In many industries, the salesforce plays a key role in defending firms’ price levels. In fact, price defense is—ceteris paribus—the sales task with the most immediate and strongest financial impact; in the short term, an additional average discount of 1% decreases a company’s profit by 11% (Marn et al. 2004). Given the potential impact of price defense on profit, a common, long-standing concern of managers is that salespeople may grant discounts too easily (Joseph 2001): almost 40% of sales managers think that their salesforce “needs improvement” in its ability to avoid discounting (CSO Insights 2011) and “avoiding discounting” is among the top five metrics used to measure the performance of and sales managers (CSO Insights 2014).

Given practitioners’ interest in this topic, marketing researchers have put an increasing emphasis on understanding the phenomenon of salespeople’s discounting or price defense. We define the intensity of salesperson price defense behavior as a salesperson’s effort invested in a price negotiation to refute a customer’s discount demand (Hüffmeier et al. 2014). For instance, if intensity of price defense is high, a salesperson may invest great effort in explaining to a customer why a specific price is justified, and he may not easily concede a discount to the customer. Conversely, if intensity of price defense is low, the salesperson may not counter a customer’s discount claim with much effort, but willingly comply with the customer’s request.

Marketing researchers have studied both consequences and determinants of salespeople’s price defense behavior. As to the first, researchers have revealed ambivalent consequences of price defense behavior: while defending prices increases the margin of a successful transaction, it may also induce customers to refrain from purchasing products they would have purchased otherwise or weaken customer relationships (e.g., Dwyer et al. 1987; Ganesan 1993; Weitz 1981; Wieseke et al. 2014). In total, whether salespeople’s price defense behavior is beneficial or harmful to firms’ overall financial performance may be subject to various contingencies, such as either a differentiation or cost-based firm strategy (e.g., Slater and Olson 2000), customer attributions of why he or she received a discount (e.g., Darke and Dahl 2003), customers’ price sensitivity (e.g., Wieseke et al. 2014), and salespeople’s customer-oriented behaviors (e.g., Alavi et al. 2016).

Second, researchers have helped companies understand the management levers that determine salespeople’s price defense behavior—hereby refraining from the normative question of whether price defense should or should not be fostered (see Fig. 1 for a summary literature overview). For example, works in this stream of literature have found that the intensity of salespeople’s price defense is influenced by the firm’s incentive scheme (e.g., Joseph 2001; Lal 1986; Weinberg 1975, 1978), the degree to which salespeople are authorized to negotiate prices (e.g., Bhardwaj 2001; Desai and Purohit 2004; Homburg et al. 2012; Stephenson et al. 1979; Wilken et al. 2010), the provision of cost information to salespeople (e.g., Wilken et al. 2010), as well as specific salespeople skills and behaviors (e.g., Alavi et al. 2016; Huang et al. 2010; Kumar et al. 2016; Wieseke et al. 2014).

Fig. 1
figure 1

Overview of relevant research streams for our study

Full size image

Interestingly, regarding determinants of salespeople’s price defense behavior, to our best knowledge, no works exist that examine how superiors’ leadership affects salespeople’s price defense behavior (see Fig. 1: Research void 1). We regard addressing this research void as important from both an academic and a managerial perspective. First, academics have argued that leadership is a key factor influencing salespeople’s behavior (e.g., Boichuk et al. 2014; Dubinsky et al. 1995; MacKenzie et al. 2001; Martin and Bush 2006; Panagopoulos and Avlonitis 2010; Schmitz and Ganesan 2014; Shannahan et al. 2013a, b). It thus seems likely that a superior’s leadership also assumes a prominent role in increasing or decreasing salespeople’s price defense behavior. If this is the case, examining this role is pertinent to contribute to a thorough academic understanding of the phenomenon of salespeople’s price defense behavior. Therefore, the unique contribution of this paper to explore the interplay of leaders’ role and salespeople’s characteristics on salespeople’s price defense. Second, if practitioners intend to increase salespeople’s price defense behavior, doing so through leadership represents a cost-efficient alternative to other, more costly approaches, such as the adjustment of the incentive scheme. Therefore, investigating the impact of leadership on salespeople’s price defense behavior may not only expand knowledge on determinants of salespeople’s price defense behavior, but may likewise provide valuable guidance to managerial practice.

Building on these notions, the goal of our study is to provide insights into the impact of leadership on salespeople’s intensity of price defense behavior.Footnote 1 We deduce our hypotheses from social learning theory (Bandura 1971) and the contingency leadership perspective (Den Hartog and Belschak 2012). First, drawing on social learning theory, prior literature finds evidence that salespeople are likely to adopt their leaders’ attitudes and behaviors (e.g., Chakrabarty et al. 2013; Lam et al. 2010), particularly if leaders are transformational (e.g., Mullins and Syam 2014; Rich 1997; Wieseke et al. 2009, 2011; see bottom of Fig. 1 as well as Table 1 for details). Drawing on these notions, we deduce from social learning theory that the degree to which salespeople defend prices may depend on sales leaders’ role modeling, that is, the degree to which sales leaders defend prices themselves. Second, and in addition, we propose that this effect is particularly pronounced if leaders exhibit transformational leadership. Hereby, transformational leadership is defined as a leader’s exhibition of idealized influence, arousing of inspirational motivation, provision of intellectual stimulation, and treatment of followers with individualized consideration (Avolio et al. 1999).

Table 1 Selected literature on social learning in sales research as conceptual basis
Full size table

Second, importantly, while we expect to find the aforementioned effects on average, they are unlikely to occur imperatively. Specifically, the contingency leadership perspective (e.g., Den Hartog and Belschak 2012; Fiedler 1978; Li et al. 2013; Yun et al. 2006) suggests that the effect of leaders’ behavior on their followers also depends on followers’ characteristics (see also Graen and Uhl-Bien 1995). This is also true for the sales context, where exchanges between sales manager and salespeople are dyadic in nature, depending on specific leader and salesperson characteristics (e.g., Schwepker 2017; Ahearne et al. 2005). However, strikingly, with few notable exceptions (e.g., Shannahan et al. 2013a, b), the interplay between sales leaders’ transformational leadership and salesperson-related variables has remained unexplored so far, also in the context of price negotiations (see Fig. 1: Research void 2). To provide an accurate and valid account of the effects of a leader’s behavior on a salesperson’s price defense behavior, we thus argue that salespeople’s characteristics need to be necessarily included in our research model. Therefore, to contribute to knowledge on the contingency leadership perspective in sales, we deduce such salesperson characteristics from the motivation–ability–opportunity (MAO) framework, which is a well-established theoretical basis for explaining human behavior (Siemsen et al. 2008) and has recently been applied to sales settings (Schmitz 2013).

