The Effect of Formal Contracts on Trust

(1)

The Effect of Formal Contracts on Trust

Empirical Analysis on the Impact of Behavioral Uncertainty on the Relationship amongst Formal Contracts on Relational Trust and Calculative Trust in Inter-organizational

Relationships

by Esmee Wiendels

S3014193

e.d.wiendels@student.rug.nl

MSc. BA - Organizational & Management Control University of Groningen

Faculty of Economics and Business June 20, 2017

Supervisor: A. Rehman Abbasi Co-assessor: Dr. M. van der Steen

(2)

Abstract

This research examines the relation of formal contracts with relational and calculative trust in inter-organizational relationships (IORs), and how behavioral uncertainty moderates this relation. For this research, I develop and test hypotheses, using a survey dataset of 131 different IORs. A multiple regression analysis found mixed results for the hypothesized relationships of formal contracts with relational trust and calculative trust and the influence of behavioral uncertainty. Formal contracts placed in a social context are positively related to relational trust, only with the interaction effect of behavioral uncertainty which makes the relation stronger. For calculative trust, formal contracts are only positive related in a direct effect. These results provide new insights in the relation of formal contracts with different types of trust. This study extends the transaction cost economics and relational exchange theory by showing that for different kinds of relational development in the IOR, the context of contracts and influence of behavioral uncertainty is important. The findings extend the contract and trust literature by illustrating that formal contracts appear to be different in relation to trust and have different influences of uncertainty. Further the research findings are helpful for managers to make decisions concerning the use of formal contracts with the development trust.

(3)

3

1. Introduction

Over the past two decades, an expanding number of firms work together to gain an increase in efficiency and creating business value. Such collaboration of firms (partnership) is called an inter-organizational relationship (IOR) and which could be joint ventures, supplier relations or strategic alliances. The sense of confidence in an IOR comes from two distinct sources: control and trust (Das & Teng, 1998). Control is a regulatory process by which the IOR have to pursue desired objectives resulting in the power to influence or direct the partners’ behavior (Das & Teng, 1998; Leifer & Mills, 1996). Control is good when informed people are reasonably confident that no great, unpleasant surprises will occur (Merchant, 1984). Most IORs are based on contracts, which are part of formal control mechanism (Minnaar, Vosselman, Van Veen-Dirks, & Zahir-Ul-Hassan, 2016). The contracts, as a control mechanism, will help IOR partners to clarify their expectations with tasks and clear communication roles (Lu & Yan, 2015). Also contracts provide a measure of expectation, through evaluating whether actions deviate from expectations with relatively complete standards (Reuer & Ariño, 2002). According to Das and Teng (1998) control mechanisms will have an impact on trust, whereas trust at the same time will moderate the effect of control mechanisms in determining control. Trust entails the positive expectation about other people’s actions or behavior in risky exchange situations (Boon & Holmes, 1991; Das & Teng, 1998). Organizations have to rely on their partners’ performance and themselves and remain vulnerable to partners’ actions, whereby the relation becomes valuable. Within trust, different types exist, such as relational trust and calculative trust. Relational trust arises from different interactions in the relationship whereby there are strong beliefs about honesty, good faith, efforts of others, and goodwill (Bromiley & Harris, 2006; Poppo, Zhou, & Li, 2016; Ring, 1996; Zaheer & Harris, 2005). On the other hand calculative trust requires calculations of benefits and costs that will outweigh self-interested opportunistic actions (Poppo et al., 2016).

(5)

5

Despite formal contracts form a basis for trust, relational trust accounts more for a basis of a social component (Woolthuis et al. 2005; Zaheer & Venkatraman, 1995). This raises questions regarding the context of formal contracts in relation to different types of trust. For calculative trust and relational trust this context can be different for the relationship development of the IOR. The empirical work of Woolthuis et al. (2005) concludes that formal contracts within relationship development to trust should be placed in a social context rather than only a formal one. This social context will lead to joint problem solving, understanding the other, and participatory decision making, and will therefore increase relational trust. Besides, the relational exchange theory (RET) accounts explicit the social context of formal contracts for relational development of trust in IORs (Heide, 1994). Furthermore, existing literature is also inconclusive about the effect of behavioral uncertainty on the relation of contracts with relational trust and calculative trust respectively (Krishnan, Geyskens, & Steenkamp, 2016; Lewicki & Bunker, 1996; Poppo et al., 2016). Behavioral uncertainty has different impacts on both contracts and trust. In addition, the impact depends also on the level of uncertainty (Krishnan et al., 2016; Lewicki & Bunker, 1996). Accordingly, the question that remains unanswered in studies on the subject is “What is the relationship amongst formal

contracts on relational trust and calculative trust in inter-organizational relationships and how does behavioral uncertainty moderate this relationship?”

To address this research question, this study will focus on formal contracts in relation with relational trust and calculative trust. Extending existing literature, this research will explore the context of formal contracts at relational development in an IOR and with forward-looking process of goal achievement with punishments and benefits (Poppo et al. 2016). In addition this research looks at the effect of behavioral uncertainty on the relationship between formal contracts and relational and calculative trust.

(6)

6

with RET. Consistent with the assumption, formal contracts are positively associated with calculative trust. On the other hand, formal contracts in a social context are not related to relational trust. However, the interaction effect of behavioral uncertainty influences this relation in a positive way. Formal contracts in a social context become more effective on relational trust. For the hypothesis that the relation between formal contracts and calculative trust becomes stronger, when behavioral uncertainty is moderate is not supported. Overall, the results of this research demonstrate strengths associated with the relation of formal contracts with calculative trust and relational trust in IORs as well as important limitations.

