent practice-turn in strategy
research (Whittington, 2006; see also the special issue of JMS, 2003, 40, 1) makes this
exploration particularly relevant.
Analysis and coding, using HyperResearch, were conducted throughout the data
collection period. Codes were both emic and etic to avoid the pitfalls of context-stripping
and contextualizing strategies (Maxwell, 1996). Event chronologies were produced on
three levels of detail. In particular, time ordered matrixes and case dynamics matrixes
were developed for each case (Miles and Huberman, 1994), and integrated as different data
points in the comparative analysis (McPhee, 1990). In line with the research question,
time ordered displays (scatterplots) provided a detailed analysis of dominant governance
mechanisms and negotiation strategies for all event histories. In order to develop
explanations, predictor–outcome consequences matrixes were developed (Miles and
Huberman, 1994).
Tape-recordings, verbatim transcriptions, and accuracy checks with respondents were
used to safeguard the credibility of the data, thereby enhancing descriptive and inter-
pretive validity (Maxwell, 1992). Theoretical validity was enhanced by using well-
established concepts (Maxwell, 1992). Reliability was enhanced by creating a systematic
database for the study and providing detailed descriptions of the analytical steps taken.
In addition, multiple data sources were used and efforts were made, throughout the
study, to build and maintain a chain of evidence ( Yin, 2003).

CASE DESCRIPTIONS
In order to ease the description of the cases and to enhance analysis, the case histories are
divided into initiation, maintenance, and dissolution/reconfiguration phases.

Esthetique–L’Oréal
When Narvesen acquired nine perfumeries in 1992, its perfume chain, Modena,
almost doubled in size. Soon after, Modena started to renegotiate its supplier con-
tracts. The plan was to reintroduce Modena, a chain of heterogeneous stores with
mediocre reputation, into Esthetique, a homogenous chain which would focus on the
selective high-end segment. To succeed in implementing the strategy, they needed
cooperation from their key suppliers, such as L’Oréal, which controlled key selectively
distributed products and brand names. However, the relationship between the two
firms was transaction-based and relied extensively on formal contracts, as an important
element in L’Oréal’s marketing strategy was to have multiple contracts with the
various entities within Esthetique (e.g. with Esthetique’s head office, with each Esthe-
tique store, and for each stand in department store shops). In 1998, Esthetique oper-
ated 45 stores, and both Esthetique and L’Oréal experienced growth above the
industry average.

Initiation: preparing for new chain introduction (spring 1992–May 1993, 14 incidents). From the
spring of 1992 to the summer of 1993, Narvesen planned to launch Esthetique. They
initiated discussions with L’Oréal (and other suppliers) to open the door to more

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 461
cooperative relationships, as the interaction was highly contentious: ‘There was not a
conversation without threats being set forth, right. Either you take this in, or else . . .’ (Esthetique
CEO, speaking of a L’Oréal director responsible for selective distribution). From the
autumn of 1992 until March 1993, several contract meetings were held between the
parties, and focused on: (1) the economic terms for future transactions; and (2) a new
contract supplement designed to regulate seven issues typically causing conflicts. The
supplement was signed, and the firms hoped to reduce conflict and improve interac-
tions by relying on standard operating procedures. Thus, the parties used authority to
mitigate conflict. Furthermore, during late winter Modena offered L’Oréal the possi-
bility to have one brand represented in the chain’s basic assortment (six brands that at
least 80 per cent of the stores were required to carry). This was instrumental to reach-
ing an agreement. ‘Naturally, we attempted to make it as attractive as possible to be in the basic
assortment’ (Esthetique CEO). During spring 1993, L’Oréal invited Modena to partici-
pate in a campaign for a key brand. However, they refused as it was perceived as a
supplier initiated activity, and ‘we were at the time interested in developing our own brand name’
(Esthetique CEO).

Maintenance: new chain introduction ( June 1993–June 1995, 59 incidents). Modena released
their plans for Esthetique and invited its suppliers to a meeting. In a memo, L’Oréal
stated that it perceived the chain concept as ‘a good foundation . . . and the best concept that is
presented in the business’. During autumn 1993, as Esthetique was introduced, L’Oréal was
invited to participate in product meetings and to present new products. L’Oréal was also
invited to draft joint budgets and sales forecasts with individual stores, making it possible
for them to influence store activities. In December 1993, the contract supplement was
adjusted, based on experience. Overall, the contract supplement was considered a
success in reducing the conflict level. As a result, interaction became less distributive than
previously. In addition, Esthetique invited L’Oréal to discuss possible changes in the
spacing of stands in a key department store shop. Department store shops presented
brand-dedicated stands that were store-specific, and paid by the suppliers. This was an
important cooperative signal, but Esthetique had been experiencing a worsening rela-
tionship with the L’Oréal director for selective distribution because of his reckless
behaviour. Esthetique’s CEO explained: ‘we tried to make it work by means of normal conver-
sations, but it did not. So I sent a letter where I stated that if we did not receive an unconditional apology
for, there were three different issues . . . and further that he and one of their salespersons were to stay away
from Esthetique, both the head office and the stores, for at least one year. We were prepared to cease all
further collaboration with L’Oréal’.
L’Oréal’s CEO complied fully: ‘You do not do these things unpunished. And he was declared
“persona non-grata” in writing. And I completely agree’. Subsequently, the two CEOs met to
work out how to preserve the relationship until the director was replaced. For almost a
year, cooperative initiatives were at a minimum. However, in the summer of 1994, the
two parties signed two new stand contracts for two key L’Oréal brands. By then, L’Oréal
had good stand spaces for three of its top priority brands in Esthetique’s most important
department store shop. In October, a new director for selective distribution arrived, ‘so
I went there and . . . we had a good contact’ (L’Oréal director). From spring to autumn 1994,
several meetings took place regarding contract terms for non-selective products. As these

© Blackwell Publishing Ltd 2009

462 H. Ness
products were unimportant to Esthetique, and the competition between suppliers was
strong, interaction was primarily focused on economic terms, and negotiations were
highly contentious.
In January 1995 a contract revision started with the new L’Oréal director, and
negotiations ended in June (1995). Several issues were discussed, particularly the eco-
nomic structure (sales terms, volume-based discounts, bonuses, etc). ‘It took time, but it was
also a discussion time, and a way to get to know each other better. In a way, because you do not know
the person well, you say “what is he trying to have me sign?” you know’ (L’Oréal director). In
particular, ‘The major issue in that agreement was the split of advertising costs. Getting an agreement
with the department store shops was important to us’ (L’Oréal CEO), and the final split ‘was tough
for them’ (Esthetique CEO).

