The Effects of Affective Trust and Suspicion in New Product Development Projects

Abstract

Relationships between functional specialists working on NPD projects can lack trust, resulting in poor communication, low levels of co-operation, and, in the extreme, the political sabotage of projects. We empirically test a hypothesized model of managerial perceptions of motives and intentions, negative project politics, and affect-based trust, as well as their effect on dysfunctional conflict and defensive behaviors, with their subsequent effect on collaboration and NPD project success. Data were from 184 projects from the technically trained manager perspective, and 145 projects were from the marketing manager perspective. We find that, despite all of the NPD process improvement and use of integration methods, NPD managerial level interactions are still fundamentally relationally-based, where the strongest effects on cross-functional managerial behavior are from perceptions of negative motives, leading to defensive behaviors and politics interfering with collaboration and NPD success. On a positive note, while evident affect-based trust leads to collaboration and NPD success, it is far more important for technically trained managers. The implications for top management lie in structuring organizational processes that are trust-rich, in order to avoid the opportunity for distrust and suspicion to arise.

1. Introduction

This paper came about as the result of an engineering manager replying to the interview question “What do you think of the marketing manager?” with “A nice guy but I just don’t trust him!” The issue of trust in cross-functional working relationships has gained significant traction in the NPD literature, and its existence is beneficial to the process [1,2,3,4,5,6,7,8,9]. The lack of trust, i.e., the suspicion relating to a counterpart manager’s motives and intentions, has not been the focal point of any of these studies [10,11].

Basic work-related interactions, such as the communication and co-operation that are necessary to facilitate effective integration and acceptance of marketing’s inputs [12], eliminate project ambiguity [13], transfer knowledge [14], and facilitate working in virtual teams [15], may be blocked when parties are suspicious of the other. Considering the complexity of the modern innovation process, where Nagano, Stefanovitz, and Vick [16] examine the functional interdependence necessary for effective NPD work within the organization, such a research gap is surprising, especially when considering the impact that a situation of distrust could have on working relationships that are so mutually interdependent [17].

Our research is relevant to all NPD active workplaces, as many of the interviewees for our qualitative phase (in-depth interviews) from both technically trained roles and marketing clearly expressed views that their successful new product projects were usually developed in a collaborative organizational environment, often by-passing formal NPD procedures and using their friends and the informal network within the organization to achieve positive NPD results. Interpersonal trust is a very important element of their working relationship with other managers. Where they trusted the other manager, they felt that most problems could be overcome through quality communication. On the other hand, where they did not trust their functional counterpart, many defensive behaviors (e.g., stalling, blocking, sabotage, documenting all actions, and open conflict) dominated the relationship and often led to a termination of all meaningful communication, with subsequent negative project affected outcomes.

We contribute to the literature by examining what happens to the project when the working relationship is not optimum or harmonious. Are there dysfunctional conflicts, defensive behaviors against opportunism, and negative project politics evident during a project? Is there suspicion and a lack of trust? Does this affect collaboration and NPD success? The traditional view of marketing’s working relationships suggests that NPD, as a key corporate activity, is still very problematic for marketing managers, often resulting in unsuccessful new products and poor relations between the functional participants [18,19,20]. Our paper explores whether this is still the case by empirically testing to see whether collaboration and NPD success is affected by negative behaviors, such as defensive behaviors, negative project politics, and dysfunctional conflict. To develop a better understanding of this, we focus on the negative relationship constructs that can emerge during NPD activities, as well as the beneficial ones (affect-based trust and collaboration).

Addressing these gaps in the literature, we make two contributions. Our first contribution is that we are the first to develop and empirically test a model of the antecedents and consequences of the negative aspects of cross-functional working relationships between marketing managers and their technically trained counterparts. Specifically, we are the first to introduce the construct of defensive behaviors, which are triggered by the perception of negative motives and intentions from a counterpart manager in an NPD setting. Our findings show that distrust and the perception of negative motives and intentions are still the strongest drivers of behavior in NPD managerial interactions. We also use data from both marketing managers and R&D managers to test our model and clearly show that the impact of key variable changes, depending on the functional background (i.e., whether from marketing or R&D).

Our second contribution is a theoretical one, in which we demonstrate the efficacy of a social exchange lens [21] and primacy of human perceptions regarding others. Blau argues that it is a fundamental aspect of human nature to seek enough social data about another person before reaching a critical point where a decision is made as to whether or not to invest any further in the relationship or engage in avoidance behaviors. While this aspect of human relationships has been examined in the NPD trust literature, the specific negative effects have not been conceptualized and empirically tested prior to our study. We clearly show that effective NPD processes are socially-based senior functional managers who rely heavily on their interpersonal assessments of each other. This is where affect and suspicion enter into cross-functional working relationships.

For top management trying to facilitate and organize effective cross-functional interactions between their managers, our research acts as an important reminder that, while the NPD literature has focused on formalized interaction methods, e.g., QFD, concurrent engineering, agile systems, and Stage Gate 2.0, all of these can break-down at the human level if there is distrust between the key actors. Top management needs to provide a climate of organizational innovation that is high trust and low personal risk, in order to mitigate the potentially disastrous effects of negative interpersonal perceptions at the managerial level.