To test our hypotheses, we conducted a field study in the automotive retailing context. Based on a dataset comprising three data sources, that is, a survey of 92 salespeople matched to 264 interaction-specific responses of these salespeople and their customers, we ran a multi-level path model to assess the interactive effects of leaders’ intensity of price defense, leaders’ transformational leadership style, and salespeople’s characteristics. Results fully corroborate our theoretical predictions. We find that the transfer of leaders’ price defense behavior to salespeople’s price defense behavior does not occur unconditionally but strongly hinges on the extent of leaders’ transformational leadership style. Moreover, this positive effect of transformational sales leaders’ on salespeople’s price defense depends on salespeople contingency factors delineated from the MAO framework (specifically, salespeople’s learning motivation, salespeople’s negotiation efficacy, and salespeople’s perceived customer negotiation lenience).

Our study makes at least three contributions to sales and marketing research. First, our study is the first to reveal that sales leadership strongly influences salespeople’s intensity of price defense, thus providing insight into this important research void (see Fig. 1). Specifically, as deduced from literature on salespeople’s adoption of their leaders’ behaviors (e.g., Lam et al. 2010; Mullins and Syam 2014; Wieseke et al. 2009, 2011) and empirically shown in our study, salespeople tend to adopt their leaders’ intensity of price defense, especially if leaders are transformational. Second, significantly extending prior literature, our study caters to calls for research on contingencies of the efficacy of leadership behavior (e.g., Avolio 2007; Hunter et al. 2007; Podsakoff et al. 1996). Specifically, we clarify that leaders’ role modeling of price defense behavior in combination with transformational leadership is not sufficient to ensure all salespeople’s price defense behavior. Instead, in the context of price negotiations, the effectiveness of role modeling in combination with transformational leadership strongly hinges on salespeople’s learning motivation, salespeople’s negotiation efficacy, and salespeople’s perceived customer negotiation lenience. Conceptualizing and empirically validating these contingencies significantly contributes to marketing research since they cannot be deduced from prior empirical studies on salespeople’s adoption of their leaders’ role modeling. Instead, we theoretically deduced these contingencies for the context of price negotiations by combining social learning theory with the MAO framework.

Third, we contribute to negotiation literature. In a recent comprehensive review of the negotiation literature, Herbst et al. (2011) demand an enhanced focus on holistic price negotiation models accounting for organizational influences on salespeople’s bargaining behavior. We address these omissions in previous research by exploring the intricate role of leaders in price negotiations among salespeople and customers.

Moreover, our findings have actionable implications for sales leaders in price negotiation–intense contexts. Most importantly, sales leaders need to be aware that their own negotiation behavior and leadership style likely influences their salespeople’s price defense behavior. Notably, this influence is unlikely to be homogeneous for all salespeople, but may strongly depend on an individual salesperson’s learning motivation, negotiation efficacy, and perceived customer lenience. Thus, if sales leaders intend to alter their followers’ negotiation behavior, they may (a) carefully adjust their own negotiation behavior and leadership style to each salesperson, or (b) aim at influencing their salespeople’s learning motivation and negotiation efficacy to match their own negotiation behavior and leadership style.

Conceptual framework and hypotheses development

In the following, we describe our conceptual framework and derive corresponding hypotheses. Figure 2 summarizes our conceptual framework, and the measurement table in the Appendix provides precise definitions and measurement levels for all constructs employed in the framework. Using a social learning lens (Bandura 1977) and the contingency perspective on leadership (e.g., Yun et al. 2006; Fiedler 1978), our conceptual model posits the following: (1) The interpersonal influence process by which leaders affect salespeople’s price defense behaviors rests on role modeling, i.e., salespeople’s imitation of their leaders’ price defense behavior. (2) This effect interacts with the leader’s transformational leadership style, reinforcing the role modeling effect. Finally, (3) the combined effect of role modeling and transformational leadership style is augmented by situational factors rooted in the MAO framework, i.e., salespeople’s learning motivation, perceived negotiation efficacy, and perceptions about customers’ negotiation lenience. In what follows, we present our hypotheses development starting with the main effect of leaders’ on salespeople’s intensity of price defense (H1).

Fig. 2
figure 2

Overview of the conceptual framework. Notes. SP = Salesperson; the model comprises three data sources: (1) salespeople data from a one-time survey (level 2); (2) salespeople’s specific responses concerning one sales encounter with a customer (level1); and (3) customers’ responses concerning the specific encounter with the salesperson (level 1). t = 1: First measurement wave, in which we conducted surveys with salespeople on components of the MAO framework and perceptions about leader behavior and style. t = 2: Second measurement wave, in which we surveyed the salespeople from t = 1 and their customers with respect to a specific sales encounter and price negotiation, immediately after the encounter ended. * As a robustness check, we aggregate salespeople’s perception of their leader’s behavior on the leader-level (level 3) and replicate our main model (see Table 3). ** For an overview of which specific control factors we included please refer to Table 3 and the model specification section

Full size image

Social learning theory: Leaders’ influence on salespeople price defense through role modeling

Research suggests that role modeling is an effective means for sales leaders to influence their salespeople’s behavior (Lam et al. 2010; Wieseke et al. 2009). Rich (1997) defines role modeling as a “behavior on part of the sales manager perceived by the salesperson as appropriate to follow that is consistent with both the values the sales manager espouses and the goals of the organization” (Rich 1997, p. 320). The reason for salespeople to adopt leaders’ behavior is rooted in social learning theory (Bandura 1969a, b, 1971). Social learning theory posits that individuals may acquire new behaviors by observing and imitating others. Social learning theory has been previously applied to the sales context and proved useful to understand sales leader–salesperson relationships (see Table 1 for an overview of respective literature). For instance, salespeople have been found to imitate their managers’ technology acceptance behavior (Homburg et al. 2010), market orientation (Lam et al. 2010), work motivation (Wieseke et al. 2011), psychological climate (Martin and Bush 2006), and even adopt their leaders’ organizational identification (Wieseke et al. 2009).

However, a core tenet of social learning theory is that individuals do not automatically and naively adopt observed behaviors but instead evaluate the consequences of the observed behavior (Bandura 1977), which is referred to as vicarious reinforcement (Bandura 1977). Importantly, individuals tend to imitate behaviors that they observed if they expect the consequences of these behaviors to be rewarding for them. As Conger and Kanungo (1987, p. 642) put it, employees adopt their leader’s behavior that they deem “worthy of imitation.”

Applied to the salesforce context, we propose that salespeople may hold particularly high reward expectations for adopting their sales leaders’ behavior and are thus likely to follow their leaders’ example (Rich 1997). Adopting their sales leaders’ behavior should be perceived as rewarding by salespeople because sales leaders occupy a “position of power and legitimacy in the sales organization” (Mathieu et al. 2007, p. 530). If sales leaders exhibit a specific behavior, it may signal to salespeople that adopting this behavior is useful to achieve success in the sales organization (Mathieu et al. 2007).

