Journal of Management Studies 45:4 June 2008
0022-2380
Trustworthiness, Risk, and the Transfer of Tacit and
Explicit Knowledge Between Alliance Partners
Manuel Becerra, Randi Lunnan and Lars Huemer
Instituto de Empresa, Madrid; Norwegian School of Management; Norwegian School of Management
The transfer of knowledge in alliances entails risk to partners, whose willingness
to accept it presumably relies on the trustworthiness that they perceive in their partners. We
investigate the extent to which the perceptions of trustworthiness and the willingness to take
risk determine the transfer of knowledge between alliance partners and their ultimate impact
on alliance success. The results show that the transfer of tacit versus explicit knowledge have
very different trust and risk profiles. Whereas explicit knowledge is closely associated with the
firm’s willingness to take risk, tacit knowledge is intimately related to high trustworthiness. The
results support the important role of trust and the transfer of tacit knowledge on the success of
learning alliances.
INTRODUCTION
Learning from a partner has been recognized as one of the key reasons why alliances may
be formed (Doz, 1996; Dussauge et al., 2000; Gulati, 1999; Hamel, 1991; Inkpen, 1998).
For instance, Teece and Pisano (1994) argue that alliances can be regarded as mechanisms for firms to learn from each other, which helps them recognize dysfunctional
routines and blindspots. Despite the benefits of learning, knowledge transfers entail some
level of risk because the partners in an alliance cannot fully control the use that the other
partner may make of the transferred knowledge. Allies may actually become strong
competitors as a result of the newly developed competencies through the alliance (Inkpen
and Beamish, 1997; Hladik, 1988; Khanna et al., 1998). Referring to this unique aspect
of learning alliances, Khanna et al. (1998) warn that any alliance will contain both
private and common scopes, which create tension in the trade-off between sharing and
protecting firm knowledge while maintaining a careful balance between competition and
cooperation (Parkhe, 1993). Thus, the transfer of knowledge between alliance partners
may be beneficial for them, but it also puts them in a situation of potential vulnerability
to the actions of the other partner.
Address for reprints: Manuel Becerra, Accenture Chair in Strategic Management, Instituto de Empresa, Maria
de Molina 12, 28006 Madrid, Spain (Manuel.Becerra@ie.edu).
© Blackwell Publishing Ltd 2008. Published by Blackwell Publishing, 9600 Garsington Road, Oxford, OX4 2DQ, UK
and 350 Main Street, Malden, MA 02148, USA.
692
M. Becerra et al.
In these circumstances of potential vulnerability in learning alliances, the perceptions
of trustworthiness of the other side become essential for the partners to be willing to take
risks (Mayer et al., 1995) and to engage in a possible exchange of knowledge with their
allies. For this knowledge transfer to happen, an adequate level of trust in their partner
seems necessary. However, we still do not know much about how the perceptions of
trustworthiness of the other party affect the transfer of knowledge between alliance
partners, the risks that they generate, and their ultimate impact on alliance success
through the facilitation of knowledge transfers.
Our study builds on the existing literature on knowledge exchange in alliances
(Inkpen and Dinur, 1998; Simonin, 2004) as well as research on trust (Mayer et al.,
1995; Rousseau et al., 1998). More specifically, we will look into the differences
between tacit versus explicit knowledge transfers with regard to their risk profile and
their associated requirements for trustworthiness, as well as their effect on alliance
success. Thus, our paper investigates the following two research questions: To what
extent do tacit versus explicit knowledge exchanges in alliances differ with regard to
the perceptions of trustworthiness between partners and their actual willingness to take
risks? How do trust, risk, and tacit vs. explicit knowledge transfers contribute to the
success of the alliance?
The categorization of knowledge into tacit or explicit has become a cornerstone in the
literature on learning and knowledge management. Our research questions are aimed at
pushing forward our understanding of how these two types of knowledge flows between
alliance partners differ with respect to their associated levels of trust and risk. We will also
study its implications for the performance of the alliance. This analysis is especially
relevant for the successful management of knowledge transfer in alliances, in particular
with respect to the levels of trust and risk that are required for different types of
knowledge to be actually transferred between alliance partners.
After formulating a set of hypotheses, we will test them in a sample of 65 alliances in
Norway for which knowledge exchange was particularly important. The results confirm
that indeed the transfer of tacit versus explicit knowledge has very different profiles for
risk, overall trustworthiness, and its different dimensions (Mayer et al., 1995). While tacit
knowledge transfer is intimately related to trustworthiness, the transfer of explicit knowledge requires a greater willingness to accept risks. The results also show that the transfer
of tacit knowledge mediates the positive effect of trust on the success of the alliance; in
contrast, the explicit knowledge transfers and the willingness to accept risks associated
with them have a negligible impact on alliance success.
TRUST AND THE TRANSFER OF TACIT AND EXPLICIT
KNOWLEDGE
Alliances are often defined as a durable, voluntary business arrangement between two
firms involving exchange, sharing, or co-development of products, technologies, and
services (Gulati, 1998). They may be based on inter-firm relations regulated by a contract
or involve some form of shared ownership, like a joint venture. The management
literature views alliances as important vehicles for learning and knowledge acquisition
and it regards this type of learning as an alternative to internal knowledge generation
© Blackwell Publishing Ltd 2008
Transfer of Knowledge
693
within a company (Almeida et al., 2002). Learning in an alliance involves accessing and
internalizing critical information, capabilities, or skills from a partner (Kale et al., 2000)
and it represents a way to short circuit the learning process compared to other alternatives for developing new competencies (Hamel, 1991).
The literature on knowledge management often makes a critical distinction between
tacit and explicit knowledge, though other dimensions of knowledge have also been
studied, such as complexity, ambiguity, and specificity (Simonin, 1999, 2004; Zander
and Kogut, 1995). Tacit knowledge is defined as knowledge that is non-verbalized,
intuitive, and unarticulated (Polanyi, 1967) and it has often been regarded as an important basis for competitive advantage from a resource-based view of the firm (Barney,
1991; Grant, 1996). In contrast, explicit knowledge can be coded and articulated, which
makes it easier to transfer (Simonin, 1999).
Both tacit and explicit knowledge transfers have been analysed in the context of
alliances. For instance, Inkpen and Dinur (1998) studied the exchange of tacit and
explicit knowledge within an American–Japanese joint venture. Tacit knowledge was
associated with culture and philosophy about business rather than knowledge covering
specific rules or guidelines, whereas explicit knowledge transfer referred to product
designs and specific manufacturing process exchanges between the partners.
Notwithstanding this important distinction between tacit and explicit knowledge
transfer, in most cases both types of knowledge exchanges are likely to occur in alliances concurrently. Lei et al. (2001, p. 217) argue that ‘in practice, the distinction
between explicit and tacit knowledge is not always very clear, since a large component
of both organizational and technical knowledge are interwoven with the firm’s culture,
product development routines, and human resource practices’. Thus, explicit and tacit
knowledge are sometimes difficult to disentangle, though their transfer between
partners usually takes place through different means and processes. While explicit
knowledge is codified and transferred verbally, tacit knowledge is more embedded in
social relations and transferred primarily through direct contact and observation of
behaviour.
For the successful exchange of knowledge to take place in cooperative relationships, it
is presumably necessary to have a trusting relationship (Ring and Van de Ven, 1994).
Highly effective collaborative relationships have been described as possessing relational
capital (Kale et al., 2000), relational embeddedness (Dhanaraj et al., 2004; Kogut and
Zander, 1992), and transparency (Hamel, 1991), which are typically associated with high
trust between alliance partners (Krishnan et al., 2006; Lane et al., 2001). All of these
partially overlapping characteristics of relationships, though they are not equivalent, are
generally considered to have a very positive effect on the relationship, particularly in
facilitating the transfer of knowledge between alliance partners.