Reconfiguration: increasing relational commitment ( June 1995–September 1998, 65 incidents). The

reconfiguration stage evolved slowly, and began with the signing in June of the new
contracts for selective distribution for 1995. An important outcome of the process was a
substantially improved relationship. In addition, joint activities and the stand agreements
paid off: one L’Oréal brand went from a traditional No. 4 to the second best performing
brand in the chain, and influenced positively the expectations for the relationship.
During winter 1996, new contract negotiations occurred and focused on the possibility
that L’Oréal could get a second brand into the basic assortment. After several meetings,
they agreed and both parties’ major concerns were satisfied: L’Oréal got their second
brand into the basic assortment, and Esthetique improved their terms. During autumn
1996, a new discussion related to stands occurred, and L’Oréal was offered a new stand
space (a standard five year contract). Esthetique’s CEO said: ‘They have done several things,
since [new L’Oréal director] came in: better terms than we previously had, and that is a development. They
certainly will have something in return. And this is more like underlying, there are no formal written
agreements on it, but more like principles for what kind of development we see, and that we agree to that
development, see joint gains from it’.
Cooperation in joint product meetings and joint activities in stores increased during
winter and spring 1997. The cooperative climate was characterized by a strong sense of
shared direction and interaction was primarily based on problem solving. Interaction in
joint product meetings became increasingly important, as they established joint six-
month plans for the relationship on the operational level. One brand manager in L’Oréal
described how the interaction in the joint product meetings had been lately: ‘I feel that
the interests to a great extent are common, and sometimes I also feel that certain things cannot be
done. . . . But we jointly attempt to find solutions’. However, interaction related to broadly
distributed products remained contentious. For instance, after having extracted major
concessions (‘We took our pants off ’ L’Oréal’s CEO commented) in negotiations related to
possible distribution of a new hair-dye product L’Oréal was about to introduce, Esthe-
tique chose a competing brand to increase supplier competition. Describing the state of
the relationship in August 1998, the L’Oréal director responsible for selective products
stated: ‘It is not like “I give you this, what will you give me?”. In fact, today it is a confidence
relationship’. Esthetique’s CEO corroborated: ‘Currently, we have an oral understanding of our
mutual commitments, and the way to proceed’.

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 463
Interpretive summary E–L. The E–L dyad emerged from recurrent contracting in the
market, and continued successfully as the data collection ended. Initially, the overriding
governance mechanism was the economic structure, and interaction was to a large extent
dominated by distributive negotiations where the integrative aspects were minor. The
initial issues addressed ‘activated’ these aspects and formed an initial relational practice
characterized as recursively distributive. Over time, additional mechanisms were intro-
duced into the governance framework, and relational practice changed as hierarchical
mechanisms reduced level and frequency of conflict and constrained use of distributive
tactics. Furthermore, as goodwill trust increased, problem solving became increasingly
important and relational practice adapted towards an integrative mode. Although the
ubiquitous importance of price underscores the importance of value appropriation
concerns, the pricing structures enabled trading on issues. However, interaction related
to broadly distributed products remained recursively distributive throughout.

COOP–Lilleborg
COOP went public in June 1997 and announced an expected annual gain of NOK100m
from automatic replenishment of stores (i.e. sales-information generated in a store is fed
back through the distribution channel and initiates machine-generated orders to replen-
ish the store). The estimate was based on experience from a cooperative project with
Lilleborg, a major supplier of detergents and personal care products. The competition
between grocery chains had significantly increased during the first half of the 1990s, and
resulted in efforts to increase efficiency. Based on common interests, common experi-
ence, and personal relationships the project was initiated as a ‘gentleman’s agreement’
(no ‘signed contract’). To succeed, substantial process innovations needed to be imple-
mented. However, test results were more promising for the flow between COOP’s
distribution centre and its stores than between COOP and Lilleborg, and COOP chose
to decrease its relational commitment and increase its internal focus. The case describes
their cooperation from 1994 until autumn 1997.

Initiation: preparing the replenishment test (April 1994–end 1995, 42 incidents). The agreement
concerned use of POS (scanner) data to implement vendor managed inventory and
automatic replenishment in the distribution chain. The parties expected this to increase
administrative and economic efficiency (e.g. simplified routines, and reduced control,
inventory, stock-outs, and transportation costs). In the initial meetings, the parties explic-
itly agreed to six principles as a basis for their cooperation: (1) trust; (2) an organizational
foundation (not only managerial level); (3) value creation for both parties in a long term
perspective; (4) open sharing of information without hidden agendas; (5) operational
simplification; and (6) mutual knowledge. With regard to the economic expectations, ‘it
was already stated that the project makes so large savings possible that calculating the costs was not
necessary’ (Project Manager (PM), Lilleborg). In late 1994 the formal decision to start the
project was made. In order to be able to coordinate their activities in this complex task,
they jointly created a ‘Project Handbook’ in early 1995. ‘We set up a project organization
where both firms participated, and the necessary governing documents. The organization was based on

© Blackwell Publishing Ltd 2009

464 H. Ness
the premises of the project because we felt that it was a joint project with common goals’ (PM,
Lilleborg).
The project handbook described, among other things, information sharing and archi-
val routines, rules and regulations for interaction, and routines for conflict resolution.
The project organization involved a joint steering group, project managers (one from
each organization working jointly), and several working groups. During the first year of
preparations (1995), the project organization was changed several times to ease coordi-
nation. During the autumn, they agreed on the budget (sharing of costs), and established
fixed prices (non-market pricing) to enhance measurement. Other important decisions
concerned the selection of test sites and the inclusion of cross-docking. Cross-docking
represented a change to more efficient operations at the COOP central terminal, as
repacking for the final destination stores would prevent actual shelving of the goods, but
required more planning and precision. Interaction on all levels was reported to be
problem solving in nature. ‘[The PM from COOP] and I had many discussions regarding the
content of the project – methods, progression, dedication of employees – and even though we did not always
agree on how to do things, I did not perceive it to be caused by political considerations’ (PM, Lilleborg;
the term ‘political’ refers to conflict of interests). A similar practice was identified in the
steering group. ‘I believe that most people that participated in the project did not perceive it like there
were people from two different organizations around the table’, and ‘at the level of the working group
leaders there was no politics at all – one just worked on a project’ (PM, Lilleborg; confirmed by
COOP’s Logistics Manager).

Maintenance: testing and planning the roll-out pilot (1996, 16 incidents). In early 1996 two test
sites were operative, and the results were promising. Compared to the initiation phase,
the test phase involved fewer joint decision points. Interaction was reported to largely
follow a stable interaction pattern, as the participants had found appropriate ways of
dealing with any issues that occurred. While the test frequently met minor operational
problems that were solved jointly, and the organization of the work groups represented
an organizational challenge, a more critical problem was insufficient resources. ‘If we are
to finish the project as planned, it is necessary to revitalize the project through agreeing on the goals, creating
renewed interest in both organizations, and not least see to that the project is provided with necessary
resources’ (Status Report to Steering Committee, May). The issue was not permanently
solved, and in early August the Steering Committee stated: ‘The main challenge currently is
to keep the project alive. Most of the project organization in COOP must be replaced shortly’. These
issues were again addressed in late August: ‘A continuance of the project and implementation of
continuous replenishment will be best secured through conducting the implementation as two separate
initiatives’ (roll-out memo). The suggestion came from COOP’s Logistics Manager, who
‘understood that Lilleborg were negative. Because of body language and reaction’. The PM from
Lilleborg explains: ‘At the moment the idea was presented, replenishment from Lilleborg [to COOP]
was in reality dead. Because COOP realized that it was very important for them to continue in the stores,
and [COOP] had insufficient resources and perhaps not sufficient faith . . . at least [COOP] did not have
any personnel interested in running the other part. We did not fight it at all’. However, it was decided
in the meeting that the pilot should proceed with focus on both flows.
As a result of test measures (that proved more promising for the internal operations
in COOP), insufficient resources, and expected turnover, COOP emphasized their

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 465
interests at the expense of Lilleborg’s, and this proved to be significant for subsequent
interaction. For the remainder of 1996, little interaction took place. At the end of
October, the steering group again decided to continue the project as a joint initiative.
Also during October, Lilleborg’s PM and COOP’s Logistics Manager were in charge of
negotiations as they changed the terms of the agreement to ExWorks (a contract stan-
dard meaning that buyer takes responsibility for goods at the supplier site; formerly
Lilleborg had delivered to buyer site) due to the increasing cross-docking practice in the
industry. These negotiations (as other ‘transactional’ issues as opposed to ‘cooperative’
issues in the relationship) took place in separate meetings (‘apart’ from the project), and
involved a detailed documentation of ‘the numbers’ as neither party accepted smaller
margins; interaction was distributive. During late autumn 1996, ‘possible solutions started to
erode. We [COOP and Lilleborg] had been too ambitious’, and ‘(formerly) one was looking forward to
working on this project, but then things started to happen as for instance personnel turnover. The
enthusiasm disappeared . . . then the development stopped’ (COOP’s Logistics Manager). When
COOP’s PM left, ‘we [COOP and Lilleborg] entered a new phase’ (PM, Lilleborg).

