Abstract
Although much research has examined human resource management (HRM), managers’ roles in HRM seem to have been ancillary to this area of research. That is, HRM theory and research largely has advanced with a focus on policies, practices, systems, and their implementation and effectiveness, with less attention focused on the managers responsible for the design, adoption, enactment, and implementation of HRM strategy and practice. The purpose of this review is to examine extant research to determine the state of knowledge of the role of managers across organizational hierarchy in HRM. Thus, we review empirical literature for studies that include aspects of the impact lower-to-middle managers, human resource managers, top management teams, CEOs, and boards of directors have on HRM content, process, and outcomes. On the basis of the findings of this systematic, multilevel review, we discuss avenues for future research at each specific manager’s level, as well as general opportunities and challenges for research on managers’ roles in HRM across all hierarchical levels.
Keywords
As organizational decision makers, managers at all hierarchical levels can influence human resource management (HRM; Guest, 1997; Lado & Wilson, 1994; Lepak & Snell, 1999; Ostroff & Bowen, 2016). For example, CEOs might outline strategic principles, and other members of the top management team (TMT) might make decisions about HRM policies on the basis of these principles and then delegate the responsibility for implementation of specific practices to human resource (HR) or lower-level managers (Russell, Steffensen, Ellen, Zhang, Ferris, & Bishoff, 2018). Indeed, managers’ roles across the organizational hierarchy make them responsible for much of the observed variability in the adoption, implementation, and effectiveness of HRM practices (Nishii & Wright, 2008).
However, research on managers’ roles in HRM has been unsystematic. Although several reviews and agendas for future research have been conducted for elements of HRM (strategic HRM, or SHRM, Becker & Huselid, 2006; high-performance work practices, Posthuma, Campion, Masimova, & Campion, 2013; governing employees, Su, Wright, & Ulrich, 2018), the field lacks holistic knowledge regarding managers’ roles in HRM. As a result, despite the many studies that have documented the variability in how HRM practices and policies are adopted and implemented—as well as in the individual- (e.g., turnover), unit- (e.g., climate), and organizational-level (e.g., firm performance) outcomes that result from HRM initiatives (Becker & Huselid, 2006; Bowen & Ostroff, 2004; Sikora, Ferris, & Van Iddekinge, 2015)—little is known about the extent to which managers at different organizational levels contribute to this variability. This has limited our knowledge of managers’ impact in HRM, despite general consensus that they are critical to the success of HRM initiatives. Given the significant influence managers can have, it is necessary to understand what we know about their effects on the adoption, implementation, and effectiveness of HRM practices and systems to advance knowledge of effective HRM in organizations.
The purpose of this article is to conduct a multilevel, systematic review of the HRM literature to determine the state of knowledge regarding managers’ roles in HRM. Our review is multilevel because we consider the following roles of managers throughout the organizational hierarchy: lower-to-middle managers (LTMMs), HR managers, TMTs, CEOs, and boards of directors (BODs). Our review is systematic in that we examine the existing literature using an organizing framework that includes the adoption of HRM policies and practices, the enactment or implementation of these policies/practices, and the resultant organizational outcomes.
We contribute to the HRM literature by synthesizing knowledge about managers’ impact on HRM content (i.e., what policies and practices are adopted), HRM process (i.e., the manner and activities through which HRM content is enacted), and HRM outcomes (i.e., what aspects the policies and practices affect and whether they are effective). Furthermore, the multilevel approach of our review enables identification of the focus and volume of extant research at each manager level, which will facilitate the development of specific and valuable avenues for future research on managers and HRM. Therefore, we provide directions for future research specific to each manager level, as well as more general future research directions that should occur across multiple levels to advance knowledge of managers’ roles in HRM.
An Organizing Framework for the Roles of Managers in HRM
To examine the roles that managers play in HRM, we conducted our review using what we refer to as the HRM content, process, and outcomes (HRM-CPO) framework. This framework, presented in Figure 1, draws from what Ostroff and Bowen (2016) identified as three key areas of HRM research: (1) the policies, practices, and systems (i.e., HRM content); (2) the implementation of these practices, policies, and systems (i.e., HRM process); and (3) the subsequent results influenced by HRM content, HRM process, or both (i.e., HRM outcomes). It long has been acknowledged that HRM content should not be considered in isolation but in conjunction with the processes by which the policies, practices, and systems are implemented (Becker & Huselid, 2006). Indeed, for practices to be more than just “isolated acts,” managers, HR professionals, and (as Ostroff & Bowen, 2016, added) scholars “need to be able to explain conceptually how and why HR practices lead to their outcomes” (Ulrich, 1997: 238).

Organizing Framework for Literature on Role of Managers in Human Resource Management Content, Process, and Outcomes
Therefore, the HRM-CPO framework provides an organized way in which we can consider the impact of managers (i.e., LTMMs, HR managers, TMTs, CEOs, and BODs) on HRM content, process, outcomes, and the relationships between them. However, beyond providing a discussion of HRM research that has been conducted at each manager level, we identify the specific manager factors (e.g., characteristics, expertise, and behaviors; also displayed in Figure 1) that influence the different portions of our framework. As we organize the knowledge of managers’ roles in HRM, we intend to provide the type of focus that is needed “to further our understanding of HR content, process, and outcomes” (Ostroff & Bowen, 2016: 207).
HRM content comprises the “what” of HRM (Ostroff & Bowen, 2016), and includes the organizational practices, policies, and systems related to the management of employees (Lengnick-Hall & Lengnick-Hall, 1988). These include high-performance work practices (Posthuma et al., 2013), systems (Lado & Wilson, 1994), or bundles of HRM content (Lepak & Snell, 2002). Our review of the literature seeks to identify what HRM content has been considered in research that pertains to managers. To expressly address our objective of understanding the roles played by managers in relation to HRM content, we are most interested in identifying manager factors that are related to organizations’ adoption of specific HRM content.
HRM process pertains to how HRM content is implemented by organizations (e.g., Sikora et al., 2015). A prevalent conceptualization of HRM process is that of Bowen and Ostroff (2004), who outline several mechanisms hypothesized to yield unambiguous HRM systems. Although these mechanisms have become a meaningful area of HRM research (see Ostroff & Bowen, 2016, for a review), we take a broader view of HRM process. That is, we consider HRM process to be the general manner and activities through which HRM content is implemented and leads to HRM outcomes, including “the way HR policies and practices are communicated to employees” (Li, Frenkel, & Sanders, 2011: 1826).
