Abstract
Based on a longitudinal, qualitative analysis of developments in the English National Health Service, we develop a process model of how organizations divest or spin off units with the aim of establishing two or more autonomous organizational entities while simultaneously managing their continued interdependencies. We find that effective organizational separation depends on generating two types of respect—appraisal and recognition respect—between the divesting and divested units. Appraisal respect involves showing appreciation for competence or the effort to achieve it, while recognition respect requires considering what someone cares about—such as values or concerns—and acknowledging that they matter. The process model we develop shows that open communication is crucial to the development of both. We also find that certain attempts to gain organizational independence and respect may unintentionally undermine the development of autonomy. Counterintuitively, we find that increasing or maintaining interorganizational links via communication may facilitate organizational separation, while attempts by units to distance themselves from one another may unintentionally inhibit it. By linking organizational separation, autonomy, independence, and respect, this paper develops theory on organizational separation processes and more generally enhances our understanding of organizational autonomy and its relations with mutual respect.
Keywords
Separation can be difficult, especially between historically highly interdependent actors. We intuitively recognize that separation between individuals, such as between two partners or a parent and child, can trigger complex dynamics. But the process of separating organizational entities from one another—via divestitures, spin-offs, or government-mandated break-ups—may also be far from straightforward, not least “due to the tight linkages that exist in today’s organizations” (Deloitte Corporate Finance, 2009: 8). Because divesting and divested units often continue to share resources such as staff and administrative processes during a transition period or for many years thereafter, “Both spinners and spun are odd hybrids: new companies with long histories; independent entities that have close ties with each other” (Economist, 2013; see also Semadeni and Cannella, 2011). In many cases, ties between divesting and divested units are never completely relinquished (Moschieri and Mair, 2012; Feldman, 2016).
Organizational change scholars often treat organizational separation as any other form of strategic change, involving a potentially difficult break with the past (e.g., Bartunek, 1984) and negotiations of how to act going forward (e.g., Gioia et al., 2010; Jarzabkowski, Lê, and Feldman, 2012). But this overlooks the fact that additive and subtractive forms of change are associated with quite distinct challenges (Albert, 1992): while organizational integration, such as a merger or acquisition, requires generating synergies by combining and standardizing practices, separation involves developing increasingly specialized organizational entities by disconnecting practices that may be highly entangled with one another.
Although organizational separation has been relatively understudied (Corley and Gioia, 2004: 174), a number of insights specific to this phenomenon are scattered across several literatures. In the strategic management literature, divestiture studies have elaborated both positive and negative performance implications for divesting or divested units of continued links between the separating entities. On the positive side, maintaining ties between separating units, such as in the form of dual directors (Feldman, 2016), gives a divesting unit continued access to resources and potential innovations (Moschieri and Mair, 2012). Maintaining ties with the divesting unit may also be beneficial for the divested unit, such as by providing access to useful social networks. Continued ties may be detrimental (Semadeni and Cannella, 2011), however, because they may allow the more powerful entity (usually the divesting unit) to impose its decisions on the less powerful entity (usually the divested unit) (Feldman, 2016). If the relationship is both highly interdependent and characterized by conflict, the more powerful organization may attempt to penalize the other, thereby harming the latter’s organizational performance (Walter, Heinrichs, and Walter, 2014).
Furthermore, the strategic management literature has suggested that continued ties between separating entities may inhibit the establishment of a new organizational identity at the divested unit (Moschieri, 2011). This builds on organizational identity scholars’ insights that subtractive change generates internal and external pressures to address questions concerning “who we are as an organization” (Corley and Gioia, 2004: 176; Gioia et al., 2010). The loss of central meanings may impose a cognitive load on the divested unit, requiring efforts to reestablish order via the rapid formation of an organizational identity (Corley and Gioia, 2004; Ferriani, Garnsey, and Lorenzoni, 2012; Sahaym, 2013). But this literature does not provide an in-depth understanding of the separation process itself—how potentially complex relationships between a divesting and divested unit are negotiated and managed.
A few studies have examined strategic change processes in contexts of organizational separation (Balogun and Johnson, 2004; Nag, Corley, and Gioia, 2007; Jarzabkowski, Lê, and Van de Ven, 2013). Though these studies have not explicitly theorized subtractive change, their empirical findings provide useful information about the types of issues that may arise in subtractive change processes. Balogun and Johnson (2004) found that questions around resource control (“who owns what”) and the separation of roles (“who does what”) may be central when organizational units become formally distinct yet remain potentially highly interdependent in practice. Diverging interpretations of roles and responsibilities can trigger acrimonious “us vs. them” feelings between newly separated units. The emergence of tension between separating organizational units is also documented by Jarzabkowski, Lê, and Feldman (2012), who noted how apparently incompatible demands to maintain a close relationship with a highly interdependent organizational unit and simultaneously treat it as no longer integral can have implications for members’ belonging and performance. These studies indicate that dynamics between separating entities concerning the management and negotiation of complex, transitionary relationships may shape the organizational separation process.
This paper develops theory on how members of interdependent organizational entities manage a process of separation from one another. We adopt a practice-based approach (Feldman and Orlikowski, 2011; Nicolini, 2012) to examine a case involving the complex separation of community care services provided by CommunityProvider (the divested unit) from MgmtAgency (the divesting unit) in the English National Health Service. 1 We examine dynamics that help us unpack how organizational independence and autonomy are interrelated and develop theory concerning the interplay of these dynamics with communication, competence, and respect. In doing so, we provide insights into how attempts to separate historically interdependent organizational units can succeed or fail.
Organizational Autonomy and Independence
The primary objective of organizational separation is to establish autonomy for the separating units, unless they were already acting autonomously prior to the initiation of structural change. Thus understanding how organizational separation is successfully implemented requires understanding how organizational autonomy is achieved. Even though autonomy is foundational to organizations (Brunsson and Sahlin-Andersson, 2000) and to workers’ satisfaction (e.g., Hackman et al., 1975; Langfred, 2005), organizational autonomy is rarely explicitly defined in management and organization theory. Instead, it is commonly associated or equated with terms such as independence, sovereignty, and self-governance (e.g., Oliver, 1991; Pfeffer and Salancik, 2003; Moschieri, 2011; Drees and Heugens, 2013).
Broadly speaking, autonomy refers to an actor’s experience of being able to act freely, without being obstructed by external interference. The concept has its roots in moral philosophy, referring to actions dictated by free will that is unhampered by oppression, manipulation, or instinctual drives (Dworkin, 1988). In political philosophy, autonomous citizens are those who are able to live their lives without oppressive control by society. In Western society, the “freedom to make one’s own choices” is regarded as a human right that all individuals enjoy and whose restriction is warranted only to protect oneself or others from harm (Mill, 2016). In psychology, autonomy refers to self-regulation, as opposed to heteronomy, which denotes control by external influences (Deci and Ryan, 1987; Ryan and Deci, 2006). Developing autonomy requires the ability to draw distinctions between oneself and others—to understand where one’s sphere of control ends and another’s begins (Bowlby, 1998).
Autonomy therefore is not just freedom from undue outside influence but also requires the capacity to exercise discretion: to be autonomous is not only to be granted autonomy by those who could interfere but to be capable of claiming it by acting autonomously. Importantly, while an actor can be characterized as autonomous, autonomy is enacted in specific practices that require varying degrees of competence (Breaugh, 1985). Even a member of a highly autonomous profession, such as a medical doctor or lawyer, may not claim or be granted autonomy concerning practices that are recognized (by the professional and/or others) as beyond that person’s expertise or ability. Applied to the organizational level, autonomy refers to having discretion over how to perform organizational practices (Brunsson and Sahlin-Andersson, 2000). An organization is expected to be held fully responsible for decisions relating to internal resource allocations (Oliver, 1991: 945) and is therefore by definition autonomous; a non-autonomous organization is an oxymoron.
While autonomy and independence are often treated in management and organization theory as synonymous, sociologists and developmental psychologists explicitly distinguish them, such that dependence on another actor does not automatically imply interference by that other actor. Durkheim (2014) highlighted that members of a society are by nature interdependent and that a society could not function properly otherwise. This, however, does not mean that its members cannot be granted autonomy or that they could not behave autonomously amidst a network of interdependent relations. This is because being dependent on contributions from others does not rescind the ability to exercise discretion, although it might reduce the range of choices or make some more or less attractive. Developmental psychology similarly highlights that children may be influenced strongly by their parents without this influence being experienced as interference (Bowlby, 1998). In other words, autonomy does not require independence.