This reasoning should specifically apply to salespeople’s adoption of their sales leaders’ price defense behavior. Salespeople’s price defense behavior should be particularly susceptible to sales leader influences because in the task environment of price negotiations, salespeople are confronted with various conflicting interests of the firm, the customer, and themselves (Alavi et al. 2016; Ganesan 1993). These conflicting demands in the realm of price negotiations can trigger salespeople’s uncertainty and consecutively their need for guidance (House 1996). Consequently, the sales leader as an experienced role model may prove a valuable source of orientation for salespeople regarding the intensity of price defense behavior they should exhibit. Specifically, salespeople should expect imitating their leaders’ intensity of price defense as rewarding because salespeople should perceive it as conducive to their own negotiation performance. Based on this reasoning, we hypothesize that:

  • H1: Salespeople’s perceptions of their leaders’ intensity of price defense behavior is positively related to salespeople’s intensity of price defense behavior.

Impact of transformational leadership on salespeople’s adoption of leader price defense behaviors

Among the potential role models to choose from, attractive models capture a learner’s attention (Brown and Treviño 2014). Transformational leaders are typically perceived as more attractive by followers and imitating their behavior as more rewarding for salespeople (Wieseke et al. 2011; Piccolo and Colquitt 2006; Martin and Bush 2006; Rich 1997). Transformational leadership in the sales context implies showing individual consideration for salespeople, intellectually stimulating as well as motivating them, and communicating a compelling vision for the organization (Shannahan et al. 2013a, b; MacKenzie et al. 2001). Seeing these facets of transformational leadership, transformational sales leader have been found to develop high quality relationships with salespeople in the sense of leader–member exchange theory (Schwepker 2017; Piccolo and Colquitt 2006; Graen and Uhl-Bien 1995). In this respect, prior research clarified that these facets of transformational leadership considerably improve salespeople’s perceptions of their managers in different areas (Martin and Bush 2006): transformational sales leaders build closer relationships with salespeople (Smith et al. 2012), are perceived as more trustworthy, are more respected, and perceived as more competent (Shannahan et al. 2013a, b; Antonakis et al. 2011; MacKenzie et al. 2001).

Seeing their strong standing with their salespeople and eligibility as role models, transformational leadership should enhance salespeople’s adoption of their leaders’ behaviors. That is, salespeople should expect particularly high rewards from imitating a transformational leader’s behavior as this leader is perceived as especially able and competent (Antonakis et al. 2011; Martin and Bush 2006; MacKenzie et al. 2001). If such a well-respected transformational leader exhibits high intensity of price defense behavior, salespeople may infer that this behavior is increasing negotiation performance, and thus be inclined to adopt it. Conversely, salespeople might view a sales leader low in transformational leadership as a less convincing role model. Thus, they might view adopting his price defense behavior as less rewarding and refrain from adopting it. Hence:

  • H2: The positive impact of leaders’ intensity of price defense behavior on salespeople’s intensity of price defense behavior is more pronounced if salespeople perceive leaders as transformational.

The key role of salespeople’s motivation, ability, and opportunity

When considering the moderating effect of transformational leadership on the role modeling process, it has to be noted that “more attention needs to be given to the followers of transformational leadership” (Bass and Riggio 2006, p. 235; Piccolo and Colquitt 2006). In fact, the contingency leadership perspective proposes that the impact of leaders’ behaviors and style on followers should fundamentally depend upon a number of follower characteristics (e.g., Yun et al. 2006; Fiedler 1978). Consistent with this perspective, in choosing such contingencies for our model, we draw on the motivation–ability–opportunity (MAO) framework, which is a well-established theoretical basis for explaining human behavior (Siemsen et al. 2008) and has recently been applied to sales settings (Schmitz 2013). Our rationale for drawing on the MAO framework is that a salesperson’s motivation, ability, and opportunity are well-established, comprehensive predictors of salespeople’s behavior (e.g., Schmitz 2013). The behavior we aim to elucidate is a salesperson’s adoption of his or her leader’s intensity of price defense. Thus, we expect a salesperson to adopt a leader’s price defense behavior more strongly if the salesperson is motivated, able, and perceives an opportunity to emulate his leader’s price defense behavior.

Therefore, and in accordance with our social learning lens, for developing the hypotheses for the moderating effects we will focus on the role of MAO variables in salespeople’s social learning processes. More specifically, we posit that the effect of transformational leaders’ role modeling on salespeople’s adoption of price defense behaviors will increase when (1) the salesperson is motivated to learn (H3), (2) the salesperson exhibits a high negotiation efficacy (H4), and (3) the salesperson perceives a customer to be lenient in a price negotiation (H5).

Salespeople’s learning motivation

Salespeople exhibiting a high level of learning motivation are highly motivated to continuously improve their abilities and view sales experiences as an opportunity to receive feedback for their further personal development (Sujan et al. 1994). Thus, such salespeople are likely to scan their work environment for learning opportunities and seek inputs for effective behaviors in the selling context. In this respect, Boichuk et al. (2014, p. 97) emphasized that transformational leadership contributes to the creation of a “guided learning environment” in which leaders give salespeople a clear road map and provide them with meaningful examples. Therefore, transformational sales leaders represent a natural source of guidance and orientation for salespeople willing to learn (Ingram et al. 2005). That is, highly learning-oriented salespeople should view adopting transformational leaders’ behaviors as particularly useful to achieve their learning goals.

With regard to price negotiations, learning-oriented salespeople aim at improving their negotiation behavior. Learning the appropriate intensity of price defense behavior is especially important for salespeople because it may determine the outcome of a negotiation. To learn which intensity of price defense behavior to exhibit vis-à-vis customers, salespeople should be particularly likely to observe the intensity of price defense behaviors of their leaders, especially if leaders are transformational. Salespeople should expect adopting transformational sales leaders’ intensity of price defense behavior as highly rewarding and useful because these leaders emanate an aura of competence, suggesting to salespeople that their price negotiation behavior is bound to be effective. Thus, they should be particularly open and sensitive to transformational leaders’ role-modeling influence. Therefore, we hypothesize the following:

  • H3: The joint effect of the intensity of leader price defense and transformational leadership style on intensity of salesperson price defense is more positive if the salesperson’s learning motivation is high.

Salespeople’s negotiation efficacy

Perceived negotiation efficacy refers to salespeople’s strength of belief in their own ability to affect the outcome of a price negotiation through their negotiation behavior (Kim et al. 2005). Consequently, salespeople’s negotiation efficacy tends to reflect the extent to which they perceive their negotiation performance to be under their own control compared with being driven by external circumstances (Mathieu et al. 2007; Den Hartog and Belschak 2012). In what follows, we present our reasoning why salespeople are more likely to adopt transformational sales leaders’ price defense behavior if they exhibit high negotiation efficacy.