The remainder of this paper is structured as follows. First, we review the background literature and present the main theoretical framework we draw upon. We then introduce, define, and justify our choices regarding antecedent and outcome variables. Next, we present our model, hypotheses, and methodology, as well as the details regarding the measurement and operationalization of our variables. Last, we discuss our results and their implications, as well as the limitations of the paper and directions for future research.

2. Theoretical Framework

To capture the complexity of cross-functional work and interdependencies that are necessary for effective new product development, numerous theories have been utilized as a theoretical base: resource-dependency [22,23], the information processing perspective [24], and the socio-political perspective [25,26,27,28,29].

While these theories are all relevant for the study of functional manager behaviors and their interactions, where project selection, resource allocations, information flows, and managerial-level political behaviors occur in the NPD process, it is social exchange theory [21] and the interaction approach [2,30,31,32] that best capture the antecedents of negative behaviors that may emerge in NPD working relationships and subsequently affect collaboration and NPD success. Specifically, it is a social exchange theory lens that focuses on interpersonal trust and its formation, which makes it relevant to all working relationships that are not strictly contractually mandated. Blau’s [21] social exchange theory is a way of understanding human exchange relationships, where trusting behaviors signal interest in, and commitment to, such relationships. In their seminal work, Morgan and Hunt [33] also identified commitment and trust as central to working relationships. When these trusting behaviors are reciprocated, they foster beneficial outcomes for the relationship, such as creating a positive atmosphere, reducing or removing barriers of task-related risk, and allowing the relationship to further develop. Interpersonal trust is seen to emerge through the successful social exchanges that influence perceptions of non-reciprocation (i.e., opportunism) and trust. The importance of trust, especially to the marketing managers who are cross-functional boundary spanners in organizations, is captured by Williams ([34], p. 25), who states that “the complexity and uncertainty inherent in managerial work and the amount of mutual accommodation it involves, effective horizontal working relationships are also critical … and that under conditions of uncertainty and complexity, requiring mutual adjustment, sustained effective coordinated action is only possible where there is mutual confidence or trust (c.f., [23], p. 1)”.

While this is the upside of trusting relationships, Dirks [17] provides a corollary: where distrust may exist between co-workers, it causes anxiety and leads to “attempts to protect their backside by monitoring partners’ actions, working to ensure personal success, and so on.” (p. 448).

It is clear, when viewed from Blau’s [21] social exchange theory lens, that Deutsch’s [10,11] perspective on relationships is essentially grounded in the interplay of trust and suspicion, regarding marketing’s NPD working relationships, which can be most effectively conceptualized when addressing negative issues. In his seminal work, Deutsch [11] wrote, “To trust another person to produce a beneficial event X (or to suspect that another person will produce a harmful event Y) an individual must have confidence that the other individual has the ability and intention to produce it.” (p. 125). By using a social exchange theory lens, we capture the fundamental building blocks of working relationships, trust, and perceptions of a counterpart’s motives and intentions through the collection of social data based on interactions.

Our second theoretical framework, the interaction approach [31,32], complements and expands upon social exchange theory by explicitly acknowledging that, through interactions over issues such as setting goals jointly, negotiating timelines, and resolving engineering and customer-need trade-offs [35,36,37,38], functional managers become integrated into the organization’s NPD processes. Importantly, the interaction approach allows us to investigate the development of working behaviors, i.e., the positive ones associated with co-operation, functional conflict, and collaboration, as well as the negative ones, such as dysfunctional conflict and defensive behaviors. Due to its versatility in capturing relationship interactions, the interaction approach has also been used to explain marketing’s working relationships with other functions [39,40,41].

3. Conceptual Framework and Hypotheses

Our hypothesized model includes the following variables: (1) trust and suspicion, (2) negative mediators (defensive behaviors, negative project politics, and dysfunctional conflict), and (3) outcome variables (collaborative behavior and, ultimately, new product success). Our hypothesized model is shown in Figure 1.

3.1. Outcome Variables

3.1.1. New Product Success

With the considerable investment in resources required in organizations’ NPD processes, measuring NPD success is an accepted practice when reviewing the effectiveness of such investments. Drawing from the literature, we conceptualize new product success as being a multi-item formed construct captured by budget, time, sales, profit aspects, and overall performance perspective [42,43]. Due to its importance in an organizational context, many studies have examined the antecedents of NPD success [43,44,45] for these reasons, we include it in our model.

3.1.2. Collaborative Working Relationship

Collaboration represents all things that are good in a working relationship. It is the antithesis of negative behavior. Collaborative behavior in working relationships consists of effective communication, trusting behavior, volitional co-operation, mutual problem solving, and esprit de corps, and it is a predictor of successful NPD working relationships [31,46,47]. We use it in our model to highlight the positive impact of behaviors that can occur in working relationships.

