A theory of HR co-creation

Abstract We introduce HR co-creation as a continuous process in which HR and stakeholders optimize value through collaborative efforts to innovate in the design and use of HR practices to better satisfy multiple stakeholder needs. This positions stakeholders as users of HR practices who co-create value in partnership with HR, and highlights that value only arises through the use of practices. We outline the conditions for HR co-creation, and argue that HR co-creation generates greater value than practices designed by HR and used by stakeholders independently by maximizing value at the level of the business system. HR co-creation is inspired by the service-dominant logic, and builds on existing multi-stakeholder perspectives of HR, HR process research, and research on employee participation and involvement. We discuss the theoretical and practical implications of this perspective on value creation.


Principles of HR co-creation
As a starting point for describing a theory of HR co-creation, we introduce some base principles, primarily drawing on the service dominant logic (SDL; Vargo & Lusch, 2004, 2016. The SDL was developed as a reaction to the goods-dominant logic that traditionally pervaded the marketing literature (Vargo & Lusch, 2004). The goods-dominant logic assumes that value is created by the firm through the generation of services or products that are sold to consumers to use. Value therefore exists in the raw products or services themselves and this value is delivered to customers. Vargo and Lusch (2004) pointed out that the goods-dominant logic downplays the role of consumers in value creation, considering them as merely recipients of value. Instead, they asserted that a product or service has no inherent value, but only potential value. Real value is created by consumers through their use of a product or service (Penrose, 1995). Value is created when consumers engage in a process of co-creation in which the firm provides the resources (product or service) to enable consumers to create value in a continuous process of resource creation (Vargo & Lusch, 2016). Although there are a handful of studies in HR that have been inspired by the SDL, testing some of its principles (e.g., Meijerink et al., 2016;Meijerink & Bondarouk, 2018;Meijerink & Bos-Nehles, 2017), we lack an overriding framework that can bring this emerging research together. Therefore, we build on the SDL (Vargo & Lusch, 2004, 2016 to elaborate the core principles of our theory of HR co-creation. First, value does not reside in an HR practice itself, but instead, value is created when the HR practice is put to use (Vargo & Lusch, 2004). Early research that has applied the SDL to the HR context has discussed that employees create value from HR practices when they are put to use (Meijerink et al., 2016;Meijerink & Bondarouk, 2018). We expand the notion of users to include more than  Table 1 Examples of HR stakeholders, types of needs, and how needs can be satisfied through independent use of HR practices.

User group(s)
Example of need Example of how independent value is created from use of an HR practice Employees Growth and development Attends performance appraisals and one-to-one meetings to receive feedback and discuss development opportunities.

Line managers
Effective employees Provides feedback and on-the-job training to ensure employees are adequately skilled for the job.

Middle managers
Motivated and high performing team Provides a one-off team bonus for exceptional team performance to motivate future teamworking.

Senior managers Responsiveness to business demands
Engages in workforce planning to identify skill and resource needs for next 1-5 years.
Trade unions Fair pay for workers Takes part in collective bargaining to agree pay increases to ensure fair and equitable pay levels for workers. Shareholders, trustees, board members Dividends; share value; indemnity Provides feedback on executive rewards to help them to better align to business goals and improve organizational performance. Clients, customers High quality service Provides feedback on front line employees' performance during interactions via suggestion box to improve service. Feedback is acted upon, thereby improving customer service. Employees' family Stable home life Partner agrees with employee how to make use of flexible-working policy to collectively manage home demands and create positive family life. Local community Employment opportunities Community members apply for jobs due to a collective positive feeling about the company because of good working conditions. Suppliers Ongoing business Follows the firm's requirements for fair working conditions, creating a safe and equitable workplace in other countries.
employees. For example, employees create value when they choose their flexible benefits, line managers create value when they select candidates for a job, as do senior decision-makers when they design the structure of work teams. External stakeholders also create value through use. This includes customers (e.g., when they provide feedback that improves customer service performance; Hong, Liao, Hu, & Jiang, 2013), family-members (e.g., when they utilize work family practices; Goodman & Kaplan, 2019), and members of the community (e.g., when they use corporate volunteering programs; Caligiuri, Mencin, & Jiang, 2013). Different practices may be relevant to different groups of stakeholders. Family members might be important users of family-friendly policies, whereas customers may be one of many relevant stakeholders who are users of performance feedback. Importantly, use in these terms does not need to involve active engagement with a practice (Vargo & Lusch, 2016). For example, meta-analytic results indicate that the availability of work-family policies had a stronger relationship with employee outcomes (job satisfaction, affective commitment, and intentions to stay) than policy use (Butts, Casper, & Yang, 2013). In this instance, employees created value simply by feeling more positive about the organization as a result of a practice. Table 1 includes examples of HR stakeholders, an illustrative need of each, and explains how users may create independent value from use of the HR practice to satiate that need.
Recognizing that value comes through use, the SDL also requires us to reposition stakeholders as active creators, rather than simply recipients or perceivers of value (Guerci & Shani, 2013;Meijerink et al., 2016;Ulrich & Brockbank, 2005). This second principle of HR co-creation recognizes that HR practices (like other products and services) do not exist in isolation from the actors involved in their design and use (van Mierlo et al., 2018), and that HR processes involve a network of stakeholders who interact with one another to create value (Ferrary, 2005;Meijerink & Bos-Nehles, 2017). This is fundamental to co-creation because it acknowledges that value creation relies on individual users' own resources (skills, knowledge, values etc.). Meijerink et al. (2016) drew on this principle of the SDL and found that employees' own competences shape the effectiveness of HR practices, thereby supporting the idea that "the beneficiary is always party to its own value creation" (Vargo & Lusch, 2016, p. 9).
The third core principle is that HR practices are only valuable if they are perceived to be so by those who use them (an idea also set out by Ulrich & Brockbank, 2005). In other words, value is "uniquely and phenomenologically determined by the beneficiary" (Vargo & Lusch, 2016, p. 8) because value is created when a practice satisfies users' needs (Hsieh & Chang, 2016;Lepak, Smith, & Taylor, 2007;Vargo & Lusch, 2004). This is underpinned by an Aristotelean perspective that value is created when individuals are "better off in some respect" (Grönroos & Voima, 2013, p. 134) as a result of an HR practice (Meijerink et al., 2016). Beneficiaries of HR practices can include a myriad of stakeholders (Colakoglu, Lepak, & Hong, 2006;Ferrary, 2005) who have different needs (see Table 1). This is important because it enables the theory to address the criticism that HR researchers and practitioners continue to equate value with organizational performance (e.g. Beer et al., 2015;Dundon & Rafferty, 2018). We argue that being better off from an organization's perspective is only one manifestation of value so, while financial performance might represent value to shareholders, our definition implies that HR has the potential to provide value to myriad stakeholders with equally countless needs (Tantalo & Priem, 2016;Tsui, 1990).
Our fourth principle builds on the third and relates to how value is created from the perspective of multiple stakeholders simultaneously. A longstanding debate in the HR literature (and indeed in business and management more broadly; Tantalo & Priem, 2016) is the extent to which firms need to temper their economic success in order to satisfy employee needs or address social or environmental concerns (Guest, 2017;Ren & Jackson, 2020). Some scholars suggest that there is an inherent tension between the needs of the firm and the needs of other stakeholders, and that gains for one necessarily entail some losses for another (e.g., Delaney & Godard, 2001;Lengnick-Hall, Lengnick-Hall, Andrade, & Drake, 2009;Porter & Kramer, 2011;Stahl, Brewster, Collings, & Hajro, 2020). Greenwood and Simmons (2004) argued that this win-lose assumption is fundamentally flawed because it represents a distributive approach, implying that HR practices create a 'fixed pie' of value that is divided among stakeholders. Tantalo and Priem (2016) highlighted that stakeholder value-creation is better represented by an integrative approach, which is not restricted by a fixed pie mentality, but instead seeks to enlarge the pie so that all stakeholders have an opportunity to gain even more value (Fisher, Ury, & Patton, 2011;Neale & Bazerman, 1992).
In an integrative approach, value is optimized by seeking synergy among stakeholders' needs and finding creative solutions to meet these needs (Tantalo & Priem, 2016). When adopting an integrative approach, individuals engage in information sharing, so that parties learn the relative importance of the multiple needs of each stakeholder group. We draw on this to suggest that, when HR and users work together in HR co-creation, they can discuss trade-offs between more or less valuable outcomes and seek synergies between different needs to enlarge the pie so that more value is created for all stakeholders. An extensive body of literature on negotiations supports the value of this approach to collective value creation (e.g. Hüffmeier et al., 2018) and that it is particularly valuable in situations where needs are incompatible or conflicting (De Dreu, Weingart, & Kwon, 2000;Harinck & De Dreu, 2004). Integrative approaches to negotiating produces satisfaction and strengthens relationships among group members (Hüffmeier et al., 2018), thereby decreasing the possibility that future conflicts arise, and benefits the larger community of which the negotiating group is a part, such as the overall organization (e.g. De Dreu et al., 2000).
Optimizing value in this way does not mean that everyone receives exactly and all of what they need. Instead, optimizing value means maximizing the shared value for all stakeholders within a business system. Our final principle is, therefore, that to evaluate the total value created by HR practices, we need to consider the sum of the value from the perspective of all individual stakeholders (Tantalo & Priem, 2016). This integrative approach represents a change in focus for HR research. For example, HR process research examines how HR practices can produce desirable outcomes for employees and the organization, but employee concerns are seen as a mediator to organizational outcomes (e.g. Jiang et al., 2012) or as independently valuable (e.g. Ogbonnaya & Messersmith, 2019;Valizade, Ogbonnaya, Tregaskis, & Forde, 2016). Instead, adopting an integrative view of value creation allows different stakeholder concerns to be valuable because together they contribute to the total value created by the HR practice.
In summary, HR co-creation is predicated on the following principles about the way value is created by HR practices: 1) HR practices only create value when they are put to use, 2) users are active agents in the process of value creation, 3) HR practices create value when they satisfy the needs of users, 4) multiple user needs can be met by approaching HR value from an integrative rather than distributive approach, and 5) the value of HR practices should be considered from the perspective of total value, which is the sum of value from the perspective of multiple stakeholders. In the following section we build on these principles by setting out how value can be created by stakeholders through the process of HR co-creation.