Reconfiguration: decreasing relational commitment ( January 1997–autumn 1997, 8 incidents). The

roll-out pilot was postponed until March, and in April it was split into two separate
projects by the Steering Group. As Lilleborg feared that their involvement would dimin-
ish, it was interested in participating in the project ‘internal’ to COOP. As a response,
it was offered a seat on the new steering committee for the pilot, which focused on the
flow from COOP’s terminal out to stores. This pilot involved several suppliers, but only
Lilleborg was given this opportunity. Lilleborg perceived this as ‘perfectly all right’, and ‘an
advantage’, and the pilot was started during spring. However, Lilleborg was still hoping to
continue the other part as well. Although personal relations did not worsen, key per-
sonnel left, and a new Director of Logistics entered the COOP system. In meetings
during summer, it became evident that the available resources did not meet the chal-
lenges in the project. ‘I am completely convinced that stopping the other part of the project was the
right thing to do. We did not pay sufficient attention, and concentrated on the first part, because there the
gains are largest’.

Interpretive summary C–LB. C–LB emerged from a socially embedded context, as well as
from previous joint experiences. The issues the cooperative initiative raised focused
predominantly on the potential for joint value creation. The overriding governance
mechanisms were norms and trust, and inclusion of authority mechanisms added to a
recursively integrative relational practice where value appropriation concerns were left
for future settlement. In particular, the hierarchical elements were several times revised
and changed. These changes were expressions of learning and adaptation to further
enhance a recursively integrative practice. As value creation, and subsequent appropria-
tion, depended on investments, the parties found ways to share cost over time, but
insufficient resources restrained sufficient value creation. In fact, when performance
measures came out in favour of COOP, they reduced resource commitment; this
resulted in an adaptation of relational practice towards a less integrative, and more
distributive, mode. Due to personnel changes, and failure to reach their goals, trust and
norms were weakened, and this added to an increasingly distributive practice. Actually,

© Blackwell Publishing Ltd 2009

466 H. Ness
when COOP decided to close down the ‘inflow’ in the roll-out, this decision was very
contentious. Similarly, a recursive distributive practice existed on related issues that were
defined as ‘commercial’ and not part of the project. However, the continued part of the
project represented a continuation of an integrative relational practice.

Scandinavia Online–Europay
The cooperation between SOL and Europay was an attempt to increase consumer
internet purchases, by providing consumers with the possibility to pay safely using their
credit card. SOL was established as a prime mover in the ‘dotcom-boom’ in Scandinavia,
and established a ‘virtual mall’ for traditional stores that wanted to pursue the potential
for internet sales. Secure online trade was perceived as being crucial to increase trans-
action volume. To be able to offer this, they sought out a new partner to SOL: Europay.
SOL’s revenues would come from charging the stores fees based on sales and software
licensing, while Europay’s revenues would come from the credit card transactions. This
way, separate streams of revenues would be created from the same transaction. The
relationship started in winter 1997, and in November 1997 one of the world’s first SET
(Secure Electronic Transaction) applications was commercially introduced to the con-
sumer market. The cooperation was founded on one general cooperative contract and
one specific SET contract. Although formal, the contracts were general and primarily
direction giving, as operational issues were not addressed. SOL suffered several years of
losses, and the transaction volume proved disappointing during the winter and spring of
1998. As a result, it revised its strategy, sold out the technology, and terminated the
relationship in July 1998.

Initiation: development of beta software (February 1997–October 1997, 32 incidents). During

winter 1997 initial talks took place between SOL and Europay. In June, a formal but
general cooperative contract was signed, followed by a formal contract for the SET
project. During initial talks, the relationship was characterized by ‘good chemistry’, and
the parties perceived each other as highly competent. As none of the contracts were
specific in operational terms, they depended on joint decision-making throughout their
relationship. A project group was assembled and also functioned as the primary joint
work group and decisive forum. To reduce conflict, they made decisions by consensus,
removing the possibility of highly contentious interactions. They also developed a system
for information sharing and project routines. As separate revenues would result from the
successful implementation of the technology, value claiming issues were removed from
the ongoing interaction. Shared costs were covered based on reciprocity, and ‘private’
costs were paid by each organization.
Trust was important in the relationship: ‘At least on our side we had great respect for those
persons in SOL. They had professional knowledge, and they had this straightforward way and openness
in the dialogue that – well, you developed trust in them. And it became evident that what they knew of
relevance, they told us’ (Europay manager). Similar statements were made by the key
informant in SOL. Immediately following the contract signing, software development
started and ended in the commercial introduction of a beta-version in November 1997.

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 467
In addition, the parties engaged in joint marketing, and discussed possible co-branding
of a credit card.

Maintenance: early production and development of 1.0 version (November 1997–February 1998, 11
incidents). After the introduction of the beta software, the development of a 1.0 version
continued. Interactions were characterized as joint project oriented: ‘The project group really
worked on the alternatives: sometimes we presented alternatives, sometimes SOL did, and that was a very
fertile process. It was much focused on solutions’, and ‘if people agree to a decision, then commitment is
much stronger’ (Europay manager). Although technical problems were encountered, they
did not evoke conflict of interest. However, the discussions regarding potential
co-branding of a card did: ‘Then again they were approached by SOL. I was not involved in that.
And their first reaction was “why bother to do this again”, but then they chose to make another attempt.
We worked out a concept, and came back to SOL with it, price tags and all’ (SOL Manager). The
negotiations were highly contentious, and SOL chose a competing offer. This did not
delay the development of the software, but impacted on related marketing: ‘It cooled our
initiatives considerably’ (Europay Manager).

Termination (March 1998–July 1998, 5 incidents). As the 1.0 version was introduced in
early March the cooperation changed as the SET contract was fulfilled. The joint
project group was dissolved, and interaction relied primarily on initiatives from the
two marketing departments. These had not been a part of the project and did not have
the same kind of ownership to it. The low transaction volume continued, and some
attempts to increase marketing were implemented. However, other SET projects in the
industry were delayed, and there was no ‘marketing push’ from the industry from
which to benefit. Interaction almost ceased, and SOL revised their strategy. This
period of inaction and half-hearted attempts at marketing contributed to the deterio-
ration of the relationship. In July 1998, SOL decided to sell the technology to other
actors.

