Paradoxes of organizational change in a merger context The reorganization of the Portuguese national healthcare system around networks of hospital centers was advanced for reasons promoted as those of effectiveness and efficiency and initially presented as an opportunity for organizational transcend

The reorganization of the Portuguese national healthcare system around networks of hospital centers was advanced for reasons promoted as those of effectiveness and efficiency and initially presented as an opportunity for organizational transcendence through synergy. Transcendence, however, failed to deliver its promise; consequently, the positive vision inscribed in it was subsequently re-inscribed in the system as another lost opportunity, contributing to an already unfolding vicious circle of mistrust and cynicism. The study contributes to the literature on organizational paradoxes and its effects on the reproduction of vicious circles.


Introduction
Reform of hospital organization and management practice has been part of a broader agenda of new approaches to managing the public sector long before the financial crisis, as part of neoliberal market measures (Saltman, Figueras & Sakellarides, 1998) but in Southern Europe the crisis intensified the demands for change.One mechanism of such change is the creation of hospital centers, a common restructuring mechanism in healthcare (Fulop, Protopsaltis, King, Allen, Hutching, & Normand, 2005).Such hospital centers either combine "previously independent hospitals formed by either the dissolution of one hospital and its absorption by another" or create "a new hospital from the dissolution of all participating hospitals" (Harris, logic for such mergers is that consolidation will be achieved and costs decrease (Starkweather, 1971).Accompanying this logic is a rationale that cost improvements can be achieved without harming the quality of patient care.The rationale argues that, given concentration and better use of resources, improved service quality that eventually reduces costs will result.In hospital management, quality and cost of patient care are the two poles that practice and its rationale respect.The trick of successful practice is to manage both simultaneously and thus transcend the need to choose one good over another: the two poles must be balanced for success.Hence, discourse in the politics of healthcare promotes economic cost benefits together with gains in clinical efficiency (Choi & Brommels, 2009;Goddard & Ferguson, 1997) in justifying hospital mergers.
Hospital organizations are more than efficient systems however, measured through the costs of transforming patient inputs into patient outputs; they are also complex organizations with complex and professionally as well as locally specific cultures.In practice, collaboration between hospital organizations is culturally difficult, resource intensive and professionally challenging (Ahgren, 2008).Organizations forced into collaboration are prone to failure and managing the quantity goals of cost and the quality goals of care simultaneously does not always lead to positive outcomes (Smith & Lewis, 2011).Nonetheless, reforms are implemented that presume it is possible to transcend the tensions between these two goals of quantity and quality, an idea that continues to captivate political and management leaders.
In this paper we investigate what happens when discourses of transcendence meet everyday practices in a context of resource scarcity.We followed a case in real time as a forced merger 1 unfolded, asking participants for their insights related to the past (initial expectations about the merger), the present (how the integration was taking place), and the future (what they foresaw as the likely future for the hospital center).We started our investigation shortly after a new CEO took charge in April 2011 and followed the process until the merger was discontinued in October 2012, although our last formal interview with the CEO took place after termination in April 2013.Follow up conversations with the CEO to crosscheck certain details were conducted until 2015.
The case was researched during the global financial crisis (GFC), which provided added impetus to the government's search for increased efficiency in the use of organizational healthcare resources as austerity gripped the southern Eurozone.The GFC and the subsequent Eurozone austerity meant that the Portuguese government was obliged to develop new policies, particularly aimed at reducing costs.It did so under difficult pressures emanating from what was known in Portugal as the 'Troika', composed of representatives of the European Commission, European Central Bank, and International Monetary Fund charged with disciplining government expenditures.In practice, different strategic goals emanated from diverse levels: the institutions of the Troika conducting the bailout, the Portuguese state via the government, the hospitals' top management teams (TMTs).Each level influenced hospital personnel's everyday perceptions of unfolding complex strategic change, a process that framed the research question: when a rhetoric of transcendence is adopted how does strategic decision-making play out if the loci for its implementation are nested at different levels of authority.
Researching the processes involved in these strategic changes demonstrated that the search for efficiency and effectiveness through strategies of transcendence often entails managing paradoxical tensions.We contribute to the organizational literature on paradoxes by empirically studying the important but under-researched theoretical link between paradox, transcendence and vicious circles.As a sequence of advances and retreats in decision-making connected with reform unfolded at the governmental level, at operational levels (organizations, teams, and individuals) mistrust and cynicism mounted in a typical process of reform fatigue.One consequence of increasing cynicism at lower levels was that as further directions for change were mandated from higher levels in practice these became interpreted as just another turn in a vicious circle of failed reform.Attempts at transcending the tension between the two objectives of quality and quantity resulted in the vortex of vicious circularity.
We studied the potential conflicts among organizational goals at multiple levels, how they interacted over time, and with what consequences.In the spirit of grounded theory, we explored processes as they happened and were articulated, witnessing the emergence of a gap between rhetorical justification and practical accomplishment.We contribute to filling this gap by asking: what happens when organizations engage in change via a discourse that seeks to transcend tight resource constraints and achieve positive outcomes in terms of both the quantity of costs and the quality of patient care?
We approach this theme in the context of a change process carried out in a Portuguese regional hospital center, contextually located in the larger setting of a bailed-out economy and consequent national budgetary pressures and governmental difficulties. 2While the GFC's causes and consequences have been much debated (Legrain, 2014) not much is known about its micro impact and the cascade of consequences at lower organizational levels, particularly on goal achievement.Achieving difficult goals, such as those in our case, becomes even more demanding when pursued under conditions of resource scarcity (Sitkin, See, Miller, Lawless, & Carton, 2011).To cut costs via efficiency gains, the government concentrated hospitals in hospital centers.In this case, then, transcendence was not implemented as something aspirational and positive, but as a part of the pressure to cut costs (exerted by the Troika) as well as the pressure to serve well (exerted by the public).We followed these processes in vivo as they unfolded in the context of complex strategic change in multiple organizations, using paradox theory as our frame of reference.