Thus organizational autonomy refers to performing organizational practices without explicit direction or approval from others, while organizational independence refers to performing practices without being influenced by others. To become a separate organization, organizational autonomy is essential, while independence is not. Clarifying this distinction may help explain contradictory findings concerning organizations’ attempts to reduce dependencies while entering into or maintaining relationships that simultaneously appear to increase them (Oliver, 1991). Moreover, explicitly distinguishing between autonomy and independence allows us to disentangle dynamics that help explain how and why organizational separation may succeed or fail.
Methods
As organizational separation processes remain undertheorized, an exploratory case study using a longitudinal, qualitative orientation to identify relevant dynamics is appropriate (Yin, 2003). Our data emerged from a broader study of strategic change implementation in the English National Health Service (NHS) that the first author conducted in conjunction with an academic research program on healthcare service procurement and healthcare systems management (see also Wiedner, Barrett, and Oborn, 2017). During this study the complexity of organizational separation between two units became apparent. After examining literature on divestitures and organizational change processes, we realized that developing an in-depth understanding of the complexity of this process was of theoretical interest (e.g., Brauer, 2006). As the study progressed, the research problem was continuously clarified by interrogating emerging findings and consulting academic literature to make sense, and guide further collection, of empirical data (Alvesson and Kärreman, 2007; Mantere and Ketokivi, 2013).
Research Setting
The NHS has undergone several waves of reform in recent decades, primarily due to increasing financial pressures. Arguably the most profound restructuring initiative during this period involved organizationally separating from one another the procurement and provision of public healthcare services (DH, 1990). As part of this purchaser/provider split that was initiated in 1991, hospitals and units providing mental healthcare services became separate legal entities that competed with one another for income from local healthcare management agencies across the country. 2 In many cases, service providers received the majority of their income from their former parent organizations, while management agencies relied heavily on their “children” organizations (i.e., hospitals and mental healthcare service providers) to meet local healthcare demands.
While hospitals and mental healthcare service providers were formally becoming more autonomous, other areas of healthcare remained organizationally integrated. For instance, the majority of local NHS management agencies continued to provide community care services—including health visits and care for housebound patients—themselves, via their own community care units that employed nurses, pediatricians, social workers, and members of many other occupations. Moreover, general practitioners (GPs or family doctors), who as “primary carers” refer patients to hospitals and community services for particular forms of treatment, were represented on their local management agency’s board and worked together with managers to plan, redesign, and monitor local healthcare services (Greener and Mannion, 2009).
Over time, several policy initiatives and guidelines explored the possibility of extending the increased autonomy that hospitals and mental healthcare service providers enjoyed to community care. Notably, the national Transforming Community Services Programme, launched in 2008, encouraged healthcare management agencies to divest their community service units. This restructuring was justified on the grounds of enabling healthcare management agencies to focus on their role of managing local health economies and allowing community services to benefit from the same degree of autonomy as hospitals and mental healthcare service providers. Shortly thereafter, in a white paper titled “Liberating the NHS” (DH, 2010), the UK government announced that all management agencies were obliged to divest community care services and procure them from “any qualified” provider organization. Figure A1 in the Online Appendix (http://journals.sagepub.com/doi/suppl/10.1177/0001839218779806) offers a simplified overview of these national developments.
One of the earliest attempts to formally separate the delivery of community care services from local healthcare system management began in 2007 at MgmtAgency, a slightly above average–sized NHS management agency in terms of number of staff (approximately 3,000). Following the MgmtAgency board’s decision to divest its community care services division, it recruited a team of managers to lead the division and its transition into a separate organization. To ensure that the change was implemented smoothly and to adequately prepare managers for their new roles, MgmtAgency established CommunityProvider as an “arm’s length provider” (Health Service Journal, 2010) for an interim period that lasted almost three years. During this time, CommunityProvider operated as a subsidiary with its own board. A timeline of key events surrounding the separation is presented in table 1.
Events in the Process of Separating CommunityProvider from MgmtAgency
MgmtAgency transferred clinical staff to CommunityProvider at the outset of the process. The division of management roles was less clear. While CommunityProvider required managers for the services it was expected to provide, MgmtAgency also required staff who were knowledgeable about these services because it was responsible for procuring and evaluating services on behalf of its local population. Thus two distinct types of community care service managers were needed following organizational separation: service managers at CommunityProvider and “commissioners”—managers in charge of procuring and monitoring services—at MgmtAgency. Managerial skills shortages, especially for highly specialized services, subsequently appeared at both organizational entities.
Moreover, in the past, changes to existing services had been designed and implemented internally and often informally within MgmtAgency. In the same manner, dealing with potentially serious, unexpected issues related to any community care service—which could trigger major service changes—had been a largely internal and informal affair. In the new context of formal, interorganizational contractual relationships, new procedures were required for notifying stakeholders, addressing issues, changing services, and potentially penalizing the organizational entity responsible for failing to comply with contractual terms and conditions. Hence at both organizational entities, adjustments, additional resources, and the development of new skills were needed to competently perform practices.
Data Collection
Our initial interviews with MgmtAgency managers highlighted the emergence of tension with CommunityProvider as it transitioned from being an arm’s-length provider to a separate legal entity. Upon learning of this tension, the first author reached out to people directly involved in this interorganizational relationship and requested interviews, permission to attend interorganizational meetings, permission to shadow community care commissioners at MgmtAgency, and access to MgmtAgency’s confidential documents regarding the procurement of community care services. Permission was granted under the condition of anonymity, and ethical clearance was obtained.
Semi-structured interviews
The first author conducted 66 semi-structured interviews with healthcare service contract managers and other individuals directly involved in aspects of community care, mental health, and acute healthcare service contract management between 2010 and 2014. This paper primarily draws on interviews conducted with members of MgmtAgency and CommunityProvider who had some involvement in shaping the emerging interorganizational relationship. We interviewed 12 CommunityProvider managers (executives and senior and middle managers), 16 MgmtAgency senior and middle managers involved in contracting community care services, and nine GP representatives who were affiliated with MgmtAgency and directly involved in co-designing its community care services strategy for the region.
We deliberately kept interviews open, allowing respondents to reflect on the interorganizational relationship, how it had developed, challenges that emerged, and how these were dealt with from their point of view. Later interviews included more-specific questions to gain further insights concerning topics and themes that appeared to be particularly salient for understanding why certain developments had occurred. With a few exceptions, all interviews were recorded, transcribed, and imported to NVivo for qualitative analysis.
During the study, MgmtAgency announced that it would not automatically renew its existing contracts with CommunityProvider and that it was putting the majority of services that it commissioned from it out to bid. This decision generated uncertainty about CommunityProvider’s viability as a separate legal entity, triggering a wave of resignations. It also provided a unique opportunity to observe emotional responses and capture managers’ reflections on why and how the level of interorganizational tension had built up over the years to the point of threatening the survival of the divested organization and of adequate community care service provision in the region. A few managers refused to participate, but others were very willing to share their views and talk about their experiences, including CommunityProvider’s CEO and other members of the executive team.
Observation
After having conducted a few interviews with MgmtAgency managers, the first author negotiated a three-month shadowing study (Czarniawska, 2007), which involved observing MgmtAgency contract managers responsible for non-hospital-based services for three to four full workdays a week. This was subsequently extended by another three months in which the study focused exclusively on the emerging relationship between MgmtAgency and CommunityProvider. Much time during the shadowing studies was spent observing both formal and informal internal MgmtAgency meetings, meetings between MgmtAgency and CommunityProvider representatives, and meetings with other stakeholders.
The shadowing study generated several types of data. First, in formal meetings, discussions were largely captured verbatim and combined with notes concerning observed dynamics. For discussions outside of formal meetings, the first author wrote or typed extensive notes and later enriched them with further recollections of events. Second, directly before and after formal meetings, as well as at other non-obtrusive moments, he conducted informal, ethnographic interviews with MgmtAgency and CommunityProvider members to gauge their reactions to recent events, understand their concerns, and clarify issues. Third, a summary of each day’s observed events was typed up. Fourth, the first author regularly captured personal reflections about witnessed events, with a particular focus on recurring themes and surprises that informed further data collection. These forms of data collection yielded several hundred pages of notes that we imported into NVivo for subsequent analysis.