Interpretive summary SOL–EP. SOL–EP emerged out of an analysis of resource comple-

mentarities, not from previous interaction. Initial meetings resulted in trust between the
parties, and together with two formal contracts and some useful hierarchical elements the
cooperation was in fact very quickly up and going. The initial issue structure was
complex, but left value appropriation concerns out, as their interactions would create
separate streams of revenues. Relational practice was recursively integrative from the
start, and the initial governance structure and negotiation behaviour enabled and rein-
forced one another. Relational practice basically stayed recursively integrative until
performance measures and personnel changes co-occurred and changed the governance
structure. Trust, norms, and authority elements were weakened, and value claiming
increasingly focused. This also changed negotiation behaviour towards a state of inaction
which created little additional value, and ultimately led to contention, as inaction only
served to reinforce the changed relevance of governance mechanisms. As in the other
cases, interactions that were not perceived as integral to the cooperation were predomi-
nantly recursively distributive.

© Blackwell Publishing Ltd 2009

468 H. Ness
DISCUSSION OF FINDINGS
I have argued that the combination and recombination of governance mechanism and
negotiation strategies used by the parties are important to understand initiation and
evolution of relational practice, and I start out with an analysis and discussion of these.
I identify four different modes of relational practice observed in the cases and discuss how
relational practice evolved and changed between the phases. Finally, I develop expla-
nations why relational practice change between different modes.

Governance Mechanisms, Negotiation Strategies, and Their Evolution

In order to analyse the evolution of governance mechanisms and negotiation strategies,
I developed scatterplots for each phase to search for patterns in the data. The integrity
of each case was preserved in this procedure. I created a 3 ¥ 4 matrix that was defined
by the different governance mechanisms (price, authority, trust) and negotiation strate-
gies (problem solving, inaction, concession making, contending). Based on the event
chronology of each case, where incidents were consecutively numbered, all incidents
were entered into the relevant cell of the matrix (according to the primary strategies in
use, and to the primary governance mechanism interaction related to) by entering the
incident number. The main patterns observed in the scatterplots are summarized in
Table II.

Table II. Observed patterns: governance mechanisms and negotiation strategies in different phases for each
dyad

SOL–EUROPAY COOP–LILLEBORG ESTHETIQUE–L’ORÉAL

Initiation
Governance mechanisms Trust and authority Trust and authority, Price, supported by
supported by price authority
Negotiation strategies Problem solving, Problem solving Contending, some
some contending problem solving
Maintenance
Governance mechanisms Trust and price, Trust and price, Price and trust, supported
supported by supported by authority by authority
authority
Negotiation strategies Problem solving, Problem solving, some Problem solving and
some contending contending and contending
inaction
Termination/reconfiguration
Governance mechanisms Price Price, supported by trust Trust and price,
and authority supported by authority
Negotiation strategies Contending and Contending and problem Problem solving, some
inaction solving, some inaction contending

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 469
The governance mechanisms and negotiation strategies changed over time in the
cases, and the findings suggest that they co-evolve. The data also show that after an initial
governance structure is established, governance mechanisms evolve and change in two
ways. First, they change per se, i.e. pricing policies and economic structures are changed
incrementally, trust and relational norms emerge and deteriorate, and authority ele-
ments are adjusted and refined to ease coordination. Second, the different mechanisms’
relative importance also changed over time and between the phases. Such structural change
took place in all cases. Research on alliance initiation has pointed to the importance of
trust and social control during initiation (Larson, 1992), and how initial conditions
foreshadow subsequent interactions (Inkpen and Currall, 2004). Hennart (2006) even
argued that the explanations provided by process research on alliance dynamics are by
far inferior to explanations of process and outcomes that rest on initial structural (in the
broad sense) conditions. However, cross-sectional research shows that ex-post changes
frequently take place in alliances as a response to misalignment, strategic change, and
when few contractual safeguards are in place (Reuer and Ariño, 2002; Reuer et al.,
2002). In addition, Doz (1996) found that early interaction in alliances had impact on
subsequent interactions, and that learning is important to understand evolution and
outcomes.
The data suggest that the imprints from the initial structure do not foreclose develop-
mental paths. Although the initial governance structures were found to have both
enabling and constraining properties that impact on early interaction patterns, initial
conditions did not predetermine outcomes. Rather, explanations of outcomes depend on
the process and how structures and interaction patterns changed over time. The data
presented here suggest that regardless of the initial combination of governance mecha-
nisms, complementary mechanisms will be incorporated as the extendedness and expec-
tations for future interaction increase, as well as when the complexity of the relationship
increases. Similarly, when complexity, extendedness and expectations for future inter-
action decrease, the governance structure starts to erode.
Furthermore, this study suggests that governance mechanisms also change as a
result of learning and aspects inherent to the process itself, particularly the use of
negotiation strategies. The findings presented in Table II show that both distributive
and integrative negotiation behaviour were integral to the alliance processes. As was
the case with governance mechanisms, the data give credibility to the argument that
the initial use of negotiation strategies has both enabling and constraining properties
that impact on the initial governance structure. Similarly, the initial use of negotiation
strategies does not foreclose particular negotiation trajectories, and regardless of
the initial use of negotiation strategies, complementary strategies will be used as the
extendedness and expectations for future interaction, as well as the complexity of the
relationship, increase. Finally, initial use of negotiation strategies does not predeter-
mine alliance outcomes. Hence, this study shows that explanations for alliance out-
comes that rest purely on initial structures oversimplify and overlook important aspects
related to process. It also suggests that explanations for alliance outcomes that rest
purely on initial behaviours oversimplify and overlook important aspects related to the
ongoing structuring. However, combining structure, action, and process might provide
powerful explanations.

© Blackwell Publishing Ltd 2009

470 H. Ness
Four Modes of Relational Practice
The theory section argued that the initial governance mechanisms represent a strong
context for interaction, and that relational practice depends on the matching of these
mechanisms and negotiation behaviour. It was further argued that two basic orienta-
tions, one integrative and one distributive, and two kinds of dynamics, recursiveness and
adaptation, formed four basic modes of relational practice. I first discuss the four basic
modes of relational practice observed in the data: recursively integrative, recursively
distributive, adaptive toward integrative, and adaptive toward distributive. Then, I turn
to discussing how structure and action form the evolution of relational practice.
The most frequently observed combination of governance mechanisms and negotia-
tion strategies in joint decisive events is ‘trust’ and problem solving contributing to an
integrative relational practice. The combination of ‘price’ and contending is also frequently
observed, contributing to a distributive relational practice. However, price is also found
relatively frequently in combination with problem solving, contributing to an integrative
practice. The explanation the data offer for this combination is that ‘price’, the economic
structure, reflects costs in addition to income. In the SOL–EP and C–LB cases, early
resource contributions were solved by taking turns in covering costs, and this enabled
value creation. This suggests that when acceptance of costs is related to creating a
foundation for future value creation, it is more easily dealt with through an integrative
practice. Furthermore, an important aspect of value claiming was removed from the
interaction in these two cases: C–LB left the value appropriation issue for future settle-
ment when performance measures and experience had accumulated sufficiently, and the
technology developed by SOL–EP created separate streams of revenues from the same
transaction. Moreover, the E–L dyad illustrates that incentive structures are sometimes
complex, involving different elements, such as bonus structures and marketing contri-
butions in addition to list prices, which have different priorities between the partners.
Thus, if they are addressed simultaneously, there is a potential for problem solving
by trading on issues. On the contrary, ‘pie-splitting’ typically involved a distributive
practice.
Processes evolved slowly and incrementally. Common to the cases was that initial
contractual negotiations influenced subsequent interaction. As many issues evolved
through ongoing processes, the parties seemed to pick up along the same tracks that
former interaction had laid out. These observed sequences of events involved interaction
patterns where dominant governance mechanisms and negotiation strategies remained
stable, suggesting that relational practice has recursive characteristics. The relative impor-
tance of different governance mechanisms changed incrementally over time, suggesting
that once a structure (or form of agreement) has ‘settled’ it is not easily changed.
Similarly, use of decisive strategies remained stable during sequences of events as the
parties gained experience and learned what kind of tactics were effective. As such
patterns developed, relational practice tended to develop as self-reinforcing spirals.
On the other hand, the patterns also changed between the different phases, suggesting
that relational practice also has adaptive characteristics. As trust and relational norms
co-evolved with information sharing, trading on issues over time, and implicit com-
pensation practices, relational practice turned either from a distributive mode to an