Theoretical framing
The idea that organizations face contradictory demands is at the core of paradox theory (Smith & Lewis, 2011), an increasingly influential approach in management and organization theory (Putnam, Fairhurst & Banghart, 2016).Paradox theory discusses how "persistent contradictions between interdependent elements" (Schad, Lewis, Raisch, & Smith, 2016, p. 10) can be used to generate organizational change via a "both/and" perspective (Smith, Lewis, & Tushman, 2016).
In this view, paradoxes can be used to revitalize organizations through transcendence, achieved by preventing them falling victim to a preference for one pole and the consequent neglect of its opposite.Transcendence is related to a "dialectic in which a thesis and its antithesis constitute the two poles of a contradiction, and the synthesis is seen as a new form that emerges from their interaction but that transcends or rises above them" (Abdallah, Denis & Langley, 2011).Bednarek, Paroutis and Sillince (2017, p. 77) defined transcendence as "the ability to view both poles of a paradox as necessary and complementary".In our case, the political discourse of transcendence represented cost reduction and quality as necessary and complementary in a synergistic way.The discourse was based on the idea that it was possible to articulate efficiency and quality in such a way that the apparent trade-off between them would be neutralized and replaced by a state of synergy, as happened with the adoption of the logic of TQM (Bodrozic & Adler, 2017).
Transcending a tension can be easier said than done.Research is necessary to understand what happens to organizations engaging with paradoxes via transcendence.Even as organizations attempt to transform paradoxical tensions into syntheses, they cannot eliminate the trade-off, as paradoxes inevitably involve both synthesis and trade-off (Li, 2016).In recent organizational theorizing, however, transcendence and contradiction have generally been presented under a positive light, as part of a "both/and" approach to paradox (Smith & Besharov, 2018;Smith, Lewis, & Tushman, 2016).Limited empirical attention has been paid to the fact that paradox may result in conflict, inconsistency and tension (Newark, 2017).In other words, it can be difficult to have your cake and eat it, to revisit the imagery of Abdallah et al. (2011).While paradox authors are alert to this risk, present research seems more concentrated on the synergy than the trade-off (Cunha & Putnam, 2017): there is a gap between how paradox is framed theoretically mostly as a positive force and its practical realization in negative terms, as we have discovered in this investigation.
An organization without paradox is a rare albeit theoretically treasured phenomena in management discourse.Such an organization would be secure in its organizational identity, producing a lived experience without tensions, ambiguities and conflict on the part of its members.Organizational identity refers to the collectively shared belief and understanding about central and relatively permanent features of an organization (Albert & Whetten, 1985).Changes in organizational formation such as a merger of previously separate entities clearly threaten stable identity.Social identity theory suggests that after a merger organizational identification is contingent upon a new post-merger sense of identity and continuity (Knippenberg, Knippenberg, Monden, & Lima, 2002), one that is emotionally stable, sharing goals, symbols, sociomaterialities and an authentic sense of purpose and commitment.

Research context and site
In 2011 the Portuguese government health ministry merged fourteen hospitals, resulting in six new hospital centers, a decision defended in the context of restructuring to promote integration, complementarity and resource concentration (i.e., human, financial, technological).The process presented a number of conceptually promising features.First, it was initiated by governmental decree with subsequent strategic decisions being made at multiple levels without local consultation and participation.Second, it involved the government, municipalities with their usual local political and institutional conditions, as well as a meta-organizationan organization whose members are organizations (Ahrne, Brunsson, & Seidl, 2016) denoted in Portuguese argot as the Troika of the European Central Bank, European Commission, and the International Monetary Fund who were the overseers of the Portuguese economy at the time.Third, the case constituted a conceptually extreme exemplar (Eisenhardt, 1989) of a merger.Extreme cases constitute relevant research objects due to their uniqueness (Flyvbjerg, 2006).
The Troika demanded cuts in public expenditure and in this context the government determined the restructuring decision but its operational execution, in practice, was in the hands of the hospital centers' top management teams (TMT).The governmental decision (by fiat), as we have noted, involved no participation at the hospital level and employees had no prior information regarding the merger until it was formally announced.No measures were adopted to minimize resistance (Kotter, 1995).In short, the decision makers neglected the fact that the successful implementation of a new strategy requires adopting a process of engagement, explanation, and expectations, especially when multiple organizations are involved: in other words a 'fair' process (Kim & Mauborgne, 2014) was not followed.
The lack of consultation and participation was particularly problematic in view of local history.
Two of the hospitals involved in the case were historical regional rivals.The antagonism between South and North (pseudonyms) was based on separate histories and identities.These hospitals rarely worked together because they referred their patients to different central hospitals separated by more than 100 km.Each hospital was wholly located in a different city between which there was regional rivalry.The majority of employees from North did not understand why the merger had occurred with South (instead of with West), raising fear that in the future the hospital would become an organizational satellite of South.
The strategic change initiative occurred in tandem with the imposition of severe efficiency measures, including pay cuts in the public sector that had an adverse impact on the members of these organizations.Hence the merger was enveloped in a context of ongoing austerity.The agenda of austerity had an impact on the strategic changes.A new facility that would physically materialize the hopes associated with the new hospital center had been promised initially when the strategic changes were first mooted.Although at the outset employees were told that a new facility would be built during the process the Ministry of Health reversed the decision due to financial measures imposed by the Troika's bailout.The merger involved four hospitals (South, East, Appendix and North, henceforth SEAN).Initially, the merger involved only three hospitals (South, Appendix, and East), two of which were already integrated and where most employees saw themselves as part of the same larger hospital (i.e., South).Rapidly, a fourth hospital was added (North). 3he merger that created SEAN was officially announced in January 2009, a result of the reorganization of the Portuguese national healthcare system.Three of the hospitals in this center were separated by approximately thirty kms.Two were co-located and already formed part of the same formal structure (South and Appendix [to South]).The initial project assumed the construction of a new building in a new location, but the government eventually reversed this decision in 2011, due to lack of funds.This reversal took place two years after the merger's public announcement, at a time when all employees were waiting for the final decision regarding the location of the new facility.The center's first top management team (TMT) was in office from January 2009 to August 2010.Two members from South, two from North and one from East composed this TMT.In August 2010 the government dismissed the TMT.According to the regional press, the reason for the collective dismissal of the TMT was the group's dysfunctions.
Each member was accused of putting its own hospital's interests above the center's goals.The former CEO defended the proposition that the creation of a common identity was necessary because the three hospitals had their own identities and working methods but failed to achieve the commonality sought.A new TMT with experience in healthcare management although eternal to the existing hospitals, thus lacking established vested interests, was nominated in August 2010.