Archival data
As part of the shadowing studies, we also collected electronic and paper documents that MgmtAgency managers were working on or that were distributed to all attendees at meetings. These included meeting agendas, draft and approved minutes, strategy presentations, service evaluation reports, service specifications, copies of e-mails, formal letters, contracts, and more. Together with interviews and observation, these documents provided insights into the issues that managers were dealing with. In total, we collected over 20GB of data from MgmtAgency, largely consisting of documents produced between 2008 and 2013. We were also given audio recordings of approximately 27 hours of MgmtAgency’s contract oversight meetings, in which issues relating to healthcare service contracts were discussed. Finally, we collected relevant news, independent inspection reports, and trade journal articles mentioning MgmtAgency, CommunityProvider, and community care service procurement in general, along with annual reports and press releases from the two organizational entities.
Data Analysis
Having continuously refined the research problem by iteratively collecting and analyzing data, we arrived at our research question that focused our analysis on explaining how and why attempts to develop organizational autonomy, and thereby effectively implement organizational separation, had been only partially successful. As our objective was to understand how separation had developed, we adopted a process approach to analyze our data (Langley, 2009). Given that we collected extensive data pertaining to only one case of organizational separation, we used intra-case comparisons to identify and explain consistency and variation (Pettigrew, Woodman, and Cameron, 2001; Ozcan, Han, and Graebner, 2017). Our focus on organizational autonomy led us to use organizational practices as units of analysis, which is consistent with an ontology that understands practices “to be the primary building blocks of social reality” (Feldman and Orlikowski, 2011: 1241), allowing the deconstruction of taken-for-granted concepts into their constituent practical enactments of recurrent activities (Nicolini, 2012).
Using the constant comparative method (Glaser and Strauss, 2000), we compared and contrasted developments within and between practices in which autonomy had developed, as shown in table 2. Our first-order analysis (Van Maanen, 1979) resulted in over 50 codes that indicated commonalities within and differences between the sets of practices and how they had developed over time. We then engaged in several rounds of grouping these codes into second-order themes and, consistent with the abductive nature of our study (Mantere and Ketokivi, 2013), consulted the literature to assist our interpretations of the emerging findings. For instance, although the data highlighted respect (and a lack thereof) as a key difference between practices in which autonomy developed and failed to emerge, we were able to distinguish between different types of respect only after having immersed ourselves in the respect literature across multiple domains, including sociology, developmental psychology, and philosophy.
Overview of Practices Identified as Requiring the Development of Autonomy
Our second-order analysis resulted in 28 second-order themes, consisting of sets of behaviors and dispositions (Bourdieu and Wacquant, 1992; Emirbayer and Johnson, 2008). Table A1 in the Online Appendix provides an overview of identified themes and representative data for each theme. Continuing to move back and forth between our data and literature, we then sought to establish linkages among the themes we had identified. After mapping out relationships, we again focused on those that were common within our categories of practices and differed between them. This allowed us to develop diametrically opposed process models, namely a model of facilitating and a model of inhibiting organizational autonomy in the process of organizational separation, and to link them to increased organizational independence and the maintenance of organizational dependencies, respectively. Once we had mapped out all dynamics we then grouped them into four sets of processes that we present in the next section. We were able to validate our interpretations of findings with members of CommunityProvider after data collection had been completed: the first author presented an analysis of the MgmtAgency–CommunityProvider dynamics to CommunityProvider’s executive team, triggering a discussion and follow-up meeting.
Findings
We structure our account by drawing a distinction between the practices that were performed autonomously five years after the initiation of organizational separation and those that were not; see table 2. In our narratives below we refer to individuals affiliated with MgmtAgency and CommunityProvider by their job title or first name, as listed in table 3. The key element in our process model, which arose through our abductive analysis, is the concept of organizational respect, defined as the worth that interdependent actors accord to each other (Rogers, Corley, and Ashforth, 2017). Our findings suggest that two quite different processes may be triggered when organizational autonomy is initially lacking in the context of organizational separation. Effectively disentangling interdependent organizational practices requires a period of heightened interorganizational communication to facilitate the development of mutual respect across newly defined organizational boundaries, minimize disruption to ongoing operations, and reduce the vulnerability that newly separated organizational entities may face. Mutual respect, in the forms of positive appraisal and recognition, enables the granting and claiming of organizational autonomy which, over time, allows for the reduction of organizational interdependencies. Avoiding or inhibiting interorganizational communication risks preventing the clarification of roles and responsibilities, the acquisition of resources to perform practices competently, and an in-depth understanding of the other organizational entity’s concerns. All of these issues—role clarifications, resource acquisition (including knowledge and skills), and mutual understanding—have the potential to shape interorganizational relationships in a profound way by contributing to increased feelings of either respect or disrespect with regard to one’s own and the other separating entity.
Main Protagonists*
This list of the main protagonists does not include all individuals who were interviewed or directly observed.
Communication Facilitates Appraisal Respect
Our model reveals dynamics based on two forms of respect: recognition and appraisal respect. Appraisal respect refers to appreciating individuals or groups for behaviors that signal their competent performance of particular activities or their efforts to achieve competence (Darwall, 1977: 38). Positive appraisal is not understood as something that should be granted automatically to everyone but as something specific that must be earned. It therefore has also been termed “particularized respect” (Rogers and Ashforth, 2017).
Communicating across organizational boundaries facilitates developing and demonstrating competence
A newly separated and dependent organizational entity may lack the ability to develop competence on its own and therefore may require access to relevant resources that others control, including knowledge and skills (cf. Oliver, 1991; Pfeffer and Salancik, 2003). It also needs to agree on roles and responsibilities with interdependent organizational entities so it can effectively coordinate with them (Balogun and Johnson, 2004). Both acquiring resources and agreeing on roles, which are beneficial for developing competence and hence positive appraisal, are facilitated by a high level of sustained communication and inhibited by a lack thereof (e.g., Gray, 2010).
In our context, a lack of competence that inhibited effective performance initially became apparent with regard to managing sexual health services. The most experienced service manager in this area, Melanie, assumed a commissioning role at MgmtAgency when the organizations first separated. Unable to find a manager with direct experience in sexual health services, CommunityProvider hired Margaret, a former veterinary nurse. Margaret recognized that she needed people to “advise and guide me to increase my learning” (Margaret, CP, interview). Hence CommunityProvider initially faced difficulties managing sexual health services autonomously and needed to acquire relevant resources.
Margaret accepted help from Melanie, who was willing to share her knowledge about the service management role. Apart from helping her deal with day-to-day issues and challenges, Melanie also encouraged Margaret to understand the “broader context” (Margaret, CP, interview) and adopt an organization-wide perspective by considering how local service improvements could be implemented across the region and interact with other CommunityProvider services to increase quality or efficiencies. By frequently communicating across organizational boundaries, MgmtAgency (via Melanie) helped CommunityProvider (Margaret and her colleagues) learn which activities they were expected to perform (i.e., agree on the distribution of roles and responsibilities), as well as how to perform them well (i.e., acquire resources, in the form of knowledge, to support competent performance).
We identified similar developments with regard to pediatric services, although here the roles were reversed, as MgmtAgency largely lacked the ability to evaluate these services. CommunityProvider’s pediatric service manager, Richard, met regularly with MgmtAgency members responsible for commissioning children’s services, including Sharon (a trained biologist) and later Jennifer (a part-time general practitioner or GP). Sharon appeared to recognize the need to develop her knowledge concerning pediatric services and was interested in meeting regularly with Richard. By contrast, her successor, Jennifer, was frequently invited to meetings by CommunityProvider but “kept canceling” (Richard, CP, interview). Richard suspected that Jennifer had some very basic knowledge of general children’s services but was not in a position to effectively evaluate the many complex and highly specialized pediatric services that CommunityProvider offered. Mark, CommunityProvider’s clinical director for pediatrics, eventually showed up unannounced at Jennifer’s GP practice to have an impromptu meeting with her: So, in the end, Mark just . . . plonked himself in the surgery and said, “I want to see you.” And she had no choice basically. Interesting point: one of the comments that [Jennifer] made [was], “Oh, do we commission [all of these] children’s services? I didn’t think we did.” (Richard, CP, interview)
After further meetings, Richard not only helped clarify Jennifer’s role but also successfully negotiated a “memorandum of understanding” (Richard, CP, interview) to improve coordination between MgmtAgency GPs and CommunityProvider clinicians. Hence interorganizational communication provided opportunities for agreeing on the distribution of roles and responsibilities and ultimately for supporting the competent management and evaluation of pediatric services.