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 471
integrative one, or became increasingly integrative. Similarly, when trust eroded and/or
value appropriation concerns became increasingly important, relational practice
adapted and became (more) distributive. As such concerns vary in degree, and over time,
it makes sense to theorize about adaptive practices in terms of increasingly integrative or
increasingly distributive. Furthermore, the interaction patterns in all cases suggest that
relational practices move between these modes, and that different relational practices
may co-exist as relationships develop.

Relationship Initiation and Emergence of Relational Practice

A relationship emerging from a socially embedded context seems to differ from a
relationship that emerges from market interaction in terms of dominant governance
mechanisms, negotiation strategies, and the relational practice that these form. In the
E–L dyad, which initially relied extensively on formal contracting and emerged from
recurrent (market) contracting, the economic structure was the fundamental mechanism.
Negotiations were to a large extent contentious, involving tactics such as holding firm
positions and aggressive argumentation, and the relational practice that emerged was
recursively distributive (predominantly value claiming in nature). However, as the con-
tract supplement was introduced, the conflict level was reduced and removed some
contentious interaction as some clear procedures became established. Thus, a problem
solving effort added elements of authority to the agreement, an incremental adaptation,
which in turn enabled a less contentious practice to emerge.
In contrast, the SOL–EP and C–LB cases were founded on agreements that were
predominantly relying on trust, and involved relatively scant formal contracting. Also in
these cases, initial problem solving interactions introduced elements of authority in order
to reduce conflicts and bring procedural certainty into the relationship. Important
aspects were the project organizations and a shared understanding of decision-making
principles (e.g. consensus). SOL–EP and C–LB represented relationships with higher
levels of operational interdependence compared to E-L, where joint decision making was
less frequently needed. In particular, this resulted in more extensive use of authority
elements as integral to the initial governance structure in the SOL–EP and C–LB cases.
Gulati and Singh (1998) found that hierarchical mechanisms are used less and not as
systematically in non-equity alliances as compared to minority investments and equity
joint ventures. However, Reuer et al. (2006) found that non-equity alliance contracts
tend to be neither more nor less complex compared to equity arrangements. The data
presented here do not make this comparison, but suggest that non-equity alliances also
benefit greatly from the introduction of authority elements into the governance structure.
Such elements enhance coordination, which in turn impacts positively on goal achieve-
ment. Although the SOL–EP dyad was terminated, they successfully developed the
software they intended to develop, and SOL sold the technology as a result of changing
strategy. Early, as well as later, experiential learning initiated adjustments in the ‘author-
ity structure’ and served to reinforce the relevance of ‘some bureaucracy’. Thus, some
hierarchical governance seems to enhance learning processes as closer and more stable
interaction is made possible, making it easier to identify solutions that improve coordi-
nation (Holmqvist, 2004).

© Blackwell Publishing Ltd 2009

472 H. Ness
In addition, the incentive structures in the two project-based relationships were long
term, and interaction related to value claiming was removed from initial interaction.
These characteristics contributed to a recursively integrative relational practice. This
suggests:

Proposition 1: Alliance initiation that emerges from: (1) a market context is likely to rely
on economic structures, and involve a recursively distributive relational practice; (2) a
socially embedded context is likely to rely on trust and norms, and involve a recursively
integrative relational practice.

Proposition 2: During alliance initiation, a predominantly integrative relational practice

will be enabled when: (1) value appropriation is a secondary concern in the initial issue
structure, as compared to situations where this is a primary concern; (2) the relation-
ship involves a shared intention to explore new (developmental) opportunities, as
compared to alliances that intend to maintain a transactional relationship.

Proposition 3: Introduction of authority elements will enhance adaptation of relational

practice in an (increasingly) integrative direction.

Maintaining Relationships: Managing the Planning Gap and the

Experience of Tensions
Although the discussed differences during initiation were important, common to the
cases was the evolving combination of governance mechanisms. In the SOL–EP and
C–LB cases, interaction over time added a growing body of written documentation to the
(largely) informal agreements. Although these were not formalized as contracts, they
were explicit documents. These involved administratively set prices to enable measure-
ment of value creation in the relationship, but also directed increasing attention to the
economic structure in the relationship (sharing of costs and distribution of value between
the parties). On the other hand, the E–L dyad involved a formal contract structure of
significance. Over time, additional formal contracting occurred (e.g. the stand agree-
ments), and extended their relationship further into the future. However, as the rela-
tionship evolved, informal agreements were made at the CEO level and became
increasingly important with regard to the direction of the relationship. As the relation-
ship became more committed, the parties relied increasingly on their ability to make it
work through constructive dialogue rather than formal contracting. Development of
trust and relational norms (in particular ‘intention to preserve the relationship’), and a
problem solving approach to issues, were instrumental to this and relational practice
became increasingly more integrative. Thus, the two ‘contractual practices’ evolved
towards similar combinations of governance mechanisms, although initial conditions
differed. The governance literature has discussed the relationship between trust and
control, and it has often been assumed that formal control undermines development of
trust (Mayer and Argyres, 2004). However, it is also assumed that over time, trust
emerges from repeated interactions and that social control increasingly substitutes formal
control and reliance on hierarchical features (Larson, 1992; Ring and Van de Ven,

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 473
1994). The case data presented here suggest that authority elements are introduced even
if there is substantial trust between the partners. Such hierarchical features do not
undermine trust, but rather add to it. Furthermore, the case data also suggest that
inclusion of authority elements when trust is low does not restrain development of trust;
rather they enable subsequent development of trust. This discussion suggests:

Proposition 4a: As relationships evolve, goodwill trust and norms will complement
incentive structures and authority in alliances that emerge from a market context, and
incentive structures will complement trust, norms and authority in alliances that
emerge from a socially embedded context.