Research strategy
We conducted a preparatory interview with the CEO coinciding with his appointment, in order to understand the core organizational issues he identified at this time.We secured permission from the ethics committee and were granted open access to the four hospitals within the boundaries established by the TMT and the committee.Contacts between the research team and the organization intensified in April 2011, eight months after the new TMT's nomination.We sought to capture the perspectives of relevant internal stakeholders, including the top management, doctors, nurses, and administrative staff through in-depth semi-structured interviews (see appendix A for more information about our central informants).By collecting data on the interpretations of employees in their natural work setting and considering their views over a twoyear period, we anchored our analysis in the members' understanding of the change (Schutz, 1967).We complemented interview data with secondary sources, including archival analysis and numerous informal interactions with members of the hospital community.The team's familiarity within the setting was thus rich and varied.We collected "process data" (Langley & Abdallah, 2011) from multiple sources in order to gain close familiarity with the case.One of the members of the research team regularly had meals in the hospital's canteen, thus gaining direct experience of the informal organization.These interactions were spontaneous, unavoidable, not arranged or recorded; they were mostly conducted with strangers and were helpful in terms of framing a sense of the place, unstructured interactions enabling us to gain intimacy with and blend into the site. 4The same researcher traveled frequently to the site in onehour rides with the CEO in order to gather information informally about the unfolding of the merger.Another author delivered a talk on change management, open to all employees who wished to participate, followed by informal interaction.The triangulation of data from multiple sources contributed to reinforce the robustness of the findings.
Methodologically, we conducted an inductive longitudinal case study (Eisenhardt, 1989) premised on an interview protocol that initially contained eight questions (see Appendix B).These questions were based on key themes related to the ongoing change.We identified these as the integration between units, the role of top management, rivalry between hospitals, service quality, influential groups and political circuitry and the major effects of the merger.

Documentary techniques of ethnographic data recording provided conceptual richness together
with interpretive analysis of subject's accounts of their lived experience that illuminated the role of everyday practices in a concrete sociomaterial context (Eisenhardt, Graebner & Sonenshein, 2016).
We conducted a total of 61 formal individual interviews that took between 15 and 90 minutes (1,873 minutes of planned conversations, in total).All interviews were tape-recorded and transcribed verbatim.They were conducted in two different time periods (first round, 46 interviews, May 2011 -September 2011; second round, 15 interviews, October 2012 -December 2012).The two rounds were defined a priori, as methodologically recommended (Francis et al.,, 2010).The first round (time 1) sought to capture individual perceptions at the beginning of the process, including advantages, problems and expectations about the change.
The second round had two finalities: to check the interpretations that emerged in the first round and to verify if the data was conceptually saturated (Glaser & Strauss, 1967).Between times 1 and 2 we maintained multiple conversations with the CEO, formally completed by a closing interview in the summer of 2015 to discuss the conceptual model and its validity.The fact that we interviewed multiple informants more than once also allowed us to test the acceptability of our interpretations, as we frequently asked for their feedback on our emerging theorizing.
Interviews were conducted in an office space in the administrative facility to make participants more comfortable, avoid interruptions and minimize data contamination.On site, some members of the staff actively avoided us, or left us waiting more than one hour before meeting.In some cases, the administrative staff impeded our entrance to their services, despite formal permission to conduct the study, mostly in South and East, while North was more receptive to our presence.
A number of factors help explain such behaviors.In South and East there was gossiping about the aim of the study and employees were afraid of the potential consequences of frank participation, namely retaliatory actions.They considered the study as being conducted for the TMT or the government.We were regularly confronted with questions indicative of lack of trust, such as "Are you working for the TMT?" and "Do you work for the ministry?"or "How many employees do you need to fire?"Our responses to these queries were quite simply to outline in simple terms the independence of our research purposes: that (1) we were studying a merger process longitudinally over time, (2) as it unfolded in real time, (3) based on the experience of being there as ethnographers of the change processes (4) who sought many views from diverse subjects to gain a textured understanding of how multiple actors framed the tensions that ensued as a result of the merger, and (5) to analyse these tensions in relation to multilevel dimensions.
In East we obtained only three interviews.This was the smallest operation within the center and the high levels of uncertainty and general lack of trust became apparent as several employees, always off the record, expressed their fear regarding the risk of hospital closure.Such avoidance also provided insight into the insecurity felt by employees.In a curious contrast, some patients voluntarily approached us when they heard about the work, as they were interested in understanding the aims of the study and in using the opportunity to pass on feedback to the TMT.
A common theme in informal conversations with patients referred to practical matters, notably the crowded emergency service in South.Additionally, we collected information from regional and national newspapers, and had access to documentation provided by the TMT, including annual reports.The research process is graphically depicted in figure 1.

Analytical strategy
Data analysis proceeded iteratively, implying the interrogation of theory and its use in the organization and interpretation of data (Clark, Gioia, Ketchen, & Thomas, 2010).We proceeded in the main steps detailed below, as indicated by the lines of open ended qualitative inquiry.
First, we organized data by transcribing interviews and taking notes of the process as it took place.Second, as we gained familiarity with the context, we started to build categories.As we progressed, existing categories, new pieces of evidence, and organizational theories were triangulated in order to make sense of practice, imbuing our grounded tentative model with theoretical sensitivity.In a third moment, we started to collapse categories into broader themes.
At this stage, we also initiated the process of theory interrogation.We did so within a team composed with the purpose of having different degrees of proximity and distance with the case (Reinecke & Ansari, 2005): some team members were distant, enriching the interpretive effort with distance and a 'devil's advocacy' orientation (Dittrich & Seidl, 2017), while others were more intimate.This allowed us to move more confidently to the phase of theoretically abstracting themes, reducing their number and reaching a higher theoretical order.As we progressed with the organization of the data around codes, four "ruptures and inconsistencies both among and within the established social arrangements (…) generating tensions and conflicts" (Seo & Creed, 2002, p. 225) began to take shape as contradictions.These contradictions identified aspiration and reality, purpose and efficiency, hope and cynicism, shared identity and multiple identities, as opposed categorization devices.