3.2. Antecedent Variables

3.2.1. Negative Motives and Intentions

Deutsch’s [11] seminal work focused on the suspicion a manager has for the perceived intention of parties in an experiment designed to examine self-serving behavior, without concern for the other participants’ well-being. This work highlighted the dilemma that many managers continue to face when they work with other managers who they may have weak social ties to or no hierarchical control over, as well as those who come from separate thought worlds, differing departments, and have differing goals. At some time, they will assess whether or not their counterpart will take advantage of them, and it will affect their subsequent behaviors towards that person, especially in regard to making oneself vulnerable to their actions via trusting behavior. Many researchers in the social science literature view having trust in another as being confident in regard to that party acting with benign or benevolent intentions [11,48]. We argue that, in contrast to Smith and Barclay’s [49] positive perspective of motives and intentions, where partners perceive the purpose or agenda behind the other’s actions as being benevolent or benign, their definition is one where there is an element of suspicion regarding a counterpart’s motives and intentions, where the others’ actions are interpreted as being malevolent or hostile. Whereas Smith and Barclay [49] found that benevolent motives and intentions had a strong predictive effect on relationship investment, we argue that all relationship outcomes are negatively affected when managers feel that negative aspects are in play. Accordingly, we hypothesize:

Hypothesis 1a (H1a).

The higher the perception that the counterpart manager has negative motives and intentions, the higher the defensive behavior against opportunism during the NPD project.

Hypothesis 1b (H1b).

The higher the perception that the counterpart manager has negative motives and intentions, the higher the perceived negative project politics during the NPD project.

Hypothesis 1c (H1c).

The higher the perception that the counterpart manager has negative motives and intentions, the higher the level of dysfunctional conflict during the NPD project.

Hypothesis 1d (H1d).

The higher the perception that the counterpart manager has negative motives and intentions, the lower the level of collaboration during the NPD project.

3.2.2. Affect-Based Trust

In contrast to negative perceptions, affect-based trust addresses many of the positive aspects of human relationships, where it involves reciprocated interpersonal care and concern for another person [50,51]. Mittal [52] argues that affect-based trust leads to the subjective feeling of security against being exploited, as well as the comfort and assurance that one’s interests are being served by another party. Jones and George [53] state that affect-based trust relates to existence of unconditional trust (i.e., the positive mood and degree of affect in the relationship). In an NPD teamwork team situation, they suggest that it has a beneficial effect on several social processes: the existence of broad role definitions leads to greater citizenship behaviors, better communal relations, high confidence in others, help-seeking behavior, free exchange of knowledge and information, subjugation of personal needs and ego for the greater common good, and high involvement in processes. McAllister [54] found that managers reporting high affect-based trust actively sought more opportunities to meet peers’ work-related needs and engaged in more productive interventions. Accordingly, we hypothesize:

Hypothesis 2a (H2a).

The higher the level of affect-based trust in the counterpart manager, the lower the level of defensive behavior against opportunism during the NPD project.

Hypothesis 2b (H2b).

The higher the level of affect-based trust in the counterpart manager, the lower the level of the negative project politics during the NPD project.

Hypothesis 2c (H2c).

The higher the level of affect-based trust in the counterpart manager, the lower the level of dysfunctional conflict during the NPD project.

Hypothesis 2d (H2d).

The higher the level of affect-based trust in the counterpart manager, the higher the level of collaboration during the NPD project.

3.3. Mediating Variables

3.3.1. Defensive Behaviors against Opportunism

Ashforth and Lee [55], in their examination of the antecedents and consequences of defensive behaviors in organizations, stated, “To defend, by definition, is to protect against attack. Most instances of defensiveness can be traced back to a perception of actual or implied threat to job security, resources, autonomy, reputation, task requirements, the self-concept, or the like.” (p. 634). For managers, once they perceive that their counterpart poses an actual or implied threat, they are likely to engage in a reactive defense of self-interest. Lewicki et al. [56] argued that this can be viewed as distrust and referred to it as a confident negative expectation, fear, propensity to attribute sinister intentions, and a desire to buffer oneself from the effects of another’s conduct. One of the main threats in a working relationship is the risk of exposing oneself to opportunism via the other party [57,58,59]. Bromiley and Cummings [60] place even more importance on its role in working relationships by defining it as the third component of trust, where a party “does not take excessive advantage of another even when the opportunity is available” (p. 303). Similarly, Rousseau [61] and Mayer et al. [48] viewed trust as opening oneself up to risk by co-operating and sharing information with another, thus providing an opening for opportunism to occur.

Once managers have determined that a threat has to be mitigated, Ashforth and Lee [55] identify a plethora of potential defensive behaviors available to them that may consist of: (1) avoiding action via over-conforming, passing the buck, playing dumb, depersonalizing, smoothing and stretching, and stalling; (2) avoiding blame via buffing, playing safe, justifying, scapegoating, misrepresenting, and escalating commitment; (3) resisting change and protecting turf. Ashforth and Lee [55] also argue that such defensive behaviors are seen as having both negative short- and long-run consequences for both interpersonal relationships and the organization’s processes. Specifically, for the individual, it can develop into defensive routines, which lose the trust and support of management; for the organization, it induces rigidity and stagnation, red tape, insularity, the politicization of the organization, a climate of distrust, and low morale. These consequences have all been identified in the NPD literature as major contributors to ineffective functional integration and NPD failure [45,62]. Accordingly, we hypothesize that:

Hypothesis 3a (H3a).

The higher the level of defensive behaviors against opportunism, the lower the level of collaboration with the counterpart manager during the NPD project.

Hypothesis 3b (H3b).

