The consequences of (not) seeing eye-to-eye about the past:The role of supervisor–team fit in past temporal focus for supervisors' leadership behavior

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers.


| INTRODUCTION
Scholars have become increasingly interested in individuals' attention to specific time frames (i.e., the past, present, or future) and how such temporal foci may help us to better understand important organizational behavior phenomena (Mohammed & Harrison, 2013;Shipp & Fried, 2014). Individuals' past temporal focus (i.e., "the attention individuals devote to thinking about the past;" Shipp, Edwards, & Lambert, 2009, p. 1), in particular, has been shown to distinctly shape their experiences and behaviors at work (e.g., Cojuharenco, Patient, & Bashshur, 2011;Zacher, 2016). For example, individuals characterized by a strong past temporal focus feel less powerful (Shipp et al., 2009), tend to delay actions and decisions (Díaz-Morales, Ferrari, & Cohen, 2008), and exhibit reduced energy and efficiency at work (Goldrich, 1967). Extrapolating these findings to leadership contexts, we anticipate that a supervisor's past temporal focus may critically shape his or her leadership behaviors. Scholars have generally assumed, in this regard, that past-focused supervisors are more likely to demonstrate an ineffective, passive approach toward leading others and to minimize their purposeful leadership efforts (Bluedorn, 2002). For example, researchers have argued that "no leader wants to be called past-oriented" (Thoms, 2004, p. 103) because a strong orientation toward the past evokes perceptions of a "counter-ideal manager" (Alipour, Mohammed, & Martinez, 2017, p. 313) who dwells on memories instead of meeting current leadership responsibilities (see also Hernández, 2017;Weick, 1979).
Interestingly, however, existing research offers little evidence to validate these widely shared notions. Whereas scholars have frequently examined supervisors' present and/or future temporal focus (e.g., West & Meyer, 1997;Zhang, Wang, & Pearce, 2014), the empirical research has largely ignored possible connections between supervisors' past temporal focus and their leadership behaviors. In fact, we are aware of only one empirical study that has investigated the role of past temporal focus in a leadership context (i.e., Nadkarni & Chen, 2014), linking a CEO's past temporal focus with his or her company's new product introductions. Hence, our understanding of the leadership implications associated with supervisors' past temporal focus is fragmented and incomplete, despite common concerns about potential drawbacks for supervisors' leadership behaviors.
Importantly, theory and research on leader-team fit (Carter & Mossholder, 2015;Cole, Carter, & Zhang, 2013) suggest that the linkage between a supervisor's past temporal focus and his or her leadership behavior may be more intricate than often assumed. This leaderteam fit perspective submits that supervisor-team interactions are not only driven by the respective supervisor's own traits, preferences, and goals, but also by the extent to which such supervisory characteristics are (mis)aligned with the respective characteristics within a supervisor's team of subordinates (see also Gibson, Cooper, & Conger, 2009). In this regard, this stream of research acknowledges that the team as a whole (rather than individual subordinates) represents a key reference point for supervisors' (mis)fit judgments (Bashshur, Hernández, & González-Romá, 2011;Cole et al., 2013). This is because one's team constitutes a highly salient and central part of a supervisor's social context at work such that supervisors often perceive the team as a collective entity and, thus, tend to direct their influence behaviors toward the team in its entirety (Hu & Judge, 2017;Oc, 2018).
In fact, research has repeatedly emphasized that incongruence between a focal individual's and his or her work team's temporal focus can trigger strong behavioral reactions and influence important work outcomes (Eldor et al., 2017;Shipp et al., 2009) because such fit (or misfit) may distinctly color a focal individual's perceptions of the team (Gibson, Waller, Carpenter, & Conte, 2007;Mohammed & Nadkarni, 2011). This reasoning is based on the notion that an individual's temporal focus manifests in tangible actions that are readily observable by others, with individuals generally preferring similarity (rather than dissimilarity) between their own and others' orientations toward time (Mohammed & Harrison, 2013;Waller, Conte, Gibson, & Carpenter, 2001). Consequently, scholars have suggested that (mis)fit between supervisors' and their subordinates' time-based personality characteristics may shape supervisors' conclusions about the adequacy of subordinates' efforts and outcomes (Alipour et al., 2017).
Specifically, we argue that supervisors may perceive incongruence between their own and their team's past temporal focus as a salient signal of task-related and/or social problems in the team, whereas congruence may signal that the team is functioning in line with supervisory preferences.
On this basis, the central proposition of this research is that the degree of incongruence between a supervisor's past temporal focus and the past temporal focus prevalent in his or her team of subordinates will affect how the supervisor leads his or her team. To specify the nature of this relationship, we enrich our focus on leader-team fit with insights from functional leadership theory (Hackman & Walton, 1986;McGrath, 1962). According to functional leadership theory, supervisors are more likely to engage in active leadership behavior when their team fails to exhibit attitudes and behaviors that supervisors consider critical for team functioning (see, e.g., Morgeson, DeRue, & Karam, 2010). In these situations, functional leadership theory argues that supervisors will respond flexibly to performance or interaction problems they perceive in their team, in an effort "to do, or get done, whatever is not being adequately handled for group needs" (McGrath, 1962, p. 5). Alternatively, when supervisors perceive their teams as functioning appropriately, they are less likely to intervene in team processes and, thus, they may adopt a more "hands-off" (i.e., passive) leadership style toward the team (Zaccaro, Rittman, & Marks, 2001).
Integrating key ideas from leader-team fit research (Cole et al., 2013) and functional leadership theory (Zaccaro et al., 2001), we therefore infer that a supervisor may view his or her team's ability to address relevant task and social issues in a distinctly negative light when supervisor and team past temporal focus are incongruent, whereas a supervisor's respective perceptions of the team may be more favorable in situations of congruence-with tangible consequences for the supervisor's specific (active and passive) leadership behaviors, as directed toward the team as a whole. Hence, as depicted in Figure 1, our key research question is how incongruence in past temporal focus between a supervisor and his or her team relates with the supervisor's task-oriented and relationship-oriented leadership (i.e., behaviors aimed at proactively promoting task execution and employee wellbeing, respectively; DeRue, Nahrgang, Wellmann, & Humphrey, 2011) and with his or her laissez-faire leadership (i.e., behaviors aimed at averting leadership responsibilities; Hinkin & Schriesheim, 2008). 1 By empirically examining this question, the present study responds to repeated calls for new insights into the leadership implications of individuals' temporal focus (Alipour et al., 2017;Thoms & Greenberger, 1995), moving beyond the examination of supervisors' 1 We conceptualize past temporal focus incongruence between a supervisor and his or her team in terms of objective (mis)fit-that is, supervisors' and their team's past temporal focus exist independently of one another (Kristof-Brown, Zimmerman, & Johnson, 2005). Moreover, we note that task-oriented and relationship-oriented behaviors represent prototypically active leadership styles (Fleishman, 1973), whereas laissez-faire behavior represents a prototypically passive type of leadership (Skogstad, Einarsen, Torsheim, Aaslang, & Hetland, 2007). A large volume of research attests to the relevance of these leadership behaviors for key organizational outcomes (e.g., supervisor and team performance as well as team members' job satisfaction and motivation; . present and future temporal focus prevalent in previous research to also investigate supervisors' past temporal focus. As such, this study broadens current theoretical knowledge on the role of time-based personality characteristics for leadership phenomena. In doing so, we aim to challenge the widely held assumption that supervisors' past temporal focus consistently evokes undesirable, passive leadership tendencies and diminishes proactive leadership efforts. Rather, we advance a novel, multiple-stakeholder perspective (see Carter & Mossholder, 2015;Cole et al., 2013) that highlights the complex interplay between supervisors' and their team's past temporal focus. This investigation therefore breaks new conceptual ground for research on subjective time frames and leadership, demonstrating that a full understanding of these linkages requires joint consideration of all relevant parties' temporal focus.