The theory of HR co-creation
Consider this example: In the spring of 2020, a major European city went into lockdown as a result of the COVID-19 pandemic. The government ordered the city's semi-public bus service to continue operations. The tragic story of Jason Hargrove, a Detroit bus driver who died from COVID-19 after a passenger coughed without covering her mouth (Levenson, 2020), was on the top of everyone's minds. The HR department was faced with meeting the overlapping yet often conflicting needs of multiple stakeholders. For instance, senior management needed to find new ways to continue operations and ensure economic viability; drivers needed to feel valued and safe, supported by their trade unions; and the community needed reliable and safe transportation for essential workers to travel to work.
To help them to feel safe, some drivers independently handmade plastic barriers that they fitted to their buses. In the first week of the crisis, HR Managers visited the bus depots and drivers shared this innovation with them. HR suggested to the senior management team that buses should only run with the screens in place. This fix was put in place quickly so that drivers felt safe, trade unions were satisfied, and senior managers were able to tell customers and the local government that essential workers could safely travel to work. Drivers also informed HR which depots had sufficient health and safety equipment (e.g., masks, sinks, soap) and the problems they foresaw in carrying out bus runs (e.g., public toilets; ensuring social distancing) and HR worked with office staff to make changes to scheduling and routes. Initially, this did not go well. Different departments were responsible for different parts of bus operations and parties did not traditionally work well together. Using their own expertise and judgement, an HR manager suggested the introduction of cross-functional 'crisis management' teams. HR approached the Chief Operations Officer and together they assigned members to different teams that focused on separate issues. These teams were given arbitrary names (goldfish, crystal, eagle) to minimize preexisting status hierarchies and they worked with drivers to test changes to the safety protocols and observed passenger responses to check their effectiveness. HR, seeing the value of these crisis teams, initiated a process to make cross-functional teams the norm across the organization, even after normal operations resume.
This example illustrates the key characteristics of HR co-creation. The process begins with users' unmet needs. When needs are unmet, users can introduce new practices or make adaptations to HR practices independently of HR, but this value is necessarily limited to the users involved in the adaptation. In our example, the drivers who created makeshift plastic barriers met their own health and safety needs and those of their own passengers, but working with HR ensured the needs of drivers, trade unions, senior management, customers, and the local authority were also satisfied. Likewise, HR made an initial suggestion to create cross-functional teams because they saw a need to adopt a multidisciplinary approach to the crisis. As such, HR built connections among stakeholders and they worked collaboratively to improve and modify the practice after user experience. In both cases, optimal value creation occurred when HR and users collaborated to design HR practices that were put to use, thereby meeting the needs of multiple stakeholders, which also encouraged more co-creation in the future. Of course, not all bus companies found such positive ways to cope with the COVID-19 crisis. For instance, one bus company in North America allegedly suspended a driver, who was also the local union representative, for demanding personal protective equipment (Brown, 2020). This implies that unmet needs only result in HR cocreation when the necessary conditions for co-creation are in place. We elaborate both the process and the conditions of HR cocreation in the following sections.