Findings
Contradictions were generated in part as a result of the web of decisions made at distinct levels by multiple actors.These produced a situation whose complexity could hardly be captured by any of those involved, lacking as they did an overall understanding of all the forces in play.The entanglement of complex decision streams created loci of inconsistency that interacted in incomprehensible ways for our informants.Decision-makers were involved in complex processes that mutually constrained each other (Michel, 2014), with no one having a full picture.
The different goals might have been rational per se but their entanglement created strange loops (Hofstadter, 2007).When the merger was formally established, resources were supposed to be channeled to the new facility.However, the discussion about the location of the new hospital consumed a significant amount of time, given the regional rivalry between South and North.
When a decision was finally reached, there were no financial resources available and priorities had been redefined.Decisions collided with decisions (Sheep, Fairhurst & Khazanchi, 2017) taking place at different nodes of responsibility.
Figure 1 highlights relevant temporal milestones.In the remainder, we contextualize participants' quotations by indicating the timing they refer to.Time 0 refers to the beginning of the merger, time 1 indicates the first round of interviews (2 years after the merger), time 2 to the second round of interviews (1 year after time 1).At time 2, employees were already expecting further governmental guidelines about a new structural arrangement, which eventually would demerge South and North, subsequently integrating them in two distinct hospital centers.At this stage, it was clear to all involved that SEAN represented just another turn of a vicious circle, turning today's 'solutions' eventually into tomorrow's 'problems' (Masuch, 1985).
We next articulate the four fundamental contradictions that emerged over time, which we identified in our informants' accounts, as well as from our observations and from the literature (see Figure 2 for information on data structure, Table 2 for representative supporting data; subsection on analytical strategy describes the composition of the categories).The four contradictions covered critical moments in the process: first, the way people interpreted the change before it started (expectations); second, the assessment of the values of its outcomes as it unfolded; third the experience of the process itself (emotions); fourth, the evaluation of organizational identity as formed through the process (how people interpreted who they were as SEAN).

Aspiration and reality: Contradictions around Expectations
A first contradiction articulated the initial aspirations and positive expectations of some employees regarding the merger with their re-interpretations, once it became clear that the merger would not lead to the expected outcome.Employees were waiting for the new facility to be built when the lack of public funds led to its cancellation.The discrepancy between expectations and outcomes resulted in cynicism and breached trust.
Aspiration.Employees initially accepting the rationale for the merger in the name of rationalization had positive expectations (Rentsch & Schneider, 1991), as they found the underlying motives for the process acceptable.Aiden explained that: "(…) we can have less resource dispersion and, in this way, provide the best service more efficiently (…) It would be good for profitability and resource allocation" (T0).Susan noted that "Financial resources will be saved.There is only one TMT and that saves money (…) I don't see any problems with this new four hospital structure" (T0).High expectations were based on promises from the Ministry of Health, publicly testified by the media: "There is a plan to build a hospital from scratch to serve the communities in different municipalities (…) The construction of the new hospital was also pointed out (…) by the Minister on TV (…)".High expectations were also present in the words of Addison: "I'm excited because it will be an asset for everyone … I look favorably to this merger.It is an opportunity to participate in projects" (T0).These informants believed that the merger would lead to a new hospital facility as reflected in Daniel and Amelia's statements: "There is the perspective of a new hospital" and "With a new hospital we will have more specialties, human resources and physical space" (T0).This expectation would be disconfirmed when, in April 2011, the Ministry announced that there were no resources to invest in a new facility.
Reality.Reality countered aspiration.Multiple problems emerged throughout the merger, raising doubt (Covin, Kolenko, Sightler, & Tudor, 1997), as reflected in matters as practical as increased costs for the professionals, for example in terms of commuting between units, as some of them started to practice in two hospitals.Amelia explained that the merged hospitals were, financially, a "bottomless pit" (T1), suggesting that a key reason for the merger, efficiency, was hardly being accomplished.The proclaimed goal might be positive but the facts were proving negative.
There were two key moments in which participants' frustration became explicit: when the Ministry of Health announced the lack of funds to build the new hospital (T1), and when the government decided to de-merge the merged hospitals and to re-merge them in different interorganizational centers.The events revealed, our informants noted, that decision-makers designed their policies without a clear understanding of their implications, creating unnecessary frustrations by making, breaking and re-making promises.Consider Susan: "decisions are too distant from the units and this distance has negative consequences.First, it demotivates employees because their needs are not attended to.Second, it shows a lack of investment (…) as we don't have supervisors or feedback.They are in another hospital" (T1).Supporting the same ideas, Addison pointed out that "the context is now different.The things that made sense in the first interview don't make any sense now … there is a system disruption.Supervisors are demotivated and employees feel the uncertainty."(T2) Our informants also highlighted physical and psychological distance.Distance explains the succession of decisions and counter-decisions.What is concrete at the ground level is abstract at the top.They described relevant problems in these domains as reflected in the following observations: "The major constraint is information and communication"; "It is complex to aggregate three organizations in one, because of the distance between them, 20 to 30 km" (Brayden); "TMT doesn't authorize that we commute in our own vehicles, only exceptionally, and the money provided doesn't pay for the gas nor the toll" (Brooklyn, T1).In summary, the collision of expectation and reality suggests that even those who received the new strategy with optimism were disappointed at this stage.