| Past temporal focus and its psychological consequences
The general construct of temporal focus describes the degree to which people tend to direct their attention toward the past, present, or future (Bluedorn, 2002;Shipp et al., 2009). Scholars have argued that such temporal foci develop through early childhood experiences and cultural socialization processes, leading to a relatively stable, traitlike pattern of temporal orientations (Shipp & Aeon, 2019). Past temporal focus, in particular, represents an individual's attention to past events and actions (Shipp et al., 2009). Individuals with a strong past temporal focus spend significant time vividly remembering past experiences, emotions, and decisions (Holman & Silver, 1998). They value history and tradition, appreciate past successes, but also dwell on earlier mistakes and failures (Zimbardo & Boyd, 1999). Individuals with a relatively low past temporal focus, by contrast, do not attach much relevance to (and, thus, easily forget about) past events and experiences (Shipp et al., 2009). As such, earlier accomplishments and failures have less influence on current decision-making and actions (Karniol & Ross, 1996).
In considering associated consequences, research and theory suggest that individuals with a pronounced focus on the past may become "psychologically locked" and passive (Rabinovich & Morton, 2012, p. 399; see also Holman & Silver, 1998). These individuals heavily rely on established routines and existing mindsets, and they prefer well-known procedures over new approaches that deviate from the status quo (Karniol & Ross, 1996). Scholars have demonstrated that highly past-focused individuals tend to choose avoidance-oriented goals over approach-oriented ones (Peeters, Schreurs, & Damen, 2014), are less likely to change their behavioral intentions (Rabinovich & Morton, 2012), and report relatively low self-efficacy (Gana, Klein, Saada, & Trouillet, 2013). Accordingly, it is not surprising that highly past-focused individuals are often characterized as indecisive and ineffective (Karniol & Ross, 1996;Shipp & Aeon, 2019).

| Integrating a leader-team fit perspective with functional leadership theory
This prevalent-and rather negative-depiction of highly past-focused individuals notwithstanding, consideration of the leader-team fit literature (Cole et al., 2013;Gibson et al., 2009) suggests that the relationships between supervisors' past temporal focus and their leadership behaviors may be more complex and context-specific. Building on the broader fit literature (e.g., Li, Kristof-Brown, & Nielsen, 2019), this leader-team perspective acknowledges that the joint attitudes, preferences, and behaviors prevalent within a supervisor's team of subordinates may critically define the supervisor's social context at work (Carter & Mossholder, 2015;Hu & Judge, 2017). Hence, a supervisor's behaviors toward the team may hinge on the degree of incongruence (i.e., fit or misfit) between relevant supervisory characteristics, on the one hand, and the respective characteristics within a supervisor's team, on the other (Cole et al., 2013;Gibson et al., 2009).
Drawing from this conceptual background, we anticipate that the degree of incongruence between a supervisor's own past temporal focus and the past temporal focus prevalent within his or her team of subordinates (i.e., team past temporal focus) may shape the supervisor's leadership behaviors toward the team. Importantly, although temporal focus was initially conceptualized as an individual-level F I G U R E 1 Conceptual model construct (Shipp et al., 2009), scholars have recently emphasized that this construct may also be used at higher levels of analysis to describe an overall work unit's (e.g., a team's) general orientation toward the past, present, or future (Shipp, in press;Shipp & Aeon, 2019). In line with research on team personality composition (e.g., Barrick, Stewart, Neubert, & Mount, 1998;Homan et al., 2008), we therefore define a team's past temporal focus as a configural property that reflects the individual team members' past temporal focus. Specifically, team past temporal focus is based on an additive composition model (Chan, 1998; see also Chen, Mathieu, & Bliese, 2004), such that individual members' past temporal focus additively combines to reflect the team's overall or general pattern of paying attention to and thinking about the past. 2 Team past temporal focus, then, describes the degree to which a team's members tend to muse about past experiences and events, rethink previous actions, value tradition and history, and rely upon strategies used in the past (Shipp et al., 2009).
As noted previously, research has shown that past temporal focus critically influences individuals' attitudes, working styles, and social behaviors (Shipp, in press;Shipp & Aeon, 2019). For example, depending on individuals' levels of past temporal focus, they will differ in how much they rely on previously used strategies, how they evaluate their own or other's job performance, and/or whether they will engage in team traditions (Shipp et al., 2009). Based on this logic, incongruence between a supervisor's past temporal focus and his or her team's past temporal focus will result in readily observable, highly salient behavioral (dis)similarities (Alipour et al., 2017) that will likely come to light across diverse tasks and work contexts. Moreover, consistent with assertions made by previous leader-team fit research (e.g., Cole et al., 2013;Gibson et al., 2009), such (dis)similarities in how a supervisor and his or her team prefer to act and interact should distinctly color a supervisor's assessments of and attitudes toward the team. Accordingly, scholars have repeatedly shown that alignment (rather than misalignment) between an individual's and his or her team's time-based personality characteristics results in more favorable work experiences (Jansen & Kristof-Brown, 2005;Slocombe & Bluedorn, 1999). Drawing on these conceptual and empirical arguments, incongruence between a supervisor's and his or her team's past temporal focus should establish an important basis from which the supervisor assesses team functioning which, in turn, has implications for how he or she leads the team as a whole.
To better explicate how supervisor-team incongruence in past temporal focus may translate into specific leadership behaviors, we draw from relevant theory and research on team leadership. Functional leadership theory, in particular, provides critical insights in this regard (McGrath, 1962;Zaccaro et al., 2001). This theoretical perspective argues that supervisors' core responsibility is to address the needs of their team that are not otherwise being handled appropriately. Following this logic, supervisors are likely to closely monitor their team's overall functioning and to intervene when they perceive relevant problems believed to be undermining team processes and effectiveness (Hackman & Walton, 1986). When supervisors perceive the team to be operating smoothly and without coordination or interaction problems, by contrast, functional leadership theory proposes that supervisors will see no need to intervene and, hence, they are more likely to adopt a passive team leadership approach (Zaccaro et al., 2001).
Integrating these arguments with our leader-team fit perspective, we anticipate that a supervisor's and his or her team's past temporal focus will jointly shape the supervisor's leadership behaviors. First, we draw from existing leader-team fit research to propose that supervisors will contrast the behaviors and attitudes resulting from their team's past temporal focus with their own respective preferences, thereby drawing conclusions about the adequacy of the team's efforts and social interactions. Second, we draw from functional leadership theory to argue that supervisors' behavioral responses toward the team will consider these conclusions, such that the supervisors will intervene with either active behaviors (i.e., task-oriented and/or relationship-oriented leadership) or react in a more passive manner (i.e., laissez-faire leadership).