We identify HRM process as an essential category of our organizing framework for two reasons. First, although it is strategically—and, often, legally—important for organizations to have HRM content “on the books” (Sikora et al., 2015: 1909), it cannot complement organizational strategy and demonstrate meaningful effects if it is not applied throughout the organization (Becker & Huselid, 2006). Second, there is general agreement by researchers that managers across all organizational levels have the potential to play at least some role in HRM process and, therefore, stand as a meaningful context researchers and practitioners can use to better understand the variability that exists in the HRM process (Nishii & Wright, 2008).
Finally, because the general goal of HRM research concerns the effective and efficient management of people in efforts to achieve organizational goals (Hogan & Martínez Lucio, 2007), we review the effect of managers on HRM outcomes. This category relates to any individual-, unit-, or organizational-level outcome of HRM content and/or HRM process. Again, our review is specifically interested in identifying the roles that managers play in influencing these outcomes. Per our framework, we note that HRM outcomes are often, though not always, a function of either HRM content and/or HRM process. However, due in part to the unsystematic way in which managers’ roles in HRM have been examined, it is often difficult to determine how HRM content and HRM process affect HRM outcomes. Thus, we hope to provide meaningful discussions about these roles in what has been referred to as the “black box” (Becker & Huselid, 2006; Messersmith, Patel, Lepak, & Gould-Williams, 2011) that accounts for the translation of HRM content to HRM outcomes.
Method
We took a multistep and iterative approach for our review, in efforts to be both thorough and systematic (Aguinis, Ramani, & Alabduljader, 2018). First, we developed search terms related to both HRM and managers. For HRM, we searched the following 21 terms: human resource*, HR, human resource management, HRM, high commitment work practice*, HCWP*, high commitment work system*, HCWS*, high investment work practice*, high investment work system*, high involvement work practice*, high involvement work system*, HIWP*, HIWS*, high performance work practice*, HPWP*, high performance work system*, HPWS*, strategic human resource*, strategic human resource management, and SHRM. For managers, we searched for the following 28 terms, which span the five categories of managers included in our review: leader*, team leader*, line manager*, middle manager*, LTMM*, human resource manager*, HR manager*, chief human resource manager*, chief HR manager*, CHRO*, top management team*, TMT*, senior vice president*, SVP*, chief information officer*, CIO*, chief technology officer*, CTO*, chief marketing officer*, CMO*, chief operating officer*, COO*, chief strategy officer*, CSO*, chief executive officer*, CEO*, board* of director*, and BOD*.
We performed searches for the above terms using the Web of Science and Business Source Complete databases. Additionally, we performed manual searches of several journals in efforts to identify relevant articles that might have been made available online prior to formal publication and that were not identified by the database searches. More specifically, we manually searched the in-press articles for journals considered outlets for top management publications or specialty outlets germane to the scope of our review. This set of journals included Academy of Management Journal, Administrative Science Quarterly, British Journal of Management, Group & Organization Management, Human Relations, Human Resource Management, Human Resource Management Journal, Journal of Applied Psychology, Journal of International Business Studies, Journal of Management, Journal of Management Studies, Journal of Occupational and Organizational Psychology, Journal of Organizational Behavior, Management Science, Organization Science, Organization Studies, Organizational Behavior and Human Decision Processes, Personnel Psychology, Research in Organizational Behavior, Strategic Management Journal, and The Leadership Quarterly. Our approach was to search each manager term with each HR term in each of these databases and journals. After an initial screening of articles that were clearly extraneous to our review (e.g., research from unrelated disciplines, such as medicine, that used the same or similar acronyms but were not related to HRM or managers), approximately 500 articles remained for possible inclusion.
Next, we reviewed the title, abstract, and full text of each article. For an article to be included, it had to meet the following criteria, which reinforce the overarching objective of this review (i.e., to examine the roles that managers play in HRM content, HRM process, and HRM outcomes): (1) it had to represent at least one of the manager levels we previously identified (i.e., LTMMs, HR managers, TMTs, CEOs, and BODs), (2) it had to include a manager factor (e.g., behaviors, traits, relationships), and (3) it had to focus on managers as actors, rather than as targets, of HRM. Regarding this last criterion, we excluded articles that involved HRM content designed for managers, such as succession plans for CEOs (Gangloff, Connelly, & Shook, 2016), compensation packages for TMTs (Carpenter & Sanders, 2002), or leadership-enhancing HRM content for LTMMs (Han, Liao, Taylor, & Kim, 2018).
This process yielded 72 empirical articles, published between 1995 and 2018, for inclusion in our review. Each article was then independently coded by two coauthors according to our HRM-CPO framework. After the initial coding, interrater agreement was high (r = .92). Discrepancies between coders primarily involved the HRM process category. After discussing the definition and understanding of HRM process, the coders reviewed all discrepancies together to resolve any disagreements, involving the two other coauthors to adjudicate a few disagreements regarding interpretation of variables. All discrepancies and disagreements were resolved, resulting in complete interrater agreement. We present the findings of our review in Table 1 and provide more detailed discussions of these findings in the sections that follow. 1
Summary of Research on Human Resource Management (HRM) at Various Manager Levels
The Roles of LTMMs in HRM Research
We begin our review by examining research that has considered the roles played by LTMMs, considered “the first level of management” (Hales, 2005: 473). These managers not only connect frontline workers with upper-level managers but also often are responsible for executing HRM strategies, policies, and practices (Nishii & Wright, 2008). Given this vital position, it was not surprising that most of the research that exists on managers’ roles in HRM-CPO is at the LTMM level (i.e., 39 of the 72 articles included in our review).
HRM Content
Research at the LTMM level has considered various types of HRM content, from individual high-performance HRM practices (e.g., Han, Bartol, & Kim, 2015; Pak & Kim, 2018; Vermeeren, 2014) to systems or bundles of HRM content (e.g., Chuang, Jackson, & Jiang, 2016; Do, Budhwar, & Patel, 2018; Jiang, Chuang, & Chiao, 2015; Neves, Almeida, & Velez, 2018; Sikora et al., 2015). However, despite the considerable breadth of HRM content considered, our review suggests that little research has been done on the roles that LTMMs play in an organization’s adoption of this content. Instead, most attention has been paid to the roles that LTMMs play in HRM process.