However, the flip side of trusting relationships and the exchange of knowledge within
them is vulnerability and the potential abuse from a partner. When a firm is able to build
a close and trusting relationship with another company, the risk potentially increases that
its own core distinctive competencies may be transferred to its partner with potentially
harmful consequences (Inkpen and Beamish, 1997; Luo, 2002). An attempt to learn
something from a partner may then end up with having trained a new competitor
(Hladik, 1988; Khanna et al., 1998). Within the literature on alliance management, this
© Blackwell Publishing Ltd 2008
694
M. Becerra et al.
potential vulnerability has been acknowledged, for instance, by Kale et al. (2000) as
protection of proprietary assets and Simonin’s (2004) concept of partner protectiveness,
when the partner may not behave in a desired manner (Das and Teng, 2001). Thus, we
may expect some level of trust and risk to be both present in learning alliances.
Earlier research has explored the positive role of trust in knowledge transfer in
alliances (Badaracco, 1991; Inkpen, 1998; Kale et al., 2000; Lane et al., 2001) and the
link between risk and knowledge transfer (Kale et al. 2000; Simonin, 2004), but researchers have not yet looked into the different characteristics of tacit and explicit knowledge
with regard to their implications for trust and risk. To the extent that explicit and tacit
knowledge are inherently different and they are transferred through different verbal and
social processes, they may likely be associated with different levels of trust and risk. We
still do not know how trust, risk, and different types of knowledge transfers are related to
each other and how they affect alliance performance. In the next section we will try to fill
this gap and formulate specific hypotheses about the effects of trust and risk in the
transfer of tacit and explicit knowledge between alliance partners, as well as the final
impact of trust, risk, and knowledge transfers on alliance performance.
HYPOTHESES
First, we will analyse the basic link between a partner’s trustworthiness and the willingness to accept vulnerability by the other partner drawing from the model suggested by
Mayer et al. (1995). These authors define trust as ‘the willingness of a party to be
vulnerable to the actions of another party based on the expectation that the other will
perform a particular action important to the trustor, irrespective of the ability to monitor
and control that other party’ (Mayer et al., 1995, p. 712). This conceptualization of trust
has been adopted by Rousseau et al. (1998), who also make a similar distinction between
the willingness to accept vulnerability, i.e. trust, and the evaluation of the partner’s
intentions or behaviour (i.e. perceptions of trustworthiness) on which trust is based upon.
Mayer et al. (1995) argued that the willingness to accept vulnerability to the actions of
another party results from the evaluations that the ‘trustor’ makes about the trustworthiness of the ‘trustee’ with respect to several specific dimensions. Building on the
trust literature, they identify three main dimensions of trustworthiness: integrity (i.e. the
overall moral character and ethical behaviour of the partner or trustee), benevolence (i.e.
the positive vs. egocentric orientation of the trustee in dealing specifically with the
trustor), and ability (i.e. general competence and expertise of the trustee). Other trust
researchers have identified different elements of trustworthiness; Butler (1991), for
example, finds ten dimensions including also consistency and promise fulfilment. Similarly, Cummings and Bromiley (1996) developed a scale based on three dimensions of
trustworthy behaviour: keeping commitments, negotiating honestly, and avoiding taking
excessive advantage of partner organizations. For these authors, perceived trustworthiness comprises the cognitive, intentional, and emotional evaluations of whether the other
party can and will fulfil expectations. In this paper we use Mayer et al.’s (1995) conceptualization of trustworthiness and its three dimensions of integrity, benevolence, and
ability. We do not make specific hypotheses for each dimension, because we would
expect the same effect for each dimension as for the overall scale of trustworthiness.
© Blackwell Publishing Ltd 2008
Transfer of Knowledge
695
However, we will also analyse empirically each dimension separately to explore the
extent to which the three dimensions of trustworthiness may not have the same effects.
Regardless of the differences in the conceptualization of trust among researchers and
the actual dimensions of trustworthiness that can be measured, there seems to be
substantial agreement that greater perceptions of trustworthiness should enhance the
willingness to accept vulnerability in dealing with another party (Rousseau et al., 1998).
This general link between trustworthiness and willingness to take risks should also be
observable in the context of knowledge exchange between alliance partners. In the alliance
context, relational risk is concerned with the probability and consequences of a partner
firm not committing itself to the alliance in the expected manner (Das and Teng, 2001). It
is for this type of relational risk, as opposed to environmental risk, that trust is particularly
important, and it has positive consequences (Krishnan et al., 2006). The risk that the
partner may behave opportunistically is an important threat for alliances, including the
partner’s potentially damaging use of the newly acquired knowledge, whose consequences
need to be taken into account by alliance partners (Heide and Miner, 1992).
Drawing from the trust literature and consistent with Mayer et al. (1995), we will argue
that only when the perceptions of partner trustworthiness are sufficiently high, will
partners be willing to accept the risks that may exist in collaborating with an alliance
partner. Only if the firm believes that its potential partner has sufficient ability, overall
integrity, and benevolent attitude towards the firm, would it be reasonable for the firm
to accept the vulnerability associated with the alliance. Otherwise if the firm considers
that the potential partner is not trustworthy, it may be a better option to avoid a risky
alliance and to look for another partner with greater trustworthiness. Consequently, we
expect a positive relation between trustworthiness and the willingness to take risk.
Hypothesis 1: Greater perception of a partner’s trustworthiness is associated with
greater willingness to take risks in the alliance.
Earlier research has shown that social relations characterized by high trust facilitate
the transfer of knowledge in organizational settings, particularly for the transfer of tacit
knowledge (Hansen, 1999). From a social learning theory perspective, intensive social
interaction facilitates the development and transfer of tacit knowledge (Brown and
Duguid, 2001). While explicit knowledge can be effectively transferred through written
documents without the need for direct contact and attachment between partners, tacit
knowledge relies primarily on direct observation and a close relationship between individuals. In this type of close social relationship, trust is also likely to exist.
Inkpen and Dinur (1998) claim that the transfer of tacit knowledge is more likely to
occur among individuals in a high trust context. They propose that the transfer of tacit
knowledge will occur through individuals rather than collectives; furthermore, initiatives of tacit knowledge transfer from the collective often fail. It is in the personal
networks between trusted individuals that tacit knowledge is gained and effectively
transferred. In contrast, trust is not really important for the exchange of explicit knowledge, which can be codified and therefore transferred independently of specific individuals. The transfer of explicit knowledge is thus less dependent on close relationships
based on trust.
© Blackwell Publishing Ltd 2008
696
M. Becerra et al.
There is already abundant empirical evidence in the alliances literature that is consistent with the idea that the transfer of tacit knowledge is more likely to take place in
a context of high trust (Inkpen, 1998; Lubatkin et al., 2001). While explicit knowledge
transfer may also be enhanced by the presence of trust, it does not require a high trust
relationship to be effectively transferred. In their study of transfer of knowledge from
a foreign partner to an international joint venture, Dhanaraj et al. (2004) found
that relational embeddedness, which is usually related to high trust, had a much
stronger influence on the transfer of tacit than explicit knowledge. They showed that
tacit knowledge is more closely connected to trust because it requires more intensity
in the transfer process than explicit knowledge, which does not need the same
socialization.
In summary, we will argue that high trustworthiness is more critical when knowledge
is tacit, and consequently transferable only through direct contact, than when knowledge is explicit. In the case of explicit knowledge, it can be effectively transferred
independently of the individuals involved in the exchange and the level of trust between
them.
Hypothesis 2: The transfer of tacit knowledge is more strongly associated with perceptions of a partner’s trustworthiness than the transfer of explicit knowledge between
alliance partners.
As we have claimed above, because of its codifiable nature, explicit knowledge needs
less effort in communication, teaching, and interactive processes to be successfully
transferred. This codifiability makes it more transparent, less sticky, and essentially easier
to transfer at greater speed (Szulanski and Capetta, 2003). In contrast, tacit knowledge is
much more difficult to identify, evaluate, and absorb since the transfer process is plagued
with inherent complexity and possible misunderstandings between the parties involved in
the process.