Given the shift from directly managing to contracting services, MgmtAgency had to develop the ability to address service issues without becoming directly involved in their management. Conversely, CommunityProvider had to establish formal methods of reporting issues and addressing them without relying on direction and support from MgmtAgency. This transition to new roles appeared to be difficult for both entities. GPs occasionally sent complaints via e-mail to MgmtAgency managers or directly to CommunityProvider managers and expected them to be dealt with. At times, however, the complaints lacked evidence or were related to issues for which CommunityProvider was not responsible. After receiving a series of complaints, Jonathan, one of CommunityProvider’s five regional generalist managers, attended a MgmtAgency board meeting to propose a formal procedure that would clarify the distribution of roles and responsibilities. The attempt appeared to be successful, as both Jonathan and MgmtAgency managers noted that unsubstantiated e-mail complaints subsequently “stopped” (Kevin, MA, interview).
Thus interorganizational communication helped both entities develop competence with regard to performing practices by providing opportunities to access relevant resources that were (temporarily) situated outside formal organizational boundaries, as well as by agreeing on and clarifying the distribution of each organizational unit’s responsibilities and roles to enable effective coordination in the context of continued interdependencies.
We also identified several examples of how a general lack, and purposeful limiting, of interorganizational communication undermined positive appraisal and contributed to negative appraisal and ultimately mutual disrespect (see Miller, 2001). This disrespect emerged from members of both organizations judging each other as unable or unwilling to competently perform particular practices. Instead of providing opportunities to address any lack of competence, a general lack of interorganizational communication limited the entities’ ability to acquire the necessary resources and agree on the distribution of roles and responsibilities as they transitioned toward separation.
Perhaps the clearest example was MgmtAgency’s inability to competently evaluate community care services. Given that MgmtAgency altered its remit from directly delivering community healthcare services for the local population to procuring them, it became responsible for explicitly specifying which services or outcomes it wanted to purchase and for negotiating relevant terms and conditions. Though rudimentary service descriptions already existed, informal and often idiosyncratic arrangements were the norm. To avoid disrupting existing services, the entities agreed that CommunityProvider would continue to perform the services it had historically provided and that an explicit description of the tasks that these services included and excluded, as well as associated costs and performance measures, would be developed over time. But MgmtAgency managers were overwhelmed with the task of making all the informal arrangements explicit and specifying exactly what the organization intended to procure for its local population.
Rather than resulting from interorganizational collaboration, community healthcare services contracting was typically enacted as follows: MgmtAgency received a total healthcare budget from the Department of Health each year; it estimated the costs of hospital services for the coming year and allocated the rest of the budget to the local mental healthcare provider, community care provider, and small voluntary organizations in the region. Formal service specifications that provided the legal basis for the contract between MgmtAgency and CommunityProvider were produced by CommunityProvider managers, not MgmtAgency. The knowledge and skills to develop community healthcare service specifications were therefore not developed at MgmtAgency.
Many CommunityProvider managers deplored MgmtAgency’s apparent inability to understand more than a “fraction” (Amanda, CP, interview) of CommunityProvider’s services, which resulted in a “parochial view” (CEO, CP, interview) by simplistically equating their organization with nurses. The lack of detailed communication and negotiation about individual services and issues inhibited MgmtAgency’s development of the knowledge required to competently evaluate services, as well as to create and implement a coherent strategy for public healthcare in the region. It also inhibited the clarification of the distribution of roles and responsibilities between the organizational entities concerning these practices. MgmtAgency managers acknowledged that their understanding of, and hence their ability to evaluate, many community care services was limited as a result of not communicating enough with CommunityProvider: When we had the commissioning–provider split it was very much: the provider floated off; we didn’t really know much about the provider or their performance. And there was a lot of conflict in the meetings and mainly that was due to our lack of understanding of the provider, and you know, we were working with a sort of bottom-line contract sum and we didn’t actually have a detailed understanding of the services and what they were delivering. (Kevin, MA, interview)
This ignorance was exacerbated by CommunityProvider managers consciously limiting the information they shared concerning staffing details (CFO, CP, interview) and not attending stakeholder events and strategy meetings set up by MgmtAgency (Dale, MA, interview). This restriction of interorganizational communication limited opportunities for developing shared understandings of which activities should be performed by whom and how, inhibiting the development and demonstration of organizational competence.
Developing and demonstrating competence generates appraisal respect
Competent performance contributes to self-confidence and positively appraising oneself, which amounts to self-respect (Grover, 2014). Continued communication provides opportunities for parties to demonstrate their (growing) competence to one another and thus to make each other aware of it. Appraisal respect is thus developed as members gain confidence in their own and the other unit’s abilities to competently perform relevant practices. For instance, all the executives we interviewed highlighted that they were proud of CommunityProvider’s achievements in sexual health services. Moreover, CommunityProvider members positively appraised Melanie’s expertise and were therefore confident in MgmtAgency’s ability to evaluate these services. Similarly, while Sharon and Jennifer at MgmtAgency gained competence in evaluating pediatric services, Richard’s continued interaction and sharing of information gave them confidence that CommunityProvider was competently performing those services. Finally, not only did MgmtAgency’s adherence to formal complaint procedures signal competence, but Jonathan’s focus on evidence and rigor appeared to impress MgmtAgency members who, in internal meetings, repeatedly highlighted the importance of being “robust” (observation notes).
By contrast, the failure to develop competence—to perform practices in ways that others expect—contributes to behavior that others may interpret as inappropriate and thus not meriting respect. If one believes that another is, or should be, in a position to competently perform a practice but fails to do so, feelings of disappointment and contempt can arise (cf. Ufkes et al., 2012). This dynamic became apparent from interviews in which members of both MgmtAgency and CommunityProvider described the other (and at times their own) organization as “incompetent” (George, MA, interview) and “immature” (CEO, CP, interview). For instance, CommunityProvider managers blamed MgmtAgency for failing to understand the complexity of its services, while MgmtAgency members allegedly likened CommunityProvider to a “fossil” (Lucy, CP, interview) that was unable to change its existing services—without appreciating the difficulty of making substantial changes in a complex system of interrelated services: There still does seem to be a view that [MgmtAgency members] can change contract definitions willy-nilly. As if I can just turn the tap and change the service focus or give it up, or I don’t know, just because they think it might be done differently. . . . That is the immaturity that I spoke of. (CEO, CP, interview)
MgmtAgency’s inability to develop a holistic understanding of CommunityProvider’s services and challenges resulted in what some CommunityProvider managers interpreted as “ill thought-through,” “knee-jerk” behavior (CEO, CP, interview) that indicated organizational incompetence, which also did not merit respect. This criticism mirrored comments from MgmtAgency members about CommunityProvider, whose management was perceived as inconsistent and “bizarre” (Allison, MA, interview). While these views generated tension between the organizations, members of both entities stressed that they attributed alleged immaturity to the other organization’s management as a whole rather than to particular individuals (CEO, CP, interview) and said that there was no animosity between individuals that would have inhibited the development of a “professional working relationship” (Allison, MA, interview). Hence incompetence was associated with—and disrespect was directed at—the respective organizational unit that was not performing in expected ways.
Appraisal respect facilitates continued interorganizational communication
The utility of communicating with others who are deemed competent—or have some knowledge or other relevant resources to share—is apparent: perceiving someone as having access to valued resources attracts interest (Bourdieu, 1986) because one prefers to talk about a subject or problem with those who know and/or can do something about it. This became evident in our study when Richard noted that although he did not approve of interference from MgmtAgency, he “really quite like[d] the fact” that Melanie, who he had “huge respect for,” frequently provided advice “because she understands the service” (Richard, CP, interview). Richard recognized that CommunityProvider benefitted from communicating regularly with MgmtAgency because the divested unit was able to gain access to Melanie’s valuable knowledge. Moreover, improvements in the provision of specialized community care services, including sexual health and pediatrics, were nominated for national awards (archival documents). This unequivocal signaling of competence further encouraged interorganizational communication, as CommunityProvider and MgmtAgency were both keen to showcase their achievements via conference presentations and brochures.