Although the distribution of value was, during initiation, a ‘future issue’ in the C–LB
case, the test measures provided observations of actual value creation and had significant
influence on the expectations for future value creation. When the test measures failed to
meet expectations and funding became problematic, interaction turned more conten-
tious and accommodating (distributive) as the incentive focus increased and competence
trust was weakened. A similar progression was observed in the SOL–EP case. However,
in both cases, interaction also followed through on former decisions in an integrative
fashion. This suggests that the increasing complexity of governance mechanisms
co-evolved with an increase in the repertoire of negotiation tactics. Interaction patterns
increasingly involved value claiming aspects, reflecting the tension created by the mixed
motives in alliances. In the E–L case, emerging trust and relational norms together with
an increasing importance of shared decision forums and routines for interaction enabled
the evolution of an increasingly integrative relational practice without missing the simul-
taneous focus on value appropriation. Furthermore, an interesting finding was the
understanding of different issues as belonging to different relational contexts. For
instance, in the E–L case the interactions related to selective and broad distribution
differed significantly: interaction related to selective distribution became increasingly
integrative, while interaction related to broad distribution remained recursively distribu-
tive. In the C–LB and SOL–EP cases, issues that were perceived as (potential) conflict
issues (such as the change to ExWorks, and the potential co-branding of a card) was
handled in separate meetings to avoid disturbance of the cooperative climate. Interaction
related to these issues was perceived as ‘traditional negotiations’ (a distributive
approach). In this way, they managed to keep the cooperative efforts separate from more
‘commercial’ exchange issues (a term used by the informants in the C–LB case). Fur-
thermore, as the two (initially integrated) parts of the replenishment project between
C–LB were increasingly viewed as separate projects, the relational practice related to the
two parts developed differently. This suggests:

Proposition 4b: As the ongoing structuring of alliances allows for increasingly complex
governance structures to emerge, these structural aspects will evolve with a similarly
complex use of different negotiation strategies.

Proposition 4c: The evolution of governance mechanisms and negotiation strategies will
give rise to different relational practices over time, and the mixed motives alliances

© Blackwell Publishing Ltd 2009

474 H. Ness
involve will allow for distinct relational practices to emerge and co-exist as
sub-processes.

Termination and Reconfiguration: Experienced and Expected Value

Creation
As new situations emerged over time (e.g. in terms of issues, opportunities, problems,
breakthroughs) the parties accumulated experience and learning. As the interaction
process and perceived outcomes were assessed, the parties developed views of the
progressive state their relationship was in, which in turn impacted on sense of direction
and future expectations. In particular, as outcome discrepancies (Kumar and Nti, 1998)
became evident in the C–LB and SOL–EP cases, the parties also changed their weight-
ing of private versus shared concerns, clearly reflecting the mixed motives in alliances.
In particular, incentives became increasingly important in these cases as performance
measures became available. Thus, relational practice changed as the incentive structure
became increasingly focused and interaction was characterized by an increased use of
contentious tactics to claim value and inaction, creating little additional value. This
added a sense of process discrepancy (Kumar and Nti, 1998) to the experienced outcome
discrepancies, reinforcing an increasingly distributive relational practice. The perfor-
mance measures weakened the intentions to harmonize conflict and preserve the rela-
tionships (important relational norms). Although trust has generally been found to enable
value creation, the case data provided here suggests that trust is not enough to sustain
relationships that fail to meet the aspiration levels for joint value creation. In addition,
both trust and the shared sense of direction were weakened by turnover among key
personnel. In the E–L case, the positive economic development contributed substantially
to the recursive integrative interaction; trust and relational norms became increasingly
important, and enhanced further exploitation of the joint value-creating potential as the
incentives were perceived as effective. This suggests:

Proposition 5a: As trends in performance measurement become evident, an increased

emphasis on the economic structure will moderate the efficiency of authority elements
and trust: (1) deteriorating performance measures will restrain the efficiency of author-
ity elements, norms, and trust; (2) improving performance measures will increase the
efficiency of authority elements, norms, and trust.

Proposition 5b: As efficiency of governance mechanisms: (1) deteriorates, relational

practice will become more distributive and enhance alliance termination; (2)
improves, relational practice will become more integrative and enhance alliance
prolongation.

Explaining Evolution of Relational Practice: Identifying Recursive and

Adaptive Forces
Perceptions related to the process of cooperation and the assessment of outcomes provide
answers to the questions of how and why relational practice evolves. Tables III and IV

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 475
Table III. Perceptions of process aspects and impact on relational practice

Recursive force on Adaptive force on relational practice

relational practice towards a more

Distributive mode Integrative mode

Frequency of events Relative stability Decrease Increase

Number of issues in an event Relative stability Decrease Increase
Conflict level of issues on events Relative stability Increase Decrease
Resource availability to deal with issues Relative stability Insufficient/decline Sufficient/increase
Progressive timeliness Relative stability Increasingly delayed On time/ahead
Procedural certainty and justice Relative stability Decrease Increase
(efficiency of authority mechanism)
Development of trust and norms Relative stability Decrease Increase
Shared sense of direction Relative stability Decrease Increase

Table IV. Perceived outcomes in evaluative events and impact on relational practice

Recursive force on Adaptive force on relational practice towards a more

relational practice
Distributive mode Integrative mode

Performance measures
Joint creation of value Satisfactory Not satisfactory and not Not satisfactory but
correctable correctable
Satisfactory but
underexploited
Distribution of value Fair Unfair, with limited Unfair, but growth
growth potential to potential to increase
increase share and/or share and/or partner
partner unwilling to willing to adjust
adjust distribution rules distribution rules
Incentive alignment Efficient Inefficient and not Inefficient but correctable
correctable Efficient but
underexploited
Learning in and from Sufficient Insufficient and Insufficient but
relationship improvement unlikely improvement likely
(exploration)
Sufficient but
underexploited
Role relations Operational and Operational and/or Operational and/or
strategic fit strategic misfit and strategic misfit but
improvement improvement realistic
unrealistic Operational and strategic
fit but underexploited

© Blackwell Publishing Ltd 2009

476 H. Ness
are developed on the basis of the theory and the presented data, and explain how
perceptions of process and outcomes impact on relational practice.
The explanations rest on developmental states; whether initial and emerging goals
have been met (equilibrium), not met or been exceeded (negative and positive discrep-
ancies), and expectations related to future interaction. The tables should be interpreted
as follows. If frequency of events is stable (Table III), this contributes to recursiveness
in the relational practice. If frequency of events decreases, this contributes to adapta-
tion of relational practice and moves it in a distributive direction. If frequency of
events increases, this contributes to adaptation of relational practice and moves it in an
integrative direction (frequency of events, in a very basic way, impacts on the potential
for trading on issues over time and thus value creation based on reciprocity consid-
erations). The same reasoning applies to the number of issues in events. Furthermore,
when perceived conflict level of issues in events is increasing, relational practice adapts
towards a more distributive mode. Table IV should be interpreted in a similar way.
Note that Tables III and IV contain a series of implicit hypotheses that, just like the
propositions offered, need to be put to critical testing (for instance, one hypothesis
that can be spelled out on the basis of Table III (Frequency of events, mid row), is: As
the frequency of events decreases, relational practice will adapt towards a more
distributive mode.) However, spelling these out in prose would unnecessarily clutter
the paper.
In order to explain the evolution of relational practice, focusing on single aspects at the
expense of others (as, for instance, focusing on development of trust and norms, or
learning, overlooking other important relational aspects) involves the danger of treating
different conditions as necessary and sufficient for recursiveness and adaptation; when
theorizing about process, conditions should be treated as necessary only (Mohr, 1982).
According to Mohr (1982, p. 45), a hallmark of process theory is that it ‘eschews efficient
causality as explanation and depends instead on rearrangement – that is, on the joining or
separation of two or more specified elements rather than on a change in the magnitude
of some element’ (‘variance’ theory typically does the latter). Thus, in order to explain the
evolution of relational practice, one should consider how perceptions of different devel-
opmental states co-occur and how they combine and recombine over time (consequently,
it would, for instance, be of limited value to understand the evolution of relational
practice on the basis of development of trust and norms alone). Alliances are multi-
dimensional, and different aspects might influence process in different ways. Conse-
quently, different aspects need to be weighted and accumulated in order to understand
evolution of relational practice. For example, if a particular set of states drives the
relational practice towards recursiveness, and these outweigh those that drive adaptation,
relational practice is likely to stay recursive (or the other way round). Furthermore, if
adaptive forces of a distributive kind outweigh integrative forces, relational practice turns
increasingly distributive (and vice versa). This suggests that theory, as well as managers,
should be sensitive to how different aspects co-evolve and impact on structures, pro-
cesses, and outcomes in alliances.
While Tables III and IV provide some specific suggestions for how different process
and outcome assessments influence the evolution of relational practice, it is not intended
as an exhaustive list as they relate to the cases studied. Together with the propositions I