The process of HR co-creation
HR co-creation involves two interconnected and interdependent processes: collaborative use and co-design. Collaborative use occurs when feedback is either sought or proactively given about user experiences, unmet needs, or challenges associated with HR practices. This can be user initiated (e.g., when the bus drivers shared feedback on safety equipment with HR) or initiated by HR (e.g., when the HR managers worked with drivers and passengers to evaluate the effectiveness of the safety measures). Feedback gathered through collaborative use can then be integrated into the design of HR practices through the process of co-design. Co-design occurs when HR and users jointly utilize their knowledge, skills, and experience to design HR practices to better meet users' needs. This was initiated by users in our example when the bus drivers put up screens and then shared their innovation with HR. Alternatively, it could be initiated by HR, as in our example when HR suggested cross-functional teams.
Collaborative use and co-design connect in a recursive learning process: in our example, the drivers initiated collaborative use when they provided feedback on the availability of safety equipment, and HR acted as the lynchpin to develop this into co-design, involving a greater number of parties (senior managers, trade unions, passengers and drivers). The process could also begin with co-design, as in our example when HR made the suggestion to begin cross-functional teams, and this was then developed by team members. After the new practice is implemented, collaborative use continues as users provide feedback and refinements on the revised practice (van Mierlo et al., 2018). Both collaborative use and co-design are necessary for co-creation. If collaborative use occurs alone it represents only a feedback process from users to HR. Likewise, co-design without collaborative use heightens the risk that practices do not adequately meet users' needs because there is no learning through practice. An example of these interconnected processes can be seen in research by May and Schwoerer (1994) who found that co-created job design enabled employees' ergonomic and physical health needs to be met as well as the organization's needs for productive healthy workforce and the need to retain the knowledge of older staff. Co-design occurred when different users (employees, managers, and medical professionals) were invited to be part of job design groups that integrated employees' experience of their work environment, and the HR professional's knowledge of job design and safety. Collaborative use was evident through the continuous review of the effectiveness of co-designed solutions.
Inspired by Grönroos and Voima's (2013) SDL model of producer/consumer co-creation, we suggest that these processes of HR value creation can be considered in three spheres: the HR sphere, the user sphere, and the joint sphere ( Fig. 1; Table 2). HR practices are often designed in the HR sphere, independently of users. At this stage, the HR practice holds potential value (Vargo & Lusch, 2004). If HR introduces a new HR practice merely because it is fashionable (a common criticism; Abrahamson & Eisenman, 2008) and this practice does not meet stakeholder needs, then it will not be used (e.g., Bondarouk, Ruël, & Roeleveld, 2019). In this case, the potential value held by the practice goes unrealized or is significantly limited.
The user sphere represents independent value creation, whereby users (individually or in groups) create value separately from HR. This occurred in our example when the drivers built their own safety screens. As such, the HR and user spheres represent the traditional assumptions about HR as designers, and users as implementers or recipients of HR practices. When their needs are met by the HR practice, users adopt the practice as it was intended by design. However, when users' needs are unmet by the existing practice, they are motivated to make adaptations. Adaptations made in the user sphere constitute independent customization and are either behavioral or cognitive in nature. Examples of behavioral customization include idiosyncratic deals (i-deals; Rousseau, Ho, & Greenberg, 2006), whereby employees (or potential employees) negotiate specific terms and conditions of employment with their (future) employer; and behavioral job crafting where individuals take active steps to shape the task or relational boundaries of their job (Wrzesniewski & Dutton, 2001) to meet their needs, such as improving work-life balance (Sturges, 2012). Behavioral customization is also seen in research that shows that line managers make local adaptations to the organization's HR practices (Kehoe & Han, 2020;López-Cotarelo, 2018). Cognitive customization is akin to cognitive crafting, whereby individuals change the way they think about how their work is designed to satisfy their need for meaning (Wrzesniewski & Dutton, 2001), or through assigning attributions about the intention behind HR practices to make sense of them (Nishii, Lepak, & Schneider, 2008).
Although customization in the user sphere is value creating because it makes a person better off, it is necessarily limited because it is bounded by the users involved in customization. In our example, the drivers created independent value by installing the screens, but this served to meet only their needs and those of their own passengers. If they had not shared this with HR (at which point they entered the joint sphere), the potential benefits to other users would not be realized. Customization by users independently of HR might also lead to undesirable outcomes that further restrict the value creation process. In the case of i-deals, for example, customization of employment conditions might create value for the employee and manager dyad involved in the i-deal but it may frustrate other users' needs (e.g. colleagues; Lai, Rousseau, & Chang, 2009, or managers;Hornung, Rousseau, & Glaser, 2009). Likewise, managers could make adaptations to HR practices to meet their needs that are harmful to others, such as failing to act on misconduct, or to provide feedback to job candidates (e.g. López-Cotarelo, 2018). Aligned to this, research has shown that HR specialists view independent customization of practices as undesirable (Guest & Bos-Nehles, 2013), leading them to monitor HR processes to ensure procedural compliance (Beer et al., 2015), frustrating the possibility of co-creation so that needs continue to be unmet. Instead, HR co-creation suggests that while users create value through use, HR capture and retain this value through collecting and sharing experiences (Lepak et al., 2007) and "facilitating the diffusion of innovations from one part of the organization to another" (Beer, 1997, p. 53), thereby increasing the value created by the practice. As such, the role of HR cannot be understated; they are the lynchpin that turns independent value creation into value co-creation.