We also observed the opposite dynamic of a perceived lack of competence hampering communication. When asked why the level of communication with certain MgmtAgency members, especially GP representatives, appeared to be low, CommunityProvider managers noted that they believed conversations were meaningless while MgmtAgency was in the midst of reorganizing itself. According to CommunityProvider’s executive team, MgmtAgency’s apparent inability to perform its most basic functions with regard to effectively monitoring, evaluating, and contracting healthcare services was “embarrassing” (CEO, CP, interview) and lacked any “sense of corporateness” (Amanda, CP, interview). The researcher also witnessed one MgmtAgency manager tell the GP representative Jennifer before an internal meeting that he believed CommunityProvider’s inability to innovate warranted measures such as helping the local mental health care provider take over its services—rather than trying to engage with CommunityProvider to collaboratively address problems. We saw evidence in both organizations that communicating with members of a seemingly incompetent or dysfunctional entity was not considered worth the effort.
Communication Facilitates Recognition Respect
In addition to appraisal respect, we also identified recognition respect as a key element that interorganizational communication (or the lack thereof) may support (or undermine). Recognition respect refers to “being aware of” or “considering” someone or something and is distinct from appraisal respect because it is not related to evaluations of competent performance or efforts to achieve competence (Darwall, 1977). One can demonstrate respect to others by signaling that their interests, values, beliefs, and/or concerns are being acknowledged—i.e., they exist and are relevant in some way (Honneth, 1992). This links to what Rogers and Ashforth (2017) termed “generalized respect,” which is the belief in modern society that everyone’s interests should be considered when making decisions. The right to act autonomously comes with the responsibility to consider others’ interests. Recognition respect is therefore directly related to the acknowledgment of coexistence and interdependence.
Communicating across organizational boundaries facilitates developing and demonstrating recognition
Developing and demonstrating recognition require communication that transcends boundaries (Carlile, 2002). Regular face-to-face interaction provides opportunities for actors to share feedback and adjust their understandings (Carlson and Zmud, 1999). This dynamic became especially apparent concerning the management and evaluation of sexual health services and pediatric services and when addressing service issues in the region that Jonathan was managing. For instance, Angela, CommunityProvider’s manager overseeing highly specialized services, including sexual health and pediatrics, noted that due to her many direct conversations with both non-clinical managers and clinicians (GPs) at MgmtAgency, she got “to know them very, very well” and “understand where they are coming from and . . . what they are trying to get at” (interview). The explication of issues provided opportunities to develop a deeper appreciation of the other organization’s concerns or, in the words of one CommunityProvider manager, to “know how someone else works; . . . what they respond to; . . . what will tick them off” (Ella, CP, interview).
A lack of communication inhibits developing recognition because opportunities to share concerns do not emerge. Hence it becomes difficult to develop an understanding of what the other cares about, how the other expects and/or wants to be treated, and on which issues the other expects or wishes to be included. This situation thus maintains boundaries of understanding. For instance, because CommunityProvider “floated off” (Kevin, MA, interview), MgmtAgency members appeared to be unaware of some of the CommunityProvider managers’ key concerns and vice versa, as highlighted by CommunityProvider’s CEO: It is still very illuminating when you have discussions with GPs [affiliated with MgmtAgency] who talk about “Well, you must’ve got money for this” or “You must’ve got money for that.” And my answer is, “I got absolutely zilch.” QIPP [quality, innovation, productivity, and prevention] money? Zilch in district nursing, etc. And they are shocked when you lay the truth of the funding arrangements out. So, we should have [clarified that we were not] cutting posts out for the sake of it [but instead] trying to live within very meager means. (CEO, CP, interview)
Recognizing others’ concerns promotes feeling respected
By recognizing what another cares about, one is able to demonstrate respect in the form of appearing to take these concerns into account. If this demonstration is interpreted as genuine and therefore successful, the other feels respected in the sense of being included (De Cremer and Tyler, 2005). In our context, demonstrating respect sometimes involved “show[ing] something tangible” (Jonathan, CP, interview). The regional manager Jonathan not only promised MgmtAgency members that he would listen to their concerns but also began to implement changes. For instance, after becoming aware that local GPs for years had not been able to admit certain patients to beds at a local community hospital managed by CommunityProvider, “within four weeks” Jonathan “ring-fenced these two beds” (Jonathan, CP, interview), which meant that from then onward two beds were reserved exclusively for patients exhibiting certain conditions that were referred by local GPs. Meanwhile, the GP rep Jennifer, who was initially accused of being unwilling to meet with CommunityProvider managers to discuss pediatric services, began to develop a reputation for being a “good listener” (CFO, CP, interview). Her newly hired colleague Tim, a senior, non-clinical manager, also gave CommunityProvider members the feeling that their organization’s concerns were being recognized by agreeing to enter into a dialogue about how CommunityProvider’s performance should be evaluated, rather than simply demanding that certain (potentially unachievable) targets be met, as had apparently been the norm in the past: When Tim attended [a performance review meeting], we talked about community matron personal health plans [and we] said, “Look, you have given us a 100% target and it is just not doable; there is a waiting list; so, let’s reduce it.” And he went, “Yeah, okay.” . . . And for me that was a valuable conversation. (Margaret, CP, interview)
In his attempts to be as open and transparent as possible, the pediatrics manager Richard shared information even when it revealed internal problems and mistakes. According to Sharon, a contract manager at MgmtAgency, “We have a good and constructive and transparent relationship, and there is no data that I have not seen” (interview). Richard’s willingness to be open and inclusive and to reveal flaws was regarded as helpful for developing a shared understanding and achieving both organizational units’ objectives. As a consequence, some actors in each organization felt that their interests were being recognized by actors in the other.
By contrast, having the impression that one’s legitimate concerns are not being considered by others triggers feelings of injustice that may contribute to anger (Miller, 2001). By not interacting regularly and failing to recognize what another cares about, it becomes difficult to manage and change this impression. Because managers’ work is concerned with effective organizational performance, beliefs that their organization is not being treated seriously may be interpreted as undermining their own work. Hence the apparent dismissal of an organization’s interests may be interpreted as illegitimately questioning the validity of, and hindering the ability to perform, one’s individual role—linking disrespect of a collective with personal feelings of being disrespected (cf. Rogers and Ashforth, 2017).
In our study, members of both organizations accused each other in interviews and internal meetings of “not thinking of how to build a consensus” (CFO, CP, interview), failing to “see the bigger picture” in terms of how their decisions affected others (Martin, MA, internal meeting observation), and behaving in ways that were “unfair” (Jonathan, CP, interview) and “unreasonable” (Helen, CP, interview). Such accusations of not demonstrating sufficient respect were especially linked to disruptions related to closing, changing, or temporarily withdrawing community care services. GP representatives at MgmtAgency believed that CommunityProvider unnecessarily triggered disruptions by making changes to general nursing services without discussing them with the affected parties first: “Suddenly there is a change, and there isn’t any explanation. It appears unilateral, cold, and unfeeling” (George, MA, interview). Mirroring this complaint, CommunityProvider managers believed that MgmtAgency managers disrupted their organization by unexpectedly announcing they would no longer fund certain services. These announcements were especially frustrating when MgmtAgency did not follow through on them, as this signaled a lack of concern that the announcements would generate preventable managerial challenges for CommunityProvider: “[Announcing the termination of a service] always ends up being around Christmas time, so that is a really bad thing to do—consulting with the workforce about whether they’re going to have a job or not” (Angela, CP, interview). Members of both entities acknowledged that they had “caused part of the problem,” contributing to disrespect by unilaterally driving changes (CFO, CP, interview) and “asking them [CommunityProvider] to do more and more” (Felicity, MA, interview) instead of trying to understand and address each other’s concerns.
Recognition respect facilitates continued interorganizational communication
Feeling that one’s concerns are being taken seriously by another party increases the willingness to communicate, because demonstrations of respect generate positive attitudes among the respected actors toward the respectful actors (van Quaquebeke and Eckloff, 2010; Blader and Yu, 2017). This dynamic became especially apparent when CommunityProvider managers were willing to meet with MgmtAgency members who they believed were interested in understanding CommunityProvider’s perspective and working together to help solve complex problems. For instance, although Sharon was not an expert on contracting pediatric services, Richard felt that her “open and honest” approach allowed him to enter into “a much broader conversation” (Richard, CP, interview) that involved sharing information about a range of issues. He consequently made efforts to have monthly face-to-face, one-on-one meetings with her, in addition to frequent telephone calls. Similarly, Jonathan became popular as he developed a reputation for listening and attempting to address concerns, as did the contracting director Tim at MgmtAgency. The first author witnessed Margaret from CommunityProvider leave her office and drive to another town at very short notice, despite a busy schedule, for an unscheduled, informal meeting with Tim at a café after he told her on the phone that he had some time to talk (observation notes).