The lower ambiguity and specificity of explicit knowledge should make it more risky
to transfer than tacit knowledge, which has inherent difficulties for replication. The more
tacit the knowledge is, the harder it is for competitors to assess the true value of this
knowledge because it is more context-specific (Chakravarthy et al., 2003; Peteraf, 1993).
Therefore, we could expect firms to heavily guard all their explicit knowledge and to
consider very risky any possible transfer of such type of knowledge, while they may see
less risk in showing their more highly-specific tacit knowledge.
Consequently, it seems reasonable to believe that it is far more dangerous for a firm to
contribute explicit knowledge to an alliance, since this type of knowledge can quickly
spread with great ease of replicability. Explicit knowledge has greater probability to leak
outside the alliance partners and be perfectly absorbed and utilized by third parties.
Sharing explicit knowledge may be considered to place partners at greater risk than the
transfer of tacit knowledge, which can only spread slowly through close interaction and
is likely to require highly-specific assets and complementary skills. Thus, we will argue
that the transfer of explicit knowledge may be perceived as inherently riskier by alliance
partners and, as a result, sharing explicit knowledge should require greater willingness to
accept risk than the transfer of tacit knowledge.
© Blackwell Publishing Ltd 2008
Transfer of Knowledge
697
Hypothesis 3: The transfer of explicit knowledge requires greater willingness to accept
risk than the transfer of tacit knowledge between alliance partners.
The possible benefits of knowledge transfers in alliances are well established in the
literature (Dussauge et al., 2000; Inkpen, 1998). For instance, in one of their earlier
studies of Hungarian joint ventures, Lyles and Salk (1996) showed that knowledge
acquisition was positively related to performance. Thus, we may expect a positive
relationship between knowledge acquisition and alliance success, at least for alliances
whose major goal is learning.
However, there is some controversy about the relative importance of the transfer of
tacit versus explicit knowledge for alliance success. Given the presumed greater relevance
of tacit knowledge for competitive advantage in the resource-based view (Barney, 1991;
Grant, 1996), it may be argued that the transfer of tacit knowledge should be more
critical than explicit knowledge for the performance of an alliance. However, Dhanaraj
et al. (2004) found a positive relation between explicit knowledge transfer and performance and failed to find a direct relationship between tacit knowledge transfer and
performance. They explained this somewhat surprising finding by arguing that it may be
due to explicit knowledge’s relative low cost of transfer, clarity, and actionability of
routines. It is possible that the transfer of explicit knowledge may be easier to detect and
evaluate and therefore deemed more important, though its possession could have a
smaller effect on the long-term success of each firm.
Lane et al. (2001) also studied the relationship among trust, learning, and performance. They found that trust is very limitedly connected to learning, but strongly related
to performance. Furthermore, despite the high correlation between tacit and explicit
knowledge, they found a negative relationship between tacit learning and performance,
but a positive correlation between explicit learning and performance. Based on these
studies, we may conclude that the empirical evidence shows that the transfer of tacit and
explicit knowledge transfer takes place through different processes and may have different effects on performance.
We believe that different alliances created primarily for learning purposes may have
different goals about the type of knowledge that they seek. In general, both tacit and
explicit knowledge transfers may be presumed to have a positive effect on performance, as knowledge acquisition is a critical goal in learning alliances. However, either
tacit or explicit knowledge transfer may be more or less closely associated with alliance
performance, depending on the goals of specific alliances. For alliances created primarily for the sake of sharing information, explicit knowledge and willingness to take
risk should be more closely associated with the success of the alliance. In contrast, trust
and tacit knowledge are likely to be more relevant for alliances created for more
complex mutual learning and development of new competencies, which require close
social interaction.
In both cases, the transfer of knowledge in learning alliances has an important
mediating role between relational context and alliance success. On the one hand, trust
needs to exist for the transfer of tacit knowledge to take place, which ultimately should
be positively associated with alliance success for those partners that seek this type of
knowledge. Referring specifically to knowledge transfer in alliances, Kale et al. (2000,
© Blackwell Publishing Ltd 2008
698
M. Becerra et al.
p. 217) showed that ‘relational capital based on mutual trust and interaction at the
individual level between alliance partners creates a basis for learning and know-how
transfer across the exchange interface’. Thus, trust acts as a behavioural lubricant and
enables partners to be more flexible, open, and tolerant in the cooperation process
(Madhok, 1995), which should result in better performance.
On the other hand, the transfer of explicit knowledge should play an analogous
mediating role in the effect of willingness to take risks on alliance success. As discussed
above, the empirical literature shows that the transfer of explicit knowledge has a
positive effect on alliance performance (Dhanaraj et al., 2004; Lane et al., 2001).
However, without the willingness to take risks, the transfer of explicit knowledge would
be less likely to occur. Therefore, we can formulate a different process for the effective
transfer of explicit knowledge. In this case, greater willing to take risks facilitates the
transfer of explicit knowledge, which in turn may result in better performance for
the alliance.
Hypothesis 4a: The transfer of tacit knowledge between alliance partners mediates the
effect of partner trustworthiness on the success of the alliance.
Hypothesis 4b: The transfer of explicit knowledge between alliance partners mediates
the effect of the willingness to take risks on the success of the alliance.
METHODOLOGY
To test these hypotheses about trust, risk, and knowledge transfer between alliance
partners, we surveyed 155 firms from a variety of industries in Norway, primarily
printing, construction, fishing, and the public sector. These firms had voluntarily participated in a Research Project on Networks and Organizations funded by the Norwegian Research Council (NRC), for which improving information technology and sharing
knowledge among participants were some of the key project goals.
First, we telephoned each company and talked directly to the CEO in almost all cases.
Getting access to the CEO was not difficult because of the small size of the companies
and the firm’s involvement in a larger government-financed project. In a few cases, the
CEO referred us to a different person who was formally in charge of managing the firm’s
alliances. We then mailed the questionnaire to the CEO or to this referred person that
we had talked to. Two weeks later, we made a reminder phone call to follow up on the
survey. Though we relied on a single informant per firm, we believe that the questions
that we wanted to address about trust, risk, and the transfer of knowledge with alliance
partners should be answered by the person that constitutes the actual link with the
external partner, who was the CEO in most cases.
We sent out a questionnaire to the 155 firms involved in this project and received
65 completed and usable questionnaires. Though they were not formally required to
participate in our study, the official support that our project obtained from the NRC
and the direct contact with the CEO probably persuaded many of them to take part
in our study, which resulted in a relatively high response rate of 42 per cent. From the
firms in the sample, 70 per cent of the firms had fewer than 50 employees, only 3 per
© Blackwell Publishing Ltd 2008
Transfer of Knowledge
699
cent had more than 1000 employees, and 80 per cent of the alliances reported by the
firms had existed for less than five years. Each company was asked a set of questions
about the closest partners (up to five alliances) in the alliances in which they participated. To avoid problems of autocorrelation of observations, we selected the first relationship they identified, except for the few situations where some firms described an
alliance to another firm in the sample. In those rare cases, we selected the second tie
that they identified.
We followed the recommendations from Churchill (1979) to design and pretest the
survey. Eleven pilot interviews were conducted early on to check the relevance of our
topic and develop some preliminary ideas about the research questions and the variables
to include in the study. Then, we developed our measurement instrument and pretested
the questionnaire on two colleagues and two CEOs from the sample, who suggested
some minor changes to the questionnaire. All independent and dependent variables,
except the two control variables (firm size and alliance years), were measured on multiitem scales. The items based on English studies were translated to Norwegian and
then translated back to English, so that the non-Norwegian researchers could check
for content changes. All items were measured on a seven-item Likert scale (strongly
disagree–strongly agree). The reliability of all scales was always above the threshold of
0.70 recommended by Nunnally (1978). The Appendix includes the items for all of the
scales. The main variables in the study are further discussed below.