HRM Process
Given our conceptualization of HRM process as the “how” of HR, research in this category addresses the means by which organizations use LTMMs to deliver HRM content to employees. Because LTMMs play a central role in the implementation of HRM content at the individual and unit level, they have considerable influence over the HRM process. For example, LTMMs have the potential to influence employee attitudes and perceptions of HRM content “through their leadership styles, personalities, and behaviors” (Nishii & Wright, 2008: 235).
One focus of LTMM research regarding HRM process is devolution, which occurs when the responsibility to enact HRM content or be involved in HR-related decisions is conferred to lower-level managers (Brewster, Brookes, & Gollan, 2015; Gilbert, De Winne, & Sels, 2011a; Op de Beeck, Wynen, & Hondeghem, 2016). Previous research has suggested that devolution is positively related to HRM effectiveness (Perry & Kulik, 2008; Ryu & Kim, 2013). Additionally, LTMMs’ involvement in the HRM process increases their opportunities to collaborate with upper-level managers or the HR department, helping build relationships that may play a vital role in effective HRM process (Alfes, Shantz, Truss, & Soane, 2013; Crawshaw & Game, 2015) or lead to meaningful horizontal collaboration (Y. Chen, Hsu, & Yip, 2011). LTMMs’ involvement, and the meaningful relationships it yields, helps LTMMs to internalize HRM content and process.
However, not all research on HR devolution indicates beneficial effects. For example, devolution has been found to hinder the strategic positioning of HR departments (Reichel & Lazarova, 2013) such that HR departments that rely heavily on devolution to LTMMs run the risk of delegitimizing themselves. That said, the causal nature of this relationship is unsettled, as Brewster et al. (2015) noted that strategically positioned HR departments are less likely to devolve responsibility to line managers. Devolution also introduces a new demand on LTMMs, complicating their jobs as their roles in HRM process increase. This can introduce stressors such as role ambiguity and role overload (Gilbert et al., 2011a). Interestingly, LTMMs’ HR competency (or their HR-related skills/experiences) has been found to provide a buffer against some of the demands LTMMs face with increased HRM responsibility (Gilbert, De Winne, & Sels, 2011b).
Existing research also provides several other insights regarding the roles of LTMMs in HRM process. Constructs related to LTMMs’ HR knowledge (Ryu & Kim, 2013), social influence effectiveness (Sikora et al., 2015), and communication behaviors (Den Hartog, Boon, Verburg, & Croon, 2013) all affect HRM process. The examination of these factors in previous research reinforces the idea that all HRM content “communicate[s] messages constantly” (Bowen & Ostroff, 2004: 206). LTMMs who can receive HR knowledge via training, who can combine that knowledge with their social influence effectiveness (i.e., political skill) to create shared perceptions among followers of HRM content, and who can deliver consistent messages with effective communication behaviors are well equipped to successfully navigate the HRM process.
HRM Outcomes
Most HRM outcomes examined at the LTMM level have been individual-level outcomes. For example, research suggests that LTMM support for HRM content is a stronger predictor of employees’ acceptance of that content when compared to TMT support (Stirpe, Trullen, & Bonache, 2013). Furthermore, implementation behaviors (broadly conceptualized as the enactment of HRM content) have been found to influence employee affective commitment (Gilbert et al., 2011b; Shipton, Sanders, Atkinson, & Frenkel, 2016) and the development of psychological contracts (Pak & Kim, 2018). Similarly, perceived supervisor support predicts key HRM outcomes such as affective commitment, turnover intentions, and work quality, as Kuvaas and Dysvik (2010) found that perceptions of investment in employee development mediated the relationship between perceived supervisor support and the previously mentioned outcomes. This seems to suggest that when carefully chosen, HRM content can help explain why and how manager factors affect various work-related outcomes.
A few studies have examined how LTMM factors affect HRM outcomes at higher levels of analysis. For instance, Den Hartog et al. (2013) found that LTMMs’ communication abilities predict unit performance, as they lower the discrepancy between LTMM and employee perceptions of HRM effectiveness. Additionally, LTMMs’ HRM role-related behaviors (e.g., HRM decision-making; Azmi & Mushtaq, 2015), LTMM gender diversity (Ali & Konrad, 2017), and the level of involvement of LTMMs in HRM process (Caza, 2011; Dany, Guedri, & Hatt, 2008; Dewettinck & Vroonen, 2017) have been connected to organizational-level performance (Darwish & Singh, 2013).
Summary and Future Research at the LTMM Level
Although LTMMs play an integral role in HRM, our review indicates their roles have less to do with strategy and decision-making and more to do with implementation. Thus, we know much more about LTMMs regarding HRM process and outcomes than on the adoption of HRM content. In particular, our review suggests that LTMMs’ competency and behaviors related to HRM play an important role in implementation, especially in terms of how employees perceive the HRM content they experience and how they evaluate its effectiveness. However, we also note that in some instances there appears to be both substitutive and masking effects in research that examined HRM content and manager factors. For example, LTMMs’ empowering leadership behaviors appeared to be a substitute for the effect of knowledge-intensive HRM content on team knowledge outcomes (Chuang et al., 2016). Alternately, empowering leadership behaviors’ effects on unit climate were masked in the presence of high initiative-enhancing HRM content (Hong, Liao, Raub, & Han, 2016).
Regarding future research at this level, one of the biggest opportunities relates to the roles LTMMs play in the adoption of HRM content. In particular, it would be interesting to understand how an organization might adopt HRM content that “plays to the strengths” of LTMMs. For example, the ability-motivation-opportunity framework suggests that LTMMs’ abilities, motivations, and opportunities predict their effectiveness in HRM (Bos-Nehles, Van Riemsdijk, & Looise, 2013; Gilbert, De Winne, & Sels, 2015; Leisink & Knies, 2011). These findings may not teach us exactly how these LTMM factors affect HRM content, but they do suggest that LTMMs bring something to the table for which organizations should account in determining what HRM content should be adopted. Therefore, we encourage future research that considers the “bottom-up” influence of LTMM factors on organizational decision-making regarding HRM content.