Efficacy beliefs increase individuals’ ability to learn effectively (Zimmerman 2000). Specifically, efficacy beliefs increase the use of self-regulatory processes while learning, such as goal setting, self-monitoring, and self-evaluation (e.g., Zimmerman and Bandura 1994; Zimmerman et al. 1992). In addition, efficacy beliefs reduce stress and anxiety while learning (e.g., Bandura 1997; Pajares and Kranzler 1995). This allows individuals to learn with higher effort and persistence, increasing their learning success (e.g., Multon et al. 1991; Salomon 1984; Schunk 1981).

Building on these findings, we expect that salespeople who exhibit high negotiation efficacy are more likely to socially learn negotiation behavior. Specifically, as outlined above, these salespeople may exhibit increased self-regulation as well as decreased stress while learning how to negotiate from their social environment, which should allow them to learn with increased effort, persistence, and thus success. In our context, social learning pertains to salespeople’s adoption of their leaders’ intensity of price defense (see H1), which we argued to be particularly pronounced if leaders are transformational (see H2). Thus, putting these pieces together, we expect the positive effect of transformational sales leaders’ on salespeople’s intensity of price defense behavior to be augmented for high perceived negotiation efficacy owing to its catalyzing effect on social learning.Footnote 2 In support of this reasoning, Pieterse et al. (2010) stated that the role modeling influence of transformational leaders should be enhanced when followers’ psychological empowerment is high, i.e., when followers feel more able to proactively influence their work role and environment. Thus:

  • H4: The joint effect of the intensity of leader price defense and transformational leadership style on intensity of salesperson price defense is more positive if the salesperson’s negotiation efficacy is high.

Salespeople’s perceived customer negotiation lenience

Perceived customer negotiation lenience reflects the extent to which a salesperson perceives a customer as yielding and docile in a price negotiation (Perdue and Summers 1991). For instance, a customer high in negotiation lenience might make little or no discount demands and might not be a tough, persistent negotiator. Consequently, salespeople’s perceived customer negotiation lenience is a situational factor, specific to each customer–salesperson interaction, and hence represents the opportunity facet in the MAO framework. Two reasons suggest that salespeople are more likely to adopt transformational sales leaders’ price defense behavior in price negotiations where they perceive high customer negotiation lenience. First, negotiating with customers who are not lenient but tough negotiators is difficult, putting high cognitive load on salespeople. Such cognitive load may in turn impede salespeople’s ability to retrieve and exhibit socially learned information in a given negotiation (Kirschner 2002; Sweller 1994; Van Merrienboer and Sweller 2005). Thus, in negotiations with tough customers, salespeople may be less likely to follow their transformational leaders’ role modeling of price defense. Conversely, if customers are lenient negotiators, salespeople may have access to cognitive resources that allow them to retrieve and imitate their leaders’ intensity of price defense.

Second, salespeople who negotiate with a tough customer face a trade-off between defending the price and securing the deal (Alexander et al. 1991; Weingart et al. 1990; Wieseke et al. 2014). To navigate this trade-off, salespeople may have to adapt their behavior flexibly to the specific negotiation situation (e.g., Alavi et al. 2016; Spiro and Weitz 1990). As a result, even if they are able to access socially learned negotiation behavior, it may be difficult for salespeople to follow their transformational leaders’ intensity of price defense as proposed in H1 and H2. Conversely, if customers are lenient negotiators, salespeople may adopt their leaders’ intensity of price defense free from situational constraints.

In summary, we argue that if salespeople perceive customers to be lenient, they are more likely to adopt their leaders’ intensity of price defense. We hypothesize:

  • H5: The joint effect of the intensity of leader price defense and transformational leadership style on intensity of salesperson price defense is more positive if the salesperson’s perception of customers’ negotiation lenience is high.

Methodology

Data collection and sample

Context description

Particularly in negotiation research, investigators have raised concerns concerning the generalizability of results obtained in laboratory settings (Alavi et al. 2016; Evans and Beltramini 1987; Zetik and Stuhlmacher 2002). Therefore, we decided to conduct our study in a field setting and collected dyadic data from salesperson–customer interactions in a B2C automobile retailing context. We chose the context of automobile retailing because discount negotiations regularly occur when customers purchase cars (Consumer Reports 2016), which is an essential requirement given the research question our study aims to answer. Moreover, salespeople are frequently exposed to their leaders’ price defense behavior in joint negotiations or in coaching sessions and can thus readily observe it (Brett et al. 2009; Deeter-Schmelz and Ramsey 1995).

To initiate the cooperation with automobile dealerships, we conducted discussions with the top management of several dealership chains within one region of Germany. Obtaining full top management commitment and support for our study was essential because our study design required thorough participation of salespeople and their customers in the dealerships and the presence of our data collection team in the dealership facilities over several weeks. Top managements of three dealership chains were willing to support our study and allow our research team to visit and stay at their dealerships. Importantly, all 28 dealerships of the chains including all salespeople were obliged to participate in our study. The automobile dealership chains represented two mainstream car brands with a very similar price positioning and product portfolio. Moreover, negotiation policies were highly similar across those dealerships chains.

In the initial meetings with the top managements of the dealership chains, one of our key goals was to discuss the dealerships’ business practices to gain an understanding of the context and be able to evaluate whether the context is compatible with our research goals. In this respect, we particularly focused on salespeople’s compensation schemes and dealerships’ negotiation policies as they constitute important influences on salespeople–customer price negotiations (e.g., Homburg et al. 2012; Joseph 2001). Discussions with the top managements of the dealership chains revealed that their compensation schemes and negotiation policies were highly prototypical for the automobile retailing industry in Germany and eligible to investigate our research question. In the following, we describe these compensation schemes and negotiation policies in detail.

Regarding salespeople’s compensation schemes, a substantial variation in compensation schemes across dealerships and across individual salespeople was prevalent for the dealership chains. This is because compensation schemes did not only depend on management policy, but also on individual arrangements with salespeople. The standard method of compensation for each salesperson was a bonus per car sold and a sales commission. Additionally, salespeople received a bonus tied to the profit of a sold car. Moreover, they received a bonus for selling of financing contracts. On average the share of variable compensation of the total compensation was approximately 70%. Seeing the variance of compensation schemes on the salesperson level, we included control variables for each of these facets of the salesperson compensation scheme in our model. Incentives specifically focusing on preventing discounts were not in place.

Regarding negotiation policies, all salespeople in the dealerships were granted full pricing authority for discounts up to 20%. When salespeople intended or were asked by customers to give a discount exceeding 20% of the list price, they were obliged to consult with their supervisor (which occurred only three times during our data collection).