The structure of this research is as follows: chapter 2 contains a literature review about formal contracts, relational and calculative trust, the TCE and the RET. It also contains the influence of behavioral uncertainty, hypothesis development, and conceptual model. Chapter 3 presents the research method which describes and justifies the chosen method, data collection, variable measurement included in the survey, and the data analysis which addresses validity and reliability concerns. Chapter 4 presents the results of this study and finally, chapter 5 contains the discussion, conclusion, limitations and directions for further research.

2. Literature Review

The literature review explains the most important concepts of this research. First, formal contracts, relational trust, and calculative trust are discussed. Thereafter, the transaction cost economics is reviewed as foundation for this research, followed by the relation between formal contracts and trust. This will be supported by the relational exchange theory, and will lead to the development of the first two hypotheses. Further the moderator behavioral uncertainty is explained followed by the hypotheses development of the effectiveness of formal contracts on relational and calculative trust under behavioral uncertainty. Finally, the conceptual model of this research is developed.

2.1 Formal Contracts

(7)

7

binding agreements between two parties with each promising to perform particular actions in the future” (Krishnan, Geyskens, & Steenkamp, 2016, p. 2522).

Contracts can be simple or complex and built a basis for shared common ground, clarify mutual expectations, and it enables goal compatibility (Das & Teng, 1998). From the perspective of the TCE, contracts serve as control tool to provide safeguards against opportunistic behavior. Due the legal forces, contracts are able to reduce the level of this opportunistic behavior (Krishnan et al., 2016). The process of control is to check (monitor) for a desired outcome through the exercise of authority or power to hold the partners to the agreements (Mellewigt et al., 2007; Zhang & Zhou, 2013). From the perspective of the resource-based view, contracts serve also as a coordinating tool (Mellewigt et al., 2007). Coordination is more to attain the desired outcome of the control process.

Firms in IORs use contracts, while at the same time they attempt to build trust (Poppo & Zenger, 2002; Wicks, Berman, & Jones, 1999). Divergence in existing literature about the relationship between contracts and trust makes it difficult to predict if formal contracts between firms will improve or inhibit the continued collaboration with trust (Malhotra & Lumineau, 2011). The relation between formal contracts and trust can be a conflict, because making a contract reduces the level of the existing trust in the IOR (Woolthuis et al., 2005). Simple contracts allow for more interpretation which could be a solution (Cao & Lumineau, 2015; Woolthuis et al., 2005). However, complex (formal) contracts are more legally binding, because more sections in a complex contract will cover more aspects of the relationship (Cao & Lumineau, 2015).

2.2 Relational Trust and Calculative Trust

Trust has been seen as the ‘glue’ that keeps firms in an IOR together (Mellewigt, 2007). While studies cover many different definitions of trust, they all agree that trust is the degree to which the firm accepts vulnerability based upon positive expectations of the intentions or behavior of others in a risky exchange situation (Das & Teng, 1998; Rousseau, Sitkin, Burt, & Camerer, 1998). Outcome of previous research shows that trust is used to gain positive expectations, reduce complexity of events and uncertainties, and results in greater relationship commitment and satisfaction (Balliet, Van Lange, 2013; Das & Teng, 1998).

(8)

8

this trust means that one firm can expect that the partner will behave as predicted and will act fairly and thereby foster the perceptions of stability (Zaheer, 1998). Also it limits performance losses that would occur because of opportunism and self-interest.

Trust can take various forms in firms, which can take place individually, but there is also a possibility that they can act together. The various forms of trust in this research are relational trust and calculative trust. Figure 1 models the two forms of trust with respect to the issue of bandwidth (Rousseau et al., 1998). Relational trust arises from social relationships where aligning core values mitigate the risk (Poppo et al., 2016). In this relationship, firms have strong beliefs about honesty, goodwill, and are faithful to each other. Calculative trust arises when a firm believes that the costs and benefits of the IOR will outweigh the self-interested, opportunistic actions (Poppo et al., 2016).

Figure 1: Trust development in Inter-organizational relationships (Rousseau et al. 1998)

“Relational trust arises when social relations evolve to a state in which each partner can expect to act according to the other’s preferences and priorities” (Poppo, Zhou, & Li, 2016, p. 726). It arises from repeated interactions. Partners have a high relational trust when they develop mutual understanding and shared identity in which they respond, feel, and think like the other (Lewicki & Bunker, 1996). A shared identity is the strongest form of social attachment.

(9)

9

negative intentions. Since relational trust provides a degree of stability, firms do not continually have to monitor the partner, continuously collect information and reassess the relationship (Poppo et al., 2016).

Calculative trust is based on rational choice, which is a characteristic of interactions based upon economic exchange (Rousseau et al., 1998). This type of trust informs expectations by forward-looking conditions: it requires calculations of benefits and costs and will depend on the relative values of cooperation and cheating (Poppo et al., 2016). When calculative trust is high, it will achieve functioning cooperation, performance goals, and will decrease behavioral uncertainty (Poppo et al., 2016). The forward-looking decision rule of calculative trust is a continual reassessment of relative payoffs to decide if firms pay to cooperate (Poppo et al., 2016). The underlying central logic is a rational assessment of rewards and punishments. When firms in an IOR believe that the rewards and punishments are well specified for a transaction, the benefits derived from excluding the specified task outweigh the costs from not doing so (Poppo et al., 2016; Williamson, 1993). If in an IOR a violation exists or one of the firms underperforms, the relationship can quickly drop down back to zero (Lewicki & Bunker, 1996). This will motivate each firm to fulfill its performance contribution.