Feeling that one’s concerns are not being taken seriously by another, by contrast, can trigger a range of negative responses. A common way of dealing with this situation is to retaliate by refusing to recognize the disrespectful party (Miller, 2001). In the context of interorganizational relations, this can amount to limiting communication with organization members who do not appear to take the interests of one’s own organization into account. For instance, in response to feeling that MgmtAgency members were treating CommunityProvider as a “junior” partner that was always at the bottom of a “pecking order” (CEO, CP, interview), CommunityProvider managers tried to “stand [their] ground” and justified their non-attendance at meetings convened by MgmtAgency and apparent lack of cooperation in this way: “Why should we have to listen and do as we’re told?” (Robert, CP, interview).
Similarly, in response to feeling excluded when decisions about service changes were made, MgmtAgency members, including executives and non-executive board members, vented their frustration concerning CommunityProvider in both internal and public meetings: It got so frustrating, and there was such negativity towards CommunityProvider by that stage. . . . And it was just getting beyond a joke. So, everybody was moaning about them . . . and it was becoming, I thought, untenable. I mean in public meetings, MgmtAgency would be badmouthing them and I felt it was unfair. . . . And it was sort of kicking the dog. (Felicity, MA, interview)
Over time, mutual disrespect among some actors escalated to “a ping-pong match of blame” (Jonathan, CP, interview) with several CommunityProvider managers believing that MgmtAgency was seeking to punish them and “get their own back” (CFO, CP, interview). Not surprisingly, very little willingness to interact and collaborate subsequently emerged.
Mutual Respect Facilitates Autonomy
Having observed how communication and respect influence each other in the context of organizational separation, we now examine how mutual respect (or disrespect) may contribute to (or undermine) organizational autonomy.
Appraisal and recognition respect enable granting and claiming autonomy
Having confidence in one’s own abilities (and thus developing self-respect) enables one to perform practices without relying on others (Grover, 2014). Believing that someone else is competent enables relinquishing control (Sennett, 2002), allowing him or her to take the lead (van Quaquebeke and Eckloff, 2010). Hence appraisal respect—in the form of respecting oneself or another—enables both claiming and granting autonomy. Furthermore, believing that another person or organization will exercise discretion in ways that recognize the interests of all interdependent parties—in other words, that the person or organization demonstrates recognition respect—also facilitates handing over control (cf. Dirks and Ferrin, 2002). Hence recognition respect enables granting autonomy.
When it came to making changes to general community care services (excluding service areas such as pediatrics and sexual health services), many CommunityProvider staff members seemed resigned to the status quo. Managers confessed that a standard response to change requests had been to provide “101 reasons” why change was “too difficult” (Jonathan, CP, interview). This attitude changed after Jonathan intervened, enabling MgmtAgency’s GPs to admit certain patients to local community care facilities. Not only did MgmtAgency members view the intervention as a signal that CommunityProvider was able to make service changes, but also the success contributed to greater self-confidence that made CommunityProvider staff more willing to “come up with new ideas” (Margaret, CP, interview) and develop a “can-do attitude” (Jonathan, CP, interview). Hence appraisal respect contributed to claiming organizational autonomy.
As CommunityProvider became increasingly confident and signaled competence, MgmtAgency was more willing to grant autonomy by expanding the portfolio of services it contracted from CommunityProvider. MgmtAgency agreed to purchase a new service designed to avoid hospital admissions, and it did not become directly involved in managing the service, such as by demanding that CommunityProvider employ a certain number of nurses (Jonathan, CP, interview; archival documents).
By contrast, not having confidence in one’s own abilities reduces willingness to claim autonomy because actors do not feel ready to assume control (Gist and Mitchell, 1992). And believing that another person or organization is not competently performing practices inhibits relinquishing control over them. Hence a lack of appraisal respect inhibits both claiming and granting autonomy. Furthermore, a lack of faith that another organization will use resources in ways that take into account one’s own organizational interests inhibits handing over control (Dekker, 2004), so a lack of recognition respect also inhibits granting autonomy.
According to Margaret, the manager of sexual health services, confidence among CommunityProvider staff was generally lacking, which inhibited efforts to make service improvements and amounted to what she termed “feeding the beast” (interview)—simply reacting to what others (including MgmtAgency) were demanding. This lack of confidence in its own competence inhibited claiming autonomy, which was particularly evident in community nursing: several CommunityProvider nurses allegedly continued to “bow down” to MgmtAgency GPs and do what they were told rather than demand that their organization’s autonomous status be respected (Lucy, CP, interview). And in at least a few instances, nurses spoke openly with GPs about what they perceived as managerial failings at CommunityProvider rather than attempting to resolve them internally first (Catherine, CP, interview).
We also identified attempts to restrict the autonomy of the other organizational unit. For instance, several CommunityProvider managers appeared to be concerned that MgmtAgency would use its powerful “monopsony” position (CFO, CP, interview) as the sole purchaser of these health services in the region to the detriment of CommunityProvider’s interests. They thus judged that it was in their interest to “have some protection for ourselves” (Amanda, CP, interview) by sharing limited information about some of its services and continuing to produce service specifications rather than let MgmtAgency specify terms and conditions. This inhibited MgmtAgency’s ability to competently monitor and evaluate services, as well as to produce its own service specifications. CommunityProvider managers restricted its autonomy based on their beliefs that MgmtAgency would not adequately take into account their organization’s concerns (i.e., demonstrate recognition respect).
MgmtAgency’s members grew concerned that their interests were not being taken into account, so they also engaged in attempts to restrict autonomy. The growing feeling that CommunityProvider was being “very deceptive” (George, MA, interview) by consciously hiding information concerning services triggered demands for it to commit to always having a certain number of staff available for each service in each region—and thus for MgmtAgency to influence CommunityProvider’s resource allocations—rather than simply commit to providing specific service outcomes. In internal MgmtAgency meetings, such restrictions were justified on the grounds that MgmtAgency’s payments to CommunityProvider apparently “got lost in a black hole” rather than being used to improve contracted services (Joe, MA, observation notes).
Claiming and granting autonomy reinforce appraisal and recognition respect
Granting autonomy signals that the other person or entity is entitled to exercise discretion and that one has some degree of confidence or faith in the other’s abilities to make valued contributions by competently performing certain tasks (Sennett, 2002). If the other accepts this autonomy, he or she feels valued (Rogers, Corley, and Ashforth, 2017), and subsequent competent performance further supports appraisal respect.
A few CommunityProvider managers, including Jonathan and Lucy, developed a reputation for addressing issues instead of complaining about them or demanding additional financial resources. The GP representative Jennifer attributed what she regarded as positive developments at CommunityProvider to efforts by “the manager in that area” (Jennifer, MA, interview), while regional manager Lucy commented that her ability to “deliver” rather than “run to mommy” was met with appraisal respect in the form of MgmtAgency members stating “you are very good” (Lucy, CP, interview).
Meanwhile, although CommunityProvider’s managers at times expressed frustration about MgmtAgency’s apparent lack of support for their organization, they also appeared to respect MgmtAgency when it did not interfere. Contract manager Amanda noted that MgmtAgency’s decision not to provide any transition funding to help CommunityProvider set up its own governance arrangements “really was a school of hard knocks from a financial perspective at the beginning of the relationship with MgmtAgency” but was also “very helpful because . . . it really focused the mind of CommunityProvider on what it needed to do to be well run and effective” (Amanda, CP, interview). In other words, by forcing CommunityProvider to manage itself, MgmtAgency recognized its need to shape its own future.
Yet dynamics that reinforced mutual disrespect when autonomy was not claimed or granted were also prevalent in our study. Just as incompetent performance can trigger feelings of disappointment and contempt, the apparent unwillingness to take full responsibility can trigger negative evaluations. Here the issue is not one of perceived incompetence but of unnecessarily relying on, and thereby disrupting, others. CommunityProvider was accused of behaving “like the adolescent who has basically left home and gone to university but keeps coming back because they want the washing done” (Richard, CP, interview) because its response to many issues was to “constantly run to mother” (Lucy, CP, interview) and “come up with the same old sob story that we haven’t got enough money and we can’t do it” (Richard, CP, interview). A notable example was CommunityProvider’s request for financial support to upgrade information systems, which MgmtAgency dismissed on the grounds that the upgrade was a “background process” that should “be happening anyway” rather than an issue requiring MgmtAgency’s support or attention (CFO, CP, interview).