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 477
have developed in the paper, they provide a new and systematic approach to under-
standing alliance dynamics, and relational practice in particular.
Due to the design choices, care should be taken in generalizing the findings. In
particular, the study was conducted in vertical non-equity alliances, and such alliances
have been argued to be distinct from equity arrangements. Furthermore, the study was
done in medium-sized firms and in a domestic context where the business practice is less
‘legalistic’ than, for instance, in the USA. As the study was conducted in one particular
cultural context, there might be characteristics of both the business contexts and the
cultural context that limit its value in other contexts. Hence, care should be taken in
generalizing to other alliance forms, international alliances, and fundamentally different
cultural contexts. It should also be noted that the propositions developed relate to theory.
Hence, further exploration, as well as deductive research that puts the propositions to
critical testing, is needed.

CONCLUSIONS
This paper has explored relational practice in strategic alliances. In particular, the paper
offers an in-depth investigation of how governance mechanisms and use of negotiation
strategies develop over time and form the evolution of relational practice. The paper
contributes to the literature in four ways. First, the paper directs attention to relational
practice as a key to understanding managerial processes in alliances. Four modes of
relational practice are identified: recursively integrative, recursively distributive, adaptive
toward integrative, and adaptive toward distributive. I also explain why relational prac-
tice is stable or shifts between modes by identifying key drivers for understanding
patterns of recursiveness and adaptation. Second, this study complements cross-sectional
research on governance by providing an account of how and why governance mecha-
nisms change over time. Third, while the research on negotiations in relationships is
limited and typically cross-sectional, this study contributes to developing an understand-
ing of how negotiation behaviour develops over time in relationships. Finally, a particu-
lar limitation in the literature on inter-firm processes has been a lack of integration of
structure and action (Sydow and Windeler, 1998). The perspective offered here provides
an attempt to overcome this limitation, and this extension is important as structures
per se do not act.
Relational practice was explored empirically in three longitudinal case studies. The
findings show that different combinations of governance mechanisms and negotiation
strategies are related to distinct relational practices, and that early interactions have
recursive characteristics. However, over time relational practice also shows important
adaptive patterns, and the study also suggests that multiple relational practices might
emerge over time and co-exist as distinct sub-processes. Finally, the study also identifies
important aspects related to process and outcome evaluations that cause continuity and
change in relational practice.
As the evolution of governance mechanisms, negotiation behaviour, and relational
practice in strategic alliances is insufficiently understood, the present study provides a
useful perspective for further exploration and theory development. For instance, can the
same patterns be observed in equity alliances and in joint ventures, in international

© Blackwell Publishing Ltd 2009

478 H. Ness
alliances, and in different cultural contexts? Are these evolutionary trajectories of gov-
ernance mechanisms, negotiation strategies, and their impact on the evolution of
relational practice common or not? Also, how do other structural characteristics not
particularly explored in this paper, as for instance different kinds of specific investments,
influence negotiation behaviour and relational practice? Further, while the concept of
trust has received much empirical attention, the authority mechanism was found to be
important in all cases. Thus, the antecedents and consequences of interorganizational
routines, and their role in dyadic stability and change, should be further explored.
Although the relationship between trust and control has been discussed in the literature,
we still need to accumulate knowledge on this relationship. Similarly, additional aspects
related to negotiation behaviour should be explored, as for instance the issue of bargain-
ing power. Also, the importance of performance measures and impact on sense of
direction in relationships should be further explored. This can, for instance, be explored
by drawing on performance feedback theory (Greve, 2003) in order to understand how
aspiration levels are set in alliances, and how dyads deal with risk in their response to
performance feedback. The study also shows that multiple relational practices emerged
and co-evolved, and how they were related to different corporate contexts. This aspect
should be further explored, as different relational contexts might provide potential for
many different relational practices to emerge. Finally, while much research on practice
has focused on either the intra-organizational level or on societies or sectors as a whole,
this paper addresses the interorganizational level. While the present study primarily relies
on explanations internal to relationships, the relationship between relational practice and
industry practice needs further exploration, as well as research that also to a larger extent
considers the impact of networks and external events. Similarly, the relationship between
intra- and interorganizational practice should be further explored.

ACKNOWLEDGMENTS
I would like to thank Dennis Arnett and Henrich R. Greve for comments on early drafts.

REFERENCES
Adler, P. S. and Kwon, S-W. (2002). ‘Social capital: prospects for a new concept’. Academy of Management
Review, 27, 17–40.
Ariño, A. and de la Torre, J. (1998). ‘Learning from failure: towards an evolutionary model of collaborative
ventures’. Organization Science, 9, 306–25.
Bell, J., den Ouden, B. and Ziggers, G. W. (2006). ‘Dynamics of cooperation: at the brink of irrelevance’.
Journal of Management Studies, 43, 1607–19.
Blois, K. (2002). ‘Business to business exchanges: a rich descriptive apparatus derived from Macneil’s and
Menger’s analyses’. Journal of Management Studies, 39, 523–51.
Boddy, D., Macbeth, D. and Wagner, B. (2000). ‘Implementing collaboration between organizations: an
empirical study of supply chain partnering’. Journal of Management Studies, 37, 1003–17.
Bradach, J. L. and Eccles, R. G. (1989). ‘Price, authority and trust: from ideal types to plural forms’. Annual
Review of Sociology, 15, 97–118.
Cannon, J. P., Achrol, R. S. and Gundlach, G. T. (2000). ‘Contracts, norms, and plural form governance’.
Journal of the Academy of Marketing Science, 28, 180–94.
Das, T. K. and Teng, B.-S. (2001). ‘Trust, control, and risk in strategic alliances: an integrated framework’.
Organization Studies, 22, 251–83.