2.3 Transaction Cost Economics

(10)

10

According to the TCE, trust reduces transaction costs by replacing contracts more with ‘handshakes’ (Poppo & Zenger, 2002). The principle by this theory is that simple (fewer complexes) contracts are more efficient than long, detailed contracts because there are fewer transaction costs. However, simple contracts can result in greater relational risks. A more complex contract is more attractive when partners in an IOR act opportunistic, because it prevents this opportunistic behavior (Mellewigt et al., 2007).

2.4 Effect of Formal Contract on Trust

According to Lu and Yan (2016), trust is determined by what firms put into in their relationship and what they receive from it. Their expectations are written in a contract which includes penalties for opportunism. The results discussed by Malhotra and Lumineau (2011), based on the TCE, show that firm decisions regarding control through contracts should embrace the effect of contract choices on subsequent trust and commitment. Poppo and Zenger (2002) suggest that a detailed formal contract provides guidelines for better cooperation between IORs and that relational trust proves if unexpected disturbance occurs in the contract details. Das and Teng (1998) support this with the argumentation that an increase in the level of trust as well as in formal contracts results in a higher level of control. According to the TCE, contracts are a basis for trust since it limits the opportunities for opportunistic behavior. The results of Dekker (2004) reinforce this with the findings that use of formal contracts in IORs drives to guarantee stability for trust and continuity in the future.

(11)

11

TCE rely on a large extent on economic and legal sanctions for the purpose of enforcing contractual obligations. On the other hand the relational exchange theory (RET) accounts explicitly for the social context of formal contracts in an IOR (Heide, 1994). The main focus of RET is sociological rather than economic (Cao & Lumineau, 2015). It is a theoretical lens to understand the situation of contracts in a social context. Firms in an IOR are expected to behave according to the shared relational norms such as information exchange and solidarity (Cao & Lumineau, 2015). The control within social context only functions well when relational norms in managing an IOR exists between the firms. Underlying the exchange process is trust, which regulates partner behavior and channel knowledge through common goals and shared understanding (Zhang & Zhou, 2013). Meaning that the TCE accounts for contracts, in this formal context, that gives better guides for cooperation and trust in the IOR, there are still some weaknesses in this relation. This research focuses on the RET to understand the situation of contracts in relational development. Therefore, relational trust seems to be more related to formal contracts when placed in a social context. For that reason the first hypothesis is presented as:

Hypothesis 1: Formal contracts are positively associated with relational trust when placed in a social context.

Prior research has not directly examined the effect of formal contracts on calculative trust. However, there is a reason to expect that formal contracts and calculative trust will have a positive relation. Calculative trust is based on an assessment of rewards and punishments, which assess the performance of the firms in an IOR (Poppo et al., 2016). According to the TCE, formal contracts can serve as a control tool to check for this performance or if firms behave opportunistically (Krishnan et al., 2016; Mellewigt et al., 2007). Contracts have a focus on the acceptable behavior in the IOR and also for the punishment in case of violation of the rules written in the contract (Lumineau, 2014). Therefore formal contracts increase the type of information that is necessary to make a benefit-punishment analysis, which is the underlying logic of calculative trust. Meaning that formal contracts serve as a control tool for the assessment of reward and punishment, this will enhance the positive outcomes of calculative trust. Therefore the second hypothesis is presented as:

(12)

12

2.5 Behavioral Uncertainty

Self-interest by firms may threaten the performance in IORs (Williamson, 1996). One of the most examined risks of self-interest that firms will defect form business agreements is behavioral uncertainty. This refers to “the potential inherent in a situation for difficulty anticipating and understanding actions” (Krishnan, Martin, and Noorderhaven, 2006 p.895). Behavioral uncertainty arises when one firm cannot effectively observe or evaluate the activities of the other firm in the IOR. (Poppo et al., 2016). This uncertainty depends on the chance that the firm will take advantage of opportunities (Fink & Harms, 2012). There is a potential for opportunistic gains, but firms are also capable of misbehaving or not disclosing, distorting, or disguising information in order to maximize their own interests (Poppo et al., 2016; Williamson, 1985). The magnitude of behavioral uncertainty varies considerably among IORs. For example the potential for opportunistic gains can be higher in an IOR when monitoring and evaluating activities are difficult (Krishnan et al., 2016). To reduce behavioral uncertainly, IORs need a control mechanism that researchers call ‘relational contracting’. One of these relational contracting can be that formal contracts are placed in a social context as mentioned earlier.

2.6 The Effectiveness of Formal Contracts under Behavioral Uncertainty

When a partner misbehaves, contracts have the benefit that they can be used to impose penalties, if the partner fails to adhere to contractual terms. Contracts reduce the partners’ ability and willingness to compromise on their performance (Krishnan et al., 2016). Partners are able to enforce the unwritten terms through private sanctions. This can take the form of a threat in the relationship of opportunistic behavior by a partner (Krishnan et al., 2016). According to the TCE, firms need the written terms to prevent this opportunistic behavior (Williamson, 1985). However, when the uncertainty is so high that it exceeds the range of privately- enforced written terms of the contract, contractual control tends to fail, and the IOR may be incapable of limiting opportunistic behavior. Also when a firm’s potential for opportunistic gains in behavior is high, contracts fail to improve the total performance of the relationship and cannot function as an effective hierarchical control mechanism (Krishnan et al., 2016; Williamson, 1996).

(13)

13

becomes useless, because firms will not use all the rules described in the contract. When they have detailed contracts, it could have a negative impact on the relation between the partners and their performance.

Relational trust is more effective when behavioral uncertainly exists because it generates positive beliefs and interpretations of each other’s actions (Krishnan et al., 2016; Poppo et al., 2016). The harmony of shared interests and goals aligns joint action, results in effective performance whereby the partners trust each other more. Relational trust enables problem solving whereby the whole relationship is emphasized rather than each individual transaction. This is more valuable when behavioral uncertainty exists, because it generates positive beliefs and interpretation of each other’s action (Krishnan et al., 2016).