| Past temporal focus and leadership behavior: Incongruence effects
When a supervisor's and his or her team's past temporal focus are aligned, our theoretical rationale (as drawn from leader-team fit research; Cole et al., 2013) suggests that the team's members will approach job tasks in ways that largely mirror supervisory preferences and inclinations. In such a scenario, it follows that the supervisor will perceive the team as relatively well-functioning. This is because the supervisor is likely to view the team as working efficiently toward goal attainment, making decisions in appropriate ways, and working to maintain a collaborative atmosphere. Hence, a functional leadership logic (Zaccaro et al., 2001) suggests that the supervisor will deem proactive leadership interventions as being unnecessary. By contrast, a supervisor's inferences about the team's overall functioning are likely to be more negative when supervisor and team past temporal focus are incongruent. The supervisor may view the team's approaches toward task-related efforts and interpersonal interactions as inefficient and misguided in this situation (because these approaches will contradict his or her preferences) such that decisive leadership interventions are needed to get the team back "on track" (Hackman & Walton, 1986).
We thus anticipate a supervisor to exhibit more task-oriented leadership when his or her past temporal focus is incongruent (rather than congruent) with the team's past temporal focus. Highly pastfocused individuals appreciate established routines and draw heavily from traditional, well-known approaches of task accomplishment, whereas less past-focused individuals often discount prior experiences and, thus, are more likely to appreciate novel means of task execution (Shipp, in press;Shipp et al., 2009). Extending these ideas to the supervisor-team interface, when a team's past temporal focus exceeds the supervisor's respective focus, it is likely that the 22 Following past research on additive composition models (Deng, Leung, Lam, & Huang, 2019;Lvina, Johns, & Vandenberghe, 2018), we argue that this configural team property is reflected in the mean value of individual team members' past temporal focus (cf. Chen et al., 2004;Kozlowski & Klein, 2000). We outline this operationalization in more detail in the Methods section.
supervisor will perceive the team to be "stuck in the past", with members spending too much time ruminating on prior experiences instead of investing efforts in current problem-solving and goal attainment. A supervisor may therefore conclude that his or her team is overly rigid and sluggish, relying on outdated solutions as opposed to challenging the status quo and tackling task requirements in new and innovative ways. Alternatively, when a team's past temporal focus is lower than the supervisor's respective focus, we believe the supervisor is likely to infer that his or her team, as a whole, is overlooking important lessons that could be learned from past successes and failures. Moreover, from the supervisor's perspective, the team may appear to be neglecting strategies and procedures that have proven effective, exhibiting a tendency to "reinvent the wheel" rather than maintaining well-established and functional routines. Accordingly, the supervisor may conclude that the team's work efforts are ineffective and/or inefficient.
With the above in mind, we therefore assume that supervisors will perceive more salient problems for their team's task-related functioning when supervisor and team past temporal focus are misaligned (in either direction). Consequently, we predict that supervisors will exhibit more task-oriented leadership behaviors if the degree of incongruence between supervisor-team past temporal focus is higher rather than lower. In an effort to maintain successful task execution, in particular, supervisors are likely to closely monitor their overall team's work behavior, to structure and coordinate the team members' joint efforts and to clearly communicate expected procedures and guidelines for team task accomplishment. Graphically, these predictions would translate into a convex (i.e., upward-curved) shaped response surface along the incongruence line (see left panel of Figure 2). Hence, we hypothesize: Hypothesis 1a Supervisors will exhibit less task-oriented leadership when their own and their team's past temporal focus are congruent rather than incongruent.
Similarly, we anticipate that a supervisor will exhibit more relationship-oriented leadership when his or her past temporal focus is incongruent (rather than congruent) with the team's past temporal focus.
Individuals with a strong past temporal focus often hold long-lasting grudges and, thus, find it difficult to overcome personal disputes (Pierro, Pica, Giannini, Higgins, & Kruglanski, 2018;Zakay & Fleisig, 2011). Also, these individuals place a strong emphasis on existing relationships and they are rather conservative in building new social connections (Park et al., 2017;Zimbardo & Boyd, 1999), whereas less past-focused individuals attach limited value to (and more easily forget about) their personal relationship histories (Holman & Zimbardo, 2009;Zimbardo & Boyd, 2008). It therefore seems reasonable to assume, on the one hand, that a supervisor in charge of leading a team that exceeds his or her own past temporal focus may view the team (as a whole) as unnecessarily reviving old disagreements and personal conflicts. Moreover, supervisors faced with this situation may perceive the team as being too focused on maintaining long-established relationships thus failing to adequately integrate new members into the team. On the other hand, supervisors whose team's past temporal focus is lower than their own may believe that team members are carelessly undervaluing each other's past accomplishments and joined histories, increasing the risk that daily hassles or misunderstandings trigger intense interpersonal conflicts. In this scenario, a supervisor may perceive the team as shortsighted, investing too little effort into the maintenance of existing relationships and the establishment of team traditions.
In sum, our theoretical rationale suggests that supervisors should perceive their team's approach toward social relations and interpersonal interactions as more problematic if their own and the team's past temporal focus are incongruent in either direction. Consequently, we propose that supervisors are more likely to exhibit relationshiporiented leadership behaviors toward the team in situations of past temporal focus incongruence (rather than congruence) to address these issues and fulfill team needs. A supervisor may try to facilitate more favorable, productive interactions within the team, for example, by creating a positive work atmosphere, encouraging trust and collaboration between team members, and acting as a friendly and approachable coach and mentor. This is illustrated by the convex shape of the proposed response surface along the incongruence line in the left panel of Figure 2. We therefore predict: F I G U R E 2 Hypothesized response surface graphs for the relationships of supervisor-team incongruence in past temporal focus with taskoriented leadership, relationship-oriented leadership, and laissez-faire leadership [Colour figure can be viewed at wileyonlinelibrary.com] Hypothesis 2a Supervisors will exhibit less relationshiporiented leadership when their own and their team's past temporal focus are congruent rather than incongruent.
We further predict that a supervisor's laissez-faire leadership toward his or her team will be less frequent when the supervisor's past temporal focus is incongruent (in either direction) with the team's past temporal focus. As outlined previously, when supervisor and team past temporal focus are incongruent, supervisors are likely to sense substantive problems with their team's approaches toward task-related efforts and interpersonal relations. Hence, we anticipate that supervisors will form a negative impression of their team's ability to function effectively without proactive leadership interventions. It appears unlikely that supervisors in this situation will resort to the passive types of behavior that characterize laissez-faire leadership (Skogstad, Einarsen, Torsheim, Aasland, & Hetland, 2007) instead exhibiting targeted behaviors to stimulate more productive processes within the team. As illustrated by the concave (i.e., downwardcurved) shape of the proposed response surface along the incongruence line in the right panel of Figure 2, we therefore suggest: Hypothesis 3a Supervisors will exhibit more laissezfaire leadership when their own and their team's past temporal focus are congruent rather than incongruent.