Data collection procedure and sources

Each salesperson completed a questionnaire providing general perceptions and attitudes concerning his or her job and supervisor. Additionally, we obtained data from salespeople and their customers directly after sales encounters with the help of questionnaires, which our research team personally administered to salespeople and customers to achieve the best possible response rates and to ensure accurate matching. Importantly, there was a time lag of one month between the initial salesperson survey (t = 1 in Fig. 2), which captured the independent variables, and the subsequent interaction-specific customer and salesperson surveys (t = 2 in Fig. 2), which captured salespeople’s perceived customer negotiation lenience and the dependent variables. We introduced this time lag to mitigate common method issues and reduce the likelihood of reversed causation. The questionnaires referred to the car that was the major subject of the sales interaction.

After approximately 30 weeks of data collection (on average, we spent one week at each dealership), we had obtained 264 salesperson–customer interactions. In this process, 92 salespeople and 264 customers participated in the survey, with a mean of 2.9 interactions recorded per salesperson. In total, 39 sales leaders were in charge of the dealerships resulting in an average control span of 2.3 salespeople per leader.

The response rate for salespeople was 100% because participation was obligatory, and the response rate for customers was approximately 45%. More precisely, in total, we contacted 587 customers of which 264 responded. For customers who did not agree to participate, we did not collect the salesperson t = 2 survey either, since we aimed for complete dyadic information in our data set. The average age of salespeople was 34.8 years with a median of 35 and a standard deviation of 10.9, and the average job experience was 11.7 years (SD = 8.8), ranging between 0 and 44 years. The customers had a mean age of 43.4 with a median of 43 (SD = 14.13).

To avoid biasing the results by only observing successful sales encounters, we collected data on both discount negotiations which were closed with a sale (45.3% of the interactions) and interactions in which no agreement was reached (54.7% of the interactions). However, all interactions ended with a final sales price the salesperson offered to the customer.

Measures

Measurement level and sources

We used measurements established in the marketing literature, with adjustments to suit the study’s context. The Appendix provides a full list of items employed in this study. In what follows, we indicate the data sources of our core constructs. In the salesperson survey in t = 1, salespeople indicated the variables leader’s transformational leadership, intensity of leader price defense, salesperson’s learning motivation, and salesperson’s perceived negotiation efficacy. We decided to measure and model salespeople’s perceptions of their leaders’ transformational leadership style and intensity of price defense on the salesperson level (t = 1) on the basis of the contingency leadership perspective (Fiedler 1978) and leader-member exchange theory (Graen and Uhl-Bien 1995). As Sin et al. 2009, p. 1048) put it: “research on supervisor–subordinate relationships has shown convincingly that leaders do not behave consistently toward all subordinates.” Accordingly, to account for potential variance in salespeople’s perceptions of their leaders, we model salespeople’s perceptions of leader behaviors on the salesperson level (level 2). However, as a robustness check, we verify our results aggregating these variables on the leader level (level 3; see robustness check section).

On the customer–salesperson interaction level (level 1, in t = 2), intensity of salesperson price defense, salesperson’s perceived customer negotiation lenience, and realized discount (discount granted to the customer, measured in % of the list price) was provided by the salesperson for each specific sales conversation with the customer.

Measurement diagnostics

To assess construct reliability, we inspected Cronbach’s alpha and conducted a confirmatory factor analysis. Regarding Cronbach’s alpha, all multi-item scales used in the study passed the recommended threshold value of .7 with the exception of the control variables salesperson intrinsic motivation and empowerment (Nunnally 1978). Regarding the confirmatory factor analysis, we initially assessed the overall fit of the measurement model before verifying the reliability of individual constructs on the basis of individual item reliabilities, composite reliability (CR) and average variance extracted (AVE). The overall fit of the measurement model was satisfactory (CFI = .97, TLI = .96, RMSEA = .04, SRMR = .05). Concerning the reliability diagnostics for the individual multi-item constructs, we report Cronbach’s alpha, CR, and AVEs in Table 2 and all factor loadings (λ) in the Appendix. Apart from the control variable salesperson intrinsic motivation, all constructs conform to prescribed thresholds for AVE and CR indicating sufficient reliability and convergent validity (Bagozzi and Yi 1988). Eventually, to assess discriminant validity of the constructs employed in our study we relied on the criterion of Fornell and Larcker (1981). According to this criterion a construct is discriminant from another construct if the construct’s AVE is larger than the squared correlation of both constructs. All constructs passed this test.

Table 2 Descriptive statistics and correlations
Full size table

Non-response bias test

In our study customers who participated in our study could have systematically varied from customers who declined to participate. Thus, to check whether such a selection or nonresponse bias might confound the results of the data analysis, we inspected whether responses from participants systematically differed from responses of non-participants. We employed special incentives (coupons for a discount on car maintenance) and a very short version of the questionnaire to collect data from customers who originally did not intend to participate. We then compared responses from participants with responses from non-participants with regard to critical customer variables (age, gender, customer satisfaction, purchase intention, realized discount; see Web Appendix 1). The results show that the groups do not differ systematically regarding these variables. Thus, a non-response bias is unlikely to be an issue for the sample.

Model specification

Our study comprises 264 customer–salesperson interactions that are nested in 92 salespeople. Because several interactions are matched to a single salesperson, the observations in the dataset are not independent from one another, which is a basic assumption of the ordinary least squares estimator. When this assumption is violated, coefficients have been found to be biased (Hox 2010). Hence, to account for the nested data structure, we employ a multilevel approach that allows the simultaneous processing of data from multiple levels. Specifically, we ran a two-level path model as specified in Fig. 2 using Mplus 7 and a full information maximum likelihood estimator. Thus, salesperson data is modeled at level 2 and interaction-specific data of salespeople (i.e., the intensity of salesperson price defense behavior, salespeople’s perceived customer negotiation lenience, and realized discount) are modeled at level 1. Moreover, as a robustness check, we estimated a three-level model comprising leadership variables on level 3, which we discuss in more detail in the robustness check section.

To assess whether a multilevel approach is factually required, we inspected intraclass correlation coefficients (ICC), which indicate the proportion of variance that resides between the groups (Raudenbush and Bryk 2002). Simulation studies show that a multilevel approach is warranted when ICCs exceed .05 to .15 (Hox 2010). In our study, the ICCs substantially exceed the recommended thresholds, providing evidence that a multilevel approach is required, and are comparable to similar studies in sales management (Hughes and Ahearne 2010; ICCintensity of salesperson price defense = .36; ICCrealized discount = .19). Given these ICCs, we conducted a multilevel regression model (Hox 2010; Raudenbush and Bryk 2002). Before running the model, we centered all predictor variables on their grand means to reduce multicollinearity issues and facilitate the interpretation of interaction effects (Hofmann and Gavin 1998). In our hypotheses development, we predicted three-way interactions between leader behavior, leadership style and salesperson factors derived from the MAO framework. For salespeople’s learning motivation and negotiation efficacy, we specified the three-way interactions in the model at the between level while the three-way interaction with salespeople’s perceived customer negotiation lenience is specified as a cross-level interaction (see equations in Web Appendix 2).