When IORs parties behave opportunistically, behavioral uncertainty is present (Cao & Lumineau, 2015). This opportunism can be reduced by formal contracts and trust, but when behavioral uncertainty is low the contractual relationship is not effective. In this situation the partners can assess each other’s behavior relative easy (Krishnan et al., 2016). Contracts fail when behavioral uncertainty is extremely high, but for both formal contracts as relational trust, behavioral uncertainty is not effective when it is low. Therefore the relationship between formal contracts and relational trust will be stronger when behavioral uncertainty is moderate to high rather than low.

Hypothesis 3: The positive relationship between formal contracts and relational trust is stronger, when behavioral uncertainty is moderate or higher rather than low.

For calculative trust, it is hard to observe the inputs of activities of the firms, when behavioral uncertainty is high (Poppo et al., 2016). Unobserved behavior is punished through expected payoffs in which rewards depend on expected outcomes. The partner’s final output can be evaluated with calculative trust and rewards and sanctions can be applied effectively. For example in a buyer-supplier relationship, the buyer may punish the supplier when he detects low-quality outputs. The forward-looking assessment will remind IORs to perform well in order to acquire additional business in the future, even though their behavior is difficult to assess (Poppo et al., 2008; Poppo et al., 2016). The work of Lewicki and Bunker (1996) concludes that calculative trust is more effective, when behavior uncertainty is moderate, because one firm gets the other firm to do what they want them to do.

(14)

14

trust. Therefore the relationship between formal contracts and calculative trust will be stronger when behavioral uncertainty is moderate rather than low or high.

Hypothesis 4: The positive relationship between formal contracts and calculative trust is stronger when behavioral uncertainty is moderate than low or high.

Based on the hypotheses the following conceptual model presented in figure 1 is developed. All the influences and pressures are proposed to be positive. The results will indicate the impact of behavioral uncertainty on the relationship of formal contracts with relational trust and calculative trust in IORs.

Figure 2: Conceptual model

H3 H4 + + H1 + H2 + Relational trust Formal contracts Calculative trust Control variables - Firm size

- Prior partner experience - Relationship type

(15)

15

3. Research Method

For this research, a theory testing approach is applied. Therefore the steps of van Aken, Berends and Van der Bij (2012) are used. The driver for this theory testing process is a business phenomenon faced by IORs that has no conclusive evidence on explanations in the academic literature. First, the business phenomenon is defined and the literature gap is found. The second step was the generation of conceptual model and hypotheses. The third step was to collect data and data analysis. Finally, analysis results are compared to the initial hypotheses. Theoretical and managerial implications are deduced and providing limitations and extensions for future research (Van Aken et al., 2012). In the following paragraphs the data collection, measurement and analysis of this research are explained.

3.1 Data Collection

This research focuses on the relation between contracts and trust in IORs. Therefore firms with a strong formal contract character in relationships with relational and calculative trust are included to provide some interesting findings. For this research, IORs of different firms are examined. The data were collected by using a questionnaire, which was provided by the thesis supervisor. The overall goal of the survey was to give insight into the workings of trust and control within IORs. The survey was developed in an English version that students translated to ensure conceptual equivalence in understanding the questions. In the survey numerous questions are asked about the variables, to collect a broad range of data. This broad range gives extensive flexibility in the data analysis.

Together with other students we randomly selected firms that have inter-organizational relationships with other firms. For each firm, an employee with enough knowledge about contracts, control and trust in the IOR was selected for the surveys. Most of them were managers or directors of the firm, because they had the most knowledge about the contracts and day-to-day activities in the relationships.

(16)

16

rate. For the final sample, 131 directors, managers, or other employees from different firms agreed to participate. This is a response rate of 23 percent. After eliminating 20 surveys with missing data, this study obtained 111 complete responses for an effective response rate of 19.5 percent.

3.2 Measurement

The questionnaire items were provided by the thesis supervisor. Most of the items were assessed using Likert-type scales, with 1 representing “strongly disagree” and 7 representing “strongly agree”.

The measure of formal contracts consists of three items. Two items asses the use of formal contracts in an IOR. The other item assesses whether the firms will have developed things over time that never needed to be expressed contractually. For the first hypothesis the measurement of formal contracts in a social context also includes the following four items: reliance on the partner to keep promises, participatory decision-making, joint problem solving and fine-grained information exchange.

Relational trust is measured with three items: thinking like the other, shared

understanding and shared identification, which each reflects that a shared identity exists, and enabling one firm to trust the other to act on its behalf (Lewicki & Bunker, 1996).

For calculative trust there are also three measurement items. Two items measure the strength of rewards and punishments in the IOR for misconduct and dealing honestly with each other. The third item asses the continuation of the relationship. The expectation of a long relationship allows exchange firms to reward or punish prior moves (Lewicki & Bunker, 1996).

Finally, this research includes the measurement of the moderating effect of behavioral

uncertainty. This variable examines the difficulty to evaluate the other firm’s activities

(Brown, Dev & Lee, 2000; Poppo et al., 2016).

(17)

17

prior experience and it is an indicator for the current presence of trust between firms (Lui & Ngo, 2004). Prior partner experience is measured with a 6-point scale how many times (never till often) the two firms worked together in the past. The last control variable is the type of the relationship with the partner firm. This means whether the relationship will be a buyer-supplier relationship, joint venture, outsourcing, research & development (R&D) or another kind of relationship (Cao & Lumineau, 2015; Pérez-Nordtvedt, Kedia, Datta, & Rasheed, 2008).