| Past temporal focus and leadership behavior: Congruence at higher versus lower levels
Our previous theorizing suggests that incongruence between supervisor and team past temporal focus (in either direction) is positively related to supervisors' active leadership behaviors and negatively related to passive leadership behaviors. Beyond these relationships, we also expect a pattern of congruence effects such that-in situations of supervisor-team congruence-a supervisor's leadership behavior should hinge on the level at which the supervisor's and his or her team's past temporal focus are aligned.
Specifically, we build on prior theory and research on past temporal focus to suggest that supervisors will exhibit fewer task-oriented behaviors when their own and their team's past temporal focus are congruent at higher (rather than lower) levels. Studies have repeatedly demonstrated that a stronger focus on the past is associated with individuals' passive behaviors (Díaz-Morales et al., 2008;Rabinovich & Morton, 2012). Leadership scholars have, accordingly, cast highly past-focused supervisors as lacking in active and goal-directed leadership efforts (Thoms, 2004). By the same token, teams with a strong focus on the past should exhibit a passive approach toward their tasks such that members are less likely to proactively seek guidance from their supervisor. Hence, when both supervisors and their teams are highly past-focused, we conclude that supervisors may be less willing-and may perceive little necessity-to actively take charge of their team's ongoing work processes and goal accomplishment.
In contrast to their highly past-focused counterparts, individuals with a relatively weak orientation toward the past are more energetic and approach-oriented (Shipp et al., 2009;Zimbardo & Boyd, 1999).
Hence, even though supervisors may not have task-related concerns when they and their team share low levels of past temporal focus, we anticipate that these supervisors will be inclined to strive for continuous improvement. Similarly, a team characterized by a low past temporal focus consists of goal-oriented, dedicated individuals (Shipp & Aeon, 2019) who may appreciate-or even proactively seek-their supervisors' task-related guidance and contributions. On this basis, we expect that, if supervisor-team congruence occurs at relatively low levels of past temporal focus, supervisors will be more motivated to set priorities for their team's tasks and organize collaborative team efforts toward the achievement of common goals. As illustrated by the downward slope of the proposed response surface along the congruence line in the left panel of Figure 2, we therefore propose: Hypothesis 1b Supervisors will exhibit less task-oriented leadership when their own and their team's past temporal focus are congruent at higher (rather than lower) levels.
Similarly, we anticipate supervisors will exhibit less relationshiporiented leadership toward the team when supervisor-team congruence occurs at higher (rather than lower) levels of past temporal focus.
Scholars have repeatedly demonstrated that highly past-oriented individuals are less sociable and more interpersonally withdrawn, compared to less past-focused individuals (Shipp, in press;Zimbardo & Boyd, 1999). Hence, if supervisors and their teams share a strong orientation toward the past, it seems likely that both parties will act rather indifferently in social settings and have little interest in developing deep interpersonal relationships. On this basis, we anticipate that supervisors will engage in fewer relationship-oriented leadership behaviors, investing relatively little effort in building high-quality relationships with team members and/or in nurturing a positive team atmosphere.
By contrast, less past-oriented individuals are outgoing and socially active, they purposefully develop ongoing interpersonal relations, and they seek new social relationships (Park et al., 2017;Zimbardo & Boyd, 1999). These individuals enjoy interacting with coworkers, and they are willing to actively invest in a positive social environment (Zimbardo & Boyd, 2008). We thus expect that with a relatively low past temporal focus, both supervisors and their teams will value high-quality interpersonal relationships in the team and desire a positive team atmosphere.
It follows that a supervisor with a relatively low past temporal focus will be more likely to personally engage with team members, acting as a coach and mentor and proactively looking out for the team's collective welfare-and a team with a similarly low past temporal focus team is likely to encourage such relationship-oriented efforts. As illustrated by the downward slope of the proposed response surface along the congruence line in the left panel of Figure 2, we therefore suggest: Hypothesis 2b Supervisors will exhibit less relationshiporiented leadership when their own and their team's past temporal focus are congruent at higher (rather than lower) levels.
Finally, we predict that supervisors will more frequently exhibit laissez-faire leadership when supervisor and team past temporal focus are aligned at higher (rather than lower) levels. Indeed, scholars have argued that highly past-focused supervisors may be prone to "inertial tendencies" (Nadkarni & Chen, 2014, p. 1812, spending their time indulging in reminiscences of the past rather than actively leading and guiding their team's efforts (Thoms, 2004). Similarly, members of a strongly past-focused team can be characterized as passive (Rabinovich & Morton, 2012), such that they may appreciate a handsoff approach toward team leadership that imposes few demands and/or expectations upon them. Hence, in situations of supervisorteam congruence at high levels of past temporal focus, it seems likely that supervisors will favor passive behaviors and shirk their leadership responsibilities.
Alternatively, less past-focused supervisors are likely to be more dedicated, goal-driven, and outgoing (Bluedorn, 2002). Similarly, a team with a relatively low level of past temporal focus consists of members who are cognizant of the need for goal-directed efforts and positive interpersonal relationships (Zimbardo & Boyd, 1999) and may therefore appreciate and expect a supervisor's active leadership efforts. Hence, we predict that supervisors will not rely on passive team leadership behaviors when both their own and their team's past temporal focus are relatively low. Even though supervisors may not perceive a pressing need for calculated leadership interventions (due to alignment with their team's past temporal focus), logic suggests that they are unlikely to neglect their team leader responsibilities in this situation. As illustrated by the upward slope of the proposed response surface along the congruence line in the right panel of Figure 2, we therefore hypothesize: Hypothesis 3b Supervisors will exhibit more laissezfaire leadership when their own and their team's past temporal focus are congruent at higher (rather than lower) levels.

| The potential roles of supervisors' and teams' present and future temporal focus
Our study's central goal is to advance a better understanding of the leadership consequences associated with supervisors' (and their team's) past temporal focus. We chose to examine past temporal focus, in particular, because (a) this type of temporal focus has received little empirical attention in leadership studies, and (b) we sought to challenge the widely shared assumption that a pronounced focus on the past is universally detrimental for a supervisor's leadership behavior. We note, however, that prior research has often jointly examined individuals' past, present, and future temporal focus to provide a more comprehensive assessment (e.g., Cojuharenco et al., 2011;Zacher, 2016). In fact, one might speculate that our assertions regarding past temporal focus could similarly apply to supervisors' and their team's present and future temporal focus. Nevertheless, previous research has shown that the psychological and behavioral consequences of individuals' present and future temporal focus differ in fundamental ways when compared to the consequences of individuals' past temporal focus (Shipp & Aeon, 2019;Shipp et al., 2009). For example, individuals with strong preferences for the past tend to favor passive forms of behavior (Peeters et al., 2014;Rabinovich & Morton, 2012), whereas neither individuals' present nor future temporal focus have been empirically linked with these behavioral tendencies. Hence, we submit that a systematic, indepth investigation of supervisor-team (in)congruence in present and/or future temporal focus extends well beyond the scope of this study. We included measures of present and future temporal focus as part of our research design and data collection to examine associated (in)congruence effects in an exploratory manner. In doing so, we follow best practice recommendations for the post-hoc analysis of scientific data (Chen, 2018;Hollenbeck & Wright, 2017) thereby enabling a clearer interpretation of the study's main findings and strengthening core inferences on the potentially unique role of supervisor and team past temporal focus.