Moreover, based on prior research on price negotiation and leadership, we included several control variables in the model. Specifically, at level 1 we added customers’ perceived price importance, customer–salesperson length of relationship (Alavi et al. 2016), a car type dummy (used vs. new) and customers’ alternative demand concession. The latter is included to account for the possibility that customers might demand non-monetary concessions or better conditions for trade-in vehicles. At level 2, we included an array of control variables potentially influencing salespeople’s price defense behavior to verify the stability of our focal leader-related effects. Specifically, we included variables capturing salespeople’s prevalent and theoretically established dimensions of compensation schemes (Zenger and Marshall 2000; share of fixed compensation, share of sales commission of total variable compensation, importance of achieving profit goals). Moreover, we included control variables accounting for salespeople’s task characteristics on the basis of Hackman and Oldham’s job characteristics model (Hackman and Oldham 1975) such as salesperson job identification, job experience, autonomy, perceived competitive intensity, and leader contingent reward and punishment for task achievement (i.e., providing positive and negative feedback on performance) (Schmitz and Ganesan 2014) (see details on the scales in the Appendix).

Results

Main results

The estimation of the multi-level path model largely corroborates the predictions of our theoretical framework (see Table 3). More precisely, the main effect of intensity of leader price defense on intensity of salesperson price defense which we postulated in H1 is confirmed (Main Effects Model: b = .47, p < .01). Furthermore, our prediction based on social learning theory presented in H2 that the positive relationship between intensity of leader and salesperson price defense should be enhanced for high levels of transformational leadership is confirmed (Two-way Interactions Model: b = .14, p < .05).

Table 3 Multi-level path model coefficients
Full size table

Regarding the three-way interactions, in H3, we suggested that the intensity of leader price defense increases the intensity of salesperson price defense if the leader exhibits a transformational leadership style and the salesperson possesses a strong learning motivation. The positive coefficient of the three-way interaction term provides support for H3 (Full Model with Controls: b = .16, p < .05). Moreover, interaction diagram A in Fig. 3 illustrates the pattern of results and further confirms our prediction in H3 (see Web Appendix 3 for a detailed discussion).

Fig. 3
figure 3

Three-way interaction plots. customer’s negotiation lenience = salesperson’s perceived customer negotiation lenience

Full size image

In H4, we argued that the intensity of leader price defense increases the intensity of salesperson price defense if the leader exhibits a transformational leadership style and the salesperson possesses a high negotiation efficacy. The positive coefficient of the three-way interaction term provides support for H4 (Full Model with Controls: b = .11, p < .05). Moreover, interaction diagram B in Fig. 3 illustrates the pattern of results and further confirms our prediction in H4 (see Web Appendix 3 for a detailed discussion).

Eventually, in H5, we argued that the intensity of leader price defense increases the intensity of salesperson price defense if the leader exhibits a transformational leadership style and salespeople perceive the customer as lenient in the price negotiation. The positive coefficient of the three-way interaction term provides support for H5 (Full Model with Controls: b = .04, p < .05). Moreover, interaction diagram C in Fig. 3 illustrates the pattern of results and further confirms our prediction in H5 (see Web Appendix 3 for a detailed discussion). Eventually, as expected, the intensity of salesperson intensity of price defense behavior exhibits a significant negative effect on customers’ realized discount in the price negotiation (Full Model with Controls: b = −.06, p < .05).

Robustness checks

We conducted four robustness checks to verify the validity of our findings. First, in the main model estimations we employed a two-level path model with salespeople variables from t = 1 residing at level 2 and customer and salespeople interaction-specific variables at t = 2 residing at level 1. Since salespeople are nested within sales leaders, it might be necessary to account for the leader level, extending the two-level to a three-level model. In particular, for the variables leader’s transformational leadership and intensity of leader price defense, it might be argued that these variables should reside at the leader level (level 3) as they pertain to leaders’ behaviors. To account for potential dependence among salespeople owing to being supervised by the same sales leader, we estimated a three-level path model, comprising a leader level (level 3), a salesperson level (level 2), and a customer–salesperson interaction level (level 1). For this purpose, we aggregated salespeople’s responses on their leaders’ transformational leadership and intensity of price defense per leader and included it as a level 3 variable while all other variables are modelled as in the main analysis. Results of this three-level path model estimation corroborate findings from our two-level analysis (please refer to Table 3) indicating that the measurement level of the leader variables does not unduly influence our results.

Second, we calculated variance inflation factors (VIFs) for the full model to assess whether results of the model execution are unduly influenced by correlations among the independent variables. Prior research suggests that multi-collinearity among independent variables in a regression model can be tolerated if VIFs do not exceed a threshold value of 5 (Menard 1995). In the model, the VIFs of all independent variables fall below this threshold value (VIFmin = 1.08, VIFmean = 1.83, VIFmax = 3.32). Based on this analysis, we conclude that correlations among the independent variables do not unduly influence our results.

Third, following a procedure recommended by Ganzach (1997), to account for the potential influence of non-linear predictors and correlations among predictors, we ran the full model and additionally included quadratic terms of the main independent variables. Results remain fully stable for this extended model. That is, the three-way interactions pertaining to salesperson’s learning motivation (b = .13, p < .10), salesperson’s negotiation efficacy (b = .10, p < .05), and salesperson’s perceived customer negotiation lenience (b = 0.4, p < .05) are all significantly positive while the quadratic terms are nonsignificant. This indicates that our findings are not unduly confounded by correlations among the predictors or by non-linearity.

Fourth, as indicated previously, 54.7% of interactions did not result in a sale. It may well be that the lack of success in these interactions was caused by a salesperson’s exaggerated price defense. If this were the case, greater price defense may have been disadvantageous, because making a discounted sale may still create more value for a dealership than not making a sale at all. To test this issue, we examined the average intensity of salesperson price defense for interactions that did and that did not result in a sale. Intensity of salesperson price defense did not differ significantly across these two group (p > .10). Thus, it is unlikely that the success or failure of an interaction was decisively driven by a salesperson’s price defense.

Fifth, to further verify our previous argument, we additionally investigated the effect of intensity of salesperson price defense on customers’ purchase intention. We replicated our main model (Table 3, Full Model) additionally including customer purchase intention as an ultimate dependent variable. In this estimation, results show an insignificant main effect of intensity of salesperson price defense behavior on customer’s purchase intention (b = −.03, p > .84; see Web Appendix 4 for the full model estimation). Consequently, salespeople’s price defense behavior on average seems to reduce realized discounts, but does not necessarily deter customers from purchasing the product.