Transfer of Tacit and Explicit Knowledge
Because of the lack of a widely accepted instrument to measure tacit and explicit
knowledge transfers, we developed several items based on earlier studies and our pilot
interviews. From the initial set of items, we selected six items, three for tacit and three for
explicit knowledge transfer. The items for explicit knowledge transfer deal with the
exchange of information and knowledge about products, industry trends, and key industry participants (i.e. customers, suppliers, and competitors), which the partners must
presumably verbalize before it can be transferred. On the other hand, the tacit knowledge items are concerned with shared activities, observation of behaviour, and direct
contact, which are associated with more complex ways to interact and acquire tacit
knowledge from the partner.
Exploratory factor analysis of the six items revealed two clear factors with eigenvalues greater than one (2.31 and 1.74), which jointly account for 67 per cent of the
total variance. All items had at least 0.65 loading coefficients and no cross-loadings
reached 0.40. However, one of the items for explicit knowledge (‘We exchange knowledge about customers, suppliers, and competitors with this company’) loaded on the
factor with the other three items for tacit knowledge. Apparently, the managers distinguish between the transfer of any kind of information (i.e. explicit knowledge about
products and industry trends) on the one side and the exchange of knowledge and direct
learning from the partner on the other side (i.e. tacit knowledge). Thus, we moved this
item to its appropriate scale. The reliability (Cronbach alpha) of the final two scales
was 0.76 for explicit knowledge (two items) and 0.77 for tacit knowledge transfer (four
items).
© Blackwell Publishing Ltd 2008
700
M. Becerra et al.
Willingness to Accept Risk
Three items were used to measure the extent to which the firm tried to minimize risks
when dealing with the alliance partner. It is important to note the distinction between the
willingness to assume risk and the actual risk-taking behaviour, which has been analysed
in the literature (Mayer et al., 1995; Sitkin and Pablo, 1992). These items do not measure
the actual assumption of risk, which in our study is reflected in the transfer of knowledge.
Instead, this variable captures the firm’s general intentions to minimize risk, defend
themselves, and avoid sharing risky projects with the alliance partner. The scale obtained
a reliability of 0.75. Because the arguments in the hypotheses section were developed
following the literature on the willingness to accept risk (rather than the minimization of
risk, which the items measure), we reversed the initial coding of the items to facilitate the
interpretation of this variable.
Evaluations of Trustworthiness
Many different dimensions of trustworthiness have been proposed in the past. The three
dimensions identified by Mayer et al. (1995) from their analysis of the trust literature
parsimoniously capture the key aspects of the expectations that one person may have
about others’ intentions and behaviour. As we mentioned earlier in the paper, these three
dimensions of trustworthiness are integrity, benevolence, and ability. The items to
measure each dimension were taken from Mayer and Davis (1999) and adapted to the
alliance context of our study. The person (i.e. the trustor) that managed the relationship
with alliance partners was asked his/her opinion about its alliance partner (i.e. the
trustee), with regard to the partner’s integrity (i.e. the overall moral character and ethical
behaviour of the partner), benevolence (i.e. the positive vs. egocentric orientation of the
partner in dealing specifically with his/her firm), and ability (i.e. general competence and
expertise of the partner).
The 12 items were aggregated to compute an overall measure of trustworthiness, which
obtained a reliability of 0.91. To further explore the effect of the three dimensions of
trustworthiness, the analysis was also conducted for each dimension separately, so that we
could see how each dimension uniquely affects the willingness to take risks and the transfer
of tacit and explicit knowledge. Each sub-scale obtained a Cronbach alpha of at least 0.80.
Alliance Success
Three items formed a scale that subjectively measured the degree to which the alliance
could be considered successful by the key informant. As alliances are entangled in other
firm activities, alliance performance can be difficult to discern from the performance of
the firm, and they are often considered to be successful in the long term (Anderson,
1990); consequently objective, short-term measures are not useful. Evaluations based on
multiple indicators, including an assessment of financial performance, are found to be
reliable proxies for alliance performance (Anderson and Narus, 1990). Thus, one item
assesses the net results of the alliance in the previous year, whereas the other two items
provide a subjective evaluation of the alliance success and its contribution to growth. The
scale achieved a reliability of 0.73.
© Blackwell Publishing Ltd 2008
Transfer of Knowledge
701
Control Variables
The empirical analysis included two control variables: firm size (in number of employees), and alliance years (i.e. years since the alliance was created). Firm size may
indicate the resource scope and slack of an organization that may influence the ability to
access resources. Young alliances are generally perceived as more risky ( Yan and Gray,
1994). Because there was substantial skewness in the distributions of the size of the firms
in the study and also in the number of years since they alliances were created, we used
their logarithm values in the analysis.
Construct Validity and Common Method Variance
All the instruments used in the survey revealed a sufficient degree of convergent validity,
as reflected by Cronbach alphas reported above. However, using perceptual data coming
from a survey raises the threat of common method variance. To see the extent to which
this could be a problem, we tested the discriminant validity of the scales through
exploratory and confirmatory factor analysis (Podsakoff et al., 2003).
First, we did a Harman’s one factor test (Podsakoff et al., 2003), which resulted in a
very poor fit (chi-squared = 541 (p < 0.001); CFI = 0.60). Consequently, we could discard
the idea that all variables, including trustworthiness and tacit knowledge transfer, were
just one evaluative image in the minds of the respondents. More than one construct
clearly emerged from the data.
Second, exploratory factor analysis shows that there are indeed five factors with
eigenvalues greater than 1 in the full set of 24 items that cover the five variables (trust,
risk, tacit knowledge, explicit knowledge, and alliance success). The items primarily load
on the expected factors after varimax rotation was conducted (principal component
analysis that explains 67 per cent of the total variance).
In conclusion, even though we are measuring perceptions of individuals with one
survey, there are several distinct and interconnected variables in the minds of these
managers. The relationships among these cognitive constructs constitute the focus of his
study, including partner trustworthiness, willingness to take risk, tacit and explicit knowledge transfers, and alliance success. Given the goals of this paper and the results from the
convergent and discriminant validity of the variables, we concluded that the potential
problems of common method variance were not significant.
RESULTS
Hierarchical regression analysis was used to test the hypotheses. Descriptive statistics for
all the variables and their zero-order correlations are shown in Table I. Several results
from the correlation table should be noted. There is significant correlation among the
variables that could be expected, for instance, between trustworthiness and willingness to
take risks, between trustworthiness and tacit knowledge transfer, between willingness to
take risk and explicit knowledge transfer, between explicit and tacit knowledge, and also
between alliance years and trustworthiness.
Only one set of correlations seems to be sufficiently high to suggest problems of
multicollinearity: the correlations among the three dimensions of trustworthiness (up to
© Blackwell Publishing Ltd 2008
702
© Blackwell Publishing Ltd 2008
Table I. Descriptive statistics and correlations table
Mean
Explicit knowledge
Tacit knowledge
Willingness to take risks
Alliance success
Trustworthiness
Integrity
Benevolence
Ability
Firm size
Alliance years
4.44
4.93
4.15
4.75
5.16
5.33
4.61
5.53
2.76
1.11
1.41
1.27
1.40
1.18
0.95
1.12
1.10
1.04
2.04
0.86
Notes: N = 65.
** Correlation is significant at the 0.01 level (2-tailed).
* Correlation is significant at the 0.05 level (2-tailed).
Pearson correlations
1
2
3
4
5
6
7
8
9
1
0.43**
0.57**
0.18
0.22
0.30*
0.05
0.27
-0.06
0.19
1
0.14
0.51**
0.56**
0.43**
0.43**
0.57**
0.03
0.22
1
-0.05
0.32*
0.41**
0.18
0.22
-0.10
0.15
1
0.48**
0.36**
0.43**
0.45**
-0.05
0.14
1
0.89**
0.84**
0.88
-0.05
0.34*
1
0.59**
0.71**
-0.10
0.39**
1
0.56**
-0.06
0.09
1
0.02
0.34**
1
0.18
M. Becerra et al.
1.
2.
3.
4.
5.
6.
7.
8.
6.
7.
Std. dev.