The higher the level of defensive behaviors against opportunism, the lower the NPD project success.

3.3.2. Negative Project Political Perceptions

Classic texts, such as Pettigrew’s [63] Politics in Organizational Decision Making, Bacharach and Lawler’s [64] Power and Politics in the Organization, Pfeffer’s [65] Power in Organizations and [66] Managing with Power: Politics and Influence in the Organization, have studied politics laden with negative connotations, from power plays, Machiavellianism, self-serving behavior, coercion, and coalition formation. Implicit in these texts is that political tactics are the tools of the powerful, or of those seeking power.

In one of the first empirical investigations of managerial level politics in organizations, Gandz and Murray [67] found that politics were common and inevitable at the higher levels of the organization and some processes are considered more political than others. When examining his four criteria for a politically-loaded process, NPD work easily meets the following criteria: (1) high levels of managerial discretion in the selection and resourcing of NPD projects; (2) a very high risk of project failure; (3) managerial relationships with top management; (4) the functional managers have to deal with other functional units to achieve their goals. Functional managers enter the political arena as they jockey for position, in order to secure the best outcomes for themselves and their departments [68].

Several NPD researchers have examined this socio-political perspective of NPD work [27,28,29,69], where political activity has both positive and negative effects in organizations. From a manager’s perspective, Allen et al. [70] argued that determining whether or not a functional counterpart’s political behavior is likely to be positive (e.g., assisting and accommodating) or negative (e.g., blocking or hindering) is of great importance for interdependent managers. In the case where managers experience what they consider negative project politics, we accordingly hypothesize that:

Hypothesis 4a (H4a).

The higher the level of negative project politics, the lower the level of collaboration with the counterpart manager during the NPD project.

Hypothesis 4b (H4b).

The higher the level of negative project politics, the lower NPD project success.

3.3.3. Dysfunctional Conflict

March and Simon [71] defined conflict as the “breakdown of the standard mechanisms for decision-making.” (p. 891). When two parties interact, there are inevitably going to be differences of opinion or conflict. In the conflict literature, Wall and Callister [72] review the literature and examine conflict causes and core processes, as well as their effects; in a particularly useful summary table (p. 519), they listed individual factors, interpersonal factors, communications, behavior, structure, previous interactions, and issues as all possible antecedents of interpersonal conflict, clearly highlighting the complexity of the issues at hand. However, in terms of interpersonal relationships, they highlight that, when an opponent is viewed as blocking a person’s goals and conflict ensues, negative emotions, such as anger, stress, and less than favorable perceptions of that party, in terms of distrust, more misunderstandings, not being able to see the other person’s perspective, and questioning the opponent’s intentions, ensue. This type of conflict in the literature is viewed as relationship conflict, and it is often referred to as destructive or dysfunctional conflict. This is the negative side of interpersonal conflict; we use Menon et al.’s [73] definitions of conflict and define dysfunctional conflict as unhealthy behaviors within an organization, such as distortion, withholding information to hurt other decision makers, hostility, and distrust during interactions. In his seminal work, Souder [18,19] found that, in the worst-case relationships between R&D and marketing, there was intense conflict with deep-seated jealousies, negative attitudes, fears, and hostile behaviors. Accordingly, we hypothesize that:

Hypothesis 5a (H5a).

The greater the level of dysfunctional conflict with the counterpart manager, the lower the level of collaboration.

Hypothesis 5b (H5b).

The greater the level of dysfunctional conflict with the counterpart manager, the lower the level of NPD success.

3.3.4. Collaboration’s Role in NPD Work

We argue that having functional counterparts collaborate on new product projects is the ultimate goal for NPD systems and processes. Jassawalla and Shashittal [46] described collaborative working relationships as ones where information sharing occurs freely and effectively, as both parties contribute positively to the working relationship and have a higher-level working relationship that is characterized by volitional behaviors (e.g., information sharing and mutual accommodation). Importantly, they found that collaborative relationships develop with experience of each, on the basis of effective cooperation supporting the role of social exchange theory in NPD work. Homburg and Jensen [74] highlighted that cooperation between sales, marketing, and R&D at the early stages of NPD projects is beneficial to NPD success. We argue that, as the NPD process is essentially informational in nature, it involves significant communication on project-related issues [24]. Collaborative working relationships, which are trust-based by definition, overcome many of the problems associated with ineffective functional integration. We adapt Kahn’s [31] measure of interdepartmental collaboration to capture the extent to which functional managers display collaborative behaviors during the NPD project. As such, the extent of collaborative behavior is a psychosocial measure we use to assess the perceptions of functional managers who interact with others from differing functional areas. Specifically, it measures whether they perceive their working relationship to be a mutual process with collective goals, informal work activity, a free exchange of ideas and information, and teamwork. This view is supported by empirical evidence, where positive relationship dynamics, such as effective communication, cooperation, and, ultimately, collaboration, result in a positive effect on NPD success rates [46,75,76,77,78]. Accordingly, we hypothesize:

Hypothesis 6 (H6).

The greater the level of collaboration, the greater the level of new product success.