3.3 Analysis

The analysis checked for construct validity and reliability with a factor analysis and Cronbach’s alpha (Van Aken et al., 2012). To perform a factor analysis, first the Kaiser-Meyer-Olkin (KMO) test and the Barlett’s test of sphericity are performed to assess the sampling adequacy (Karlsson, 2009). The KMO score of .647 exceeds the conventional threshold of .500 and the Bartlett’s test of sphericity is highly significant (p = .000). These results indicate sufficient sampling adequacy and allows performing a factor analysis.

An exploratory factor analysis (EFA) is conducted for each multiple-item variable. For the EFA, an analysis on the basis of fixed values of factors and excluding cases listwise was used. Each item that was coded to multiple variables, loaded to wrong variables or did not load to any variables was removed. Appendix A, describes the final results of the EFA after item deletion. In these results, one item of formal contracts and two of calculative trust are removed because they were double loaded or not at all. The factor loadings provide evidence of construct validity.

(18)

18

4. Results

A multiple regression analysis is preformed to examine the hypothesized effects of formal contract on relational and calculative trust, and with the moderator effect of behavioral uncertainty (Aiken & West, 1991). For the effect of formal contracts on relational trust, formal contracts are placed in a social context. To get the results for formal contracts in a social context, the items of formal contracts and social control are added together.

Before the regression analysis is conducted, the variables are tested for normal distribution and homoscedasticity (Osborne & Waters, 2002). Therefore the P-plot and scatterplot are used as controls. The P-plot shows that the residuals generally follow the normality line, so it can be assumed that the data is normally distributed. The scatterplot of the data in this research shows that residuals are randomly scattered across the horizontal line and therefore it can be assumed that the data are homoscedastic.

The Durbin-Watson statistic is used to assess that the observations are independent. When they are, the outcome is between 1.5 and 2.5 (Durbin & Watson, 1971). For the regression model with relational trust the Durbin-Watson is 1.902 and with calculative trust it is 1.883, which implies that in both models the observations are independent. Finally the data are checked for outliers with the residual plot, because these outliers can lead to misinterpretation of the model (Cook, 1977). The plot shows that there is one outliers for relational trust in this research (lower than 3 and bigger than -3), this one is removed from the sample.

To pass the multicollinearity test and interpret the results, the variance inflation factor (VIF) and tolerance levels are used as controls. To avoid the mulitcollinearity problem in the regression analysis, the variables are mean-centered before the interaction terms are constructed (Aiken & West, 1991). With the mean-centered variables, the highest value of the VIF is 1.166, well below the 10.0 benchmark. The tolenance levels are all higher than 0.1, which indicates that multicollinearity is not an issue with mean-centered variables.

(19)

19

included in formal contracts in a social context which corresponds to each other. Further the highest correlation is between formal contracts and calculative trust (r = .440, p <0.1). It is interesting that the correlation between formal contracts and behavioral uncertainty is low (r = -.046, p > .10 & r = -.066, p > .10), despite the fact that the effect of behavioral uncertainty is better when it is not low (Krishnan et al., 2016; Williamson, 1979). More surprisingly is the low correlation between behavioral uncertainty and calculative trust (r = -.009, p > .10). Previous research has concluded that calculative trust is hard to observe when behavioral uncertainty is high and that it becomes effective when behavior uncertainty is moderate (Lewicki & Bunker,1996; Poppo et al., 2016). The correlation in this research does not support this conclusion.

Table 1. Descriptive and correlations

Variable Mean S.D. 1 2 3 4 5 6 7 1 Formal contracts 5.00 1.47 2 Formal contracts in a social context 5.21 1.028 .672*** 3 Relational trust 4.70 1.22 .156 .155 4 Calculative trust 3.65 2.07 .440*** .358*** .279*** 5 Behavioral uncertainty 3.10 1.43 -.046 -.066 -.197** -.021 6 Firm size 9,999.39 34,091.66 .022 -.049 -.032 -.009 -.232** 7 Prior partner experience 4.02 1.96 -.081 -.126 .209** .099 -.160 -.016 8 Partnership type 2.03 1.49 .139 .122 .087 .274*** .020 -.058 -.182* Notes: * p < .10, ** p < .05, *** p < .01 (two-tailed) Listwise N=105

(20)

20

Table 2. Results of multiple regression

Variable Relational trust Calculative trust

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

Intercept 3.963***(.332) 4.038*** (.330) 4.178*** (.330) 2.109*** (.561) 2.169*** (.516) 2.315*** (.532)

Control variables

Firm size -.000 (.000) -.000 (.000) .000 (.000) .000 (.000) .000 (.000) .000 (.000)

Prior partner experience .139** (.059) .133** (.059) .119** (.058) .160 (.100) .189** (.092) .172* (.093)

Relationship type .103 (.079) .088 (.078) .056 (.078) .422*** (.134) .346*** (.123) .315** (.126)

Main effects

Formal contracts in a social context Formal contracts Behavioral uncertainty .179 (.111) -.138* (0.83) .243** (.113) -.133 (.081) .584*** (.121) .034 (.129) .626*** (.126) .022 (.129) Interaction effects

Formal contracts, social context x Behavioral uncertainty

.159** (.072)

Formal contracts x Behavioral uncertainty .097 (.088)

(21)

21

Model 2 (table 2) shows that the direct effect of formal contracts in a social context not relate to relational trust and insignificant (β = .179, P > .10). Hypothesis 1, which predicts that formal contracts in a social context are positively associated with relational trust, is therefore not supported. However, model 3 shows that this direct effect of formal contracts in a social context is positively related (β = .243, P < .05).