Conditions of HR co-creation
This begs the question: When do individuals respond to unmet needs by making independent customizations, and when do they engage in HR co-creation? HR co-creation is a complex relational process, involving the effort of dyads or groups with goals that may or may not always be compatible (Guest & Bos-Nehles, 2013). HR co-creation also operates within a larger social context, including established HR practices and inevitable power disparities among parties. Therefore, we integrate the insights from the SDL (Vargo & Lusch, 2004, 2016 with relevant research from the organizational sciences about factors that help build relational ties, foster collaboration, information sharing and initiative-taking in groups, and create integrative outcomes (Gittell & Douglass, 2012;Prahalad & Ramaswamy, 2004;Ranjan & Read, 2016;Rouse, 2020;Tantalo & Priem, 2016), which are all characteristics of HR co-creation. We synthesize these insights to identify three conditions that provide a nurturing environment for HR co-creation; ability-based trust, psychological safety, and when parties use and appreciate power based on knowledge and/or relationships. 3 Combining insights from the SDL (Vargo & Lusch, 2016) and HR scholarship (e.g., Bowen & Ostroff, 2004;Guthrie, 2001;Kehoe & Collins, 2017), we also acknowledge that structural features are antecedents of the conditions for HR co-creation.
First, parties are only likely to engage in HR co-creation if they trust in their own and others' ability to co-create effectively. Trust that the other party (or parties) has complementary knowledge is particularly important because individuals are more likely to seek co-3 Through our review of this literature, we found multiple typologies with overlapping characteristics. We based ability-based trust on suggestions that knowledge (Ranjan & Read, 2016) transparency (Prahalad & Ramaswamy, 2004), and mutual trust are necessary for co-creation (Rouse, 2020); psychological safety reflects conditions of safety, affection, cohesion (Rouse, 2020), relationship coordination and coproduction (Gittell & Douglass, 2012); and power is reflected in the recognized importance of relational leadership (Gittell & Douglass, 2012), and complementary utilities (Tantalo & Priem, 2016). Potential value is created through the creation of HR practices before they are used.
Actors interact through a continuous process of design and customization to enable practices to meet diverse user needs and enhance commitment to practices.
Users create value through independent adoption or customization of HR practices.
An integrative approach to value creation occurs through a virtuous cycle of HR co-creation, originating from stronger relational ties and increased perspective taking thereby maximizing value at the level of the business system. Use (including evaluation) and independent adaptation of HR practices to meet individual (or group) needs. Reactively takes part in co-design of HR practices to meet needs. Examples 1) HR designs a new appraisal process and communicates process via self-service portal to employees, line and project managers. 2) HR organizes wellness day and invites employees and their family to attend.
1) Customers provide feedback about their experience of sales process that informs training needs of sales assistants. 2) HR seeks feedback from employees about how pay decisions could be communicated to enhance feelings of procedural fairness.
1) Employees select options from flexible benefit plan to support their, and their family's, wellbeing. 2) Current and past employees write company reviews online, and applicants read them to inform decisions. created solutions if believe they have insufficient knowledge or influence to bring about the desired changes on their own (Budjanovcanin, 2018;Kellogg, 2019). In HR co-creation, this is most likely to be a combination of users' knowledge about their own needs and experience of using HR practices, and HR's specific expert knowledge, which is their "knowledge of the principles, practices, and functions of effective human resource management" such as workforce planning, managing disciplinary processes, or carrying out training evaluations (SHRM, 2019). This ability-based trust (Mayer, Davis, & Schoorman, 1995) partly depends on individuals' self-efficacy about whether they have sufficient knowledge about HR practices to make a meaningful contribution. For example, individuals are more open to improvementrelated suggestions from others when they believe in their own expertise (Fast, Burris, & Bartel, 2014), and users are more likely to engage with HR practices if they have confidence in their abilities to make use of HR practices (Bos-Nehles, Van Riemsdijk, & Kees Looise, 2013). Ability-based trust is also other-focused, in that specific knowledge about HR practices signals to others that one is able to contribute to co-creating HR practices. For example, evidence suggests that users engage with HR practitioners if they trust in HR's expert knowledge (Guest & King, 2004;Pritchard & Fear, 2015) and HR are more likely to involve users in HR practices if they perceive that they have HR-related capability (Björkman, Ehrnrooth, Smale, & John, 2011;Op de Beeck, Wynen, & Hondeghem, 2016). Professional expertise is also needed from a practical perspective to enable HR professionals to effectively enact changes to practices that better meet users' needs (Aldrich, Dietz, Clark, & Hamilton, 2015;Parker, Andrei, & Van den Broeck, 2019).
Proposition 1a. Trust in one's own and others' ability to co-create increases the likelihood of HR co-creation.
HR co-creation is also more likely to occur when users and HR feel psychologically safe to take interpersonal risks (Edmondson, 1999). This is because HR co-creation involves sharing new ideas, learning from one another, and evaluating one's own and others' ideas objectively to find ways to improve HR practices. These actions require individuals to take risks and make themselves vulnerable: HR are vulnerable to criticism of the HR practices they are responsible for, and users are vulnerable when they provide feedback that leaves them open to retaliation (Bies & Tripp, 1996). Although psychological safety is an individual level perception, it reflects "essentially a group-level phenomenon" (Edmondson & Lei, 2014, p. 30) because it relates to safety within a specific group. In the context of HR co-creation, this means that individuals engage in co-creation when they feel psychologically safe in the network of HR and users who are involved in co-creating practices to better meet users' needs.
Psychological safety is important for HR co-creation because it enables individuals to work collaboratively across boundaries to solve complex tasks (Edmondson, 1999;Edmondson & Lei, 2014). Research shows that those who work in psychologically safe environments are open to learning from one another (Siemsen, Roth, Balasubramanian, & Anand, 2009), suggest ideas to make improvements (Liang, Farh, & Farh, 2012), and take initiative on new endeavors (Baer & Frese, 2003;Nembhard & Edmondson, 2006). At the group level, a psychologically safe environment enables teams and individuals to learn (e.g. Carmeli & Gittell, 2009), and be more creative (e.g., Hu, Erdogan, Jiang, Bauer, & Liu, 2018), which are all characteristics of co-creation. Likewise, research shows that individuals suggest improvements or innovations, collaborate with others, and exert discretionary effort when others are open, inclusive, and supportive of new ideas (e.g. Detert & Burris, 2007;Scott & Bruce, 1994).
Proposition 1b. Psychological safety in the network of HR and users increases the likelihood of HR co-creation.
Many SDL researchers have suggested that notions of command, control and exploitation are irrelevant to the collective production and cooperation that are central to value co-creation (Cova, Dalli, & Zwick, 2011;Prahalad & Ramaswamy, 2004;Vargo & Lusch, 2004). Yet critics attest that the idea of a harmonious co-creation of value is naïve, simplistic, and overly unitarist (Chowdury, Gruber, & Zolkiewski, 2016;Peñaloza & Venkatesh, 2006). This critique is particularly relevant in the context of HR where parties such as employees, managers, unions, customers, and shareholders have varying degrees of legitimate power (i.e. hierarchy-based and exercised through coercion and reward) due to their proximity to the organization, the strength of their interests, or the structures that provide power (e.g. Ferrary, 2005). Moreover, whereas shareholders and managers typically hold legitimate power, HR often lacks it (John & Björkman, 2015;Trullen & Valverde, 2017), which limits their ability to engage in the "deviant innovation" that is required for HR co-creation (Guest & King, 2004;Legge, 1978). Gittell and Douglass (2012, p. 719) argued that co-creation is possible even when there are differences in legitimate power and interests between two stakeholders by recognizing that "[b]oth are necessary, neither is sufficient, and neither is intrinsically more important than the other". However, legitimate power is only one base of power from which people may draw (French & Raven, 1959). Gittell and Douglass (2012) suggested that co-creation is more likely when individuals "exercise power and influence based on their personal qualities rather than positional authority" (p. 719). Therefore, power can be harnessed to enable co-creation, but this depends on the type of power and how power is enabled. Supporting this idea in the context of HR implementation, Budjanovcanin (2018) found that employees utilized expert power (originating from specific HR knowledge), information power (from knowledge of HR practices), referent power (from interpersonal relationships) and connection power (connections with those with legitimate power) to actively adapt and change HR practices.
Examining when these sources of power can be harnessed for co-creation, Kellogg (2019) found that junior staff in medical institutions were able to co-create changes to working practices by utilizing the informal power that originated from their unique competencies. Those with high positional power enabled this process by, for example, facilitating learning and rewarding innovation. Wiewiora, Chang, and Smidt (2020) found that managers in project-based organizations were able to implement new ideas when they had forged relationships with those who held positional power. Likewise, although power asymmetries can stifle learning (Bunderson & Reagans, 2011), research suggests that learning can occur when parties have expertise and build relationships with one another, thereby reducing power-related barriers to learning (Kehoe & Tzabbar, 2015). This is aligned to Ferrary's (2005) perspective that success in implementing HRM-related changes is more about how relationships with parties with different sources of power are managed, rather than the instrumental content and process of the change. Therefore: Proposition 1c. Individuals are more likely to engage in HR co-creation when they are able to exert influence over the design and use of practices through power that is based on knowledge or relationships.
Efforts to co-create are embedded, shaped by, and reproduced within existing structural and social systems (Edvardsson, Tronvoll, & Gruber, 2011). Vargo and Lusch (2016) theorized that value creation is embedded in a web of social networks and associated institutions. They emphasized the importance of "institutions", which are practices, norms or meanings, and "institutional arrangements", defined as "inter-related sets of institutions that together constitute a relatively coherent assemblage that facilitates coordination of activity in value co-creating service ecosystems" (p. 18). Drawing from the SDL, we assert that organizational and social structures have the potential to support HR co-creation by enabling the three main conditions for HR co-creation. By recognizing the shared values, beliefs, practices, and norms (i.e., institutions as per SDL) that surround and influence HR and users, this provides a fuller picture of how value in the HR context is created. Although there are likely to be several manifestations of structural systems that are important in influencing the conditions for HR co-creation, we focus here on two: organizational forms and systems of HR practices.
Some organizational forms are inherently more conducive to HR co-creation. Holacracies or project-based organizations, for instance, distribute power throughout the organization where information flows through linked working groups, individuals exchange roles as needed, everyone has a voice, and lines of accountability (rather than hierarchy) organize work (Romme, 2015;Wiewiora et al., 2020). Traditional command and control hierarchies, on the other hand, are more likely to impede the conditions needed for HR co-creation. In bureaucratic forms of organizing, employees work within siloes, are often inattentive to the needs of others, and this mode of organizing hampers trust among members (Gittell & Douglass, 2012).
Even when organizational forms do not inherently support co-creation, systems of HR practices themselves can facilitate the conditions needed for HR co-creation. Systems of HR practices can support co-creation by enabling relational ties and fostering a collaborative environment between users and (representatives of) the organization. For example, high-involvement HR systems (Guthrie, 2001) focus on "practices that develop and support a workforce that is self-programming and self-managing" (p. 181), and employee perceptions of high-involvement work practices are positively related to organizational innovation because they elicit collective interactions for knowledge exchange (Li, Wang, van Jaarsveld, Lee, & Ma, 2018). In a similar vein, specific configurations of HR practices engender collaboration (Mossholder, Richardson, & Settoon, 2011), co-operation (Kang, Morris, & Snell, 2007) and relational goals (Kehoe & Collins, 2017). These bundles of practices send signals that proactive, collaborative, and improvementfocused behaviors are accepted and rewarded (Bowen & Ostroff, 2004), so act as an antecedent to the conditions needed to foster HR co-creation. Although a particular organizational form or bundle of HR practices are not necessary ingredients for HR co-creation, these structural conditions are levers for facilitating trust, psychological safety, and the recognition of multiple forms of power. Therefore: Proposition 2. Certain organizational structures or systems of HR practices enable the conditions for HR co-creation.