4. Research Method

4.1. Sample and Data Collection

As the purpose of the study was to seek a key technical informant who had been involved in a working relationship on NPD projects with a marketing manager or key marketing decision-maker, the first step was to obtain a suitable mailing list from a commercial supplier with the following criteria: (1) a list of companies that had the job titles of R&D manager, engineering manager, or manufacturing manager; (2) that these companies also employed a marketing manager. From here, the companies were screened for NPD activity, and the technically trained managers were contacted to seek their permission to participate. A commercial supplier provided the final mailing list comprising 813 companies, with an annual turnover greater than AUD $10 million and minimum of 50 employees, predominantly from the manufacturing sector. Interestingly, in many cases, the managers only agreed to participate if the dyads were not going to be used. They did not want their counterpart knowing that they had filled out a survey about them. While frustrating the researchers, the sensitivity that the managers displayed to the topic minimized the possibility of social desirability bias in their responses. After permission was granted, a self-reported mail-out questionnaire was sent out, with call-backs from a team of research assistants. The first mail out sample collected was the R&D manager’s view of the working relationship (Sample 1), and this was replicated from a marketing manager’s perspective in the second mail out Sample 2.

Sample 1: Technically trained managers as respondents (R&D, engineering, and manufacturing managers). These managers were contacted to determine: (1) if they had participated in any NPD projects over the last 3 years; (2) whether they had significant involvement with the marketing manager during this project. In total, 184 managers returned a questionnaire from the 337 managers contacted, for a net response rate of 54.6%. R&D managers comprised 40.2%, engineering managers comprised 20.1%, manufacturing managers comprised 31.5%, and other technically trained managers comprised 7.1%.

Table A1. Respondents by industry type, Samples 1 and 2.

Most Respondents	Industry Type	Sample 1 Technically Trained Manager Study n = 184 Percentage of Sample %	Sample 2 Marketing Manager Study n = 145 Percentage of Sample %
1	Building materials	12.0	19.0
2	Food	10.9	15.0
3	Machinery manufacturer	9.8	8.0
4	Automotive components	7.6	2.9
5	Electrical equipment and components	4.3	6.0
6	Medical/ pharmaceutical	6.5	0.01
7	All other manufacturers	48.9	49.1
Total		100.0	100.0

provides the industries from which both samples came from.

As we summated their responses for SEM data analysis, in order to ensure that there were no distinct functional differences in the manager’s patterns of responses, a multivariate analysis of variance (MANOVA) was performed across all of the key variables posited in the conceptual model. We also tested for Type I error, using the Bonferroni correction at the conservative level of α = 0.5 [79], and found no statistically significant differences between the groups.

Sample 2: Marketing managers as respondents. In total, 145 managers responded, from the 294 marketing managers that agreed to participate, resulting in a net response rate of 49.3%. Respondent titles included marketing manager (50.3%), sales and marketing manager (34.5%), and sales manager (4.1%); a total of 11.1% were in the other category. Most firms were goods producers (95.8%), while the remainder (4.2%) were software producers. Again, we tested for Type I error using the Bonferroni correction, and the results indicated that there were no statistically significant differences found between the groups.

4.2. Measurement Refinement

We included four reflective multi-item (negative project politics, affect-based trust, dysfunctional conflict, and collaboration) and three formative (negative motives and intentions, defensive behaviors, and NPD success) constructs to test our model. All but negative project politics were existing scales. A listing of the final items is shown in

Table A2. Operational measures.

Construct	Items	Adapted from
Negative motives and intentions of the counterpart manager	Seven-point scale anchored by 1 “Completely Disagree” and 7 “Completely Agree.”A (1) there were few hidden agendas in our work*; (2) neither of us had to wonder about the purpose behind the other’s behavior*; (3) s/he acted with good intentions*; (4) s/he often had ulterior motives; (5) s/he would use me if it benefited him/her.	Smith and Barclay, [49]
Perceptions of negative project politics by the counterpart manager	Seven-point scale anchored by 1 “Completely Disagree” and 7 “Completely Agree.” Respondents were asked the extent to which: (1) the other manager and I often played politics against each other; (2) I spent a lot of my time “covering my back” because of the other manager’s politics.	New Scale
Affect-based trust	Seven-point scale anchored by (1) “completely disagree” and (7) “completely agree”. McAllister [54]: (1) ours is a relationship in which we both freely share our ideas, feelings, and hopes; (2) I can talk openly to him/her about difficulties that I’m having at work and know he/she will want to listen; (3) if I shared my problems with him/her, I know that he/she would respond constructively and with understanding.	McAllister [54]
Defensive behavior against opportunism	Seven-point scale anchored: 1 “Completely Disagree” and 7 “Completely Agree.” (1) I documented all aspects of my discussions with the other manager regarding this project; (2) I monitored changes in the project situation because the other manager would definitely take advantage of such changes to my detriment; (3) I worked openly with the other manager because s/he would not take advantage of me *; (4) I shared information cautiously with the manager to avoid it being used against me.	Bromiley and Cummings [60]
Dysfunctional conflict	Seven-point scale anchored: 1 “Completely Disagree” and 7 “Completely Agree.” When the two of us got together in group meetings, tensions between the two of us frequently ran high; (2) I generally disliked having to work with him/her; (3) there were no disagreements between myself and the other manager over the running of this project *; (4) throughout the project, there was little interpersonal conflict between myself and the other manager *.	Menon and Jaworski [96]
Collaboration	Seven-point scale anchored by 1 “Completely Disagree” and 7 “Completely Agree.” Respondents were asked the extent to which: (1) the other manager and I achieved project goals collectively; (2) the other manager and I had a mutual understanding about the project development process; (3) the other manager and I informally worked together on project matters; (4) the other manager and I freely shared ideas, information, and/or resources on project matters; (5) the other manager and I work together as a team.	Kah [31]
New product success	Seven-point scale anchored by 1 “Completely Disagree” and 7 “Completely Agree.” Respondents were asked: (1) the NPD project achieved its budget objectives; (2) the NPD met its time schedule objectives; (3) in terms of contribution to sales, the new project was successful; (4) in terms of contribution to profit, the new project was successful; (5) the overall performance of this NPD project met our objectives.	Griffin and Page [43]