Mirroring these complaints, CommunityProvider’s managers expressed their disappointment about MgmtAgency “lacking any appetite for risk” (CEO, CP, interview) and apparently being “frightened” of making any radical changes to the local health economy that could upset other stakeholders, such as the local hospital, despite increasing financial pressures (Lucy, CP, interview). MgmtAgency did not appear to be claiming autonomy in addressing healthcare system management challenges, despite being responsible for doing so. This further inhibited the development of appraisal respect.
Refusing to grant autonomy when it is expected risks triggering resentment because it may be interpreted as illegitimately curtailing one’s freedom (Shera and Page, 1996). CommunityProvider’s managers frequently complained about MgmtAgency routinely interfering in operational matters concerning community care services and compared it with an estranged father who “every so often . . . comes in and is critical and gives us a good kicking and says: ‘You’re not doing it well enough! You need to do this, this, and this!’” (Richard, CP, interview). CommunityProvider’s CEO complained about MgmtAgency being “paternalistic” and “controlling,” such as by questioning the establishment of an autonomous community care organization even several years after the decision had been made (CEO, CP, interview). Interestingly, CommunityProvider’s CFO noted in an interview that his organization, by refusing to agree to certain requests, could be seen as “helping to disempower” MgmtAgency. Both organizational units appeared to be disrespecting each other by restricting each other’s expected degrees of freedom.
Autonomy Enables Increased Independence
As we have noted, autonomy and independence are not synonyms: autonomy refers to exercising discretion in particular practices (Breaugh, 1985) and recognizing differences (Bowlby, 1998), while independence refers to the lack of interdependencies or connections. Based on our findings, however, we argue that over time autonomy can enable increased independence. In our study of organizational separation, we identified two relevant dynamics. First, the competent performance of practices can result in an organization beginning to experiment in ways that require fewer inputs from the hitherto dominant provider and to establish connections with previously independent others. Second, as interdependent organizational entities become more knowledgeable about the other and learn which interests they are expected to acknowledge (and how), the necessity of heightened and sustained interorganizational communication diminishes. Lack of autonomy, by contrast, inhibits the ability to develop alternatives and maintains the need to communicate extensively so as not to inadvertently trigger feelings of disrespect.
Autonomy enables substituting organizational dependencies over time
CommunityProvider’s confidence in its own competence helped managers bid for contracts from other management agencies and persuade them of its ability to manage specialist services. Successful bids resulted in receiving income from multiple sources, reducing (but not eliminating) CommunityProvider’s dependence on MgmtAgency. CommunityProvider won a contract to provide sexual health services in a neighboring region within a couple of years of its establishment—and in three other regions across the country shortly after. It also successfully expanded its provision of pediatric services outside the region administered by MgmtAgency (archival documents).
Despite such successes, less than three years after CommunityProvider became a separate legal entity MgmtAgency announced that it was putting the majority of its services (excluding sexual services and pediatrics) out to bid. And after an unsuccessful attempt to merge these services with the largest local hospital, MgmtAgency again assumed direct responsibility for managing them less than a year later. This triggered feelings of failure among CommunityProvider’s managers for “not keep[ing] a good enough relationship” with MgmtAgency, “never prov[ing] that we are different,” and not “doing the smart things and really demonstrating quick changes” (CFO, CO, interview). CommunityProvider’s failure to develop autonomy was associated with the disrespectful interorganizational relationship and its inability to signal competence.
Autonomy reduces the need to communicate extensively over time
The perceived development of CommunityProvider’s competence in some areas and the impression that MgmtAgency’s concerns were being acknowledged with regard to specialized services and addressing service issues in the region that Jonathan was responsible for reduced the need for the organizations to interact concerning these practices. Tim, the contract director at MgmtAgency, emphasized the benefits of frequent interorganizational communication in interviews and meetings but acknowledged that its necessity decreased when the organizations competently performed the practices expected of them and in ways that acknowledged each other’s concerns: “if you’ve got a community care provider that’s getting on and sorting it out, you leave it until there’s a row [conflict]” (Tim, MA, interview).
Only after it became apparent that MgmtAgency would not renew contracts for general community care services with CommunityProvider did conversations take place to clarify roles and responsibilities and understand concerns and complexities related to the competent performance of certain practices. Our notes from the last meeting observed (in 2013) indicate a stark contrast with previously formal and often tense interorganizational performance review meetings that had resulted in participants allegedly “crying in the corridor.” Attempts to understand and support each other now seemed to emerge instead:
Fred (CP) laughs, stands up, walks over to Tim, smiles, and hugs him. Several meeting participants also laugh.
While the eventual reintegration of most local community services into MgmtAgency suggests that both executive teams failed in their mandate to divest CommunityProvider from MgmtAgency, CommunityProvider did survive as a (much smaller) legal entity with operations in multiple regions. Although it lost the majority of its services, it ultimately became a provider of specialized community services predominantly related to pediatrics and sexual health—the practices in which mutual respect, autonomy, and increased independence had developed.
Dealing with a Lack of Autonomy during Organizational Separation: Approaching and Distancing
The inability to competently perform a practice can be dealt with in at least two ways: (a) extensively communicating with others, which risks exposing oneself and signaling incompetence but enables developing competence and mutual respect over time, and (b) limiting communication so as to avoid demonstrating incompetence while attempting to overcome it independently. Though the latter approach appears to minimize the risk of undermining appraisal respect, because a lack of direct communication reduces opportunities to signal incompetence, it risks undermining recognition respect by limiting opportunities for sharing organizational concerns. This can lead organization members to feel excluded and inhibit developing mutual understanding and adjusting to each other’s expectations.
Limiting interorganizational communication also risks undermining appraisal respect because organizations do not learn from each other, gain potentially valuable resources, or agree on how roles and responsibilities should be distributed. We observed this when CommunityProvider maintained MgmtAgency’s ignorance concerning the specification and evaluation of community care services. And while limiting communication limits opportunities to demonstrate both competence and incompetence, it also limits entities’ opportunities to demonstrate efforts to improve performance by signaling their efforts to achieve competence. This undermines respect because the pursuit of competence can also attract appraisal respect (Darwall, 1977).
Our analysis does not suggest, however, that interorganizational communication directly causes mutual respect or that the lack thereof causes mutual disrespect. As our findings show, members of newly separated organizational entities not only have to frequently interact or communicate across organizational boundaries but also have to actively use the resulting opportunities to negotiate, listen, and demonstrate their understandings to develop organizational respect. They need to engage in dialogue rather than mere debate or discussion (Tsoukas, 2009). But a high level of sustained interorganizational communication is needed for dialogue to be possible.
Our findings also indicate that attempts by members of one organization to increase their autonomy do not, by themselves, appear to trigger disrespect and ensuing attempts by members of the other organization to restrict autonomy. Instead, how such attempts are made and interpreted is key: unless they are accustomed to having their autonomy severely curtailed, interdependent parties expect to be acknowledged when decisions are made that may affect their interests. Withholding this acknowledgment can trigger emotional, defensive responses in the form of active resistance against, or withdrawal from, the disrespectful entity.
Discussion
This paper enhances our understanding of subtractive change by theorizing links among organizational separation, autonomy, independence, and mutual respect and disrespect, shedding light on the process of subtractive change rather than on its antecedents (e.g., Feldman, 2013) or performance outcomes (e.g., Vidal and Mitchell, 2015). Our process model contributes to the strategic divestiture literature by moving beyond questions of how many and what types of interorganizational ties are beneficial for divesting and divested units (Feldman, 2016) and highlighting that the negotiation and management of ties over time is critical for newly separated units to develop organizational autonomy. We present our process model in figures 1a and 1b. Figure 1a shows what can occur when organizations approach each other with open communication during the separation process, and figure 1b illustrates what occurs when divesting and divested entities distance themselves instead.
Reducing/maintaining organizational dependencies during organizational separation: Approaching during organizational separation.
Reducing/maintaining organizational dependencies during organizational separation: Distancing during organizational separation.
Our model also clarifies the distinction between organizational autonomy and independence—terms often used interchangeably (e.g., Oliver, 1991)—by demonstrating that developing and maintaining autonomy is a necessary intermediate stage that has to be reached for an entity to reduce its dependence on a particular part of its environment, such as its former parent organization or subunit. As our findings show, by acting independently without being able to competently perform certain practices autonomously or without signaling respect to others who control valuable resources, an organization risks unintentionally triggering responses that may limit the granting and claiming of autonomy. This finding supports Chia and Holt’s (2011) argument that a direct and forceful strategy is often less effective than more subtle, indirect action. It also highlights that autonomy is not static: it is developed and maintained via ongoing practices involving interdependent entities—a view compatible with a process perspective of organizing that treats practices as the primary building blocks of social reality (Tsoukas and Chia, 2002; Feldman and Orlikowski, 2011).