While the findings of our previous robustness checks may seem surprising at a first glance, an explanation might be that a salesperson’s price defense only decreases a customer’s purchase intention under certain circumstances. For instance, key contingencies may be a salesperson’s communication skills to effectively deploy price defense behaviors without deterring customers (Blanchard et al. 2016; Kwon and Weingart 2004) and a customer’s level of trust (Schurr and Ozanne 1985). Furthermore, the importance that customers attribute to receiving price discounts represents a further key contingency which is known to strongly vary between different customers (e.g., Wieseke et al. 2014). This reasoning is likewise supported by an additional robustness check in Web Appendix 5, which shows that intensity of salesperson price defense is also unrelated with a salesperson’s objective annual sales.

Discussion

Summary of findings

Results of our empirical analysis fully corroborate our theoretical predictions based on social learning theory, the contingency perspective of leadership, and the MAO framework. We find that the transfer of leaders’ price defense behavior to salespeople’s price defense behavior does not occur unconditionally, but rather strongly hinges on the extent of leaders’ transformational leadership style and, importantly, on the variables delineated from the MAO framework. Specifically, results show that leaders constitute particularly viable role models for salespeople’s price defense behavior if leaders exhibit a transformational leadership style and salespeople possess a high learning motivation, negotiation efficacy, or perceive that customers exhibit high negotiation lenience.

Research issues

Our study contributes to academic research in at least three ways. First, we provide first insight into the research void of how sales leadership influences salespeople’s price defense behavior. Notably, prior research has already examined various management techniques that influence salespeople’s price defense behavior, such as incentivization (e.g., Joseph 2001; Lal 1986; Weinberg 1975, 1978) and delegation of pricing authority to salespeople (e.g., Bhardwaj 2001; Homburg et al. 2012; Wilken et al. 2010). However, the omission of leadership variables in prior research limited academic knowledge on the phenomenon of salespeople’s price defense and neglected a potentially important lever that is easily actionable for practitioners. In addressing this omission, our study also confirms that findings of prior literature on salespeople’s adoption of leaders’ behaviors hold for the price negotiation context. More specifically, as Table 1 illustrated, prior research has established that certain sales leader behaviors and attitudes are socially learned by salespeople, such as market orientation (Lam et al. 2010), adaptive selling (Chakrabarty et al. 2013), and organizational identification (Wieseke et al. 2009)—especially if leaders are transformational (Mullins and Syam 2014; Wieseke et al. 2009, 2011). We add to this literature by showing that such social learning also pertains to behavior in price negotiations, and by confirming that transformational leadership may act as a catalyst in this respect.

Admittedly, the aforementioned findings merely transfer an established mechanism (social learning catalyzed through transformational leadership) to a new context (price negotiations). However, our second and unique contribution is the conceptualization and empirical confirmation of the fundamental moderating influence of salespeople characteristics (see H3 through H4). Specifically, a key tenet of prior sales leadership research centers on the notion that role modeling and transformational leadership alone are sufficient to inspire employees to imitate leaders (e.g., Mullins and Syam 2014; Rich 1997; Wieseke et al. 2009, 2011). However, this perspective has been questioned by the contingency leadership perspective and leader–member exchange theory, which suggest that the effect of leaders’ behavior on their followers also depends on followers’ characteristics (e.g., Den Hartog and Belschak 2012; Fiedler 1978; Graen and Uhl-Bien 1995; Li et al. 2013; Yun et al. 2006; see also Graen and Uhl-Bien 1995). For example, Podsakoff et al. (1996) noted that transformational leaders’ impact on followers’ behaviors needs to be appropriately specified and contextualized, and Hunter et al. 2007, p. 439) emphasized a “lack of examining potential moderators” in the analysis of the consequences of transformational leadership on followers (see also Avolio 2007).

To address the dearth of research on contingencies of sales leadership in leader–member exchanges, we deduced a set of salesperson characteristics (learning motivation, negotiation efficacy, perceived customer lenience) from the MAO framework that in the context of price negotiations affects the degree to which leaders’ role modeling and transformational leadership induces salespeople’s adoption of leaders’ price defense behaviors. Conceptualizing and empirically validating these contingencies significantly contributes to marketing research since they cannot be deduced from prior empirical studies on salespeople’s adoption of their leaders’ role modeling. Instead, we theoretically deduced these contingencies for the context of price negotiations by combining social learning theory with the MAO framework.

Our findings bear implications for leadership research on role modeling in sales contexts because respective works might not only focus on the leadership style but must take into account its interplay with salespeople’s characteristics. Specifically, as our findings illustrate, salespeople endowed with different characteristics might perceive and react to leader behaviors in different ways, corroborating the contingency leadership perspective and leader–member exchange theory (e.g., Den Hartog and Belschak 2012; Fiedler 1978; Graen and Uhl-Bien 1995; Li et al. 2013; Yun et al. 2006). Omitting to include such fundamental contingencies in research on sales leader–salesperson dyads may lead to overly simplified or even erroneous conclusions (Grewal et al. 2013). We hold that future research should carve out further mechanisms and contingencies that determine how sales leaders can steer salespeople’s negotiation behavior. Two directions seem particularly worthwhile in our view. First, it might be interesting to examine which leader influence tactic is most effective for different types of salespeople to facilitate salespeople’s adoption of leaders’ price defense behavior (Furst and Cable 2008). Second, future research may delve deeper into the role of leaders’ contingent rewards (e.g., MacKenzie et al. 2001), which has been found to positively affect leader–member exchange (e.g., Wayne et al. 2002). While we controlled for rewards both in terms of leaders’ leadership style and salespeople’s incentive structure, it may be interesting to examine how such rewards facilitate salespeople’s social learning of price defense behaviors. For example, certain types of contingent rewards may positively affect salespeople’s learning motivation, which as our study indicated acts as a catalyst of social learning.

Third, our study contributes to negotiation literature. In a recent comprehensive review of the negotiation literature, Herbst et al. (2011) demanded an enhanced focus on holistic price negotiation models to account for organizational influences on salespeople’s bargaining behavior. Thus, to date, the role of sales leaders in price negotiations and how sales leaders affect salespeople’s negotiation performance is not well understood. We addressed this omission in previous research by exploring the role of sales leaders in price negotiations among salespeople and customers and thus integrated two major previously disparate research areas that are naturally related. Therefore, our study conceptually and empirically carves out the pivotal influence of sales leaders on salespeople’ negotiation behavior and performance.