Additionally, future research should continue to explore the devolvement of HRM to LTMMs. Given that research suggests devolution increases HRM effectiveness (Perry & Kulik, 2008; Ryu & Kim, 2013), but at a potential cost to the strategic legitimacy of HR departments (Reichel & Lazarova, 2013) or to LTMMs who face extra job demand (Gilbert et al., 2011b), future research needs to consider its implications more thoroughly. Furthermore, data presented by Brewster and colleagues suggest that devolution may not occur as frequently as researchers have previously thought, “raising questions about the value” (2015: 593) of this practice. Thus, it is important to examine how, when, and under what circumstances devolution is best used.
Finally, much of the previous research at the LTMM level has considered bundles of HR practices. This is intuitive, as LTMMs are responsible for more than one HR practice. However, because explaining the variability in effective implementation is a central concern of this area of research, we agree with Kuvaas, Dysvik, and Buch’s (2014) contention that future research should examine LTMMs’ roles in the implementation of individual HR practices, as being responsible for one type of HRM content may require a different set of competencies than others (e.g., selection vs. training). Although this approach might be more laborious, the precision of studying LTMMs’ responsibility for single HR practices allows for a more accurate understanding of the LTMM factors (e.g., styles or behaviors) related to the effectiveness of these practices.
The Roles of HR Managers in HRM Research
Despite their potential roles in both HRM content (e.g., in creating and designing HR practices) and HRM process (e.g., in developing activities to help HR practices get implemented), we found limited research that has examined the roles that HR managers play in the HRM-CPO framework. In fact, only eight articles involving HR managers met our inclusion criteria. Although HR managers were referenced in many of the articles we examined, their involvement most often was in a capacity of reporting on the nature or status of their organizations’ HRM-CPO (e.g., Perry & Kulik, 2008), rather than as the focal subjects of research. Despite this, we identified several interesting findings related to HR managers’ roles in HRM-CPO.
Several articles examined HR manager factors related to interpersonal processes, such as shared views and collaboration. For example, although consensus regarding HRM effectiveness between HR managers and LTMMs was related to firm performance (Y. Chen et al., 2011), consensus between HR managers and CEOs was not (Guest & Conway, 2011), suggesting that these shared views are more important the closer they occur to where HRM content is implemented. Relatedly, the collaboration that occurs as HR managers are involved in strategic decision-making affects HRM outcomes such as turnover and financial performance, indicating that organizations should consider involving HR managers in the HRM process (Y. Chen et al., 2011; Darwish & Singh, 2013; Sheehan, Cooper, Holland, & Cieri, 2007).
Other research has examined individual-level characteristics of HR managers, such as business-related capabilities and commitment. HR managers whose capabilities extended beyond HRM-related skills (Bjorkman, Ehrnrooth, Smale, & John, 2011; Huselid, Jackson, & Schuler, 1997) and who had affective commitment for specific HRM content (Lievens & Corte, 2008) influenced the effectiveness or approach to HRM process. Finally, Kulik, Cregan, Metz, and Brown (2009) found that in the case of complaint and grievance HRM content, HR managers whose formalized roles included the handling of such events experienced less emotional exhaustion, which then increased perceptions of HRM effectiveness. This study has implications for HRM process and suggests that organizations can play a role in easing the HRM process of HR managers through the formalization of procedures.
Summary and Future Research at the HR Manager Level
We found it surprising that only eight articles involving HR managers examined their role in HRM-CPO. Of these articles, three (i.e., Huselid et al., 1997; Kulik et al., 2009; Lievens & Corte, 2008) focused solely on HR managers. The remaining five studied HR managers in conjunction with other managers, including LTMMs (Bjorkman et al., 2011; Y. Chen et al., 2011; Darwish & Singh, 2013), CEOs (Guest & Conway, 2011; Sheehan et al., 2007), and BODs (Sheehan et al., 2007). Given their function within organizations, we suggest that increased research attention be given to HR managers. For example, future research could follow what has been done at the LTMM level by considering HR manager factors, such as communication behaviors (Den Hartog et al., 2013) or relationships with other managers (Alfes, Shantz, et al., 2013). Future research should also extend the work of Kulik et al. (2009) by considering how the formalized HRM procedures of HR managers play a role in HRM process.
The Roles of TMTs in HRM Research
Firms’ TMTs are charged with making strategic decisions, which are influenced by their personalized interpretations, experiences, values, and personalities (Hambrick, 2007). Given this, TMTs should have significant influence on the type of HRM content firms adopt, the manner in which it is implemented, and its impact on outcomes. However, our review found limited empirical research related to TMTs and HRM. Additionally, many of these studies considered TMTs as the targets of HRM (e.g., with executive-focused HRM systems), rather than the roles they played in HRM. Thus, 14 studies met our inclusion criteria of examining the impact of the TMT on the HRM content adopted by the firm, on the processes through which content is implemented, or on the outcomes of the HRM initiatives. 2
HRM Content
Five studies have examined TMT effects on HRM content. The evidence, although limited, indicates that TMT composition, beliefs about the value of HRM content, and flexible leadership affect the type and volume of HRM content adopted. More specifically, research has found that firm adoption of HRM content is related to TMT commitment to equal employment opportunity–affirmative action values (Konrad & Linnehan, 1995) and TMT value-based beliefs (Arthur, Herdman, & Yang, 2016), as well as their beliefs about the economic benefit of HRM investments (Arthur et al., 2016; Milliken, Martins, & Morgan, 1998). Ali and Konrad (2017) found a positive relationship between TMT gender diversity and the adoption of gender-diversity policies, diversity leadership practices, and work-life programs. Similarly, Milliken et al. (1998) found that gender diversity among the TMT was positively related to work-family HRM content (e.g., leave programs, flexible work options, dependent care programs, financial benefits). Finally, flexible leadership styles of TMT members were positively related to the adoption of high-involvement work practices (Ordiz-Fuertes & Fernández-Sánchez, 2003).