Purpose and performance: Contradiction around goals
Contradiction between purpose and performance characterized the ambivalent role of the TMT in radical change, as it had to split its focus between leading employees in the direction of positive aspirational change and managing the center according to governmental dictates of efficiency.
Leadership entails an inspirational core (March & Weil, 2005) whereas management implies a pragmatic approach to the daily duties of administering, especially in a crisis-ridden, bailed-out economy.The two roles of leader and manager may collide and oppose one another, confronting managers with paradoxical choices (Choi, Holmberg, Löwstedt & Brommels, 2011;Smith, Lewis & Tushman, 2016).It was difficult to inspire employees when the TMT's actions were severely constrained or neutralized by the government and when promises were not fulfilled.
Purpose.The role of TMTs is crucial in complex organizational change as it in principle instills the process with a sense of meaning (Empson, 2000).Managers' tasks include explaining why, what, and how transcendence makes sense.In this case, the main task of the TMT was put forth by the Ministry of Health, as can be read in the legal documents formalizing the decision: "By closing these hospitals, their rights and duties become the responsibility of the hospital center".
The CEO tried to infuse change with meaning, as described by informant Elijah, well aware of the effort: "This TMT is doing a lot of good things, such as meetings with all employees (…) sharing the goals and their vision of the merger" (T1).
Informants recognized the TMT's actions as an attempt to affect employees' attitudes and behaviors positively towards change.Abigail considered that "The TMT is concerned about employees and is committed to change in order to achieve a peaceful and continuous integration (…) The TMT has made a major effort to avoid feelings of defeat or subjugation" (T1).Other participants, such as Brooklyn, were also appreciative of the work conducted by the TMT: "After this TMT arrived, there was an attempt to improve service integration with the nomination of a director per service, but for all units" (T1).In other words, there could be reasons on top of efficiency to advance the change process.The management team was trying to make purpose real.
Performance.The merger was a political decision designed with an overarching goal of efficiency.Efficiency gains, however, became very problematic for the TMT as relevant stakeholders (e.g., politicians, media, population, employees) vigorously started to pursue their own sectional interests.Professionals were concerned with service quality, communities with ease of access, municipalities with their local political agendas.Under the austerity measures imposed by the Troika, budgetary pressures increased the severity of the situation and constrained management decisions, affecting the integration process (Meyer & Lieb-Doczy, 2003).Resource scarcity led to intense political action (Pfeffer & Salancik, 1978).Our informants identified all these issues.Carter noted that "These decisions [about mergers] are above top management team's decisions", whereas Chloe, referring to politics, suggested that "There was an external influence to the health area that harmed the integration" (T0).Isabella pointed out that "Motivation is complicated because we see things that are not fair… other employees are able to make pressure and get what they want…" (T1).The constraints also contributed to what was perceived as the centralization of decisions in the management team, reducing the influence of middle managers.Participants had a negative view of the action of the latter: "There are supervisors in the paper, but not in practice.The TMT is seen as God Almighty" (Abigail, T1); "There is a need for more intervening administrators who talk actively with the TMT (…) Middle managers are too distant" (T1), Aaliyah added.
Powerless middle managers, a necessarily absent TMT (due to geographic dispersion of the hospitals in the center), an intervening government, and a determined Troika, all were involved in the shaping of the processes of change.Purpose and efficiency collided, with the collision inhibiting both purpose and efficiency while increasing political unrest.Attempts at mobilization by the CEO constituted expressions of episodic power in a context of systemic power that neutralized these episodes via institutionalized rules and routines that often evolved in ways different from those indicated by the CEO (Clegg 2014;Hargrave & Van de Ven, 2017).
Episodes were neutralized by the system at large, as illustrated by the following observation: North had a treatment room fully equipped with advanced technology and ready for use in 2012.
The room was designed to reduce costs to the order of thousands of Euros per day and was therefore fully aligned with the overarching and transcending goals of efficiency and service quality.The room, however, was closed for months because of the lack of a formal document approving its inauguration.The collision of purpose and bureaucracy expressed process inconsistencies and reduced the credibility of the overall process, illuminating a Kafkaesque dimension to the overall process (Clegg, Cunha, Munro, Rego, & Sousa, 2016).Managerial agency is critical for organizations to hybridize policy and practice (Cloutier, Denis, Langley & Lamothe, 2015) but in this case the power circuitry made hybridization difficult.The several power circuits (Clegg, 2014) were never integrated and the lack of integration short-circuited change.

Hope and cynicism: Contradictions around emotions
Trust and hope are critical change ingredients due to the tensions and fears that change initiatives provoke.Our informants experienced change ambivalently.Some were skeptical about the process, whereas for others it was a thoughtful move, based upon the belief that positive outcomes could be expected.The unfolding of the process, however, turned hope into cynicism.
Hope.Hope defines the overall perception that objectives can be met (Lewin, 1935).It offers the sense that one can cope with a challenge and actively respond to a situation because one has the ability to direct energy towards the goal and to know how to plan the pathway to achieve it (Snyder, Irving, & Anderson, 1991).The apprehension caused by transformative change can be reduced by hope-inducing mechanisms such as the clarification of a vision and purpose and the definition of the goals to be met (Luthans, Youssef & Avolio, 2007).For example, Alexis hoped for and believed in the creation of a new facility: "I thought that the Hospital Center would have a new facility (…) I thought it was an opportunity for growth and I was blown away" (T0).
Ella added that "[I saw the restructuring] with a positive expectation.Like the majority of employees, [I saw it] with a favorable perspective (T0).(…) [Before the change] I felt limited and now with a [new] service I expect more team rotation, sharing of equipment and more services available, but nothing is happening" (T1).Some participants also felt that the change was an opportunity for personal and team growth.Among them, Addison noted: "I was anxious [about the merger] because it was an opportunity for all (…) it was an opportunity to build new projects (…) [and since] we must work in a network (...), working as a team is vital" (T0).
In time 1, we asked employees about their expectations with regards to the future of the center.
The answers led to different interpretations: "I would like to say that in five years this will be better, because we always have hope" and "There is a lot of fear (…) [since] they can close the small hospitals"; "I would like to see a new hospital; it would be a great challenge" and "I do not know if the hospitals will still exist because we are changing"; "I have hope and I am expecting a new reality", or "five years from now I think it will be the same" (T1).
Cynicism.Cynicism may be defined as "a negative attitude toward one's employing organization, comprising three dimensions: (1) a belief that the organization lacks integrity; (2) negative affect toward the organization; and (3) tendencies to disparaging and critical behavior toward the organization that are consistent with these beliefs and affect" (Dean, Brandes, & Dharwadkar, 1998, p.345).Cynicism results from a pessimistic outlook in regard to change and is associated with attributing blame to "those responsible" for lacking the motivation or the ability to achieve a successful outcome (Wanous, Reichers & Austin, 2000).Some employees felt confused with the transition, a feeling that is common in mergers (Mirvis & Marks, 2003).As the vision derailed, cynicism mounted.People either evaluated their managers as incompetent and insensitive or they believed that something was going on behind their backs (Mirvis & Marks, 2003).When there is lack of continuity and employees feel that they are losing something, they question the credibility, the intentions or the actual power of their managers.
The following quotes, at times 1 and 2, confirm the high levels of cynicism: "I am seeing a lot of confusion and conflicts (…) there will be quarrels" (Aubrey) (T1) and "When you make a structural merger of this nature, you cannot play around with the institutions" (Aaliyah; T2)."The merger was bad (…) the people involved in the merger were also bad.No one understood the merger.The population did not understand it, neither did the professionals" (T1).Amelia also expressed a pessimistic opinion: "It is difficult to point out the advantages (…) People did not agree with the merger.They think they have the best [procedures, practices, services] and they don't want to lose it (…) I'm seeing this Hospital Center closing in the short-term.This is a failure (…) a fruit of the crisis and the poor administration" (T2).Other informants also observed that "The choice of service directors was not related to their competence, but with other characteristics" (Addison) and "the reason for [early] retirements was related to the nomination of services' directors" (Alexis).Anthony concluded that "the merger is an outrage" (T1 + T2).

