When Calculative Practices are No More: On the De‐Accountingization of the Operational Level of a Public Sector Agency

This study reports on an attempt to remove management accounting's calculative practices at the operational level of a Swedish Public Agency. Using a Habermasian perspective, the study shows how the agency has attempted to replace the previous accounting practice, which involved target setting, performance management and measurement with a new leadership philosophy, and accounting practices that aim at generalizing the individual's private interest toward organizational interests. The result is interpreted as an attempt to make the individual responsible for the welfare of the collective in which, in its absence, the kind of validity that accounting's calculative practices enable is very much present as a longing to soothe the anxiety and uncertainty brought about by the responsibility to lead oneself.

In this vein, NPM reforms have been characterized as processes of accountingization (Campanale & Cinquini, 2016;Kurunmäki, Lapsley, & Melia, 2003;Power, Laughlin, & Cooper, 2003) or, as Power and Laughlin (1992) put it, the colonization of the public sector by financial imperatives and measures. Such colonization works by fragmenting and controlling the definition of the problems that public organizations address both internally through management accounting, which steers the decisions of managers, and externally through financial accounting, which steers the decisions of those external to the organization (Power et al., 2003). In effect, as Miller (1992, p. 76) writes, accounting brings forth "ways of seeing, calculating and managing" and in doing so it directs attention, but it also subjectifies individuals as both objects and relays of calculation: individuals are made comparable, assessable, and governable in terms of performance but also enabled as calculating selves to compare, assess, and regulate themselves and others (Miller, 1992).
Hence, as a way of seeing, accounting may curb our field of vision by redescribing reality, that is, organizational selfunderstanding, into terms of calculable economic reason. It does so by "provid[ing] the very definitions of the areas it regulates" (Power et al., 2003, p. 144) and giving ontological primacy to calculable and economic facts, ends, and interests. In that sense, it can become a totalizing or colonizing force, as it were. A force that "may eclipse broader questions of accountability" (Power & Laughlin, 1992, p. 133) and affect core values. This is in turn highly problematic because economic value, especially in the public sector, is not the sole value of interest (Osborne, Radnor, Kinder, & Vidal, 2015).
Against the backdrop of such outlook Power et al. (2003), p. 150) write: "what may be needed is therefore less accounting in its traditional sense and a recovery of the social and subjective dimensions of accountability that Arrington and Puxty (1991) have articulated." In effect, viewed as an interested pursuit, "accounting is always at least (1) a means toward economic ends that (2) are the norm-governed ends of some particular polity and that (3) are consequential for individuals with unique needs and desires" (Arrington & Puxty, 1991, pp. 43-44). So, even though accounting can fragment and reduce organizational self-understanding into its objective dimension, as in the case of accountingization, it can always be contested from its social and/or subjective dimensions through resistance and reform.
Plausibly, then, an alternative to contest the dominance of the objective dimension of organizational selfunderstanding and recover the social and the subjective dimensions of accountability could tentatively be to, at least partially or temporarily, remove accounting from the equation, pun intended, in what could be understood as the opposite of accountingization, namely de-accountingization. Insofar as "accounting information […] determines how certain aspects of a system […] should work" (Kurunmäki et al., 2003, p. 114), the question is then: how do those "certain aspects" of a system work or what are they replaced with when accounting is removed and the system is deaccountingisized, as it were. And, what does that say about accounting's calculative practices?
In order to pursue such avenue of research, that is, the notion of de-accountingization, we turn in what follows to the Swedish Public Employment Service (PES), which in an effort to renew itself started a reform that in its initial steps has involved the removal of management accounting practices such as target-setting and performance management and measurement from the field of vision of the agency's operational level. In its stead, self-leadership has been advanced as a means to steer oneself as an organizational member. The PES provides thus a setting where employees at the operational level of the organization, who were previously exposed to targets, performance management and measurement, now have to lead themselves, and where the absence of accounting's calculative practices, given their previous experiences at the agency, is very much present.
Studying the removal of management accounting and the calculative practices it enables, in this case at the operational level of the PES, is of particular relevance in view of the widespread critique of NPM and its consequences (Diefenbach, 2009;Lapsley, 2009), especially in relation to research on accounting as an infiltrating or colonizing phenomenon (Broadbent, 1998;Broadbent, Laughlin, & Read, 1991;Campanale & Cinquini, 2016;Kurunmäki et al., 2003;Lapsley, 2007;Oakes & Berry, 2009;Power et al., 2003) because through its absence, in this case the absence Electronic copy available at: https://ssrn.com/abstract=3601126 of targets, performance management, and measurement, we may locate alternative ways of understanding the role of accounting. As Catasús (2008Catasús ( , p. 1005) puts it, "by starting an investigation from accounting absence we gain the opportunity to make statements of what accounting is not and thereby examining the qualities that make up accounts." In what follows, we turn first to research on accountingization and the thesis of internal colonization (Habermas, 1987). The following sections present the method, the results of the empirical study at the Swedish Public Employment Service (PES), and a discussion about the role of accounting's calculative practices and what we learn from their absence in more general terms.