HRM Process
Although TMTs’ positions distance them from direct implementation of HRM content (i.e., relative to HR managers and LTMMs), four of the TMT studies we reviewed examined the effect of TMTs on the process of translating HRM content into a beneficial outcome. However, consistent with expectations about the role of the TMT, these studies primarily explored how TMTs affect the translation of content into outcomes due to the “big picture” they provide for the organization through values, beliefs, trust, and support (Arthur et al., 2016; Farndale & Kelliher, 2013; Stirpe et al., 2013). More specifically, previous research found that firms’ ability to effectively implement HRM content was affected by things such as value-based HRM beliefs held by TMTs (Arthur et al., 2016) or trust in senior management (Farndale & Kelliher, 2013). Similarly, Stirpe and colleagues (2013) found that TMT support for HRM strengthened employee acceptance of HRM content that is championed by HR departments. Thus, extant literature generally lacks information about any direct TMT involvement in implementation. In one exception, Barrick, Thurgood, Smith, and Courtright (2015) found that TMT strategic implementation—which they conceptualize as TMTs’ willingness to define, pursue, and measure strategic objectives—strengthens the positive relationship between HRM practices and collective organizational engagement.
HRM Outcomes
Finally, several studies have examined direct relationships between TMTs and HRM outcomes. In one study, TMT support of HR innovations was positively related to employee acceptance of these HR innovations; however, these effects were not as strong as those of LTMM support (Stirpe et al., 2013). In another study, Konrad and Linnehan (1995) found that TMT values were positively related to the employment status of people of color. Research has also found that TMT behavioral integration (i.e., “a meta-construct that refers to the interactions within the TMT and encompasses elements of information sharing, collaboration, and decision-making”; Carmeli, 2008: 713) and HRM-related capabilities (Carmeli & Tishler, 2006) were positively related to both firm and HRM performance.
Finally, there is evidence of a relationship between HR representation on TMTs and firm performance for fast-growing and smaller firms (Welbourne & Cyr, 1999). However, Ali and Konrad’s (2017) findings indicated the presence of a positive indirect effect from TMT gender diversity to corporate social responsibility, but not return on assets or net profit margin, via work-life programs and gender-diversity policies. In sum, although there are some encouraging results for HRM research at this manager level, a lack of overlap across the studies in this area make it difficult to reach any general conclusions about the effects of TMTs on HRM Outcomes.
Summary and Future Research at the TMT Level
Overall, most knowledge regarding TMTs’ involvement in HRM addresses their effects on the process of translating HRM content into outcomes. This primarily has taken the form of examining the big picture items of TMT values and visions. However, it appears that investigations of the effect of TMT behaviors (e.g., Carmeli, 2008) and characteristics (e.g., Cogin, Sanders, & Williamson, 2018) also hold promise for informing HRM process.
We echo the assertion by Boada-Cuerva, Trullen, and Valverde (in press) that HRM research at the TMT level is ripe for advancement in several directions and following such lines of inquiry, is needed to advance our knowledge in this area. First, given that TMTs are responsible for setting organizational policy, it is surprising that more research has not examined their roles in the selection and adoption of HRM practices and systems. Although some recent research (i.e., Ali & Konrad, 2017) explored a link between TMT gender diversity and the adoption of diversity and inclusion-related systems, more work is needed that extends beyond diversity-related HRM content.
In addition, it would be interesting to explore whether the above (and other) relationships are nonlinear. That is, as diversity increases, is there an increasingly strong relationship between TMT diversity and the adoption of diversity policies and practices? If not (i.e., the relationship is nonlinear and the slope flattens or turns negative), this could suggest that token status TMT members (Kanter, 1977) lack the sufficient power to facilitate the adoption of policies to encourage more diversity. Alternatively, a moral licensing perspective (Miller & Effron, 2010) might suggest that this occurs as a result of majority TMT members “cashing in” moral credits accrued from adding a minority TMT member, such that they feel justified blocking diversity policies. Finally, very diverse TMTs might view the firm as diverse and not see policies designed to increase diversity as a need.
Interestingly, research that found support for TMTs’ roles in HRM outcomes tended to overlook mechanisms that would address how and why TMTs affect HRM outcomes. Therefore, research is needed in this area. For example, although research suggests that TMT composition (e.g., HR representation) affects firms’ long-term financial performance (Welbourne & Cyr, 1999), the mechanisms explaining these effects remain unexamined. Similarly, Cogin and colleagues (2018) recently found that the relationship between work-life support practices and customer satisfaction was stronger when higher proportions of TMT members had children. Although they argued that this signaled the importance of family and helped create a culture where employees felt more comfortable taking advantage of the practices, they did not measure culture in this study.
The Roles of CEOs in HRM Research
CEOs play a prominent role in organizational strategy, decision-making, and multilevel organizational outcomes (Busenbark, Krause, Boivie, & Graffin, 2016; Hambrick, 2007). Given this, it might be assumed that CEOs have a substantial impact on HRM. However, the roles of CEOs in the HRM-CPO framework have not been a prominent focus of research. In total, we found 11 empirical studies that met our inclusion criteria.
HRM Content
Most of the studies examining CEO impact on HRM have addressed the degree and type of HRM content that organizations adopt. For example, research has found that CEOs with more general and international management experience are more likely to make HRM investments (Khavul, Benson, & Datta, 2010). In a Chinese context, college-educated CEOs were more likely to adopt Western-style HRM content to encourage employee participation and empowerment (Frear, Cao, & Zhao, 2012: 4011). Additionally, Chadwick, Super, and Kwon (2015) found support for a relationship between CEO emphasis on SHRM and firm use of commitment-based HR systems. Finally, Michiels (2017) found that family-owned firms with nonfamily CEOs tended to adopt more formal compensation practices.
Some research has examined how CEO leadership styles and behaviors (e.g., transformational vs. transactional) affect the adoption of HRM content. Interestingly, evidence indicates the effects of these behaviors are relatively balanced, and it appears that using CEO behaviors to predict adoption of HRM content depends on the types of content considered. Transformational CEOs, for instance, are likely to adopt HRM content that is skill based (Lopez-Cabrales, Bornay-Barrachina, & Diaz-Fernandez, 2017) and human-capital enhancing (Zhu, Chew, & Spangler, 2005) in nature. On the other hand, transactional CEO behaviors are more likely to adopt function-based HRM content (Lopez-Cabrales et al., 2017). Additionally, research suggests that firms led by CEOs who engage in contingent-reward behaviors (a dimension of transactional behaviors) are more likely to adopt telecommuting HRM content (Mayo, Pastor, Gomez-Mejia, & Cruz, 2009). Finally, whereas Ng and Sears (2012) found that transformational CEOs were more likely to adopt diversity-related HRM content, transactional CEOs who were older and who had higher social values also adopted them (albeit to a lesser degree).