Shared identity and multiple identities: Contradictions around identity
In terms of organizational identity employees needed to feel that "despite all the changes, [this] is still their organization (Knippenberg et al., 2002: 235).In this case, the merger ended up creating a persistent sense of liminality, with employees kept "betwixt and between" identity contradictions, expressing identity ambiguity, pitting hospital against center (Turner, 1969).
Messages were ambiguous: they proclaimed a new, shared identity while not denying the coexistence of multiple, previously stabilized, identities.Shared identity.A sense of identity "serves as a rudder for navigating difficult waters" (Albert, Ashforth & Dutton, 2000, p.13).The adoption of a new organizational identity in a merger constitutes a demanding exercise, involving coordination and integration across organizations and calling into question employees' assumptions and beliefs (Clark et al., 2010).Changes in identity are required for the process to move forward (e.g., Corley & Gioia, 2004;Gioia & Thomas, 1996).Different initiatives were designed by the TMT in order to build an integrated identity at the level of the center.These initiatives sought to convey a sense of unity and stronger bonds and relationships between services via the appointment of a general manager.In addition, they encouraged members of each service in different hospitals to meet and discuss the goals for their services in search of the synergies necessary for transcendence.Participants were aware of the actions aimed at the creation of an identity and the integration of the hospitals.In Andrew's description: "There was integration in the financial and administrative services (…) I hope that functional and informational services will be integrated by the end of the year.Therefore, the work conditions as well as patients' conditions will be better.(…) I think the greatest challenge is the service integration that is going on.Furthermore, we are integrating the support services in order to provide the best service" (T1).Isabella added, "There is a concern to make more services available in the other units.It was good for the patients (…) There was an increase in the quantity of work.(…) The three institutions are represented when we need to decide something related to our work.(…) Yes, we showed 'love to our colors'" (T1).Anna observed that "The idea that North and East are against South is not true (…) I defend not only North but the hospital as a whole.Even if I don't agree with some objectives, my team works towards those goals, because if they exist, they are important" (T1).

Multiple symbolic and sociomaterial identities.
A key challenge for employees was the shift from identification with an organization (their original hospital) to a new and more diffuse multiorganization entity (the new multiunit center).Knippenberg et al. (2002) explained that the relationship between pre-merger identification and post-merger identification is not positive when employees perceive a lack of continuity.Participants from North noted that several processes and procedures were interrupted because South maintained an old-fashioned way of doing things.They had to wait for that hospital to update its processes before being allowed to update their own processes, which, in their view, undermined the standard of service quality, thereby neutralizing attempts at transcendence.Employees found themselves in a liminal state located at the intersection of discarded and unformed identities and their symbols.There was sociomaterial evidence of the partial maintenance of the original identities.Because of cost control, all the apparel, including uniforms, linens and sheets, maintained the name of each hospital instead of the name of the center; internal documents still had the symbols of each original hospital.All formal and public documents, however, showed the name and stamp of the center.These seemingly minor but relevant sociomaterial signs of discrepancy symbolized identity contradictions in the merged organizations.As our informants observed "We are moving on as three hospitals, each one with autonomy" (Elijah) and "For me, there is no difference between working in North or in this hospital center (…) nothing has changed" (Avery; T1).For Ethan there was not "any big difference, unless some professionals change attitudes towards the new identity as a single institution.But I can see some obstacles and difficulties.People are not willing to do it.This has a negative influence.If there is someone who wants to do something more innovative and standardized, there is someone who is against it and who will try to delay the processes" (T1).
Anna indicated that "We are one single institution and all professionals must know and understand that.[If they resist this new institution], we will lose all the good things.We have to do something to avoid this [resistance].It is a lack of identity with the new institution" (T1).Supporting this idea, "Most people dedicate themselves to the service but not to the institution" (T2), explained Ava, in line with Amelia who previously expressed being "sad with the merger.
It would only make sense if we merged with West Hospital because people [patients and professionals] identify themselves with West" (T1).Alexis concluded that "It is lacking a home that everyone can feel as their own.I feel like South and North are unable to grow.There is no us.Most of people do not feel [this new hospital as their home], but I do."(T2).