An * indicates reverse coded item, italicized indicates it was deleted from the analysis.

. Means, standard deviations, correlations, and internal consistencies of the constructs are shown in

Table A3. (a) Sample 1, technically trained managers (TTM) viewpoint of marketing manager (MM). (b) Sample 2, marketing managers (MM) viewpoint of technically trained managers (TTM). Means, standard deviations, correlations, and internal consistencies of constructs.

(a)
			Internal									Consistency
Correlations of Constructs
	No. of Items	Mean	Std Dev.	Alpha	Composite Reliability	Neg Motive and Intentions	Affect-Based Trust	Negative Project Politics	Defensive Behavior	Dysfunct Conflict	Collab	NPD Success.
Negative motives and intentions a	5	2.12	1.66	n.a	n.a	1
Affect-based trust	3	4.83	1.54	0.92	0.95	−0.55	1
Negative project politics	2	4.7	1.47	n.a	0.95	−0.20	−0.06	1
Defensive behav a	4	2.3	1.55	n.a	n.a	0.66	−0.05	0.25	1
Dysfunctional conflict	3	3.06	1.68	0.73	0.83	0.65	−0.44	−0.19	0.58	1
Collaboration	5	4.9	1.28	0.92	0.94	−0.64	0.69	−0.19	−0.63	−0.58	1
NPD success a	5	5.1	1.53	n.a	n.a	−0.40	0.50	0.04	−0.43	−0.44	0.55	1
(b)
			Internal
Correlations of Constructs												Consistency
	No. of Items	Mean	Std Dev.	Alpha	Composite Reliability	Neg Motive and Intentions	Affect-Based Trust	Negative Project Politics	Defensive Behavior	Dysfunct Conflict	Collab	NPD Success.
Negative motives and intentions a	5	2.4	1.54	n.a	n.a	1
Affect-based trust	3	5.0	1.67	0.92	0.93	−0.55	1
Negative project politics	2	1.8	1.12	n.a	n.a	0.53	−0.36	1
Defensive behav a	4	2.5	1.59	n.a	n.a	0.61	−0.45	0.55	1
Dysfunctional conflict	3	2.3	1.70	0.73	0.84	0.65	−0.45	0.50	0.55	1
Collaboration	5	5.6	1.34	0.92	0.93	−0.48	0.71	−0.48	−0.57	−0.58	1
NPD success a	5	5.0	1.55	n.a	n.a	0.02	−0.19	0.04	0.38	0.02	−0.06	1

^a denotes a formative measure.

.

The negative project politics construct was developed by using Churchill’s [80] procedure, where four items were used in the survey. As part of the data testing process (e.g., checks for multicollinearity, skewness, and kurtosis), we performed principle components analysis and confirmatory factor analysis to ensure that the four scale items were highly correlated and uni-dimensional ([81], p. 601). The results indicated that the four items were not a single construct, with two items each loading cleanly on what appeared to be two separate constructs, i.e., positive and negative project politics. An examination of the split items showed that both constructs were theoretically viable and had face validity; we used the two items measuring negative project politics in our model.

For all our measures, reliability analysis revealed that the composite reliability and alpha coefficients for all of the scales exceeded 0.73, which is above the generally accepted benchmark of 0.70 [82]. Two methods were used to establish convergent validity. Examining the t-values of each item from the partial least squares (PLS) output found them all statistically significant [83], and the average variance extracted (AVE) for each of the reflective constructs exceeded 0.50 [84]. Details for each scale are in Table A3. We also examined discriminant validity by first using Fornell and Larcker’s [84] approach, where discriminant validity is established if the squared multiple correlation of two variables is less than the average variance extracted (AVE). Then, we used Chin’s [85] approach, where the individual factor loadings of the items are examined to determine that all factors loaded cleanly on the construct of interest and no other items had a higher cross-loading with them. No concerns relating to cross-loading were raised, with all items loading cleanly on their constructs.

4.3. PLS Structural Model Results

We used partial least squares (PLS) to estimate our structural model, as our primary concern was the prediction of exogenous variables [85,86]. PLS was appropriate, since we used both formative and reflective measures; no assumptions were made about multivariate normality, and our final sample sizes were not large (Sample.1, n = 184) and (Sample 2, n = 145). We used SmartPLS 2.0 [87] to estimate our structural model. In order to establish the stability and significance of our parameter estimates, we computed the t-values using 200 bootstrap samples.