HRM Process
Three studies have examined CEOs and HRM process; however, only one found significant effects. Xi, Zhao, and Xu (2017) found that CEO relationship-focused behaviors were positively related to employee relations climate, which has been conceptualized as the overall agreement among organizational members of the “practices that deal with the governance of the relationship between employees and the employer” (Posthuma et al., 2013: 1195). Employee relations climate subsequently predicted firm performance via employee attitudes. The Guest and Conway (2011) hypothesis that consensus between CEOs and HR managers would strengthen the relationship between HRM content and HRM outcomes was not supported. Likewise, CEO support did not strengthen the HRM process as was predicted by Sheehan and colleagues (2007).
HRM Outcomes
Research involving CEOs and HRM has focused on firm-level HRM outcomes. These articles examined different aspects of firm performance, including employees’ attitudes (Xi et al., 2017), productivity (Chadwick et al., 2015), and absenteeism (Zhu et al., 2005), as well as financial performance (Chadwick et al., 2015; Zhu et al., 2005). Another study examined firms’ dynamic capabilities, conceptualized as the ability that a firm has to respond to the inherent volatility of work environments (Lopez-Cabrales et al., 2017). Interestingly, all four of these articles included CEOs’ behaviors in their examinations of HRM outcomes. These behaviors included emphasizing and supporting HRM (Chadwick et al., 2015; Sheehan et al., 2007), transformational and transactional behaviors (Lopez-Cabrales et al., 2017; Zhu et al., 2005), and relationship-focused behaviors (Xi et al., 2017). Compared to transactional CEOs, transformational CEOs were more likely to adopt skill-based HRM content, which allowed their firms to have sensing and seizing—as opposed to reconfiguring—dynamic capabilities (Lopez-Cabrales et al., 2017). However, there is no indication how this, in turn, might affect performance-related HRM outcomes.
Summary and Future Research at the CEO Level
Because the amount of research on CEOs and HRM was limited, it is difficult to draw broad conclusions about their impact. However, given that several studies have found relationships between CEO characteristics and the adoption of HRM content (e.g., Frear et al., 2012; Khavul et al., 2010; Michiels, 2017), future research should build upon these findings by considering other CEO characteristics or the boundary conditions of their effects on HRM-CPO. For example, does CEO experience (e.g., as CEO at a previous firm or in terms of tenure at the current firm) dictate the adoption of HRM content? It would be interesting to see whether firms with more experienced CEOs are more likely to adopt HRM systems or bundles of practices rather than individual practices, as well as whether these bundles are better matches with firm strategy than those adopted by firms with less experienced CEOs. Similarly, researchers could examine whether the functional background of a CEO affects the firm’s choice of HRM practices, as experience in an HR function might better equip the CEO to recognize appropriate matches between firm strategy and HRM practices.
Although several studies have investigated the effects of CEO leadership behaviors and HRM (e.g., Lopez-Cabrales et al., 2017; Mayo et al., 2009; Zhu et al., 2005), little attention has been paid to their effects on HRM process. Findings in the leadership literature regarding the influences of leadership on followers suggests scholars should investigate the role that CEOs’ varying leadership styles play on TMT and LTMM implementation of, and employee attitude towards, HRM content. For example, does CEO transformational leadership make it more likely that LTMMs will “buy in” to the value of HRM practices and therefore implement them at a higher rate or implement them more effectively? Similarly, does CEO charismatic rhetoric (Baur et al., 2016) make it more likely that employees will view new HRM practices and policies as viable solutions to firm issues or as benefits to themselves?
Finally, although SHRM and strategic management research tends to focus on the effects of CEOs on firm performance, it is surprising that more studies have not examined the effects of CEO characteristics and behaviors on the link between HRM content and outcomes. Similar to the potential research directions outlined above for CEO effects on HRM process, it would seem valuable to investigate whether CEO leadership behaviors or styles enhance the effects of adopted or implemented practices on financial performance outcomes, as well as employee satisfaction and well-being outcomes. Additionally, future research should examine these effects using objective, rather than subjective, measures of performance.
The Roles of BODs in HRM Research
Only six studies that addressed BODs and HRM met our inclusion criteria. This small body of work was largely about how BOD factors influence the selection of specific organizational HRM content. At the BOD level, this focus is intuitive. Given BODs’ governance roles, they are less likely to be involved with the implementation of HRM content at the front line and more likely to influence things such as decision-making or monitoring decision-making. A common theme from several of these articles was how a BOD factor affects the adoption and subsequent “trickle down” of related HRM content. For example, the percentage of women on BODs has been identified as a factor that affects the likelihood that firms adopt LGBT-friendly (Everly & Schwarz, 2015), gender-inclusive (Gould, Kulik, & Sardeshmukh, 2018), and work-family friendly (Mullins & Holmes, 2018) HRM content. Likewise, firms with BOD members who have HR expertise are likely to adopt diversity management HRM content (Mullins, 2018). Such research is in line with resource dependence theory and the capital (e.g., expertise, experience) of BOD members (Hillman & Dalziel, 2003).
However, we know little about how BOD factors might influence HRM outcomes. Sheehan et al. (2007) explored the link between HR representation on the board and perceived organizational performance, arguing that having an HR presence on the board would give it legitimate power with which to influence the organization such that HRM receives greater support. However, Sheehan and colleagues did not find support for this relationship. Additionally, Tsao, Chen, and Wang (2016) found evidence that the presence of independent directors in family firms affected HRM content (i.e., high-performance work system adoption). Their findings also suggest that HRM content mediates the relationship between board independence and subjective performance.
Summary and Future Research at the BOD Level
Although there are a limited number of articles at the BOD level, our review suggests BOD factors do play a role in HRM-CPO, primarily regarding HRM content. For example, HR and underrepresented minority representation on BODs both showed some impact on content adoption. However, we also note that some research found no effects for HR representation on BODs. This could potentially be due to the varying ways in which the concept of “HR representation” was operationalized. In a study that found this relationship, HR representation was measured as both current and former HR expertise/responsibilities (Mullins, 2018), whereas a study that found no effect operationalized HR representation as only current HR responsibilities being present on BODs (Sheehan et al., 2007). As research in this area advances, more thought should be given to how HR representation, and other variables, are operationalized.