Discussion
The case illustrates how decisions at one level produced a puzzling whole.Participants at the hospital operational level were forced to revise the way they represented the merger as it unfolded; what they experienced was tension rather than transcendence.Borrowing from Pors (2016), they saw the strategy narrative falling apart: aspirations were disconfirmed by reality; purpose was neutralized by the centralized, bureaucratic metrics; hope was undermined by cynicism, while identity issues produced a liminal space between the new, incomplete shared identity and the former hospital identities.In summary, as represented in table 3 and figure 2, unaddressed or unresolved contradictions belonging to different domains transformed an attempt at reformation into another turn of a vicious circle.Vicious circles operate when solutions end up aggravating the problems they were intended to solve (Tsoukas & Cunha, 2017).
- --------------------------------- At the level of the hospitals, one can anticipate that the failed merger will be incorporated in future change efforts, as part of the organizational culture that contains memories of solutions that worked but also of those that failed.The planned, top-down initiatives conducted for a period of more than three years generated a persistent experience of liminality and contradiction, eventually resolved when individuals realized that the new hospital was just a mirage and that SEAN was about to be discontinued with the next wave of change.Even the faithful felt disappointed at the end.Consistent with the literature on organizational paradox (e.g., Lewis, 2000), we observed nested tensions at multiple levels, which resulted from goals drifting over time, as different stakeholders acted at multiple levels.A full catalogue of the potential triggers of paradox was present: inconsistent demands, shifting boundaries, complex relationships and quarrels over identity (Tian & Smith 2015).These were tackled as if contradictions did not matter: as decision-makers created contradictions they remained unaware of the consequences of their decisions.Employees throughout the process were confused by a proliferation of mixed messages forcing them to recreate their understanding of the merger as it unfolded.Some members developed an ideal perspective on the merger: a vision of diverse professionals working together in a new facility, an improved "common home".They conducted their identity work envisioning an integrated hospital center, which would symbolically mobilize a new identity.After the reversal, they felt disappointed.Positive expectations and beliefs grew negative and the ensuing tension promoted cynicism.Professionals considered that the efforts were a waste of time, because change would always arrive from decision makers who do not understand how difficult it is to integrate work in and between "real" organizations.
Our research, as often happens with inductive work, started by being an "agnostic" (Garg & Eisenhardt, 2017) exploration of a framework of transcendence.We ended up with a study on how transcendence breeds vicious circles, a theme theorized (Smith & Lewis, 2011) but lacking empirical treatment.We observed that contradictions can constitute essential driving forces of positive institutional change (Seo & Creed, 2002) but they can also reinforce the vicious circularity of interrupted reforms (Cunha & Tsoukas, 2015), with waves of change promoting no transformation but merely reiterating the organizational habits of the past even when a discourse of transcendence is used by organizations to articulate the poles of a paradox.The institutional contradictions emerging over this three-year period trapped the change process in a circle of unstable priorities: macro level actors defined and controlled the change with formal orientations and re-orientations but their contradictory actions reduced the credibility of management teams at the ground level.Reduced credibility further reinforced mistrust, deepening the vicious circle (Masuch, 1985).Synergies demand "purposeful iterations" between contradictory elements (Hargrave & Van de Ven, 2017, p.324), i.e. attending to and equilibrating the poles in tension.
Those iterations did not take place.The result was the emergence of unintended consequences resulting from the incapacity of decision makers at several levels to grasp the complexity of the system as a whole, as well the interactive effects produced by their decisions (Merton, 1936).Murdoch's (2015) finding about European Union meta-organizations found similar results: complex solutions typically engender unexpected problems and both/and approaches to paradoxes can produce disillusionment.
For practice, the study has insights for managers and policy makers.It suggests that the coexistence of multiple institutional logics without cross-level coordination and competent temporal articulation (Granqvist & Gustafsson, 2016), may lead to cynicism, hopelessness and a lack of responsiveness to directives which, in turn, amplifies execution difficulties for top managers tied to decision processes beyond their control.Organizational members learned that directives from the top are inconsistent over time; that top managers lack the power to get things done; that the best answer to change is to protect oneself against fluctuating priorities via the maintenance of a cynical distance.Well-intentioned managers, especially in the public sector, sometimes learn that managerial intentions that are not necessarily aligned with external meso logics (at the level of the government) can be neutralized by macro-level logics (in this case, those of the Troika).Changes can be reversed at any time and leaders impeded from leading (Bennis, 1989).When this process is institutionalized through successive feedback loops (Tsoukas, 2017), environments replete with references to reform and reformism may be reinterpreted, at the ground level, with cynicism and irony.Change is never inscribed on blank institutional pages (Cunha & Tsoukas, 2015) and past failures increase the likelihood of future failures, by encapsulating participants in vicious circles that are difficult to understand and to escape.In fact, the process triggered concurrent tensions in expectations, goals, emotions and identity.These issues interacted in complex ways, rendering the process immune to understanding.At the end, transcendence was no more than a mere rhetorical motto, an intention neutralized by the circumstance and by the power of opposing forces.
Methodologically, the study has some implications for paradox scholars.We were able to uncover the process of transcendence because of a number of methodological choices.First, we accompanied the case longitudinally.This allowed us to trace the construction of contradictions as process unfolded and to conclude that their management reinvigorated a vicious circle of repeated claims of change without true consequences.Organizational contradictions often take time to materialize and to emerge and the fact that we have approached the case longitudinally, allowed us to collect variegated angles of analysis, composing a multifaceted picture of an unfolding reality.The above characteristics also allowed us to explore the reasons why paradoxes persist (Smith & Lewis, 2011).Persistence results from the fact that any new change process is inscribed over existing institutional solutions.The interactional effects of such a friction of new and old are not immediately obvious.In some cases, they take time to gain shape and to be processed by individuals, which means that following them in real time allowed us to observe a contradiction in the making rather than its retrospective justification.To understand the emergence of paradox, therefore, it is important to follow processes in real time, as they unfold.
It took time for our interviewees to make sense of the ongoing change.Only when they realized that some critical elements of the process failed to materialize, were they able to conclude that there was a sense of déjà vu, typical of vicious circularity: they were back where it all started, even though every return is necessarily a variation on a theme (Langley & Tsoukas 2010).
Because paradox persists, methods that track phenomena over time are particularly suited to study them.Paradoxes are puzzling; therefore, approaching them from diverse angles will be helpful to understand how the clash of interpretations may explain their metamorphosis over time.
While vicious circles are by no means exclusive to public organizations (Garud & Kumaraswamy, 2005), the persistence of vicious circles may be more prevalent in state bureaucracies such as the Portuguese, with its detailed rules, limited accountabilities and history of failed reforms (Cunha & Tsoukas, 2015).
Further research may consider the nested dimension of transformative change in the complex institutional ecologies of the state, involving individuals and teams, organizations and metaorganizations.Exploring how meta-organizations penetrate the everyday life of organizations and the citizenry is an urgent task.We offered only a glimpse of the potential it contains.The lack of decision articulation, the proliferation of goals and threat of strategic reversals all conspired against the reform that actors claimed to defend.The study shows that actors can destroy the same process they vigorously defendtheir practice destroys their ideas, a finding which counters functional views of paradox as management tool.