As the data collected was self-reported from single informants for both studies, common method bias (CMV) may have been an issue [88]. In a recent paper, Tehseen et al. [89], reviewing the way in which CMV is dealt with in the entrepreneurship literature, suggested the use of either procedural or statistical procedures. For a single respondent, they argued for the use of Chin et al.’s [90] measured latent marker variable approach, which uses a latent marker variable from an unrelated theoretical construct to test whether its addition to the specified path model would result in a change of greater than 5% in the R² of the original model. In both studies presented here, the R² change was below 1%, indicating that the data was not affected by CMV.

• Sample 1: The effect of variables—technically trained manager’s viewpoint of the marketing manager

Of the 15 hypothesized relationships, 14 hypotheses were supported (refer to

Table 1. PLS model testing results.

			Sample 1	Sample 2
			TTMs	MMs
Hypotheses			Std. Beta (t-Value)	Std. Beta (t-Value)
Negative motives and intent	→ Defensive behav.	H1a (+)	0.607 (16.665) ***	0.526 (2.943) ***
	→ Neg project politics	H1b (−)	0.233 (3.212) ***	0.474 (8.757) ***
	→ Dysfunct. conflict	H1c (+)	0.578 (18.506) ***	0.573 (11.839) ***
	→ Collaboration	H1d (−)	−0.177 (5.768) ***	−0.224 (4.295) ***
Affect-based trust	→ Defensive behav	H2a (+)	−0.273 (9.122) ***	−0.164 (2.237) *
	→ Neg project politics	H2b (−)	0.083 (1.077) ns	−0.102 (1.932) **
	→ Dysfunct. conflict	H2c (+)	−0.122 (3.596) ***	−0.139 (2.943) *
	→ Collaboration	H2d (+)	0.417 (13.411) ***	0.451 (11.638) ***
Defensive behavior	→ Collaboration	H3a (−)	−0.187 (5.780) ***	−0.112 (2.307) *
	→ NPD success	H3b (−)	−0.101 (2.507) *	0.631 (3.482) ***
Negative project politics	→ Collaboration	H4a (−)	−0.052 (1.976) *	−0.075 (1.675) *
	→ NPD success	H4b (−)	0.178 (3.762) ***	−0.183 (2.116) *
Dysfunctional conflict	→ Collaboration	H5a (−)	−0.178 (6.284) ***	−0.138 (2.934) *
	→ NPD success	H5b (−)	−0.169 (4.304) ***	−0.183 (2.296) *
Collaboration	→ NPD success	H6 (+)	0.430 (12.373) ***	0.104 (2.229) *
R2	TTMs	MMs
Def. behavior	0.484	0.397
Dysfunct conflict	0.427	0.434
Collaboration	0.620	0.651
NPD success	0.364	0.222

*** Sig. ≤ 0.001 (one-tailed); ** ≤ 0.01; * ≤ 0.05.

). Our structural model showed good predictive ability, where the R² for defensive behaviors was 0.484, for dysfunctional conflict the R² = 0.427, the R² for collaboration was 0.620, and the effect of collaboration between the R&D manager and marketing manager on new product success resulted in an R² of 0.365. These results suggest that the model has good explanatory power.

• Sample 2: The effect of variables—marketing manager’s viewpoint of the technically trained managers

Of the 15 hypothesized relationships, all were supported (refer to Table 1). Our structural model showed a good predictive ability, where the R² for defensive behaviors was 0.397, for dysfunctional conflict the R² = 0.434, the R² for collaboration was 0.651, and the effect of collaboration between the R&D manager and marketing manager on new product success resulted in an R² of 0.222.

5. Discussion

Our study examines the effects of the negative aspects of working relationships in a critical area of organizational work, i.e., new product development. It is of value for engineering, manufacturing, and R&D (technically trained managers) working relationships with marketing, as it finds that, while negative relationships are in the minority, when they do arise, there are significant negative consequences. Both samples find that our models do have high predictive validity, and the variables do explain much of the variance in the models. There are counter-intuitive findings and results that do differ, depending on the functional perspective examined.

Our first finding is that when a peer manager is perceived to have negative motives and intentions, not surprisingly, nothing good eventuates from this perception. While not a surprising result, the strength of the relationship with defensive behaviors and dysfunctional conflict was surprisingly high, when examined against negative project politics and collaboration. It seems that the subjective assessment of another manager is still important in an era where formalized processes (quality functional deployment and stage-gate) try to ensure co-operation and interaction between functional specialists [91,92]. This finding emphasizes that for some managers, NPD work is still a fundamentally relationally-based process, where risk is to be avoided in working relationships.

Of interest, when examining the effect of negative motives and intentions, is that, while technically trained managers and their marketing counterparts both saw a strong significant negative effect on negative project politics, it is with the marketing managers that its effect was stronger. This finding supports the argument that, for marketing, breaking down the walls to fortress R&D and functional silos [75] is politically difficult, as technically trained managers have longer terms of tenure and often more established social and power networks within the firm. Adding to this effect may be the fact that, as boundary spanners, marketing managers are more sensitive to the resistance from other areas and more cognizant of what is required to work with a hostile counterpart.