How does HR co-creation create value?
The first way that HR co-creation generates value is through the satisfaction of needs; HR practices that are co-created are better able to satisfy users' needs compared to practices that are developed by HR alone. However, there are potentially multiple users of each HR practice, and each user may have a set of needs, some of which may be incompatible with the needs of other users. Although HR research has recognized that multiple user needs can be met when the needs are complementary (Peccei, 2004), HR co-creation provides a more detailed explanation of how HR can manage conflicting needs. To do so, the level of analysis needs to shift from individual users to the business system level (Tantalo & Priem, 2016). The goal of HR co-creation is therefore to maximize the value that is created by all stakeholders within the system, not necessarily to maximize the value for an individual user.
Strategy scholars have identified several ways that multiple goals can be accommodated simultaneously. First, since each stakeholder has a set of economic and noneconomic needs and weigh each differently depending on the time and circumstance, stakeholders can trade-off with one another to optimize the total value created. Even when parties have ideological differences, research suggests that stakeholders may emphasize activities to reach a mutually beneficial solution in the interests of long-term survival (Thompson & Bunderson, 2003). A second way to meet more than one stakeholder's needs is to generate an innovate solution to create new sources of value (Neale & Bazerman, 1992;Tantalo & Priem, 2016). This is evidenced in research that finds that collective participation in problem solving creates more creative solutions that better fit multiple parties' needs (Hargadon & Bechky, 2006). For example, in response to a financial crisis, organizations adopting a distributive approach could respond by making employees redundant to save immediate bottom-line costs to ensure that the business stays afloat; employees lose and the firm wins. Through an integrative approach they could reduce labor costs by implementing a 4-day working week, long-term hours flexibility, or inviting employees to take periods of unpaid leave (Cascio, 2002;Porter & Kramer, 2011). This satisfies the needs of shareholders, managers, employees, trade-unions, family members and customers. This is a manifestation of the integrative assumptions that underpin HR co-creation in that individuals seek innovative solutions to increase the "size of the pie" available for different user groups thereby creating greater total value at the level of the business system (Neale & Bazerman, 1992;Tantalo & Priem, 2016).