Transfer of Knowledge
703
Table II. Regression analysis of the willingness to take risks
Dependent variable: Willingness to take risks
Independent variables
Model 1
Model 2
Model 3
Alliance years
Firm size
Overall trustworthiness
Integrity
Benevolence
Ability
0.16
-0.14
0.04
-0.10
0.33*
-0.04
-0.05
R2
Change in R2
R2 change F test
0.56**
-0.08ns
-0.07ns
0.039
1.07
0.136
0.097
5.85*
0.213
0.174
3.69*
Notes: Standardized coefficients reported.
** Significant at the 0.01 level (2-tailed).
* Significant at the 0.05 level (2-tailed).
ns
This coefficient was not significant when the other two dimensions of trustworthiness were dropped from the analysis.
0.71) and also with its overall aggregate value (up to 0.89). We computed the variance
inflation factors for all the variables; all of them were well below the level of 10 used to
test for multicollinearity. However, when the three dimensions of trustworthiness were
included in the same step, the overall increase in R2 was usually highly significant, but
none of the three variables had a significant coefficient, which is clear evidence of some
multicollinearity among this set of variables. To deal with this problem, we did the
analysis for the overall measure of trustworthiness and for the three dimensions as a
group, but also each dimension of trustworthiness was included separately without the
other two and their significance is reported in the results.
Table II investigates how trustworthiness perceptions between alliance partners are
associated with the willingness to take risks within the alliance; more precisely, the effect
of the trustor’s perceptions of the trustee’s trustworthiness on the willingness of the trustor
to take risks in dealing with the trustee. First, the two control variables are included in
Model 1. In the next model, we can see that indeed greater perceptions of trustworthiness
are clearly associated with greater willingness to take risk, after controlling for alliance
years and firm size. The results are significant for the overall measure of trustworthiness
in Model 2 as well as for its different dimensions in Model 3. When the three dimensions
are included in the last step in a hierarchical regression analysis, the increase in R2 from
Model 1 to Model 3 is 0.174, which is a significant increase at the 5% confidence level.
Thus, there is empirical evidence in favour of Hypothesis 1.
It is interesting to see that integrity is the only dimension of trustworthiness with a
significant impact on the willingness to take risks. The other two dimensions, ability and
benevolence, are insignificant in this model, even when only one of them is left as the
independent variable. It seems that the moral character and integrity of the partner
© Blackwell Publishing Ltd 2008
704
M. Becerra et al.
Table III. Regression analyses of the transfer of tacit and explicit
knowledge
Dependent variable: Tacit knowledge transfer
Independent variables
Model 4
Alliance years
Firm size
Overall trustworthiness
Integrity
Benevolence
Ability
Willingness to take risks
0.21
0.00
R2
Change in R2
R2 change F test
0.044
1.20
Model 5
Model 6
Model 7
0.03
0.06
0.55**
0.01
0.02
0.01
0.02
-0.02s
0.14s
0.51**
-0.04s
0.14s
0.52**
0.05
0.355
0.311
7.85**
0.356
0.001
0.72
0.310
0.266
19.64**
Dependent variable: Explicit knowledge transfer
Independent variable
Model 8
Model 9
Model 10
Model 11
Alliance years
Firm size
Overall trustworthiness
Integrity
Benevolence
Ability
Willingness to take risks
0.21
-0.09
0.15
-0.07
0.17
0.02
-0.06
0.05
-0.03
0.33ns
-0.29ns
0.17ns
-0.01ns
-0.23ns
0.20ns
0.59**
0.145
0.101
1.91
0.422
0.277
23.09**
R2
Change in R2
R2 change F test
0.046
1.27
0.070
0.024
1.35
Notes: Standardized coefficients reported.
** Significant at the 0.01 level (2-tailed).
* Significant at the 0.05 level (2-tailed).
s
This coefficient was positive and significant when the other two dimensions of
trustworthiness were dropped from the analysis.
ns
This coefficient was not significant when the other two dimensions of trustworthiness were dropped from the analysis.
makes alliance members willing to accept risks within their relationship, rather than their
perceptions about the partner’s ability or benevolence towards them.
The analyses of the relative impact of trustworthiness perceptions and willingness to
take risks on the transfer of tacit and explicit knowledge are shown in Table III. With
regard to tacit knowledge, the control variables are included first in Model 4, then overall
trustworthiness is included in Model 5 while its different dimensions substitute for it in
Model 6, and finally willingness to take risks is added in Model 7. The results provide
strong evidence that the transfer of tacit knowledge is significantly associated with higher
© Blackwell Publishing Ltd 2008
Transfer of Knowledge
705
perceptions of trustworthiness (significant coefficient of 0.55 for overall trustworthiness
and a significant increase in R2 of 0.311 for its three dimensions), though not with the
willingness to take risk. These results support Hypothesis 2.
With respect to the three dimensions of trustworthiness, the partner’s ability is the
most important dimension of trustworthiness for the transfer of tacit knowledge, though
the other two dimensions would also be significant if they were the only dimension left in
the model. Thus, all the three dimensions of trustworthiness are associated with greater
transfer of tacit knowledge, though the association is stronger with the partner’s
perceived ability.
Exactly the opposite situation takes place in the transfer of explicit knowledge. As we
can see in Models 9, 10, and 11, overall trustworthiness and its dimensions (either
independently or as a group) are not associated with the transfer of explicit knowledge.
In this case, only the willingness to take risk determines the transfer of explicit knowledge,
which is the last variable introduced in Model 11 (significant coefficient of 0.59). Therefore, there is also support for Hypothesis 3.
Finally, the results provide mixed support for the last two hypotheses, as Table IV
shows. With regard to Hypothesis 4a, the regression analysis confirms that the transfer of
tacit knowledge partially mediates the positive effect of trust on alliance success, for both
the overall measure of trustworthiness and each of its different dimensions. Following the
Baron and Kenny (1986) procedure to test for mediation, overall trustworthiness had a
significant coefficient of 0.50 on alliance success in Model 12. Trustworthiness also had
a significant positive effect (0.55 coefficient) on the regression on tacit knowledge in
Model 5 reported earlier. When both trustworthiness and tacit knowledge are introduced
in Model 13, the highly significant coefficient for trustworthiness decreases from 0.50 to
an insignificant 0.28 coefficient, while the tacit knowledge variable has a highly significant coefficient of 0.39. Thus, in support of Hypothesis 4a, the tacit knowledge transfer
variable captures a significant part of the explanatory power that was previously attributable to overall trustworthiness, whose coefficient is significantly reduced when tacit
knowledge transfer is included in the analysis. Trust, however, does more for the success
of the alliance than just facilitating tacit knowledge transfers.
Models 17–19 also provide support for Hypothesis 4a, now based on the different
dimensions of trustworthiness. Each of the three dimensions is significant when introduced alone in Model 17, though they are not when introduced at the same time because
of the high correlation among them. When tacit knowledge is introduced in the next step
for each dimension separately, the coefficients for all of them are substantially reduced,
becoming insignificant for integrity and ability and reducing the coefficient for benevolence from 0.43 to 0.26, both significant (the results from these regressions are not shown,
but their significance is reported in Table IV). Again, tacit knowledge transfer partially
mediates the effect of the dimensions of trustworthiness on alliance.
In contrast, Hypothesis 4b is not supported. Neither willingness to take risk nor explicit
knowledge are significant in Models 14 and 15. The limited effect of these two variables
on alliance success is also shown in Models 16 and 19, where all the variables are
included. In comparison, overall trustworthiness and tacit knowledge transfer are positively associated with alliance success in Model 16, though only benevolence remains
significant in Model 19.