To add contrast to the negative elements of working relationships, we included affect-based trust in our model and found that, generally, affect-based trust did reduce the negative effects in working relationships, and it had a strong positive relationship with collaboration. Again, there were interesting functional effects, where, for technically trained managers, affect-based trust had a stronger effect on their defensive behaviors, thus indicating that the care and concern shown in affect-rich relationships is important for facilitating cooperation. Interestingly, for marketing managers, affect-based trust reduced negative project politics; however, this was not the case for technically trained managers (see Table 1).

The construct of defensive behaviors against opportunism was expected to have uniform negative effects on our outcome variables for both functions, as Ashforth and Lee [55] provide considerable theoretical support for these hypothesized relationships. Collaboration was negatively affected by defensive behaviors against opportunism for both marketing managers and their technically trained counterparts. However, for marketing managers, defensive behaviors against opportunism leads to NPD success. When examined closely, this result is supported by the literature, where many of the criticisms of marketing have been regarding their lack of professionalism [20], poor quality of information received by R&D [30,93], and lack of understanding of what engineering, manufacturing, and R&D constraints are [94,95]. So, it seems that when marketing managers feel threatened by their functional counterparts, their defensive behaviors lead to ensuring that their project commitments are met, thus trying to avoid any potential criticisms. For the technically trained managers, defensive behavior on their part may mean that many of the key customer inputs (e.g., customer information and adoption forecasts) that marketing bring to a project are not utilized, as there is a lack of trust evident; thus, a subsequent negative effect on NPD project success.

Dysfunctional conflict between functional managers behaved exactly as expected, with negative effects on both collaboration and NPD success, suggesting that eliminating it is still a necessary goal for effective working NPD relationships. Collaboration’s effect on NPD success was also as expected, with a positive effect on NPD success. Interestingly, marketing managers do not view collaboration as important as technically trained managers do, who reported a much stronger positive effect on NPD success.

6. Implications and Conclusions

Our goal was to determine whether or not the negative side of marketing’s working relationships still held theoretical salience and managerial importance for an NPD process, which has evolved considerably from the 1970s and 1980s, where it was characterized by departmental disharmony, inter-functional rivalry, and conflict to a more sophisticated process, as evidenced by formalized NPD processes limiting the risk inherent in the process and allowing for greater opportunity for collaborative behaviors. The good news is that, on the whole, marketing’s working relationships with their technically trained counterparts are not severely impacted by negative factors, but, in contrast, are affect-laden and collaborative in nature.

While, in general, the working relationships are good, in the instances when they were not, we found that there were some very interesting differences in perceptions between marketing managers and their technically trained counterparts. We empirically confirmed an intuitive finding that negative motives and intentions, as well as suspicion and uncertainty in a counterpart’s motives, in a working relationship are not a good thing. The belief that a counterpart will act opportunistically is still a concern to some managers involved in joint NPD work.

By taking a deliberately negative perspective in our modeling, we show that, despite this evolution in NPD sophistication in many firms, for some managers, having to cooperate with other functional managers on NPD projects makes them very sensitive to their counterpart’s motives and intentions. Our results highlight that a social exchange lens does still have theoretical relevance as managers try to obtain enough social data to determine whether the relationship will be worthwhile to pursue or they should adopt defensive behaviors against opportunism. The concept of trust and suspicion are relevant today. However, our results suggest that NPD work is handled in a more professional, cooperative, and collaborative manner than the previous literature has suggested.

Our study also highlights the complementarity of social exchange theory and its use of trust, as well as the interaction approach, as appropriate theoretical lenses to study the working relationships in organizations. Lewicki et al. [56] argue that, for managers in modern organizations, there are significantly more reasons to distrust, “where the challenges of speed, quality, global reach, which require trust, also have precipitated distrust, through corporate restructuring, downsizing, and fundamental violations of the psychological contracts connecting individuals with organizations” (p. 439).

For marketing managers, realizing that their technically trained managers respond to affective laden gestures during NPD interactions and appreciate collaboration as a work mode can help them quickly overcome any perceptions that they bring to the relationship regarding negative motives and intentions. This is very important, as it is often marketing managers who are the new kid on the block and have to establish that they come in peace. Additionally, understanding that technically trained managers do react very strongly when feeling threatened or that they are in conflict situations suggests that preventing a slide into negative activities is mutually beneficial.

While the thought of negative behaviors affecting NPD activity in the firm holds an intrigue, our findings suggest that the decades of research, process improvement, and benchmarking studies into effectively managing and integrating functional specialists have had a beneficial effect. The very nature of NPD work makes the case for functional specialists to work together in a manner where they are free from the risk of opportunism and decision-making can be fast, efficient, and unanticipated problems are dealt with quickly, without second guessing the motivations and intentions of their counterparts. It is imperative that top management ensure that NPD processes continue to support collaborative practices where functional managers feel that they are being supported in a non-threatening manner in their NPD activities.

As with any study, limitations exist. Cross-sectional studies are only a snapshot in time. It would be useful to conduct a longitudinal study to examine trust development at various stages of a project. Similarly, a greater examination of what types of defensive behaviors are used by the differing managers would possibly shed light on their perceptions of exactly what are considered as negative actions by their counterpart.