Hypothesis 2 examines the positive relationship between formal contracts and calculative trust. Models 5 and 6 show that formal contracts relate positively to calculative trust (β = .584, P < .01; β = .626, P < .01) and therefore in support of hypothesis 2.

The full model (M4 & M6 in Table 2) is used to test the interaction hypotheses. Hypothesis 3 assesses the moderating role of behavioral uncertainty on the relation between formal contracts in a social context and relational trust. The interaction effect of formal contracts in a social context x behavioral uncertainty is positive and significant (β = .159, p < .05). This finding supports hypothesis 3.

Hypothesis 4 examines the moderating effect of behavioral uncertainty on the relation between formal contracts and calculative trust. This interaction effect of formal contracts x behavioral uncertainty is insignificant (β = .097, p > .10). This finding does not support hypothesis 4.

To gain more insight into the interaction effect in hypothesis 3 and 4, a process analysis is conducted for the interactive terms (Hayes, 2013). For hypothesis 3, the interaction effect of behavioral uncertainty is split into low, moderate, and high to estimate the effect of each level as seen in figure 3. When behavioral uncertainty is moderate the relation is significant (β = .246, p < 0.5) and when it is high the relation becomes even stronger (β = .475, p < 0.1). On the other hand, when behavioral uncertainty is low, the relation is not significant (β = .016, p > .10).

Hypothesis 4 cannot be supported. For the interaction effect of formal contracts x behavioral uncertainty on calculative trust, there are no statistical significance transition points within the observed range of the moderating effect of behavioral uncertainty.

(22)

22

Figure 3. The interaction effect of behavioral uncertainty on the relationship between formal

contracts in a social context and relational trust.

As table 2 shows, there are also effects of the control variables on relational and calculative trust. First, prior partner experience is positive related to relational trust in model 1 till 3. This implies that prior partner experience fosters better relational trust in an IOR. In model 5 and 6, prior partner experience is also positive related to calculative trust. Also relationship type is positively related to calculative trust, however this is not related to relational trust.

5. Discussion and Conclusion

In this last chapter the discussion and conclusion of this research is presented. First, a summary of the results and research implications are presented. Second, the managerial implications are given and finally the limitations and directions for further research.

5.1 Research Implications

(23)

23

of the relationship (Rousseau et al., 1998). Previous research seldom examines formal contracts in relation to different types of trust. This is an important research gap because there is a debate which effect of formal contracts is most reliable with relational trust and calculative trust (Cao & Lumineau, 2015; Poppo et al., 2016; Woolthuis et al., 2015). The results of this research contribute to the literature by demonstrating the positive value of formal contracts placed in a social context in relation with relational trust, the positive value of formal contracts with calculative trust, and the moderation conditions of behavioral uncertainty, which influence the effectiveness of formal contracts and relational and calculative trust.

First, the results of the positive direct effect of formal contracts when it is placed in a social context on relational trust (H1) is not supported. However, when the interaction effect of behavioral uncertainty is added, the direct effect of formal contracts in a social context does become significant. This result is consistent with the findings of previous studies showing that relational trust becomes more effective when behavioral uncertainty is present (Krishnan et al., 2016; Poppo et al., 2016). As shown in figure 3, the degree of behavioral uncertainty is relevant for the relation between formal contracts and relational trust. Low behavioral uncertainty does not influence this relation much. The higher behavioral uncertainty becomes, the stronger the effect will be on the relation between formal contracts in a social context and relational trust.

Second, the effect of the social context of formal contracts influences the relation with relational trust. Previous studies show that contracts have weaknesses in a formal context in relation with relational trust (Das & Teng, 1998; Dekker, 2004; Heide, 1994; Woolthuis et al., 2005; Zaheer & Venkatraman, 1995). To confirm this literature, the direct effect of formal contracts on relational trust gives the result that this is not significant (β = .122, p > .10). When the interaction effect of behavioral uncertainty is added, the direct effect is marginally significant (β = .147, p < .10), and the interaction effect is low correlated and not significant (β = .059, p > .10). This concludes that formal contracts results in a damaging effect on relational trust (Das & Teng, 1998; Dekker, 2004). Therefore the effect of the social context with shared interest results in effective performance and relational trust of the firms (Cao & Lumineau, 2015).

(24)

24

have the ability to increase the information necessary for a benefit-punishment analysis, the underlying logic of calculative trust (Lumineau, 2014; Poppo et al., 2016; Williamson, 1993). Fourth, in the results, behavioral uncertainty is very low correlated with calculative trust and also the interaction effect with formal contracts has a very low correlation. This might signal that the results show that there are no statistical related points for all levels of behavioral uncertainty. Therefore this research fails to find support for the hypothesis that formal contracts relate stronger to calculative trust when behavioral uncertainty is moderate. According to Lewicki and Bunker (1996), calculative trust should be most effective when behavioral uncertainty is moderate (see also Krishnan et al., 2016; Poppo et al., 2016). For formal contracts, behavioral uncertainty should not be low or extremely high (Williamson, 1979; Williamson, 1996). However in balance, the potential positive and negative effects cancel each other out, resulting in an insignificant effect.

This research contributes to the theoretical arguments of TCE and RET regarding the relationships between formal contracts, relational trust calculative trust, and behavioral uncertainty. This study enriches TCE by the result of the positive relation of formal contracts and calculative trust. The use of formal contracts is consistent with the underlying logic of calculative trust (Lumineau, 2014; Wallenburg & Schäfffer, 2014; Williamson, 1985). The results supports RET with the relation of formal contracts in a social contact with relational trust (Heide, 1994). Formal contracts in a social context had not a direct effect on relational trust in the results, but with the influence of behavioral uncertainty it does have. Also in comparison with only formal contracts (without the social context) the social context does matter, which is explicitly accounted by the RET.