Proposition 3a.
Total value is optimized when parties make trade-offs or innovate.
Users who co-create an HR practice also experience higher commitment to the practice. This is supported by research that shows that individuals are more committed to team decisions if they are involved in the development of the process to reach specific outcomes (Dooley, Fryxell, & Judge, 2000). Likewise, research shows that involving employees in change interventions ensures commitment to change and learning (Nielsen & Randall, 2012) and results in less cynicism towards organizational change (Brown & Cregan, 2008). Interestingly, research on co-creation in product development has found that consumers identify more with and become more committed to a co-created service or product (Atakan, Bagozzi, & Yoon, 2014), a phenomenon known as "the IKEA effect" (Norton, Mochon, & Ariely, 2011). Greater commitment to HR practices is value creating because users are more likely to utilize practices to which they are committed (Sikora & Ferris, 2011), thereby realizing the potential value inherent in the practice. Proposition 3b. Users feel more committed to HR practices that are co-created.
HR co-creation also generates greater value by fostering the conditions that encourage more co-creation; a virtuous cycle that is represented in Fig. 1. The reciprocal and beneficial exchanges that typify co-creation help to form a set of common values, conferring a shared purpose between HR and users resulting in stronger relational ties (Gittell & Douglass, 2012;Rouse, 2020). This is important because there is often a sense of psychological distance between HR and users of HR practices (Guest & King, 2004). In particular, when HR is not visible (e.g. off-site or outsourced) or users are more distal or secondary (e.g. communities' engagement with industrial relations; Rainnie, Herod, & Champ, 2007) interactions between HR and users are limited (Glaister, 2014). Co-creation activities generate opportunities for interaction thereby strengthening the relationship between these potentially disconnected parties.
This, in turn, enables greater opportunities for co-creation because individuals are more willing to trust in each other's knowledge and abilities (Kim, Su, & Wright, 2018) and feel more psychologically safe in closer relationships (Edmondson & Lei, 2014). Likewise, individuals with legitimate power are more likely to relinquish control, enabling those with alternative sources of power to influence outcomes (Budjanovcanin, 2018;Wiewiora et al., 2020). As a result, parties are more likely to cooperate to achieve integrative solutions when they have a closer relationship with a shared purpose (Bundy, Vogel, & Zachary, 2018). In other words, the stronger relationship between HR and users that arises from co-creation activities further fosters the conditions needed for HR co-creation.
Proposition 4a. The process of HR co-creation leads to stronger relational ties between HR and users, which in turn enables future HR co-creation.
HR co-creation also represents a virtuous cycle of value creation through increased perspective taking; a process that enables "individuals [to] adopt others' viewpoints in an attempt to understand their preferences, values, and needs" (Grant & Berry, 2011, p. 79). Individuals are able to form a more accurate impression of others' perspectives when they gather new information through interpersonal interactions (Eyal, Steffel, & Epley, 2018). This is supported by research conducted by Parker and Axtell (2001) who found that employees were better able to take the perspective of external suppliers when they interacted with them, had a greater understanding of the suppliers' job, and how their own job contributes to the whole (knowledge that is fostered through engaging in co-creation). Therefore, engaging in HR co-creation enables individuals to form a more accurate understanding of other stakeholders' needs and how they can be met.
This represents a virtuous cycle because perspective taking in turn creates the conditions that enable HR co-creation to take place: Perspective taking inherently enables individuals to have a better appreciation of each other's abilities (Parker & Axtell, 2001); predicts feelings of psychological safety (Post, 2012); and perspective taking enables those with legitimate power to better recognize, and enable, the influence that arises from others' knowledge and relationships (Kellogg, 2019).

Table 3
Comparison of HR co-creation to other stakeholder-led perspectives of HR.

Perspective
Stakeholder role Perspective on control Level of agency permitted to stakeholders Which stakeholders are considered?

Source of value
Multi-stakeholder perspectives (e.g. Beer et al., 1984;Beer et al., 2015) Stakeholder needs inform design, and stakeholder outcomes should be considered.
HR or management controls other stakeholder inputs.
Passive recipients. All stakeholders. Satisfied users' needs.
Management retains control but permits employee input when invited.
Passive as standard, but active within permitted mechanisms or interactions.
Employee need satisfaction within the bounds of senior decision-maker priorities.
Process approaches to HR (e.g. Bredin & Söderlund, 2011;Meijerink et al., 2016) Role-based focus (e.g. HR as designers, managers as implementers) although increased empirical emphasis on active, dynamic roles of stakeholders in design and implementation.
HR or management controls other stakeholder inputs.
Traditionally passive recipients, increasingly active users.
Primarily employees and managers (at different levels).
Organizational performance mediated by satisfied users' needs. Customers' concerns and shareholders as a secondary outcome. HR co-creation Stakeholders are actively involved in design and use, which can be initiated by HR or users.
HR and those with positional power relinquish centralized control of HR practices.
Active users. All stakeholders. Satisfied user needs, greater commitment to co-created HR practices, and enabling future HR co-creation.
Proposition 4b. The process of HR co-creation leads to perspective taking between HR and users, which in turn enables future HR cocreation.

Implications of HR co-creation
The theory of HR co-creation extends at least three areas of scholarship: multi-stakeholder approaches to HR, which point to the need to consider all stakeholder needs, not just the business (e.g. Beer, Spector, Lawrence, & Mills, 1984;Paauwe, 2004); HR process research, which focuses on explaining how HR practices add value through their implementation and use (e.g. Nishii & Wright, 2008;Sanders, Shipton, & Gomes, 2014); and employee participation, which emphasizes the importance of two-way interactions between employees and decision-makers (e.g. Marchington, 2015;Strauss, 1998). Despite these substantial areas of scholarship, stakeholder concerns are still predominantly prioritized insofar as they benefit the organization. Pfeffer (2016) pointed out that scholars have forgotten that there is a moral imperative to consider the impact of HR practices on multiple stakeholders "not because these things affect costs and profits, although they surely do, but as important outcomes in their own right" (p. 668). We suggest that HR co-creation theory extends these existing lines of research by reevaluating the assumptions that underpin the role of multiple stakeholders in HR value creation. In particular, by recognizing that HR practices only create value if they satisfy users' needs, and that this goal can be best achieved when stakeholders are active agents in creating value, HR co-creation offers opportunities to further our understanding of stakeholders in value creation. The key differences between these three bodies of research and HR co-creation are summarized in Table 3.

Multi-stakeholder perspectives of HRM
Multi-stakeholder approaches to HR have been particularly influenced by the ideas set out by Beer et al. (1984) in their "Harvard Model" of HR. Beer et al. recognized that HR practices create value such that they have a positive influence on multiple, long-term outcomes of individual wellbeing, organizational effectiveness, and societal wellbeing, and that these outcomes are of equal importance. There have been several theoretical extensions to the Harvard Model that emphasize the importance of legitimacy, fairness, and morality (e.g., Aust, Matthews, & Muller-Camen, 2020;Boselie, 2014;Paauwe, 2004), and there have been attempts to map and measure indicators of performance for multiple stakeholders (e.g., Colakoglu et al., 2006;Ferrary, 2005;Stahl et al., 2020). The multistakeholder perspective therefore represents a more collective approach to HR with a definition of value that is more than, yet inclusive of, firm performance (Cleveland et al., 2015;Dundon & Rafferty, 2018;Greenwood & Simmons, 2004) and recognizes that HR practices are only valuable when they satisfy users' needs (Ulrich, 1989). In this regard, it is aligned to the philosophy of HR cocreation.
The importance of this perspective has been widely discussed, but it has not been extensively implemented in either practice or research. In their reflection on 30 years of the multi-stakeholder perspective of HR, Beer et al. (2015) expressed frustration that "with some notable exceptions, the field has become successful by developing a narrow focus on economic performance [and]… has lost sight of the wider group of stakeholders" (p. 428). One potential reason for the lack of theoretical and empirical development is that the multi-stakeholder perspective lacks substantial detail on how these ends can be achieved (Jackson et al., 2014). Another potential reason is that the field has been underpinned by the dominant assumption that when change occurs, there is a polarization of stakeholders that results in alliances and win-lose relationships (Ferrary, 2005). Theory is needed to address how different stakeholder needs, which can at times be at odds, can be managed.
There have been recent efforts to address this, particularly in a Special Issue of Human Resource Management Review (Westerman et al., 2020). In one paper, Podgorodnichenko, Edgar, and McAndrew (2020) suggested that to balance incompatible goals of multiple stakeholders, organizations need to develop ethical leaders, create a culture that values sustainability, and build a structure that promotes collaboration, creativity and innovation. While these factors may help to create a context that recognizes tensions among stakeholders, they do not (as the authors acknowledge) resolve paradoxical tensions. HR co-creation addresses this by approaching value creation with an integrative approach and repositioning total value from the perspective of the business system. This moves us beyond the dominant perspectives that consider stakeholder outcomes as either exclusive of one another (e.g. Ogbonnaya & Messersmith, 2019;Valizade et al., 2016) or as a mediator of value for the business (e.g., Jiang et al., 2012;Nishii & Wright, 2008). Instead, HR co-creation allows us to evaluate multiple stakeholder outcomes simultaneously in a single equation. To do this, it is first necessary to identify an organization's stakeholders and their respective needs (that are self-identified), then measure the extent to which these needs are satisfied. This approach, inspired by Tantalo and Priem (2016), involves weighing stakeholder needs depending on the focus of the practice. For example, the needs of family members may be considered in measuring the value of flexible working practices but would be weighted far less (if at all) when evaluating the value of recruitment practices. The result is that HR can ascertain which stakeholders may be under-served, and how this contributes to overall value.
In the same special issue, Ren and Jackson (2020) argued that resolving the paradox inherent in the multi-stakeholder approach requires "HR institutional entrepreneurs" to transform the traditional logic that HR is meant to increase firm performance to a logic that defines performance as the achievement of business, natural environment, and community goals simultaneously. Stakeholder goal incompatibility is assumed to diminish when HR institutional entrepreneurs are able to align people and processes to the three stakeholders simultaneously. Although we agree with Ren and Jackson that the dominant HR logic requires an update, their perspective positions stakeholders as passive actors to be influenced by the HR institutional entrepreneur. Instead, we argue (on the basis of the SDL) that a better reflection of the role of stakeholders in value creation requires us to see them as actively involved in the process of value creation.