Additionally, future research on the intersection of BODs and HRM should focus on HRM content or HRM outcomes, as they are not likely to be involved with, or in a position to affect, HRM process. Regarding content, it would be interesting to extend the investigations of HR representation on boards and the adoption of HRM practices. Resource-based perspectives (Hillman & Dalziel, 2003) would suggest that HR-specific board capital would be beneficial to organizations attempting to match policy with strategy. While several of our identified articles did examine such representation, future research should move beyond gender representation. For example, are organizations with racially and ethnically diverse boards more likely to adopt diversity and inclusion policies? Does the enactment of such policies mediate the relationship between the respective board characteristics and firm performance?
Furthermore, researchers could investigate the impact of other BOD characteristics, including whether BOD ownership relates to HRM policy adoption. Identity theory (Hillman, Nicholson, & Shropshire, 2008), as applied to BODs, argues that board members often identify more strongly with either shareholders or with management (i.e., CEOs) on the basis of their ownership stake in the company. Thus, it would be interesting to see whether firms with high BOD ownership concentrations are hesitant to spend money on HRM initiatives as a result of high identification with shareholders. Conversely, boards with low ownership concentrations (i.e., those that identify more with CEOs) might be more receptive to the adoption of HRM practices. We argue such examinations can help address the SHRM “black box” conundrum (Becker & Huselid, 2006; Messersmith et al., 2011) and may also shed light on how and why BODs influence firm outcomes (e.g., through the effects on HRM content and process).
Challenges and Opportunities for Future Research
From strategic decision-making to implementation at the front line, managers have considerable (if not ultimate) influence over what HRM content is adopted; the activities, manner, and mechanisms by which the content is implemented; and, subsequently, the individual-, unit-, and organizational-level HRM outcomes. Given their crucial roles in each aspect of the HRM-CPO framework, we made recommendations for how research at each manager level should advance to continue this important work. We close our review by considering more generalized challenges and opportunities for future research to include how the roles of managers in HRM-CPO can help advance HRM theory and be more carefully integrated with theory related to HRM and leadership, how research can span multiple hierarchical levels, and the refinement of the HRM process construct.
Advancement of HRM Theory
Our systematic review has clear implications for HRM theory, and we implore researchers to carefully consider and integrate the roles of managers to further advance such theory. For instance, Wright and McMahan’s early theorizing conceptualized SHRM as “the pattern of planned human resource deployments and activities intended to enable an organization to achieve its goals” (1992: 298). Per our review, it is easy to see how managers would play a role throughout this entire definition. For example, organizational decision makers—such as CEOs or HR managers—are responsible for “planned HR deployments” (i.e., adopting or developing HRM content), while TMTs and LTMMs are responsible for the “activities” (i.e., HRM process) that accompany these deployments. By integrating the roles of managers into this conceptualization of SHRM, it becomes clear that the latter could easily benefit from and grow because of the former. Furthermore, we suggest researchers give more attention to the inclusion of managers’ roles as they develop, refine, and/or test HRM theory. For instance, if a model does not explicitly examine the role of managers, we encourage researchers to somehow capture the underlying roles that managers play as part of the mechanisms that affect the adoption and implementation of HRM content. This should be done either to explicitly measure the effects of managers or to more precisely isolate the effects of HRM content.
Two recent perspectives have made progress at moving managers to the foreground of HRM theory. Nishii and Wright (2008) include leaders in their process model as those who are responsible for executing organizational HRM content, therefore playing a key role in the implementation of intended HRM practices. Similarly, Jackson, Schuler, and Jiang (2014) consider the roles of managers in their aspirational framework of SHRM. Jackson et al. acknowledge that LTMMs, together with HR professionals and target employees, are (or should be) involved in “formulating, communicating, and responding to elements of the [HR] systems” (4). Although these are important advancements, we encourage researchers to move beyond just HRM process as it occurs on the front line, as addressing other manager levels will extend these theoretical perspectives of HRM in meaningful ways.
Integrations With Leadership Theory
Surprisingly, a relatively small number of studies in our review invoked traditional leadership theory, such as transformational and transactional leadership (e.g., Han et al., 2015; Lopez-Cabrales et al., 2017; Vermeeren, 2014; Zhu et al., 2005). Therefore, the integration of HRM and leadership research represents a prime area for future inquiry, as leadership theory can be applied to the roles of all managers. In addition to LTMMs’ abilities, motivations, and opportunities and transformational leadership behaviors (Vermeeren, 2014), factors such as LTMMs’ ethical leadership (Brown, Treviño, & Harrison, 2005), authentic leadership (Gill, Gardner, Claeys, & Vangronsvelt, 2018), or abusive supervision (Tepper, 2000) could affect their adoption and implementation of HRM content, ultimately resulting in desirable or undesirable HRM outcomes. More specifically, LTMMs high in ethical leadership (Brown et al., 2005) would be expected to consistently and indiscriminately implement HRM content toward all followers. In contrast, abusive supervision research indicates that highly abusive LTMMs may exploit their power when implementing HRM practices.
Furthermore, we believe the multilevel framework of leadership theories in units (G. Chen & Kanfer, 2006) may serve as a promising theoretical framework to integrate leadership and HRM and fully reveal the role of leadership in HRM. This framework generally suggests that leadership behaviors directed toward individual employees are “discretionary” and can influence individual employees’ experience in a work unit. Moreover, the aggregated leadership behaviors toward all employees in the work unit are “ambient,” form the work unit’s social universe, and affect all employees above and beyond “discretionary” leadership behaviors. It follows that managers signal “what is important and what behaviors are expected and rewarded” (Bowen & Ostroff, 2004: 204) to not only individual employees but also the whole work unit.