5.2 Managerial Implications

(25)

25

uncertainty should be present to improve the relational trust when they use formal contracts in a social context. When firms are relying on calculative-based trust, then they should use formal contracts with detailed rules for a benefit-cost analysis.

5.3 Limitations and Further Research

This study has some limitations that need to be addressed in future research. First, the focus of calculative and relational trust in this research is based on the work of Poppo et al. (2016). While this research has had some interesting findings, there are some differences with earlier studies. The sample size is much smaller, which could influence the significance of the relations. Therefore, future research on formal contracts, trust and behavioral uncertainty in IORs is advised to test the hypotheses with a lager sample size consisting of firms in different countries and industries. National cultures can play a big part on behavioral uncertainty and affecting trust in IORs (Fink & Harms, 2012). Also the legal system of countries may influence the legal terms of formal contracts (Poppo & Zenger, 2002). Hopefully future work will extend this inquiry.

A second limitation is that calculative trust is tested with only one item. Because there was no consistency between the three items, two items were removed for the analysis. Therefore it was, among other things, not possible to calculate Cronbach’s alpha. Because of the short period of doing research, there was no time left to add new items or re-collect the data to increase its validity. Future research may extend the variable calculative trust with adding other items to come up with different results.

Third, the results provide evidence with the effect of behavioral uncertainty on formal contracts using only two types of trust: relational and calculative. Trust is present in multiple levels (individual, group, firm and institutional), it can be within and between organizations, can have different roles (cause, outcome, and moderator), and can be in different types (relational, calculative, goodwill, institutional) (Rousseau et al., 1998). Future research efforts may extend similar inquiries as this research but by using other types of roles of trust.

(26)

26

Fifth, this research did not made a distinction between firms in another level of their relationship length. Trust and behavioral uncertainty could be different in the early stages of the formation of the IOR than in a later stage. While this study only used a fixed moment in time of an IOR, future research could focus on longitudinal data to more closely investigate how the relation of formal contracts and trust, with the influence of behavioral uncertainty evolves over time.

Despite these limitations, the analysis suggested in this paper has identified the relation between formal contracts and relational and calculative trust in IORs. This study has investigated the interaction effect of behavioral uncertainty on this relation, which results in different outcomes. The research enhances understanding of the role of formal contracts, trust and the effect of behavioral uncertainty in inter-organizational relationships.

Acknowledgements

(27)

27

References

Aiken, L. S., & West, S. G. (1991). Multiple regression: testing and interpreting interactions. Newbury Park, CA: Sage Publication.

Arranz, N., & Arroyabe de, J. C. (2012). Effect of formal contracts, relational norms and trust on performance of joint research and development projects. British Journal of Management , 23, 575- 588.

Balliet, D., & P.A.M., V. L. (2013). Trust, conflict and cooperation: a meta-analysis. Psychological

Bulletin , 139 (5), 1090-1012.

Boon, S., & Holmes, J. G. (1991). The dynamics of interpersonal trust: Resolving uncertainty in the face of risk. In R. A. Hinde, & J. Groebel, Cooperation and prosocial behavior. (pp. 190-211). Cambridge, England: Cambridge University Press.

Bromiley, P., & Harris, J. (2006). Trust, Transaction cost economics, and mechanisms. In R. &. Bachmann, Handbook of Trust Research (pp. 124-143). Edward Elgar.

Brown, J., Dev, C., & Lee, D. (2002). Managing marketing channel opportunism: the efficacy of alternative governance mechanisms. Journal of Marketing , 64, 51- 65.

Cao, Z., & Lumineau, F. (2015). Revisiting the interplay between contractual and relational governance: A qualitative and meta-analytic investigation. Journal of Operations

Management , 33, 15-42.

Cook, R. (1977). Detection of Influential Oberservation in Linear Regression. Americal Statistical

Association (19), 15-18.

Das, T., & Teng, B. (1998). Between trust and control: developing confidence in partner cooperation in alliances. Academy of Management Review , 23 (2), 491-512.

Dekker, H. (2004). Control of inter-organizational relationships: evidence on appropriation concerns and coordiation requirements. Accounting Organizations and Society , 29, 27-49.

Dillman, D. A. (2000). mail and internet surveys: the tailored design method (2 ed.). New York: Wiley.

Durbin, J., & Watson, G. S. (1971). Testing for serial correlation in least squares regression.lll.

Biometrika , 1 (58), 1-19.

Fink, M., & Harms, R. (2012). Contextualizing the realitonship between self-commitment and performance: Environmental and behavioural uncertainty in (cross-border) alliances of SMEs.

Enterpreneurship & Regional Development , 24 (3-4), 161-179.

Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: a

(28)

28

Heide, J. B. (1994). Interorganizational Governance in marketing channels. Journal of Marketing ,

58, 71-85.

Karlsson, C. (2009). Researching operations management. New York: Routledge.

Krishnan, R., Geyskens, I., & Steenkamp, J. B. (2016). The effectiveness of contractual and trust-based governance in strategic alliances under behavioral and environmental uncertainty.

Strategic Management Journal , 37, 2521-2542.

Krishnan, R., Martin, X., & Noorderhaven, N. (2006). When does trust matter to alliance performance? Academic of Management Journal , 49 (5), 894-917.

Leifer, R., & Mills, P. (1996). An information processing approach for deciding upon control strategeis and reducing control loss in emerging organizations. Journal of Management , 113-137.

Lewicki, R., & Brinsfield, C. (2011). Framing trust: trust as a heuristic. In W. Donahue, R. Rogan, & S. Kaufman, Framing Matters: Perspectives on negotiation research and practice in

communication (pp. 110-135). New York: Peter Lang Publishing.