Managerial implications

Approximately 70% of companies delegate some extent of pricing authority to their salesforce (Lim and Ham 2014; Hansen et al. 2008; Frenzen et al. 2010). Seeing this influence of salespeople on their companies’ price setting, managers are often concerned with ensuring their salespeople’s price defense behaviors (Joseph 2001). As mentioned previously, almost 40% of sales managers think that their salesforce “needs improvement” in its ability to avoid discounting (CSO Insights 2011) and “avoiding discounting” is among the top five metrics used to measure the performance of and sales managers (CSO Insights 2014). Interestingly, our study shows that it is sales managers themselves whose behavior strongly influences their salespeople’s price defense. Notably, this influence is unlikely to be homogeneous for all salespeople, but may strongly depend on an individual salesperson’s learning motivation, negotiation efficacy, and perceived customer lenience. In this respect, at least three major conclusions can be drawn from our study.

First, when willing to stimulate price defense, leaders are recommended to display consistent price defense behavior to set appropriate examples to their teams, and to adopt a transformational leadership style. This is particularly important if salespeople are highly motivated to learn and are able and endowed with the opportunity to defend prices vis-à-vis customers. Our findings suggest that in this case, salespeople are particularly likely to internalize their leaders’ price defense. To this end, firms can ensure high levels of leaders’ price defense intensity through installing appropriate incentive systems and internal marketing measures to convince leaders of the importance of these behaviors. Furthermore, seeing that adjusting their leadership style may be challenging to some sales leaders, firms may invest into trainings that help leaders develop such flexibility.

Second, leaders who lead transformationally and are able to set a good example of price defense are recommended to consider the aforementioned MAO factors in their personnel decisions. Specifically, leaders should focus on hiring salespeople who are motivated to improve their selling skills and perceive themselves as having a high negotiation efficacy. Moreover, leaders can train and coach their teams accordingly to improve their perceived ability to negotiate and therefore ultimately foster the adoption of price defense behavior. Thus, to promote salespeople’s learning motivation, leaders can create a culture that values learning and personal development. Such a culture might be installed through regular feedback discussions among leaders and salespeople to set and monitor learning goals. Promoting and creating awareness for a learning motivation should furthermore assume a key role in salesforce training. Similarly, we suggest that firms might foster salespeople’s negotiation efficacy through the sharing of best practice behaviors in price negotiations. Additionally, sales leaders and their companies may try to influence salespeople’s perceptions about customers’ lenience, for example by presenting research demonstrating that price discounts are actually less important than salespeople think in affecting customers’ ultimate choices.

Third, leaders who do not lead transformationally are required to carefully monitor and evaluate their salespeople’s price defense behavior. Our results suggest that in this case, setting an example to salespeople who are motivated to learn and perceive themselves as able and endowed with the opportunity to defend prices does not easily yield the desired effects. Conversely, salespeople in this case can exhibit reactance to their leaders’ role modelling attempts, requiring leaders to resort to other means of control. For example, leaders can use other leadership influence tactics, such as rational persuasion or exchange (Yukl et al. 2008), or price defense incentives to foster their employees’ price defense behaviors.

In addition to these implication for sales leaders, salespeople should likewise be aware that their current negotiation behavior is likely to be influenced by their supervisors—whether intended by their leaders or not. To negotiate with customers more strategically, salespeople are advised to proactively consult with their leaders on what intensity of price defense is required and then adjust their negotiation behavior accordingly. Hereby, salespeople should carefully reflect their own behavior and aim at implementing requested policies rather than adopting or rejecting their leaders’ negotiation behavior in a non-reflective manner.

Limitations and avenues for future research

Our study has several limitations that provide fruitful avenues for future research. First, our study does not answer the question what the right level of price defense is. Instead of adopting such a normative stance, following prior literature (e.g., Wieseke et al. 2014) our work is of explicative nature and aims to improve our understanding of the factors that factually drive salespeople’s price defense. Future research on salespeople’s price defense as well as managerial practice may benefit immensely from studies providing further guidance on which intensity of price defense salespeople should exhibit. As explicated before, this question is not trivial since price defense may exhibit ambivalent effects on desired outcomes, increasing the margin of a successful transaction but potentially weakening customer relationships (e.g., Dwyer et al. 1987; Ganesan 1993; Weitz 1981; Wieseke et al. 2014).

Second, worth noting is that our study focused on one industry, that is, automobile dealerships. We chose this industry because of price negotiations are common in automobile retailing (Consumer Reports 2016) and it has therefore been examined in studies on salespeople’s price defense, securing our alignment with prior literature (e.g., Galinat and Müller 1988). This being said, future studies may replicate and extend our findings in other contexts to establish their generalizability beyond automobile retailing. For example, it may be interesting to examine the role of leadership for price negotiations in a B2B field sales context. Two reasons suggest that in this context the effect of a leader’s on a salesperson’s intensity of price defense may be weaker than in automotive sales: (1) Salespeople in B2B field sales may act more autonomously and thus have less contact with their leaders. Thus, salespeople may be less prone to socially learning negotiation behaviors from their leaders. (2) Salespeople in B2B field sales often face professional purchasing organizations that are geared to realizing price discounts. Thus, in this context salespeople are more likely to face customers with lower negotiation lenience, which our results suggest to reduce salespeople’s adoption of their leaders’ negotiation behavior.

Third, all core constructs in our study, apart from several control factors are measured from salespeople’s perspective either in t = 1 or t = 2. However, it may be worthwhile extension to our model to include variables from customers’ perspective to the model such as customers’ factual negotiation lenience or other customer characteristics such as interaction orientation (McFarland et al. 2006) or loyalty (Wieseke et al. 2014). Web Appendix 6 presents a corresponding robustness check, showing that our model also holds for a customer measure of negotiation lenience. Relatedly, prospective works might employ more comprehensive measurement of salespeople’s learning orientation and negotiation efficacy, addressing the limitation of our study that we measure these constructs through two two-item scales.

Fourth, a limitation of our model is its indirect measurement of trade-in vehicles. Specifically, in automobile retailing customers may forego monetary discounts if they instead achieve a higher selling price for their trade-in vehicle. To take such effects into account, we controlled for a variable indicating whether customers demanded concessions from salespeople other than monetary discounts, additional equipment, or warranties. However, we acknowledge that this variable is not a direct measurement of the involvement of trade-in vehicles.

Fifth, an avenue for future research pertains to the unit of analysis in the multi-level model. Our conceptual model comprises three levels: leader, salesperson and customer. However, the leadership measures were rated by salespeople and not by the leaders themselves. This is common practice in leadership research, though (Jaramillo and Mulki 2008; Mathieu et al. 2007; Ahearne et al. 2005; MacKenzie et al. 2001; Shoemaker 1999; Tyagi 1985) because self-reported data of sales leaders on their own leadership behaviors can be biased (Gramzow et al. 2003). In this respect, future research might additionally include leaders’ perceptions of their behavior in the model and investigate the consequences of leader–salesperson perceptual discrepancies on price negotiation outcomes (see Kraus et al. 2015 for a similar approach).