| Sample and procedures
We collected data from various organizations and industries across Germany, in an attempt to increase the generalizability of our findings (Demerouti & Rispens, 2014). We approached potential participants (i.e., supervisors and their direct team members) using personal and university contacts (for similar procedures, see Bunderson, Van der Vegt, Cantimur, & Rink, 2016;Pundt & Venz, 2017). To be included in the present sample, our contact person in each team had to confirm that the potential participants worked as a team such that they shared mutual goals, interacted regularly, cooperated toward joint goal achievement, and directly worked with a common supervisor (Arrow & McGrath, 1995;Kozlowski & Bell, 2003). After initial contact, potential participants received general information about the study and, depending on organizational constraints, they received either a web-based or a (otherwise identical) paper-and-pencil survey.
The supervisor survey assessed supervisors' past temporal focus, potential control variables, and demographics. The team member survey captured team past temporal focus, potential control variables, and demographics. Additionally, it asked team members to assess their immediate supervisor's task-oriented, relationship-oriented, and laissez-faire leadership. We informed participants about the voluntary nature of the study and assured confidentiality. Neither supervisors nor subordinates received financial compensation for their participation.
After initial contact, 94 supervisors (with 602 direct team members) indicated their interest to participate in the study. From this initial sample, 91 supervisors (97%) and 307 of their team members (51%) provided data. To be included in the study, (1) supervisors had to provide their own, complete the survey, and (2) two or more team members needed to complete the team member survey (cf. Rubin, Munz, & Bommer, 2005). Based on these criteria, we excluded 7 supervisors and 11 team members. Our final sample therefore comprised 84 supervisors and 296 of their direct team members. The number of member responses per team ranged from 2 to 11 (M = 3.52, SD = 1.76), for an average within-team response rate of 55%. Supervisors were, on average, 44 years old (SD = 10.41), and 54% were male. Their mean organizational tenure was 11.53 years (SD = 8.85). The team members in our sample were, on average, 38 years old (SD = 11.84), and 63% were female. Their mean organizational tenure was 8.27 years (SD = 8.84). The participating supervisors and team members came from a wide variety of industry sectors, including services (30%), public administration (21%), manufacturing (17%), health care (15%), sales (11%), and finance (6%). 3 We believe the present sample is appropriate to test our hypotheses for several reasons. First, we approached supervisors and teams who interacted directly and on a regular basis. Second, the team members in our study worked on shared, interdependent tasks that required frequent, cooperative, and coordinated interactions for joint goal achievement. Hence, the team's actions, as a whole, were likely to represent a salient reference point for supervisors' judgments of members' task-related functioning and interpersonal relations. Finally, we deliberately used a sample of supervisors and teams with varied industry and task backgrounds. With prior research emphasizing that temporal aspects are relevant for leader-follower and team interactions across diverse contexts (Bluedorn & Jaussi, 2008;Shipp & Cole, 2015), we thereby avoid limiting our conclusions toward an overly narrow array of work situations.

| Measures
We translated all measurement instruments to German following a back-translation procedure (Brislin, 1980).

| Supervisor past temporal focus
We asked supervisors to self-rate their past temporal focus using a four-item measure developed by Shipp et al. (2009). Example items are: "I replay memories of the past in my mind" and "I think back to my earlier days." Supervisors used a 1 (rarely or never) to 5 (very often or constantly) response scale for these items. Cronbach's alpha was .86.

| Team past temporal focus
Team members rated their own past temporal focus using the same four items developed by Shipp et al. (2009). As noted before, team past temporal focus conceptually reflects a configural unit property (Kozlowski & Klein, 2000) derived from individual members' past temporal focus ratings. Because individuals' past temporal focus represents a trait-like characteristic (Shipp & Aeon, 2019), strong within-team consensus is unlikely. Therefore, we drew on an additive composition model (Chan, 1998) to operationalize teams' overall past temporal focus. When additive composition is applied, one creates a "summary index" that does not rely on within-unit agreement but, rather, uses the arithmetic average (or, alternatively, the sum) of a lower-level variable to depict the unit-level construct (Chen et al., 2004, p. 282; see also Chan, 1998). In this way, team past temporal focus is functionally equivalent with (although structurally different from) individual members' past temporal focus (see, e.g., Kozlowski & Klein, 2000;Morgeson & Hofmann, 1999). Cronbach's alpha for team past temporal focus was .88.

| Task-oriented leadership
Team members rated their supervisors' task-oriented leadership directed at the team using five items from Stogdill (1963). Example items are: "My supervisor tries out his or her ideas in the team" and "My supervisor lets his or her subordinates know what is expected of them" (1 = strongly disagree; 5 = strongly agree). Cronbach's alpha was .87.

| Relationship-oriented leadership
Team members assessed their supervisors' relationship-oriented leadership toward the team using five items from Stogdill (1963). Example items are: "My supervisor does little things to make it pleasant to be a member of the team" and "My supervisor looks out for the personal welfare of his or her subordinates" (1 = strongly disagree; 5 = strongly agree). Cronbach's alpha was .88.

| Laissez-faire leadership
Team members rated their supervisors' laissez-faire leadership directed toward the team using a three-item measure developed by Van Dierendonck, Stam, Boersma, De Windt, and Alkema (2014).
Sample items are: "My supervisor is not concerned with the team's results" and "My supervisor is absent when his or her subordinates need him or her" (1 = strongly disagree; 5 = strongly agree). Cronbach's alpha was .80.

| Control variables
We considered a number of variables as potential controls. First, alternative temporal foci (i.e., present and/or future focus; Shipp et al., 2009) might influence the behavioral consequences associated with past temporal focus (Alipour et al., 2017;Gibson et al., 2007) Additionally, we considered supervisors' gender (0 = female; 1 = male) and age (in years) as potential covariates because theory and research suggest that role expectations might bias our proposed relationships. Task-oriented leadership, for example, may be more congruent with stereotypically male (rather than female) role expectations (Eagly & Johnson, 1990) and with stereotypes toward younger (rather than older) individuals (Buengeler, Homan, & Voelpel, 2016). Relationship-oriented leadership may, however, be more congruent with stereotypes toward females (Eagly, Makhijani, & Klonsky, 1992) and toward older individuals (Buengeler et al., 2016), whereas laissez-faire leadership may be more accepted among males (Eagly, Johannesen-Schmidt, & Van Engen, 2003) and older persons (Zacher, Rosing, & Frese, 2011). Moreover, we considered the teams' organizational tenure (in years) as a possible covariate. Team members' average organizational tenure may bias a supervisor's perceptions of the team's expertise (Bell, Villado, Lukasik, Belau, & Briggs, 2011) and may therefore alter his or her leadership style.
Finally, it is important to consider possible dispersion effects when using the team mean to operationalize a configural unit property (Kozlowski & Klein, 2000). Hence, we incorporated the within-team standard deviation of individual members' past temporal focus as a control variable when testing our hypotheses (see also Gibson et al., 2009;Homan et al., 2008).