© Blackwell Publishing Ltd 2009

 Governance, Negotiations, and Alliance Dynamics 479
Das, T. K. and Teng, B.-S. (2002). ‘The dynamics of alliance conditions in the alliance development
process’. Journal of Management Studies, 39, 725–46.
Doz, Y. L. (1996). ‘The evolution of cooperation in strategic alliances: initial conditions or learning
processes?’. Strategic Management Journal, 17, Special Issue, 55–83.
Dyer, J. H. and Chu, W. (2003). ‘The role of trustworthiness in reducing transaction costs and improving
performance: empirical evidence from the United States, Japan, and Korea’. Organization Science, 14,
57–68.
Dyer, J. H. and Singh, H. (1998). ‘The relational view: cooperative strategy and sources of interorganiza-
tional competitive advantage’. Academy of Management Review, 23, 660–79.
Eisenhardt, K. M. (1989). ‘Building theories from case study research’. Academy of Management Review, 14,
532–50.
Fryxell, G. E., Dooley, R. S. and Vryza, M. (2002). ‘After the ink dries: the interaction of trust and control
in US-based international joint ventures’. Journal of Management Studies, 39, 865–86.
Ganesan, S. (1993). ‘Negotiation strategies and the nature of channel relationships’. Journal of Marketing
Research, 30, 183–203.
Ghosh, M. and John, G. (1999). ‘Governance value analysis and marketing strategy’. Journal of Marketing, 63,
Special Issue, 131–45.
Greve, H. R. (2003). Organizational Learning and Performance Feedback. A Behavioral Perspective on Innovation and
Change. Cambridge: Cambridge University Press.
Gulati, R. and Singh, H. (1998). ‘The architecture of cooperation: managing coordination costs and
appropriation concerns in strategic alliances’. Administrative Science Quarterly, 43, 781–814.
Hagedoorn, J. and Hesen, G. (2007). ‘Contract law and the governance of inter-firm technology partnerships
– an analysis of different modes of partnering and their contractual implications’. Journal of Management
Studies, 44, 343–66.
Heide, J. B. (1994). ‘Interorganizational governance in marketing channels’. Journal of Marketing, 58, 71–85.
Hennart, J-F. (2006). ‘Alliance research: less is more’. Journal of Management Studies, 43, 1621–8.
Holmqvist, M. (2004). ‘Experiential learning processes of exploitation and exploration within and between
organizations: an empirical study of product development’. Organization Science, 15, 70–81.
Inkpen, A. C. and Currall, S. C. (2004). ‘The coevolution of trust, control, and learning in joint ventures’.
Organization Science, 15, 586–99.
Jarzabkowski, P. (2004). ‘Strategy as practice: recursiveness, adaptation, and practices-in-use’. Organization
Studies, 25, 529–60.
Kumar, R. and Nti, K. O. (1998). ‘Differential learning and interaction in alliance dynamics: a process and
outcome discrepancy model’. Organization Science, 9, 356–67.
Larson, A. (1992). ‘Network dyads in entrepreneurial settings: a study of the governance of exchange
relationships’. Administrative Science Quarterly, 37, 76–104.
Larsson, R., Bengtsson, L., Henriksson, K. and Sparks, J. (1998). ‘The interorganizational learning dilemma:
collective knowledge development in strategic alliances’. Organization Science, 9, 285–305.
Lax, D. A. and Sebenius, J. K. (1986). The Manager as Negotiator. New York: The Free Press.
Lui, S. S. and Ngo, H.-Y. (2005). ‘An action pattern model of interfirm cooperation’. Journal of Management
Studies, 42, 1123–53.
Macneil, I. R. (1980). The New Social Contract. New Haven, CT: Yale University Press.
Madhok, A. (1995). ‘Opportunism and trust in joint venture relationships: an exploratory study and a
model’. Scandinavian Journal of Management, 11, 57–74.
Maxwell, J. A. (1992). ‘Understanding and validity in qualitative research’. Harvard Educational Review, 62,
279–300.
Maxwell, J. A. (1996). Qualitative Research Design. Thousand Oaks, CA: Sage.
Mayer, K. J. and Argyres, N. S. (2004). ‘Learning to contract: evidence from the personal computer
industry’. Organization Science, 15, 394–410.
McEvily, B., Perrone, V. and Zaheer, A. (2003). ‘Trust as an organizing principle’. Organization Science, 14,
91–103.
McPhee, R. D. (1990). ‘Alternate approaches to integrating longitudinal case studies’. Organization Science, 1,
393–405.
Miles, M. B. and Huberman, A. M. (1994). Qualitative Data Analysis, 2nd edition. Thousand Oaks, CA:
Sage.
Mohr, J. and Spekman, R. (1994). ‘Characteristics of partnership success: partnership attributes, commu-
nication behaviour, and conflict resolution techniques’. Strategic Management Journal, 15, 135–52.
Mohr, L. B. (1982). Explaining Organizational Behavior. San Francisco, CA: Jossey-Bass.

© Blackwell Publishing Ltd 2009

480 H. Ness
Ness, H. and Haugland, S. A. (2005). ‘The evolution of governance mechanisms and negotiation strategies
in fixed-duration interfirm relationships’. Journal of Business Research, 58, 1226–39.
Pruitt, D. G. and Carnevale, P. J. (1993). Negotiation in Social Conflict. Buckingham: Open University Press.
Reuer, J. J. and Ariño, A. (2002). ‘Contractual renegotiations in strategic alliances’. Journal of Management, 28,
47–68.
Reuer, J. J., Zollo, M. and Singh, H. (2002). ‘Post-formation dynamics in strategic alliances’. Strategic
Management Journal, 23, 135–51.
Reuer, J. J., Ariño, A. and Mellewigt, T. (2006). ‘Entrepreneurial alliances as contractual forms’. Journal of
Business Venturing, 21, 306–25.
Ring, P. S. and Van de Ven, A. H. (1992). ‘Structuring cooperative relationships between organizations’.
Strategic Management Journal, 13, 483–98.
Ring, P. S. and Van de Ven, A. H. (1994). ‘Developmental processes of cooperative interorganizational
relationships’. Academy of Management Review, 19, 90–118.
Spekman, R. L., Forbes, T. M. III, Isabella L. A. and MacAvoy, T. C. (1998). ‘Alliance management: a view
from the past and a look to the future’. Journal of Management Studies, 35, 747–72.
Stinchcombe, A. L. (1985). ‘Contracts as hierarchical documents’. In Stinchcombe, A. and Heimer, C. (Eds),
Organization Theory and Project Management. Bergen: Norwegian University Press, 121–71.
Sydow, J. and Windeler, A. (1998). ‘Organizing and evaluating interfirm networks: a structurationist
perspective on network processes and effectiveness’. Organization Science, 9, 265–84.
Van de Ven, A. H. (1992). ‘Suggestions for studying strategy process: a research note’. Strategic Management
Journal, 13, 169–88.
Vlaar, P., Van Den Bosch, F. and Volberda, H. (2007). ‘Towards a dialectic perspective on formalization in
interorganizational relationships: how alliance managers capitalize on the duality inherent in contracts,
rules and procedures’. Organization Studies, 28, 437–66.
Walter, J., Lechner, C. and Kellermans, F. W. (2008). ‘Disentangling alliance management processes:
decision making, politicality, and alliance performance’. Journal of Management Studies, 45, 531–60.
Whittington, R. (2006). ‘Completing the practice turn in strategy research’. Organization Studies, 27, 613–34.
Williamson, O. (1996). The Mechanisms of Governance. New York: Oxford University Press.
Windeler, A. and Sydow, J. (2001). ‘Project networks and changing industry practice – collaborative content
production in the German television industry’. Organization Studies, 22, 1035–60.
Wright, M. and Lockett, A. (2003). ‘The structure and management of alliances: syndication in the venture
capital industry’. Journal of Management Studies, 40, 2073–102.
Yin, R. K. (2003). Case Study Research, 3rd edition. Thousand Oaks, CA: Sage.
Zaheer, A. and Venkatraman, N. (1995). ‘Relational governance as an interorganizational strategy: an
empirical test of the role of trust in economic exchange’. Strategic Management Journal, 16, 373–92.
Zajac, E. J. and Olsen, C. P. (1993). ‘From transaction cost to transaction value analysis: implications for the
study of interorganizational strategies’. Journal of Management Studies, 30, 131–45.

© Blackwell Publishing Ltd 2009