HR process research
Research on HR processwhich focuses on the mechanisms through which HR practices inform organizational performance (Sanders et al., 2014) has traditionally been governed by the assumption of a linear value chain whereby HR design practices and managers implement them, thereby shaping employees' attitudes and behaviors that inform performance (e.g. Nishii & Wright, 2008). A growing body of research is challenging this linear assumption by examining the processes involved in HR practices in nontraditional organizational forms and in the context of increasing digitization of HR, finding that this process involves more than this triad of actors (Bredin & Söderlund, 2011;Keegan & Den Hartog, 2018;McDermott, Fitzgerald, Van Gestel, & Keating, 2015;Meijerink et al., 2016), and that the processes are often nonlinear (Bondarouk et al., 2019;Trullen et al., 2020;van Mierlo et al., 2018). While this body of work offers valuable insights into experiences in a range of contexts, it lacks a guiding theoretical framework (Jackson et al., 2014) and the existing HR value chain is unable to fully account for the HR processes that this body of work highlights (Bondarouk, Trullen, & Valverde, 2018).
Several papers have also emerged to reposition users as more active agents in HR process. In their review article, Kehoe and Han (2020) discussed ways that line managers enact agency in HR process and highlighted a number of unresolved questions. In particular, they pointed out that we know little about when line manager adaptations might help or hinder organizational outcomes and suggested that line manager expertise is often a limiting factor. HR co-creation responds to this by suggesting that the kind of independent adaptations discussed by Kehoe and Han are limited in their ability to create value; line managers (and other users) create more value when they work in collaboration with HR, thereby pooling expertise and experience. Meijerink et al. (2016) likewise drew on the SDL to empirically demonstrate that employees' own competences shape the effectiveness of HR practices, thereby supporting the idea that users co-create value with HR. We build on these ideas by explaining how the process of co-creation occurs, and the mechanisms through which employees' characteristics shape HR experience.
In a similar vein, van Mierlo et al. (2018) and Trullen et al. (2020) highlighted that HR practices are not passive entities but are shaped by the actors who use them. In their theoretical models they suggest, like us, that employee behavior that deviates from the intended use of the HR practice serves to shape the practice itself in a recursive process of evolution (van Mierlo et al., 2018) until it becomes routinized (Trullen et al., 2020). We take this a step further by explaining how this deviation can best add value; not through independent customization (which is the focus of Van Mierlo et al.'s model) but in concert with HR who do more than "monitor the progress of implementation…modifying the intended HRM practice to provoke desirable behavioral reaction from managers and employees" (van Mierlo et al., 2018, p. 3036). In HR co-creation, HR practitioners (or their proxies) have a key role to play in enabling and harnessing value creation. Our conditions likewise contribute to Van Mierlo et al. and Trullen et al.'s perspectives by identifying when deviation from intended practices can enable co-creation of value, which is not the focus of their models.
The theory of HR co-creation therefore provides a much-needed theoretical framework to guide this growing theoretical and empirical interest in a more dynamic approach to HR value creation. HR co-creation suggests that a better representation of multistakeholder HR process involves understanding user needs and designing practices to meet multiple user needs, rather than involving user needs only when they are useful to the organization. This means that scholars should place the user as more central in explaining these value creating processes, as this better represents how users and HR engage in the process of co-creation and the conditions necessary for this to occur.

Employee participation
Rooted in the industrial relations literature, employee involvement and participation (EIP) traditionally referred to formal mechanisms whereby employee representatives (such as trade unions) and managers met to discuss employment conditions not covered by collective bargaining, in, for example, joint consultative committees (Marchington, 2015), the Germanic model of codetermination (Harcourt, Gall, Wilkinson, Croucher, & Lam, 2020), or partnership agreements that grew out of the UK (Ackers & Payne, 1998). The decline of institutions such as trade unions and works councils has increasingly led to the adoption of informal systems and interactions (Harcourt et al., 2020;Marchington, 2015). Informal EIP incorporates the "day to day relations between supervisors and subordinates in which the latter are allowed substantial input into decisions…a process which allows workers to exert some influence over their work and the conditions under which they work" (Strauss, 1998, p. 15). EIP is represented by practices including formal problem-solving groups, suggestion schemes, and engagement surveys, as well as informal interactions, and proactive invitations by managers to involve employees in decisions affecting their employment conditions and work activities. As such, many of the practices included in the definition of employee participation are the same as those involved in HR co-creation.
Although EIP is predicated on the importance of two-way interactions between employees and decision-makers, it diverges from HR co-creation because it largely adopts a top-down perspective whereby management (which may be represented by HR) "allow" (Strauss, 1998) employees to be involved in decision-making. For example, Harcourt et al. (2020) explained that the decline of codetermination in Germany is the result of management vying to retain control over decision-making. This is corroborated by research showing that managers reject expressions of employee voice because of status quo and confirmation bias (Li, Barnes, Yam, Guarana, & Wang, 2019), and that the positive potential outcomes of EIP were unrealized because employees felt frustrated that, despite management rhetoric, they had little influence over decisions (Leana, Ahlbrandt, & Murrell, 1992). A longitudinal study on the effects of a partnership among workers, management, and a trade union showed that although the partnership produced mutual gains (e.g., more training and development, better safety climate), the issues that were collectively discussed focused on business needs only, and not the needs of various stakeholders (Glover, Tregaskis, & Butler, 2014). Furthermore, research has also found that employees are often reluctant to engage in EIP due to the top-down nature of the process (Townsend, Wilkinson, & Burgess, 2013). Concerns have therefore been raised that EIP is just another form of management control.
A key difference between EIP and HR co-creation is the assumptions that underpin them. EIP is representative of the distributive 'win-lose' assumption that we discussed earlier. The integrative assumptions that underpin HR co-creation explicitly require parties operating in the joint sphere to work together on an equal footing by placing users' needs central, whereas EIP is controlled from above. Therefore, while some of the mechanisms that enable formal or informal EIP are likely to be the same as HR co-creation, the key difference is therefore the approach by which these exchanges are approached. In other words, while HR co-creation is unlikely to occur without EIP, EIP can occur without co-creation. Furthermore, by specifying that HR co-creation requires the user-centric processes of co-design and collaborative use we make it clear that HR practices need to continuously evolve through use to better suit users' needs. This is distinct from EIP, where these processes are controlled by managers who choose when to invite users to participate, how much to listen to their perspective, and when and how to adapt HR practices. We therefore see HR co-creation as way to better understand the beneficial and reciprocal exchanges that are evidenced between HR and users, whereas EIP focuses on a oneway process of control. Furthermore, the conditions that enable HR co-creation are aligned to the factors that have been found to limit the utility of EIP, highlighting the importance of these to facilitate these beneficial, reciprocal exchanges.