| Aggregation statistics
We conceptualized team past temporal focus (and the control variables, team present temporal focus, and future temporal focus) as a configural team property such that the aggregation of individual members' scores to the team level does not require within-team consensus (cf. Chan, 1998;Chen et al., 2004). In contrast, task-oriented, relationship-oriented, and laissez-faire leadership are best characterized as shared team properties (Kozlowski & Klein, 2000). These three constructs consider a supervisor's behavior toward the team as a whole thus representing ambient stimuli that similarly shape all members' leadership perceptions (Courtright, Colbert, & Choi, 2014;Hartnell, Kinicki, Lambert, Fugate, & Doyle Corner, 2016). Hence, aggregation of these variables to the team level follows a consensus composition model (Chan, 1998) such that team members' average ratings can only be used to meaningfully reflect a supervisor's behavior if there is sufficient convergence among individuals' scores (Chen et al., 2004).
Following recommended best practices, we assessed within-team agreement regarding individual members' task-oriented, relationshiporiented, and laissez-faire leadership ratings. Consistent with Biemann, Cole, and Voelpel's (2012) suggestions, we computed r WG (J) scores for these variables using a rectangular (uniform) null distribution as well as several alternative null distributions (see also James, Demaree, & Wolf, 1984). The alternative null distributions represented (a) a slight skew for all three variables (allowing for some leniency bias), (b) a normal distribution for task-oriented and relationshiporiented leadership (because a bell curve might reflect the "true" distribution of these variables), and (c) a triangular distribution for laissez-faire leadership (because respondents may lean toward the middle, neutral response option for this "undesirable" type of leadership). We also computed intraclass correlation coefficients (ICC1 and ICC2) to further justify our aggregation decisions (Bliese, 2000). Table 1 reports the estimates for these aggregation statistics. As shown, the obtained r WG(J) values revealed moderate to strong withinteam agreement for all three leadership styles (LeBreton & Senter, 2008). Furthermore, the ICC(1) values (along with the F-values from the associated one-way analyses of variance) illustrated that a statistically significant and considerable proportion of the variance in individual members' leadership ratings was attributable to their team membership. Finally, the ICC(2) values for task-oriented and relationship-oriented leadership indicated acceptable reliability, although we note that the respective value was lower than desirable for laissezfaire leadership. In sum, this overall pattern of aggregation statistics suggests that it was justified to aggregate individual members' assessments of their supervisors' leadership behavior to the team level of analysis (Bliese, Maltarich, & Hendricks, 2018).

| Analytic strategy
We utilized polynomial regression techniques (Edwards & Cable, 2009;Edwards & Parry, 1993) to test our incongruence hypotheses using the following equation (to simplify, control variables are not displayed): where LSB represents the respective leadership behavior (i.e., taskoriented, relationship-oriented, and laissez-faire leadership) and S and T represent supervisor and team past temporal focus, respectively.
Prior to conducting the analyses, we scale-centered supervisor and team past temporal focus to facilitate the interpretation of the results (Edwards, 1994;Edwards & Parry, 1993). In a first step, we then regressed our respective outcomes on the control variables. In a second step, we added the set of five polynomial terms, as depicted in Equation 1 (Edwards, 2002). Finally, we used the resulting coefficient estimates to generate three-dimensional response surface plots that we subsequently used to evaluate the overall pattern of the relationships (Edwards & Parry, 1993). is positive and significant. Support for Hypothesis 3a (i.e., laissez-faire leadership will be higher when supervisor and team past temporal focus are congruent rather than incongruent) would require the curvature along the incongruence line to be negative and significant.
Furthermore, to support Hypotheses 1b and 2b (i.e., task-oriented and relationship-oriented leadership will be lower when supervisor and team past temporal focus are congruent at higher rather than lower levels), the slope along the respective congruence line (calculated as a 1 = b 1 + b 2 ) should be negative and significant. Finally, support for Hypothesis 3b (i.e., laissez-faire leadership will be higher when supervisor and team past temporal focus are congruent at higher rather than lower levels) would require the slope along the congruence line to be positive and significant. 7 4 | RESULTS present and future temporal focus from all further analyses to reduce the number of parameters to be estimated and, thus, provide increased power for the following statistical tests (Becker et al., 2016). Despite a lack of association with other study variables, we continued to control for within-team dispersion (i.e., standard deviation) of team past temporal focus because partialling out withinteam variability is recommended when estimating supervisor-team (in)congruence polynomials (e.g., Gibson et al., 2009). Notably, excluding all of the covariates or including all considered control variables did not unduly influence the pattern of the hypotheses tests we report below.

| Confirmatory factor analyses
We conducted multilevel confirmatory factor analyses (CFAs) to examine our focal measures' viability. First, we estimated a multilevel CFA for the hypothesized five-factor measurement model, with supervisor and team past temporal focus as well as task-ori- Moreover, the hypothesized five-factor model demonstrated significantly better fit (p < .001), compared to a series of alternative four-factor models that subsequently forced the items for any two of the three leadership styles to load onto a common latent factor.
Similarly, the five-factor model showed significantly better fit (p < .001) compared to a three-factor model with all leadership items loading on a single latent factor. Overall, these results support our measures' convergent and discriminant validity. Detailed results for all CFAs are available upon request.

| Task-oriented and relationship-oriented leadership
Our theorizing suggests that supervisors will engage in fewer task-oriented and relationship-oriented leadership behaviors when supervisor and team past temporal focus are congruent rather than incongruent (Hypotheses 1a and 2a). Moreover, we predicted that these leadership behaviors will be less pronounced when supervisor and team past temporal focus are congruent at higher (rather than lower) levels (Hypotheses 1b and 2b). As shown in Table 3, the 7 We conducted several analyses to examine whether our sample contained an adequate distribution of incongruence combinations regarding supervisors' and their team's past temporal focus. Using z-standardized past temporal focus scores, these analyses revealed that supervisors exceeded their team's past temporal focus by more than 0.5 SD in 38.1% of the present cases, whereas teams exceeded their supervisor's respective focus by more than 0.5 SD in 41.7% of the cases. In the remaining 20.2% of the cases, there was relatively high supervisor-team similarity in past temporal focus (i.e., difference less than 0.5 SD). Similarly, visual scatterplot inspections (available upon request) showed that our sample adequately covered the possible range of supervisor-team incongruence combinations, although extreme past temporal focus values were rare. Finally, we screened for multivariate outliers based on leverage, studentized residuals, and Cook's D statistics (Aguinis & Edwards, 2014).
Depending on the outcome variable, we identified and dropped three to four outliers (i.e., less than 5% of the sample) and subsequently re-estimated the hypotheses tests. The results and conclusions from these additional analyses did not meaningfully differ from the fullsample findings reported.
T A B L E 1 Aggregation statistics for supervisors' leadership behavior Note. n = 84 teams. Shape = shape of the alternative null distribution; σ 2 E = variance of the alternative null distribution. Abbreviation: ICC, intraclass correlation coefficient.