Contributions to Our Understanding of Respect
Respect, in the form of appraisal and recognition (Darwall, 1977), helps us understand processes of developing organizational autonomy and thus of effectively separating organizational entities from one another. Recognizing all involved organizations’ interests and interdependencies appears to be vital to maintain functioning practices and avoid disruptions during the subtractive change process. And because successful organizational separation depends on transitioning from one to two or more autonomous entities, issues of organizational competence and their appraisal are central. Developing mutual respect across organizational boundaries is a way to gain and maintain the resources necessary for one or both organizations. This co-dependency facilitates access to resources from other sources via the development of increased organizational competence and self-confidence over time, thereby contributing to the potential for increased independence (cf. Pfeffer and Salancik, 2003). Hence developing mutual respect between separating entities appears to be an effective way to increase the chances for the organization’s survival.
But the salience of respect in shaping organizational separation processes has wider implications for interorganizational relations in general. The novel construct of organizational respect requires shifting the dominant view of respect as operating between individuals toward considering (1) collective (i.e., organizational) interests, values, and concerns, (2) the granting of autonomy to a collective (i.e., organizational members), and (3) collective feelings and beliefs of being respected or disrespected by members of another organization. It highlights that, as members of separate organizational entities communicate with one another, they act as organizational representatives who may gain (dis)respect for their own organizations and demonstrate (dis)respect to the organizations they interact with. It thus complements related concepts that have been studied at the interorganizational level, such as trust (Bachmann and Zaheer, 2008).
Subtractive Change and Identity Formation
Our findings suggest that separation and identity formation (Gioia et al., 2013) are parallel avenues of research with fruitful interconnections. They indicate that developing a shared understanding of “who we are as an organization” (Corley and Gioia, 2004) is complementary to the question of how separating organizational entities delineate, understand, and perform their key practices while managing organizational interdependencies. We argue that having a shared identity is not equivalent to an organizational entity having achieved autonomy, not least because autonomy is related to the enactment of specific practices and is therefore based on different units of analysis.
In the separation process that we witnessed, identity formation appeared to play a relatively limited role, as actors were preoccupied with challenges related to the competent enactment of key organizational practices. The limited salience of identity formation in our empirical case may have several explanations. First, identity formation is founded on the creation of a meaning void that is aroused by internal and external pressures (Gioia et al., 2010). In our case these pressures were relatively modest, not least because viable alternatives for identification, such as professional identity (Ashforth, Harrison, and Corley, 2008), existed. Moreover, the reorganization was not very visible to clients and other key external stakeholders, limiting their need to understand and categorize the separating entities. In fact, based on our observation notes from multi-stakeholder meetings, several patient representatives did not appear to know or care about the structural changes at all.
The second explanation concerns the temporal relationship between processes of separation and identity formation. The process of identity formation in Corley and Gioia’s (2004) classic study took place after the formal separation of a unit that appeared to have already established organizational autonomy. This suggests that identity formation follows separation rather than interacting with it: once we are able to act autonomously, we can ask ourselves what we should do with this autonomy and how we see ourselves. Hence questions of identity may become especially salient when an apprentice is no longer an apprentice—when he or she has cut ties with his or her former master (Sennett, 2009). Whether identity formation generally follows rather than occurs during separation merits further research.
A further question that arises from our study is the impact of separation on the parent organization. To date, studies that have examined identity issues in the context of organizational separation have focused on the divested unit (Corley and Gioia, 2004; Moschieri, 2011; Ferriani, Garnsey, and Lorenzoni, 2012; Sahaym, 2013). In our case it appears that coming to terms with changes in roles and responsibilities was at least as difficult, if not more so, for the parent organization. While the divested unit’s purpose remained relatively unchanged, as it continued to provide community care services, members of the divesting unit had to adjust to moving away from providing front-line services toward exclusively contracting services. This increased distance from actively participating in front-line services directly associated with healthcare appeared to be an issue that required more managerial visioning and sensegiving (Gioia et al., 2010) than did helping members of the divested unit understand what their organization was or stood for.
Scope Conditions and Future Research
Our findings emerged from a case in which the competent enactment of practices in newly separated organizational entities was hampered, at least during a transition period, due to the lack of clear organizational boundaries and/or access to relevant resources (see also Balogun and Johnson, 2004; Jarzabkowski, Lê, and Van de Ven, 2013). The more highly interdependent (or entangled) practices are, and the more they are core to the functioning of at least one of the organizational entities involved, the more likely issues such as the ones we identified will arise and the more likely that high levels of sustained communication across organizational boundaries can enable their resolution. By contrast, we would not expect the development of organizational autonomy, and hence of disentangling practices, to be precarious when a divested (or divesting) unit is already operating relatively autonomously before formal separation and does not lose access to resources as part of the restructuring. In such cases there is no need to expend effort to separate, because the units, according to our definition, are separate in practice.
The potential competition that existed around certain practices in our study, most notably with GPs offering or wanting to offer certain community care services themselves rather than referring them to CommunityProvider, may cause concerns about the transferability of our findings. But other cases of organizational separation may involve such potential competition and require effective communication to avoid and resolve disputes. Even in the case of an organization (such as a university) outsourcing auxiliary services, such as catering or travel booking, it is possible that members may attempt to circumvent the organization’s preferred supplier or interfere when they believe their needs are not being met, and their underlying concerns could be successfully addressed via interorganizational communication.
Finally, our findings also have methodological implications for studies of subtractive change: understanding separation processes and the development of organizational autonomy requires looking not only at organizational entities but also within and across formal organizational boundaries. As our case highlights, a perspective that is both broad and deep enables identifying the work involved in the day-to-day disentangling of distinct yet potentially highly interrelated organizational practices as new working relationships between previously integrated organizational units are developed. It also helps us recognize that organizational autonomy may emerge with regard to some practices and not with others, or at different times. A process approach that adopts practices rather than organizations, groups, or individuals as units of analysis allows us to discern conditions and dynamics that together shape how the organizational entities involved develop autonomy and, in the process, mutual respect.
Supplemental Material
DS_10.1177_0001839218779806 – Supplemental material for Cutting the Cord: Mutual Respect, Organizational Autonomy, and Independence in Organizational Separation Processes
Supplemental material, DS_10.1177_0001839218779806 for Cutting the Cord: Mutual Respect, Organizational Autonomy, and Independence in Organizational Separation Processes by Rene Wiedner and Saku Mantere in Administrative Science Quarterly
Footnotes
Acknowledgements
We thank Michael Barrett, Kevin Corley, David Deephouse, Mosche Farjoun, Nina Granqvist, Christian Hampel, Ruthanne Huising, Paula Jarzabkowski, Matthew Jones, Rajiv Nag, Eivor Oborn, Wanda Orlikowski, Fabio Petani, David Seidl, Trin Thananusak, and Paul Tracey, as well as colleagues at Warwick Business School, Cambridge Judge Business School, and Desautels Faculty of Management, McGill University. We are also grateful to Associate Editor John Wagner and three anonymous reviewers for their helpful comments. Additionally, this paper benefited from the constructive feedback of audiences at the Montreal Organizations Writing Workshop (MOWW), Strategy as Practice Interest Group at the 2017 Academy of Management Conference, Organization Theory Research Group (OTREG), and a SoSe research seminar at the WU Vienna University of Economics and Business. Last but not least we thank the participants in this study who so generously provided their time and insights. This project was supported by the U.K. National Institute for Health Research (NIHR) Collaborations for Leadership in Applied Health Research and Care (CLAHRC). Any opinions, findings, and conclusions expressed in this material are those of the authors and do not necessarily reflect the views of the National Health Service in England, the NIHR, or the U.K. government’s Department of Health.
Supplemental Material
All organization and individual names have been changed in this paper to maintain anonymity.
These agencies, which are responsible for ensuring the provision of adequate public healthcare services for their local population, have been variously referred to as “health authorities,” “primary care groups,” “primary care trusts,” and “clinical commissioning groups.” The term “primary care” indicates that they represent local general practitioners (GPs).
Authors’ Biographies
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.