© Blackwell Publishing Ltd 2008
706
M. Becerra et al.
Table IV. Regression analysis of alliance success
Dependent variable: Alliance success
Independent variables
Model 12
Model 13
Model 14
Model 15
Model 16
Firm size
Alliance years
Overall trustworthiness
Tacit knowledge transfer
Willingness to take risks
Explicit knowledge transfer
-0.04
-0.04
0.50**
-0.06
-0.05
0.28
0.39**
-0.10
0.15
-0.09
0.10
-0.08
-0.24
0.29
-0.07
-0.06
0.39*
0.31*
-0.30*
0.14
R2
Change in R2
R2 change F test
0.238
0.341
0.103
7.83**
5.30**
0.028
0.51
0.084
0.055
3.09+
0.395
5.23**
Dependent variable: Alliance success
Independent variable
Model 17
Model 18
Model 19
Firm size
Alliance years
Integrity
Benevolence
Ability
Tacit knowledge transfer
Willingness to take risks
Explicit knowledge transfer
-0.07
-0.02
-0.07s
0.28s
0.36s
-0.08
-0.02
-0.05ns
0.23s
0.15ns
0.37*
-0.08
-0.03
0.05ns
0.27s
0.13ns
0.30
-0.30
0.18
0.267
0.355
0.088
6.56*
0.404
0.049
1.90
R2
Change in R2
R2 change F test
3.56**
Notes: Standardized coefficients reported.
** Significant at the 0.01 level (2-tailed).
* Significant at the 0.05 level (2-tailed).
s
This coefficient was positive and significant when the other two dimensions of trustworthiness were dropped from the
analysis.
ns
This coefficient was not significant when the other two dimensions of trustworthiness were dropped from the
analysis.
It is interesting to note that the coefficient for willingness to take risks is actually
negative and significant (or marginally so) in the models with all the variables. It seems
that once the benefits of trustworthiness and the transfer of explicit and tacit knowledge
have been taking into account, more willingness to take risks is actually associated with
less successful alliances. Our interpretation of this result is that the willingness to take
risks is not inherently good or bad; in fact, it has an insignificant zero-order correlation
with alliance success. However, it may provide some benefits for the partners, which in
these alliances is primarily channelled through the transfer of explicit knowledge. Once
the possible benefits of explicit knowledge are controlled for, accepting greater risk may
turn into gullibility and have negative consequences.
© Blackwell Publishing Ltd 2008
Transfer of Knowledge
707
DISCUSSION
The results provide evidence that tacit and explicit knowledge are indeed very different
with regard to their relationship to trustworthiness, risk, and alliance success. The
transfer of explicit knowledge is closely associated with the willingness to take risks and
both of them have little impact on the evaluations of the success of the alliance. In
contrast, the transfer of tacit knowledge is highly related to the perceptions of the
partner’s trustworthiness and both have a significant effect on alliance success. These
findings have important implications for the management of knowledge in alliances and
also for the literature on trust that we would like to discuss below. We will also look at the
limitations of our study.
The Transfer and Protection of Knowledge in Alliances
In contrast to some of the existing empirical research (Dhanaraj et al., 2004), we have
found that the transfer of tacit knowledge is more closely associated with alliance
performance than explicit knowledge transfers, whose positive effect on the success of the
alliance is negligible. This is fully in line with the resource-based view of the firm and the
presumed critical importance of tacit knowledge for competitive advantage (Barney,
1991), but it has not received clear empirical support in the alliances literature. Though
we have not found performance consequences for the transfer of explicit knowledge in
alliances, we believe that this may not be generalizable to all types of alliances. Quite
possibly, the firms in our sample may have been primarily concerned with establishing
strong social links to the partners rather than just facilitating the transfer of information
data among firms. We believe that probably for this reason only the exchange of tacit
knowledge had an effect on alliance success, but not the exchange of explicit knowledge.
Our results also show that the transfer of explicit knowledge requires greater willingness to take risk than the transfer of tacit knowledge; therefore, it seems that explicit
knowledge is more carefully guarded by managers. This finding represents an interesting
paradox when it is connected with the results about the greater impact of tacit knowledge
transfer on alliance success. Tacit knowledge constitutes a source of competitive advantage for many firms and they frequently participate in alliances to exchange this type of
knowledge, but they seem to be more concerned with protecting their explicit knowledge
instead. Given the limited effect of explicit knowledge transfer on performance, we
should probe further into why managers protect it so much.
It is possible that managers may have a somewhat distorted perception about the
relative risks of tacit versus explicit knowledge transfers driven by their differences in
observability and replicability that we highlighted earlier. In this sense, because explicit
knowledge is immediately identifiable but tacit knowledge is harder to detect and to
absorb, they may protect themselves against those knowledge transfers whose risks are
readily observable and more easily avoided. In other words, transferring explicit
knowledge involves a ‘clear and present danger’ that needs to be carefully managed,
even though it may be relatively less important for the ultimate competitive advantage
of the firm. In any case, the high willingness to accept risk that explicit knowledge
transfers require, coupled with its lower effect on alliance success, deserves further
research.
© Blackwell Publishing Ltd 2008
708
M. Becerra et al.
Implications for Research on Trust
Mayer et al. (1995) claim that we need to distinguish among the antecedents of trust (i.e.
evaluations of the trustworthiness of the other party with regard to its integrity, ability,
and benevolence towards the trustor), trust itself (i.e. the trustor’s willingness to take risks
when dealing with the other party), and actual risk-taking actions (like the exchange of
knowledge between alliance partners). They argue that the need for trust only arises in
a risky situation, in which the perceived risk is compared to the level of trust. Based on
their model, the trustor will actually engage in risk-taking actions only ‘if the level of trust
surpasses the threshold of perceived risk’ (Mayer et al., 1995, p. 726). Though this model
seems theoretically reasonable and inspired our research, the role of trust in facilitating
risk-taking activities (such as the transfer of knowledge in alliances) needs to be qualified,
as the results show with respect to the different mediating roles of perceived trustworthiness and the willingness to take risk.
We did find that greater perceptions of a partner’s trustworthiness are positively
associated with the willingness to take risks, as Mayer et al. (1995) and Rousseau et al.
(1998) suggest. However, the relationship between trustworthiness and actual risk taking
(i.e. tacit and explicit knowledge transfers) is not mediated by the willingness to take risks,
as the results in Tables III and IV show. In other words, trust facilitates the transfer of
tacit knowledge regardless of the willingness to take risks of the managers involved; in
contrast, the transfer of explicit knowledge requires greater willingness to take risks, but
it is not significantly associated with trustworthiness. Though we should be cautious
about drawing causality inferences from cross-sectional research, trustworthiness and the
willingness to take risks, though positively correlated, do not seem to form a causal chain
that leads to the transfer of knowledge.
Contrary to Mayer et al. (1995), our findings suggest that risk does not necessarily need
to be present for trust to exist, to be meaningful, and to exert its lubricant effect within
relationships identified by the literature on trust and alliances. Thus, we may question
the definition of trust as ‘the willingness of a party to be vulnerable to the actions of
another party’ (Mayer et al., 1995, p. 712). In our opinion, saying ‘I believe that you are
trustworthy’ is equivalent to ‘I trust you’, regardless of whether there is more or less risk
involved in the relationship. In fact, this conceptualization of trust as the perception of
another party’s trustworthiness about specific dimensions has actually been used by
abundant research on trust (Butler, 1991; Cummings and Bromiley, 1996; JohnsonGeorge and Swap, 1982; Krishnan et al., 2006). The conceptualization in Mayer et al.
(1995) and Rousseau et al. (1998) of trust as willingness to take risks may place too much
emphasis on what can be regarded as just one consequence of trust (i.e. willingness to take
risk), which sometimes may not be relevant or characteristic of the relationship in which
trust is studied. However, willingness to take risk may be very important in other
circumstances.
Trust, defined as perceptions of high trustworthiness, is driven by variables like the
generalized propensity to trust of individuals (Rotter, 1971) and communication frequency (Becerra and Gupta, 2003), and it has many consequences, such as increasing the
willingness to take risks that Mayer et al. (1995) stressed. However, it has many other
consequences, like facilitating open communication (O’Reilly et al., 1987), reducing
© Blackwell Publishing Ltd 2008
Transfer of Knowledge
709
conflict (Zaheer et al., 1998), and enhancing relational opportunities (Huemer, 1998),
which are not necessarily related to willingness to take risks. This is also the case for the
transfer of tacit knowledge in our study, which is not associated with greater willingness
to take risks.