Measure r WG(J)uniform r WG(J)first-alternative-distribution r WG(J)second-alternative-distribution ICCs
T A B L E 2 Means, standard deviations, and correlations Note. n = 84 teams. Coefficient alphas are shown along the diagonal in parentheses. # p < .10. *p < .05. **p < .01. ***p < .001. curvatures of the response surfaces along the incongruence lines were positive and significant for both task-oriented (a 4 = 0.31, p < .01) and relationship-oriented leadership (a 4 = 0.23, p < .05). Examination of the respective response surface plots (see Figures 3 and 4) illustrates these convex (i.e., upward curved) shapes such that taskoriented and relationship-oriented leadership were higher when supervisor and team past temporal focus were incongruent in either direction, as opposed to when they were congruent. Hence, both Hypotheses 1a and 2a were supported.

T A B L E 3 Polynomial regression results of leadership behavior on supervisor-team incongruence in past temporal focus
As further shown in Table 3, the slope of the response surface along the congruence line was negative and significant for task-oriented leadership (a 1 = −1.10, p < .05), whereas this coefficient was close to reaching a traditional value of statistical significance for relationship-oriented leadership (a 1 = −1.01, p = .06). Accordingly, as

| Laissez-faire leadership
Regarding supervisors' laissez-faire leadership, our theorizing suggests that the frequency of such behavior will be higher when supervisor and team past temporal focus are congruent rather than incongruent (Hypothesis 3a) and when supervisor and team past temporal focus are congruent at higher (rather than lower) levels (Hypothesis 3b). As shown in Table 3, the curvature of the respective response surface along the incongruence line did not reach statistical significance (a 4 = −0.15, p = .15). As such, Hypothesis 3a was not supported. As predicted, however, the slope along the congruence line was positive and significant (a 1 = 1.17, p < .05). Supporting Hypothesis 3b, Figure 5 shows that laissez-faire leadership was higher when congruence between supervisors' and their team's past temporal focus occurred at higher (rather than lower) levels.

| Exploratory analyses
Although the primary goal of our research was to uncover the leadership consequences associated with supervisors' and their team's past temporal focus, we conducted a number of supplementary analyses to explore the potential roles of supervisor-team incongruence in present and future temporal focus. In doing so, we repeated our polynomial regression analyses, but we replaced supervisor and team past temporal focus with present temporal focus in the first set of reanalyses and with future temporal focus in the second set of reanalyses. Results for all three leadership styles yielded no significant slope or curvature estimates regarding either present temporal focus or future temporal focus. It therefore appears that incongruence between supervisors' and their team's present as well as future temporal focus is not associated with a supervisor's task-oriented, relationship-oriented, or laissez-faire leadership. Detailed results for these additional analyses are available upon request.

| DISCUSSION
By integrating a leader-team fit perspective with insights from functional leadership theory, the present research illustrates how incongruence between a supervisor's and his or her team's past temporal focus may shape the supervisor's leadership toward the team. Our findings revealed that supervisors more frequently exhibited task-oriented and relationship-oriented leadership behaviors when their own past temporal focus diverged from their team's respective focus, whereas such proactive leadership behaviors occurred less frequently when supervisor and team past temporal focus were congruent. Moreover, when supervisors' and their team's past temporal focus were aligned at a relatively high level (rather than a lower level), task-oriented and relationship-oriented leadership were less frequent whereas laissez-faire leadership was more pronounced. of leadership, thereby addressing scholars' repeated calls for the further investigation of this important yet long-neglected aspect (e.g., Alipour et al., 2017;Bluedorn & Jaussi, 2008;Thoms & Greenberger, 1995). Despite theoretical advances on the possible role of supervisors' temporal focus (Alipour et al., 2017;Thoms, 2004), empirical research is sparse and has largely been limited to the roles of present and future temporal focus (e.g., West & Meyer, 1997;Zhang et al., 2014). Adding to this literature, our study illustrates that supervisors' past temporal focus may critically affect their leadership behavior as well (over and above their present and future temporal focus). To fully understand the relevance of supervisors' temporal focus for different types of leadership behavior, it therefore seems necessary to adopt a wider perspective, considering the extent to which supervisors direct attention toward past, present, and future events and experiences.

| Theoretical implications
Moreover, the present study adopts a multiple-stakeholder view on the leadership consequences of past temporal focus, thereby shedding new light on the important role of supervisor-team incongruence in this regard. Our findings illustrate that the behavioral consequences of supervisors' past temporal focus are more complex and multifaceted than previous theory has suggested. The existing literature has described highly past-focused supervisors in distinctly negative terms, arguing that these supervisors will often shirk their leadership duties and adopt a passive, ineffective approach toward leadership (Bluedorn, 2002;Thoms, 2004

| Limitations
Despite notable strengths (e.g., multiple data sources and multiple raters of supervisors' leadership behavior), this study has limitations that should be considered when interpreting its findings. The sample was drawn from one country (i.e., Germany), which may raise concerns about cross-cultural generalizability. Prior research has indicated, for example, that a strong focus on the past is more highly appreciated in other cultural contexts (e.g., some Asian countries), compared to Europe or North America (Gao, 2016;Ji, Guo, Zhang, & Messervey, 2009). Hence, although our theorizing is not bound to a specific cultural setting, future research that aims to constructively replicate our findings in alternative cultures and/or cross-cultural teams would help to better understand any cultural contingencies.
In a similar vein, our focal measures were originally developed in a different cultural context (i.e., the United States), raising potential concerns about data equivalency. In an effort to enhance semantic and measurement equivalence, we used common back-translation procedures (Brislin, 1980) to transfer these measures into German. Additionally, we followed suggestions by Hult et al. (2008) to further examine data equivalency. We compared our focal variables' descriptive statistics (i.e., means, standard deviations, and coefficient alphas) against prior studies using U.S. samples and identical survey measures (Courtright et al., 2014;Hartnell et al., 2016;Shipp et al., 2009;Thoroughgood & Sawyer, 2018). We observed similar values across this set of studies, providing additional confidence that the respective measures were similarly interpreted (details on these results are available upon request). Moreover, our survey-based data collection procedures were designed to closely mirror prior organizational research on temporal focus and/or leadership, as conducted both in the United States (e.g., Hartnell et al., 2016;Thoroughgood & Sawyer, 2018) and in Germany (e.g., Blickle et al., 2013;Strobel, Tumasjan, Spörrle, & Welpe, 2013). We believe these considerations alleviate data equivalency concerns to some extent (Hult et al., 2008)-although we acknowledge that cross-cultural studies with subsamples from diverse cultural contexts would be required to conclusively address this issue.
We note that our cross-sectional study design precludes causal conclusions. Longitudinal investigations could help to address this concern. To be sure, an investigation into supervisors' and their team's subjective time orientations over the course of objective time would enable a dynamic, "completely temporal" perspective (Shipp & Cole, 2015, p. 250) that may promote a better understanding of the role of time-based characteristics for processes of leadership. Similarly, because team constellations can change over the course of a supervisor's tenure, another fruitful extension of our research would be to investigate the behavioral consequences of "retrospected, current, and anticipated fit" (Shipp & Jansen, 2011, p. 76) between supervisor and team past temporal focus. Finally, scholars have generally assumed that past temporal focus is a relatively stable, trait-like construct; however, this assumption has recently been challenged (Shipp, in press;Shipp & Aeon, 2019 Furthermore, our supervisor sample size (supervisor n = 84) and average within-team response rate (i.e., 55%) are less than optimal.
Importantly, however, we ensured that at least two members rated each supervisor's leadership behaviors (as well as team past temporal focus), with an average of 3.52 raters per team. Also, the supervisor sample size is comparable to existing studies on leader-team (in)congruence effects (e.g., Carter & Mossholder, 2015;Lam, Lee, Taylor, & Zhao, 2018), and our within-team response rate mirrors the response rates commonly observed in organizational field research (Anseel, Lievens, Schollaert, & Choragwicka, 2010). In a supplementary analysis, we reran our hypotheses tests but statistically controlled for within-team response rate. The respective results were identical to our main analyses, providing additional confidence in the robustness of our study's findings.
Although the pattern of empirical results corroborated our theorizing, another limitation is that we did not directly test the theoretical mechanisms assumed to underlie the proposed relationships. Integrating the leader-team fit perspective (Cole et al., 2013) with functional leadership theory (Zaccaro et al., 2001), we argued that incongruence between supervisor and team past temporal focus may influence a supervisor's leadership behavior by shaping his or her perceptions of the team. Future research could benefit from directly capturing these mechanisms, for example by measuring supervisors' assessments of their team's task and interpersonal processes. Finally, the response surface for task-oriented leadership indicated a significant curvature along the congruence line (a 2 = −0.17, p < .05; see Table 3). Hence, although supervisors exhibited less task-oriented leadership when congruence between supervisor and team past temporal focus occurred at higher rather than lower past temporal focus levels (as predicted in Hypothesis 1b), this association was more complex than initially assumed (i.e., curvilinear rather than linear). We encourage future research to further examine this unexpected finding.