Conclusion
The study explains how a complex process of institutional change aimed at transcendence ended up creating and reinforcing vicious circles via the emergence of multiple contradictions.When contradictions persist, they can become circular, with every cycle making the circle more pronounced and difficult to evade.The supra-national goals of the Troika, the national goals of the government, the local sensitivity of the municipalities, the personal motivations of the participants, all contributed to a frustrated change that unfolded hesitantly and inconsistently to general disappointment.At the ground level, participants felt puzzled by the inconsistencies: leaders who did not lead, efficiency measures neutralized by bureaucracy, political agencies interfering with and neutralizing other political agencies.Overall, it seemed a system held captive in traps of its own making.
Conflicting strategic priorities produced waves of measures and counter-measures that overwhelmed base level participants, even those with decision-making capacity.Inconsistencies strengthened a vicious circle of mistrust and deepened bureaucratic control, disempowering decision-makers.Paradox research suggests that contradictions can be a force for change and renewal (Clegg, Cunha & Cunha, 2002;Lewis, 2000;Lewis & Smith, 2014) but they can also be a motive for stagnation.When transcendence fails to equally engage the two poles in tension, reformist rhetoric does not necessarily lead to transformation thus aggravating the problems that it sought to tackle in the first place.A failure of this magnitude typically escalates the call for   "Now the problem is the uncertainty"  "This brings insecurity for the staff and teams"  "There is a normal relationship, but as separate hospitals"  "The communication should be improved"  "There are difficulties with transportation [to the other hospitals] (…) the cost it has for professionals" (c) Purpose  "We had 3 major objectives: functional consolidation or integration; work conditions and motivation improvement; and, quality of service improvement"  "It seems that the top management team have done a big effort to avoid a feeling of defeat or subjugation because, in fact, this is not the intention"  "Now [the hospital] has several attractive factors because there are several things that don't need money, but need organization"  "Also, the sensitivity has been demonstrated with the units' problems [by the top management team]"  "The top management team show worries and commitment in order to achieve a calm, peaceful and continuous integration, reducing some difficulties" (d) Performance  "When we got there everything was a shame in financial terms, and it still needs a lot of work"  "There are supervisors in paper, but not in practice.Therefore, the top management team was seen as God Almighty."  "We need money from the Ministry"  "There is a feeling of abandonment and neglect"  "The decision center is too far away from the functional units"  "There are several things that don't depend on the top management team, they have the will, but if there is no money, it's over" (e) Hope  "I hope that the integration is complete"  "I hope that it will be better.I have hope, I have faith"  "(…) we are waiting for something to happen"  "I hope it will be better because everyone has that hope" (f) Cynicism  "This should be extinct (…) You cannot make a silk purse out of a sow's ear"  "They don't know what they did… it is so confusing [about mergeing 3 Hospitals]"  "We are afraid… they may close the small hospitals"  "The merger was bad… the people in the merger were also bad"  "The previous TMT didn't know how to make people respect the new things" (g) Shared identity  "There were several actions in that way [integration]"  "The creation of one service in the 3 hospitals"  "There was a gradual process of rapprochement"  "The teams went to the other hospitals"  "Each service is managed by one Director (…) service union"

Identities
 "There is one thing: they are from South Hospital"  "They've never received us in a correct manner: they are from North Hospital"  "We rarely communicate with East Hospital"  "The North Hospital's culture is different from East's culture"  "The North Hospital is somehow abroad"  "There is an exaggerated concentration [of services] in South Hospital"  "We feel solidarity with them" [North Hospital about East]  "I am in charge here and for that reason I don't have to integrate anything.My employees just have to improve the service"  "No one knows better than me, so there are no arguments [with other hospitals]"

Goals
Public interest vs efficiency  Purpose and efficiency were constructed as compatible oppositesto some extent because of the new facility but also due to the logics of collective synergy at level of the center. But in some cases bureaucracy visibly neutralized both purpose and efficiency.
 The system is captured by bureaucratic dysfunction.This prevails over purpose and efficiency. Bureaucracy rules.

Hope and cynicism
Temporal work, psychological capital, change history Future orientation vs past memory  The change process was initiated as a combination of logics (efficiency, quality). Reception was mixed  Over time, cynicism prevailed, as the new construction of facility was cancelled.
 Cynics were, in retrospect, better interpreters of the process. In the next change wave, cynicism equals realism.

Identity and culture
Organizational identity versus center identity  Identities influenced the unfolding of the process. Confusion over identity was not temporary.
 Protecting local identity guards the organization against successive change. Avoid identity dissolution by not mixing.• Heath Ministry decides to reverse the decision about the new facility.• Hospital center must work as a unit but using the different hospitals sites.
• Employees are waiting for the new facility.• Aspirations and dreams about the new facility (location, development opportunities, etc.)

Macro Level
Individual Level • Management problems due to self-interest based actions.
• Low credibility and trust in managers.
• Questions about how they should work with such distance between the hospitals • Employees feel that they are not working as one.
• Reduction of public health expenses.
• 2 nd Decree Law: Separation of North from the hospital center.
• Tw o new mergers.
• Employees become cynical about everything involving change towards a unique hospital functioning.
• It is clear the lack of an unique identity.
• "Flavor of the month" -Employees do not understand why so much effort for a merger that ends up with a new separation.
• 1 st round of interviews • 2 round of interviews • Mixed reactions to the new top management team.

2013
• Last interview with CEO the beginning, we thought that [the merger] would be good for profitability and resources' appreciation"  "I had a positive expectation as well as most [colleagues]"  "It was what was expected and it seem that… one new hospital (…) At the time, we thought it [merger with different sites] would be a temporary thing"  "The document conveyed that the 3 hospitals were extinct and the Hospital Center was created.(...) There was the perspective of a new hospital." "The North Hospital was the surprise [in the merger]" (b) Reality  "This [Hospital] is a financial bottomless pit and now we have the dismissals"  "Most people were afraid of what would happen" Figure 1

Table 2 :
Representative supporting data for each 2 nd order theme

Table 3 :
How unresolved tensions reinforced vicious circles