THEORETICAL FRAMEWORK
NPM has been be characterized as an endeavor to reform the public sector along a rationalistic line that adheres to a market and private sector logic and which has also been related to managerialism (Pollitt, 1990), economicization (Gorz, Handyside, & Turner, 1989), etc. In essence, this means that the public sector should be reformed to be more result-, cost-, efficiency-, and audit-oriented (Diefenbach, 2009). NPM practices focus thus on management accounting and control practices that prioritize economic value and matters of efficiency through the implementation of market logics, cost-oriented, and performance management (Hasselbladh, Bejerot, & Gustafsson, 2008) that presuppose management accounting measurements to be truthful representations of reality.
The function of accounting as a representational resource and a basis for decision making is however deeply problematic. In effect, as Power et al. (2003)) argue, there is no way to ascertain what is a true income or true cost in a representational sense because in the end their determination boils down to a matter of organizational power and politics. From such vantage point, an alternative is to view accounting's implication in the creation of economic facts, and therefore of social reality, and to function as a symbolic resource that not only enables decision making but more often than not serves to rationalize decisions ex post as premises to drive arguments in organizational politics. To that effect, accounting, understood as a set of calculative practices, has significant effects on people, who in turn can be said to inhabit calculable spaces that not only make "visible the hierarchical arrangement of persons and things" (Miller, 1992, p. 75), but also constitute them into calculable and calculating subjects (Miller & O'leary, 1987) amenable to control by virtue of being "subject to the neutrality of and objectivity claims that calculative expertise brings with it" (Miller, 1992, p. 79).
In effect, as Miller (2001, p. 380) writes, "management accounting seeks to affect the conduct of individuals in such a way that they act freely in accordance with specified economic norms." Put differently, accounting does not act upon individuals to tell them what to do, but what to achieve in economic terms. Accounting practices such as targets or budgets imply the freedom to choose how to achieve them and in doing so serve to "fabricate and extend practices of individualization and responsibility […] and to establish mutuality or reciprocity between forms of personal identity and the realm of economic calculation" (Miller, 2001, p. 381). The individual is thus not only made operable and calculable but also turned in turn into a calculating self (Miller, 1992;Miller & O'leary, 1994a) that is constructed as an economic citizen (Miller & O'leary, 1994a(Miller & O'leary, , 1994b responsible for her own performance.
Following this line of thought, NPM reforms can be characterized as processes of construction of economic citizenship (Miller & O'leary, 1994a), economicization (Gorz et al., 1989, managerialization (Pollitt, 1990), and, specifically in this context, accountingization (Power & Laughlin, 1992) of public sector organizations in which individuals are made into calculable and calculating subjects by way of accounting's calculative practices. In order to lay forth a frame of reference to understand de-accountingization in the form of the partial removal and absence of accounting's calculative practices at the operational level of the PES, we turn now to Habermas (1984Habermas ( , 1987. According to Habermas (1987), the social world consists of two interrelated spheres or levels, which he denominates the "lifeworld" and "systems." The lifeworld resides in the taken for granted, and it represents the lived experiences that give meaning to everyday life and reflects the norms and values of society. Systems are in turn expressions Electronic copy available at: https://ssrn.com/abstract=3601126 of the lifeworld in terms of definable and palpable organizations that should operationalize the lifeworld's desires and which are steered through steering media such as power and money (Habermas, 1987), accounting (Lawrence, 1999;Power & Laughlin, 1992), or medicine (Jones, 2001). In operationalizing the lifeworld's desires, steering media may however steer systems into an autonomous logic of their own that does not reflect the lifeworld anymore. In that case systems may grow apart from the lifeworld and even colonize it, which essentially means that communicative discourse is replaced as the coordination mechanism of the lifeworld and that steering media, such as accounting, are used instead. In effect, steering media are needed in order for the lifeworld and systems not to grow apart and in an ideal world, such coordination should be achieved through communication, debate, and dialogue-seeking consensusbetween actors. Over time, however, discursive skills have evolved and become more refined. In consequence, the differences between the natural, social, and subjective worlds, which together form the lifeworld, have grown and along such differentiation increasingly complex systems have been created (Habermas, 1987). As complexity increases the distance between the lifeworld, steering media and systems also increases and it becomes more difficult to direct the systems' behavior (Broadbent et al., 1991). The result of such process may in turn distort the lifeworld and its symbolic reproduction and bring about loss of meaning and anomie.