In addition to overall trustworthiness, we also explored its three main dimensions. As
we focused primarily on the comparison between explicit and tacit knowledge transfers,
we leave for trust researchers to probe further into the differences among the three
dimensions of trustworthiness. However, some of the results deserve to be highlighted.
First, it seems that the willingness to take risks in alliances is only associated with the
perceptions about the partner’s general integrity, but not with his/her ability or benevolence towards the trustor. In other words, risks (like those involved in the transfer of
explicit knowledge) are avoided with those partners regarded as having low integrity,
regardless of their expertise or their attitude towards the firm. Second, though the three
dimensions of trustworthiness are associated with greater transfer of tacit knowledge,
more tacit knowledge is exchanged with partners perceived as having greater ability.
This seems very reasonable because learning should preferably be done from knowledgeable experts, rather than well-intended partners. Finally, we have also detected that
benevolence, probably because it is usually associated with interest congruence, is significantly higher in those alliances with greater success. The results confirm that trustworthiness is indeed a multidimensional construct and each of them has unique effects on
relationships. Future research may study the interaction among these dimensions of
trustworthiness and the ultimate reason why they have these different effects.
Limitations
There are some limitations that should be identified and their potential impact assessed.
Most of them deal with the data used in our study, including a relatively small sample of
Norwegian alliances and the perceptual data coming from one side of the dyad. As we
explained in the Methodology section, the perceptual measurement of most of the
variables does not seem problematic to us due to their inherent perceptual nature. More
specifically, we measured the perceptions about the partner’s trustworthiness as well as
the trustor’s willingness to take risks from the perspective of the firm making such
evaluations through the person involved in managing the alliance relationship. Similarly,
this person evaluated the transfer between alliance partners of explicit and tacit knowledge, as well as the overall satisfaction with the results obtained from the alliance. The
construct validity analysis showed that these are distinct concepts. However, they are
perceptual variables and we should acknowledge that we are truly investigating how
these variables, most of them cognitive-emotional, relate to each other in the minds of the
people in charge of these alliances.
We should also note the limitations about the explicit knowledge transfer variable,
which relies on just two items after moving another one to the tacit knowledge transfer
scale. Though this variable achieved an acceptable level of validity (Cronbach alpha of
0.76) and in the exploratory analysis was clearly distinguishable from the tacit knowledge
scale, we wanted to explore how much of the paper’s results depended on this particular
variable. We did the entire analysis without the explicit knowledge transfer variable and
© Blackwell Publishing Ltd 2008
710
M. Becerra et al.
obtained very similar results for the transfer of tacit knowledge. Very briefly, Hypothesis
1 is still supported (the greater the trust, the greater the willingness to take risks), the part
of Hypothesis 2 that deals with tacit knowledge is still supported (trust and tacit knowledge transfer are positively associated), Hypothesis 3 also holds (willingness to take risks
is not associated with the transfer of tacit knowledge), and Hypothesis 4a still receives
support (transfer of tacit knowledge significantly mediates the positive effect of trust on
alliance). Thus, though the results from the transfer of explicit knowledge only build on
two items, the findings are quite robust, even when the variable for explicit knowledge
transfer is dropped from the analysis.
Overall, though we need to take into consideration these limitations, the results
provide empirical evidence that the transfer of tacit vs. explicit knowledge are quite
different in terms of their profiles regarding trust and risk. Both types of knowledge are
associated with different relational contexts, which have different effects on alliance
performance. These results have important implications for both research on trust and
the management of knowledge in alliances.
ACKNOWLEDGMENTS
We would like to thank Mark Kriger for comments received on an earlier version of this article, and the
Norwegian Research Council for financial support for this research.
APPENDIX: VARIABLES IN THE STUDY AND THEIR RELIABILITY
(CRONBACH ALPHA)
Explicit knowledge transfer (0.76)
a. We do not provide this company with information about our current products and
services (reverse coded).
b. We do not exchange ideas and information about industry trends with this
company (reverse coded).
Tacit knowledge transfer (0.77)
a. We share activities that provide learning with this company.
b. We regularly visit each other’s facilities and observe onsite how operations are
conducted.
c. This company and ours have learned much from the direct contact between our
two organizations.
d. We exchange knowledge about customers, suppliers, and competitors with this
company (moved from the explicit knowledge transfer scale).
Willingness to take risks (0.75)
a. We try to minimize risk when cooperating with this firm.
b. We try to defend ourselves against the influences of this firm.
c. We avoid sharing risky projects with this company.
© Blackwell Publishing Ltd 2008
Transfer of Knowledge
711
Alliance success (0.73)
a. Last year’s net results from our business with this partner were satisfactory.
b. Our cooperation with this partner must be characterized as a success.
c. Our cooperation with this partner has contributed to growth in our firm.
Trustworthiness (0.91) – scale adapted from Mayer and Davis (1999)
a. Integrity (0.83)
– This firm has a strong sense of justice.
– We never have to worry about whether this firm will keep to its word.
– The management team in this firm tries hard to be fair in dealings with others.
– Sound principles seem to guide the behaviour of this firm.
b. Benevolence (0.83)
– This firm really looks out for what is important for us.
– This firm is very concerned about our welfare.
– Our needs and desires are very important to the management in this firm.
– This firm will go out of its way to help us.
c. Competence (0.89)
– This firm is very capable of performing its job.
– We feel very confident about this firm’s skills.
– This firm is known to be successful at the things it tries to do.
– This firm has much knowledge about the work it does.
REFERENCES
Almeida, P., Song, J. and Grant, R. M. (2002). ‘Are firms superior to alliances and markets? An empirical
test of cross-border knowledge building’. Organization Science, 13, 147–61.
Anderson, E. (1990). ‘Two firms, one frontier: on assessing joint venture performance’. Sloan Management
Review, 31, 19–30.
Anderson, J. and Narus, J. (1990). ‘A model of the distributor firm and manufacturing firm working
partnerships’. Journal of Marketing, 54, 42–58.
Badaracco, J. L. (1991). The Knowledge Link: How Firms Compete Through Strategic Alliances. Boston, MA: Harvard
Business School Press.
Barney, J. B. (1991). ‘Firm resources and sustained competitive advantage’. Journal of Management, 17,
99–120.
Baron, R. M. and Kenny, D. A. (1986). ‘The moderator-mediator variable distinction in social psychological
research: conceptual, strategic, and statistical considerations’. Journal of Personality and Social Psychology,
51, 1171–82.
Becerra, M. and Gupta, A. K. (2003). ‘Perceived trustworthiness within the organization: the moderating
impact of communication frequency on trustor and trustee effects’. Organization Science, 14, 32–
44.
Brown, J. C. and Duguid, P. (2001). ‘Knowledge and organization: a social-practice perspective’. Organization
Science, 12, 198–213.
Butler, J. K. (1991). ‘Towards understanding and measuring conditions of trust: evolution of a condition to
trust inventory’. Journal of Management, 17, 643–63.
Chakravarthy, B., McEvily, S., Doz, Y. L. and Rau, D. (2003). ‘Knowledge management and competitive
advantage’. In Easterby-Smith, M. and Lyles, M. A. (Eds), Handbook of Organizational Learning and
Knowledge Management. Oxford: Blackwell, 305–23.
© Blackwell Publishing Ltd 2008
712
M. Becerra et al.
Churchill, G. A. (1979). ‘A paradigm for developing better measures of marketing constructs’. Journal of
Marketing Research, 16, 64–73.
Cummings, L. L. and Bromiley, P. (1996). ‘The organizational trust inventory (OTI): development and
validation’. In Kramer, R. M. and Tyler, T. R. (Eds), Trust in Organizations: Frontiers of Theory and Research.
Thousand Oaks, CA: Sage.
Das, T. K. and Teng, B. S. (2001). ‘A risk perception model of alliance structuring’. Journal of International
Management, 7, 1–29.
Dhanaraj, C., Lyles, M. A., Steensma, H. K. and Tihanyi, L. (2004). ‘Managing tacit and explicit knowledge
transfer in IJVs: the role of relational embeddedness and the impact of performance’. Journal of
International Business Studies, 35, 428–42.