Practical implications
Although HR co-creation may seem like an idealized form of practice, we find evidence that it already occurs in a range of organizational forms (e.g., Burrell, 2018;Coolen, 2019;May & Schwoerer, 1994;Ulrich, 1989). Yet this is only possible when the conditions for co-creation are in place. Together with HR, all users who engage in HR co-creation are responsible for creating these conditions. To engender ability-based trust, both HR and users need to have, and be seen to have, expert knowledge. This knowledge is particularly important because the design and use of HR practices are bounded by legislative and institutional requirements that must be navigated. This requires HR to balance a user-centric approach with an adherence to prescribed rules (Fu, Flood, Rousseau, & Morris, 2020). This means that, rather than architecting rigid and over-prescribed HR practices, practices need to be loosely designed, so that they are applied flexibly to meet more users' needs (aligned to Kuvaas, Dysvik, & Buch's, 2014 concept of 'enabling' HR practices). HR's involvement in co-design establishes the boundaries of what is essential due to external constraints, and users' experience informs how those ends can be achieved. From a user's perspective, expert knowledge is derived from their experiences of HR practices, so users need to make use of the practices available. To demonstrate their expert knowledge, users need to provide feedback to HR about the effectiveness of HR practices, and any adaptations they make to create value. If each party recognizes these complementary contributions, ability-based trust is more likely to thrive.
Research on the antecedents of team psychological safety suggests that role modelling on the part of the leader, through coaching (Edmondson, 1999), and demonstrating integrity (Leroy et al., 2012) and humility (Hu et al., 2018) builds safety, roles that are particularly important for HR and organizational decision-makers. Although HR can facilitate psychological safety in the joint sphere by setting up formal team structures, organizing shared team rewards, and celebrating successful co-creation outcomes (e.g., Bresman & Zellmer-Bruhn, 2013;Chen & Tjosvold, 2012), all users, such as employees, trade union representatives, or middle managers, need to play an active role in ensuring that it is safe to speak up in the joint sphere. Furthermore, psychological safety varies with levels of familiarity (Koopmann, Lanaj, Wang, Zhou, & Shi, 2016) and the quality of social relationships among team members (e.g. Gu, Wang, & Wang, 2013) so allowing time for social interaction as part of formal co-creation activities (e.g. focus groups) would be beneficial.
Users who hold legitimate power need to relinquish centralized control when operating in the joint sphere. Those without legitimate power are encouraged to utilize their knowledge and relationships to identify opportunities to improve HR practices to benefit more than one individual or group. One way to harness power to generate support for HR co-creation is through coalition building; creating a temporary alliance of individuals focused on joint action for a specific purpose (Stevenson, Pearce, & Porter, 1985). Bruch and Sattelberger (2001) explained how Lufthansa made a radical turnaround, from near bankruptcy in 1991 to their best results in over 70 years in 1999. Part of the turnaround is attributed to HR's role in initiating and nurturing a network of change agents throughout the company and building coalitions with employee representatives and works councils within and across Lufthansa's businesses. Coalitions help parties to overcome conflicting needs by focusing attention on a shared goal (e.g. designing practices that meet users' needs) and are especially effective when legitimate power is low (Brass & Burkhardt, 1993). This is also supported by Trullen and Valverde (2017) who found that establishing coalitions was most strongly correlated with successful HR implementation over and above other types of influence tactics.
HR co-creation also has practical implications for HR professionals, or anyone who holds responsibility for the design and diffusion of HR practices. This theory may empower HR because their involvement is necessary for HR co-creation; they are the lynchpin in turning independent value creation into value co-creation. As such, HR must be aware of users' experiences, to keep an eye out for innovations, and to create opportunities for the interactions that are necessary for co-creation to occur. HR analytics, for example, provides an opportunity for HR to engage in "continuous listening" (Coolen, 2019) about how practices are experienced by multiple stakeholders. This means more than simply being aware of who HR's stakeholders are, but also understanding their unique needs, how different stakeholders relate to each practice and, when evaluating the value created by a practice, weighting each stakeholder per practice in order to measure value at the level of the business system. For example, in their deployment of agile cross-functional teams, global banking group ING uses quarterly business reviews to track whether teams are innovating in a way that meets customer, business, and employee needs (Barton, Carey, & Charon, 2018).

Concluding remarks
There is an emerging body of research that questions some longstanding assumptions about value in the context of HR. The time is right for HR scholarship to develop theory that frames and guides this research. We see HR co-creation as a way to meet this end because it provides a more accurate account of how HR practices are used to create value in many organizations. It is also an aspirational framework for how users can be placed as more central in a truly stakeholder-focused approach to HR. HR co-creation provides considerable insight into HR value creation by reframing the design and use of HR practices as a continuous process that derives from the collaborative efforts of HR and users to innovate HR practices that meet multiple user needs. Our conceptualization also offers a way to balance the interests of multiple stakeholders, by focusing on value as a sum of multiple stakeholder interests. HR co-creation theory opens up new possibilities for HR scholarship, practice and education, and we hope that it acts as a catalyst for new conversations on how HR can truly facilitate value.

Author statement
We declare that there are no conflicts of interest associated with this manuscript. The paper is not under consideration for any other publication and it has not appeared in print in any form.

Declaration of Competing Interest
None.