More specifically, managers’ leadership behaviors (e.g., transformational leadership behaviors; Vermeeren, 2014) may simultaneously influence individual subordinates’ interpretation and all subordinates’ shared understanding of the availability and reinforcement of HRM practices and policies. For instance, given that transformational leaders are known for caring about followers’ career development needs by showing individualized consideration (e.g., Bass, 1998), it stands to reason that transformational leaders are likely to use training and development opportunities to help their followers meet their career goals. As such, G. Chen and Kanfer’s (2006) framework suggests that highly transformational leaders would not only make training and development programs available to individual followers and encourage them to take on appropriate training and development opportunities but also form a shared understanding among all followers that training programs are available for everyone’s development.
Additionally, we identified a pattern from several articles that suggest either substitutive or compensatory effects between LTMM factors and HRM content. For example, empowering leadership acted as a substitute for the effects of HRM systems on team knowledge acquisition and sharing (Chuang et al., 2016) and initiative climate (Hong et al., 2016). Likewise, the positive effects of service leadership on customer knowledge and service climate were stronger when service-oriented HRM content was low (and vice versa; Jiang et al., 2015). Finally, commitment-based HRM practices were less likely to be resisted by employees when their managers engaged in ethical leadership behaviors (Neves et al., 2018). Collectively, these points reinforce our argument that it will be important for researchers to investigate whether, when, and under what circumstances LTMMs’ factors affect the HRM-CPO framework.
Multilevel Nature of HRM
There are many opportunities to improve our multilevel knowledge of the overlap between managers and HRM. For example, we identified no empirical examinations of the flow of HRM content from where it originates (e.g., top managers or organizational HR decision makers) to where it is implemented (e.g., with LTMMs and frontline employees). The concept of devolution seeks to capture this but often is measured at the LTMM level. Therefore, we reaffirm Ostroff and Bowen’s call that “attention be devoted to a trickle-down effect among leaders” (2016: 206) and call for research that captures the entire devolution process.
In terms of other types of multilevel explorations of managers’ roles in HRM-CPO, we note two specific studies that provide useful templates for conducting such research. Farndale and Kelliher (2013) found that the equitable use of performance appraisals was related to higher employee commitment but that this relationship was stronger when employees trusted senior managers. Thus, employee perceptions of HRM content were affected by the LTMM level (i.e., via procedural justice) and upper-level managers (i.e., via trust in senior managers). These findings show that the manager factors of two different levels of managers were relevant to a favorable HRM outcome. This demonstrates the complexity of the HRM process, as factors at multiple levels need to be considered to tell the complete story. It is also a useful study to model in that it provides a parsimonious look at a multilevel picture.
The second example is Han and colleagues’ (2015) work examining the moderating effects of LTMMs’ contingent-reward leadership behaviors on the linkages between individual pay for performance, expectancy, and job performance. Their model includes three levels (i.e., organization, group, and individual) in conjunction with one HRM practice. We encourage future researchers to consider models like this, as we argue that multilevel models contribute significantly to our understanding of the overlap between managers and HRM.
What Do We Mean When We Say “HRM Process”?
In recent years, researchers have placed increasing emphasis on the need to understand HRM process (Becker & Huselid, 2006; Bowen & Ostroff, 2004; Sikora et al., 2015). They have suggested that we can develop a better understanding of the variability that is often found in HRM outcomes by considering concepts such as how HRM content is implemented (Ostroff & Bowen, 2016), which, in turn, can help address some of the elusive “black box” questions of SHRM (Becker & Huselid, 2006). Therefore, we believe that a great opportunity exists for researchers to explicate further the meaning of HRM process.
To do this, we encourage researchers to take a broader view of this concept by considering HRM process in a more general manner by looking at all mechanisms, activities, and procedures used by organizations to implement HRM content to their employees. A way to understand the importance of this is to compare HRM process to what we know regarding HRM content. Research on HRM content has identified different types of content, taxonomies for individual practices (Posthuma et al., 2013), and concepts such as systems and bundles (Lepak & Snell, 2002). We believe an important next step in fully understanding HRM, and the role that managers play in it, is for researchers to give HRM process a similar degree of attention.
Some progress has been made at mapping this domain. As previously mentioned, Bowen and Ostroff’s (2004) HRM system strength construct has played a key role in laying the foundation for what researchers mean by HRM process. They used climate and attribution theory to argue that the intended messages of HRM content are clearer when they are distinct, consistent, and agreed upon. However, we argue that the process mechanisms related to HRM system strength are likely one of many parts of what might constitute HRM process and, thus, call for scholars to delineate more fully this area of research. For instance, researchers could build on the work of Wooldridge, Schmid, and Floyd (2008) by extrapolating their framework for middle managements’ general strategic roles to their specific roles in HRM. Wooldridge and colleagues argue for research that examines the subprocesses of idea generation, initiative development, and integration/execution. We encourage researchers to overlay these principles onto facets of the HRM-CPO framework by considering the roles that managers play in deciding which HRM practices/strategies get adopted (idea generation), developing the means for how the HRM content will be executed (initiative development), and contributing to the implementation of the HRM plan (integration/execution).
Conclusion
Bowen and Ostroff argued that “HR content and process must be integrated effectively in order for prescriptive models of strategic HR to actually link to firm performance” (2004: 206). As organizational leaders, managers have the ability to determine what type of content is adopted and considerable say over the process through which it is implemented. Thus, understanding managers’ roles in HRM is critical to understanding the impact of HRM in and on organizations. To this end, we have synthesized research on the roles that LTMMs, HR managers, TMTs, CEOs, and BODs play concerning HRM content, HRM process, and HRM outcomes. Our hope is that this review of the current state of knowledge and suggestions for future inquiry will advance HRM research by facilitating increased and specific exploration of the roles managers play in HRM.
Supplemental Material
JOM816179_SM – Supplemental material for Putting the “Management” Back in Human Resource Management: A Review and Agenda for Future Research
Supplemental material, JOM816179_SM for Putting the “Management” Back in Human Resource Management: A Review and Agenda for Future Research by David S. Steffensen, B. Parker Ellen, Gang Wang and Gerald R. Ferris in Journal of Management
Footnotes
Acknowledgements
We would like to acknowledge the support and guidance of our action editor, Brian Hoffman, and two anonymous reviewers. We would also like to thank Jeffrey Ling and Jennifer Sexton for their comments on previous versions of this manuscript. Finally, we wish to acknowledge the research assistance we received from Marcus Atherholt, Marly Bressler, Serena LeMand, and Pam Weir.
Supplemental material for this article is available with the manuscript on the JOM website.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