Lewicki, R., & Bunker, B. (1996). Developing and maintaining trust in work relationships. In Trust

in Organizations: Fronties of theory and research (pp. 114-139). Thousand Oaks, CA: Sage

Publication.

Lu, S., & Yan, H. (2015). Contractual control, the propensity of trust, active trust development: construction industry. Journal of Business & Industrial Marketing , 31 (4), 459-471.

Lui, S. S., & Ngo, H. (2004). The role of trust and contractual safeguards on cooperation in non-equity alliances. Journal of Management , 30 (4), 471-485.

Lumineau, F. (2014). How contracts influence trust and distrust. Journal of Management , 1-25. Macneil, I. R. (1978). Contracts: adjustment of long-term economic relations under classical,

neoclassical and relational contact law. Northwestern University Law Review , 72, 854-905. Malthora, D., & Lumineau, F. (2011). Trust and collaboration in the aftermath of conflict: the effects

of contract structure. Academic of Management Journal , 54 (4), 981-998.

Mellewigt, T., Madhok, A., & Weibel, A. (2007). Trust and formal contracts in inter-organizational relationships: Substitutes and complements. Managerial and Decision Economics , 28, 833-847.

Merchant, K. (1984). control in business organizations. Marshfield, MA: Pitman publishing.

Minnaar, R., Vosselman, E., Van Veen-Dirks, P., & Zahir-Ul-Hassan, M. K. (2016). A relational perspective on the contract-control nexus in an interfirm relationship. Management

Accounting Research .

(29)

29

Osborne, J., & Waters, E. (2002). Four assumptions of multiple regression that researchers should always tests. Practical Assessment, Research & Evaluation .

Pérez-Nordtvedt, L., Kedia, B., Datta, D., & Rasheed, A. (2008). Effectiveness and efficiency of cross-border knowledge transfer: an empirical examination. Journal of Management Studies ,

45 (4), 714-744.

Poppo, L., & Zenger, T. (2002). Do formal contracts and relational governance function as substitutes or complements? Strategic Management Journal , 23, 705-725.

Poppo, L., Zhou, K. Z., & Li, J. (2016). When can you trust "trust"? Calculative trust, relational trust and supplier performance. Strategic Management Journal , 37, 724-741.

Poppo, L., Zhou, K. Z., & Ryu, S. (2008). Alternative origins to interorganizational trust: an interdependence perspective on the shadow of the past and the shadow of the future.

Organization Science , 19 (1), 39-55.

Reuer, J., & Ariño, A. (2002). Contractual renegotiations in strategic alliances. Journal of

Mangement , 28 (1), 47-68.

Ring, P. (1996). Fragile and resilient trust and their roles in economic exchange. Business and Society

, 35 (2), 148 - 175.

Rousseau, D. M., Sitkin, S. B., Burt, R. S., & Camerer, C. (1998). Not so different after all: A cross-discipline view of trust. Academy of Management Review , 23, 393-404.

Uzzi, B. (1997). Social structure and competition in interfirm network: The paradox of embeddedness. Administrative Science Quarterly , 42, 35-67.

Van Aken, J., Berends, H., & Van der Bij, H. (2012). Student Projects. In Problem solving in

organizations: A methodological handbook for business and management students.

Cambridge University Press.

Vosselman, E. G., & Van der meer-Kooistra, J. (2009). Accounting for control and trust building in interfirm transactional relationships. Accounting, Organizations and Society , 34 (2), 267-283. Wallenburg, C. M., & Schäfffer, T. (2014). The interplay of relational governance and formal control in horizontal alliances: a social contract perspective. Journal of Supply Chain Management ,

50 (2), 41-58.

Weeds, S. E., & Mitchell, T. R. (1980). The role of environmental and behavioral uncertanty as a mediatior of situation-performance relationships. Academy of Management Journals , 23 (1), 38-60.

Wicks, A. C., Berman, S. L., & Jones, T. M. (1999). The structure of optimal trust: Moral and strategic implications. Academy of Management Review , 24, 99-116.

(30)

30

Williamson, O. E. (1996). The Mechanisms of Governance. Oxford, UK: Oxford University Press. Williamson, O. (1979). Transaction cost economics: the governance of contractual relations. Journal

of Law and Economics , 22 (2), 233-261.

Woolthuis, R., Hillebrand, B., & Nooteboom, B. (2005). Trust,contract, and relational development.

Organizations Studies , 26 (6), 813-840.

Zaheer, A. M. (1998). Does trust matter? Exploring the effects of interorganizational and interpersonal trust on performance. Organization Science , 9, 141-159.

Zaheer, A., & Harris, J. (2005). Interorganizational trust. In O. Shenkar, & J. (. Reurer, Handbook of

Strategic Alliances (pp. 169- 197). Thousand Oaks, CA: Saga Publications.

(31)

31

Appendix

Appendix A: Rotated Exploratory Factor Matrix & Cronbach’s Alpha

Alpha 1 2 3 4 5 Social context .829 SOCIALCONTEXT_1 .802 SOCIALCONTEXT_2 .839 SOCIALCONTEXT_3 .813 SOCIALCONTEXT_4 .775 Relational trust .715 RETLTRUST_1 .830 RETLTRUST_2 .825 RETLTRUST_3 .725 Formal contracts .738 CONTRACT_1 .856 CONTRACT_2 .862 Behavioral uncertainty .730 BHVRUN_1 .823 BHVRUN_2 .917 Calculative trust - CALCTRUST_2 .936

Notes: Rotation method: Varimax.

The Effect of Formal Contracts on Trust