Doz, Y. L. (1996). ‘The evolution of cooperation in strategic alliances: initial conditions or learning
processes?’. Strategic Management Journal, 17, Special Issue, 55–84.
Dussauge, P., Garrette, B. and Mitchell, W. (2000). ‘Learning from competing partners: outcomes and
durations of scale and link alliances in Europe, North America and Asia’. Strategic Management Journal,
21, 99–126.
Grant, R. M. (1996). ‘Prospering in dynamically competitive environments: organizational capability as
knowledge integration’. Organization Science, 7, 375–87.
Gulati, R. (1998). ‘Alliances and networks’. Strategic Management Journal, 19, 293–17.
Gulati, R. (1999). ‘Network location and learning: the influence of network resources and firm capabilities
on alliance formation’. Strategic Management Journal, 20, 397–420.
Hamel, G. (1991). ‘Competition for competence and interpartner learning within international strategic
alliances’. Strategic Management Journal, 12, 83–103.
Hansen, M. T. (1999). ‘The search-transfer problem: the role of weak ties in sharing knowledge across
organization subunits’. Administrative Science Quarterly, 44, 82–111.
Heide, J. B. and Miner, A. S. (1992). ‘The shadow of the future: effects of anticipated interaction
and frequency of contact on buyer-seller cooperation’. Academy of Management Journal, 35, 265–
91.
Hladik, K. J. (1988). ‘R&D and international joint ventures’. In Contractor, F. J. and Lorange, P. (Eds),
Cooperative Strategies in International Business. Lexington, MA: Lexington Books, 187–203.
Huemer, L. (1998). Trust in Business Relations: Economic Logic or Social Interaction? Boréa: Umeå.
Inkpen, A. C. (1998). ‘Learning and knowledge acquisition through international strategic alliances’. Academy
of Management Executive, 12, 69–80.
Inkpen, A. C. and Beamish, P. W. (1997). ‘Knowledge, bargaining power, and the instability of international
joint ventures’. Academy of Management Review, 22, 177–202.
Inkpen, A. and Dinur, A. (1998). ‘Knowledge management processes and international joint ventures’.
Organization Science, 9, 454–68.
Johnson-George, C. and Swap, W. C. (1982). ‘Measurement of specific interpersonal trust: construction and
validation of a scale to assess trust in a specific other’. Journal of Personality and Social Psychology, 43,
1306–17.
Kale, P., Singh, H. and Perlmutter, H. (2000). ‘Learning and protection of proprietary assets in alliances’.
Strategic Management Journal, 21, Special Issue, 217–38.
Khanna, T., Gulati, R. and Nohria, N. (1998). ‘The dynamics of learning alliances: competition, cooperation and relative scope’. Strategic Management Journal, 19, 193–210.
Kogut, B. and Zander, U. (1992). ‘Knowledge of the firm, combinative capabilities and the replication of
technology’. Organization Science, 3, 383–97.
Krishnan, R., Martin, X. and Noorderhaven, N. G. (2006). ‘When does trust matter to alliance performance?’. Academy of Management Journal, 49, 894–917.
Lane, P. J., Salk, J. E. and Lyles, M. A. (2001). ‘Absorptive capacity, learning, and performance in
international joint ventures’. Strategic Management Journal, 22, 1139–61.
Lei, D. J., Slocum, W. J. and Pitts, R. A. (2001). ‘Building cooperative advantage: managing strategic
alliances to promote organizational learning’. Journal of World Business, 32, 203–23.
Liebeskind, J. P., Oliver, A. L., Zucker, L. and Brewer, M. (1996). ‘Social networks, learning and
flexibility: sourcing scientific knowledge in new biotechnology firms’. Organization Science, 7, 428–
43.
Lubatkin, M., Florin, J. and Lane, P. (2001). ‘Learning together and apart: a model of reciprocal interfirm
learning’. Human Relations, 54, 1353–82.
Luo, Y. (2002). ‘Building trust in cross-cultural collaborations: towards a contingency perspective’. Journal of
Management, 28, 669–94.
© Blackwell Publishing Ltd 2008
Transfer of Knowledge
713
Lyles, M. A. and Salk, J. E. (1996). ‘Knowledge acquisition from foreign parents in international joint
ventures: an empirical examination of the Hungarian context’. Journal of International Business Studies, 27,
877–904.
Madhok, A. (1995). ‘Revisiting multinational firm’s tolerance for joint ventures: a trust-based approach’.
Journal of International Business Studies, 26, 117–37.
Mayer, R. C. and Davis, J. H. (1999). ‘The effect of the performance appraisal system on trust for
management. A field quasi-experiment’. Journal of Applied Psychology, 84, 123–37.
Mayer, R. C., Davis, J. H. and Schoorman, D. (1995). ‘An integrative model of organizational trust’. Academy
of Management Review, 20, 709–34.
Nunnally, J. C. (1978). Psychometric Theory, 2nd edition. New York: McGraw-Hill.
O’Reilly, C.A., Chatman, J. A. and Anderson, J. C. (1987). ‘Message flow and decision making’. In Jablin,
F. M., Putman, L. L., Roberts, K. H. and Porter, L. W. (Eds), Handbook of Organizational Communication:
An Interdisciplinary Perspective. Newbury Park, CA: Sage.
Parkhe, A. (1993). ‘Strategic alliance structuring: a game theoretic and transaction cost examination of
interfirm cooperation’. Academy of Management Journal, 36, 794–829.
Peteraf, M. (1993). ‘The cornerstones of competitive advantage: a resource based view’. Strategic Management
Journal, 14, 179–91.
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y. and Podsakoff, N. P. (2003). ‘Common method biases in
behavioral research: a critical review of the literature and recommended remedies’. Journal of Applied
Psychology, 88, 879–903.
Polanyi, M. (1967). The Tacit Dimension. Garden City, NY: Anchor.
Ring, P. S. and Van de Ven, A. (1994). ‘Development processes of cooperative interorganizational relationships’. Academy of Management Review, 19, 90–118.
Rotter, J. B. (1971). ‘Generalized expectancies for interpersonal trust’. American Psychologist, 26, 443–
52.
Rousseau, D. M., Sitkin, S., Burt, R. and Camerer, C. (1998). ‘Not so different after all: a cross discipline
view of trust’. Academy of Management Review, 23, 393–404.
Simonin, B. (1999). ‘Transfer of marketing know-how in international strategic alliances: an empirical
investigation of the role and antecedents of knowledge ambiguity’. Journal of International Business Studies,
30, 463–90.
Simonin, B. (2004). ‘An empirical investigation of the process of knowledge transfer in international strategic
alliances’. Journal of International Business Studies, 35, 407–27.
Sitkin, S. B. and Pablo, A. L. (1992). ‘Reconceptualizing the determinants of risk behavior’. Academy of
Management Review, 17, 9–38.
Szulanski, G. and Capetta, R. (2003). ‘Stickiness: conceptualizing, measuring, and predicting difficulties in
the transfer of knowledge within organizations’. In Easterby-Smith, M. and Lyles, M. A. (Eds), Handbook
of Organizational Learning and Knowledge Management. Oxford: Blackwell, 513–34.
Teece, D. and Pisano, G. (1994). ‘The dynamic capabilities of firms: an introduction’. Industrial and Corporate
Change, 3, 537–55.
Yan, A. and Gray, B. (1994). ‘Bargaining power, management control, and performance in United States–
China joint ventures: a comparative case study’. Academy of Management Journal, 37, 1478–517.
Zaheer, A., McEvily, B. and Perrone, V. (1998). ‘Does trust matter? Exploring the effects of interorganizational and interpersonal trust on performance’. Organization Science, 9, 123–42.
Zander, U. and Kogut, B. (1995). ‘Knowledge and the speed of the transfer and imitation of organizational
capabilities: an empirical test’. Organization Science, 6, 76–92.
© Blackwell Publishing Ltd 2008