| Directions for future research
Research could extend our model by examining the consequences of supervisor-team incongruence in other temporal foci. Although our study only revealed significant associations between supervisor-team incongruence in past temporal focus and supervisors' task-oriented, relationship-oriented, and laissez-faire leadership, prior studies have demonstrated linkages between supervisors' present and future temporal focus and other leadership behaviors (e.g., transformational leadership; Zhang et al., 2014). Hence, scholars could examine whether supervisor-team incongruence in present and/or future temporal focus may shape such alternative leadership styles. Although supervisors with a pronounced future temporal focus may generally lean toward transformational behaviors, for example, they might find such leadership to be superfluous when their team shares this orientation toward the future.
Another fruitful extension would be to examine the consequences of supervisor-team incongruence in past temporal focus across different contexts. As noted above, temporal aspects are likely to be relevant across most work settings (Shipp & Cole, 2015). At the same time, it is possible that past events, experiences, and actions are more salient in some situations than in others (e.g., during after-action reviews and performance appraisals; Murphy & Cleveland, 1995;Villado & Arthur, 2013). Hence, the behavioral consequences illustrated in the present study might be particularly pronounced in such situations. Moreover, supervisors may perceive past temporal focus incongruence especially negative in such circumstances, potentially resulting in autocratic (cf. Harms, Wood, Landay, Lester, & Lester, 2018) or abusive types of leadership (cf. Tepper, 2007).
Furthermore, research could draw on our findings to expand existing knowledge on other time-based personality constructs. Chen and Nadkarni (2017), for example, demonstrated that supervisors with pronounced time urgency exhibit more temporal leadership (i.e., scheduling deadlines and making sure these deadlines are met). Following our rationale, this relationship might be attenuated for supervisors working with a similarly time-urgent team. By examining such notions, scholars could offer a more comprehensive understanding of the leadership consequences associated with supervisors' and their team's time-based characteristics.
Future research could also consider the consequences associated with past temporal focus (mis)fit between supervisors' and individual subordinates (i.e., at the dyadic level). Our interest in supervisor-team past temporal focus incongruence is in line with existing research on leader-team fit (e.g., Cole et al., 2013) and functional leadership theory (e.g., Zaccaro et al., 2001). Nevertheless, it may be worthwhile to examine linkages between dyadic past temporal focus (mis)fit and more "individualized" forms of leadership, such as individual-focused dimensions of transformational leadership (e.g., individualized consideration; Avolio & Bass, 1995) or leader-member exchange (Graen & Uhl-Bien, 1995).
Finally, scholars could investigate relationships between supervisorteam (mis)fit in time-based personality characteristics and important organizational outcomes other than leadership behavior. For example, incongruence between team members' time-based personality traits can reduce job satisfaction (Jansen & Kristof-Brown, 2005). Besides contributing to proactive leadership behaviors, incongruence between supervisor and team past temporal focus may therefore be accompanied by lower levels of job satisfaction among both parties. Such findings would complement our study's rather positive depiction of supervisor-team incongruence in past temporal focus with a more nuanced and critical perspective.

| Practical implications
This study's findings yield important insights for organizational practice. As noted before, the existing literature has described supervisors who devote substantial attention to the past as "counter-ideal" and passive managers (Alipour et al., 2017, p. 313; see also Thoms, 2004).
Our results, however, demonstrate that this depiction of highly past-focused supervisors tells only part of the story. Indeed, such supervisors may favor more passive behaviors (i.e., laissez-faire) and less frequently engage in active forms of leadership (i.e., task-oriented and relationship-oriented) when matched with a similarly pastfocused team-and research has repeatedly shown the detrimental consequences associated with such a hands-off approach toward leadership (e.g., Judge, Piccolo, & Ilies, 2004).
Nevertheless, our findings also illustrate that excluding highly pastfocused individuals from supervisory positions might be premature.
After all, even highly past-focused supervisors may be willing and able to exhibit more proactive forms of leadership when faced with a less past-focused team. Hence, rather than using past temporal focus as a selection criterion for supervisory positions, we believe organizations are well-advised to consider their supervisors' temporal focus in their leadership development efforts. Organizations may, for example, implement leadership training programs that address (highly pastfocused) supervisors' tendencies toward passive (laissez-faire) leadership. Such programs could likewise educate supervisors on the benefits associated with accurately understanding their team's past temporal focus and the potential consequences for their own behavioral choices.

| CONCLUSION
This study advances a multiple-stakeholder perspective on the role of past temporal focus for processes of leadership, demonstrating that incongruence between supervisors' and their team's past temporal focus may critically shape supervisors' proactive and passive leadership behaviors toward the team. As such, the present research extends our knowledge about the relevance of subjective time in leadership situations, illustrating that the role of supervisors' focus on the past critically hinges on the social context in which the respective consequences unfold. We hope that our findings can provide an impetus for further research on this issue that will advance a deeper understanding of the role of subjective time for organizational behavior in general and leadership in particular.

ACKNOWLEDGMENT
This work was supported by a grant from the Deutsche Forschungsgemeinschaft (German Research Foundation, DFG) to Frank Walter (392382642).