The whole process can be understood as a matter of specialized practices of complex systems that favor the system's own and limited operational imperatives, which in the case of accountingization as a process of colonization may mean both how "management accounting 'steers' the economic decision-making of managers [and co-workers]; and financial accounting 'steers' the economic decision-making of those external to the organization, typically investors" (Power et al., 2003, p. 130) or, in the public sector, the government. In doing so, accounting steers as a regulator of a system that marks what is and is not according to itself, and functions as a supplement to the discursive coordination and reproduction of the lifeworld. With greater complexity, as it were, accounting may replace communication, debate, and dialogue; constitute the areas it regulates; and penetrate into the symbolic reproduction of the lifeworld.
In the case of accounting, instrumental reason is paramount, but as Arrington and Puxty (1991) express, accounting is a means that is also always related to the ends of a particular social context and is consequential to the individuals it engages. In effect, instrumental reason is not the only reason: alongside an objective world of instrumentality, Habermas (1984) identifies also an intersubjective world and a subjective ditto. These worlds correspond to the practices of science, politics, and art and build upon different rationality complexes, that is, instrumental reason, practical reason and affective reason, respectively, which inform consequently different forms of action and build upon different validity claims.
At this juncture, it should be stated that Habermas (1984Habermas ( , 1987 borrows from speech-act theory and does not make a distinction between speech and action. Through speech and action, he argues, we establish shared intersubjective consensus. In effect, for Habermas (1987), every speech act involves making implicit validity claims to truth, rightness, truthfulness, and comprehensibility. Validity, in this sense is not limited to its formal logical sense of a truth conditional, but understood in a pragmatic sense where "the meaning of an utterance depends on the consensus bringing reasons that can be adduced for it by its speaker" (Finlayson, 2005, p. 35). Habermas (1987) was however mostly interested in what a theory of meaning can do for a social theory and not the other way around. He opens thus an avenue to study how social order is not only maintained but also constituted by way of the reasoning built into speech and action.
In effect, a theory of meaning is for him a theory of understanding and insofar as understanding is intersubjective to understand the social world we need to understand its meaning and how it is communicated.
As Broadbent et al. (1991) note, however, Habermasian thought is difficult to put to practice because it is a social theory that remains at an abstract and impracticable level. It must therefore be operationalized and to do that we use Arrington and Puxty (1991), who have adapted the rationality complexes from Habermas (1984) in their discussion of accounting as interested action. Accounting, they write, stands in relation and potentially impacts three worlds: "(1) the objective world of the existing state of affairs in which accounting intervenes; (2) the social world of a polity for which accounting is done; and (3) the subjective world of the individuals influenced by accounting. It follows then that the kinds of interests that can be grounds for accounting action can perhaps be differentiated into these three worlds-the objective, social, and subjective interests" (1991, p. 38) (see Table 1 below).

Objective Teleological Success
Generalizable Norm-conformative Rightness Private Dramaturgical Truthfulness By objective interests, Arrington and Puxty (1991, p. 39) mean "the need or desire to intervene in the world of the existing state of affairs." Such need is materialized through teleological action, that is, action toward an end, and evaluated against criteria of truth, "whether the actor has succeeded or failed in bringing his perceptions and beliefs into agreement with what is the case in the world" and a criterion of efficacy, "whether (the actor) succeeds in bringing what is the case in the world into agreement with his desires and intentions." The role of accounting from this type of interest is to function as a marker of what is the case in the world, that is, a representation about the state of affairs that may reduce uncertainty, and as such it is a means to an end. From such vantage point, if it becomes totalizing, it tends to exclude the reasonableness of the end which it is a means to achieve, as well as the acceptability of the means employed to achieve it. It becomes, in other words, exclusively instrumental.
Generalizable interests refer in turn to a normatively regulated action in such a way that "the social world […] brackets […] action inside generalizable norms and expectations" (Arrington & Puxty, 1991, p. 41). An action's merit is evaluated instead by the criteria of rightness, which is collectively recognized to be the right thing to do according to norms and expectations. Introducing such intersubjective outlook of norm-conformative action, the contestability of means and ends excluded by the objective interest is reappropriated in relation to the norms and expectations about what is legitimate action. Norms and expectations may however vary from organization to organization and, although they reflect societal norms and expectations, they may also differ from them, which makes the adjudication of rightness a complicated matter.
Subjective interest refers to the preferences, beliefs, wills, desires, and motives of the individual. These are portrayed through dramaturgical action, that is, social action is understood as "an encounter in which participants form a visible public for each other and perform for one another […] a performance enables the actor to present himself to his audience in a certain way; in bringing something of his subjectivity to appearance, he (sic!) would like to be seen by his public in a particular way" (Habermas, 1984, p. 90). Accounting, then, plays an important role because it represents individuals and puts them onstage, as it were, so that the subject, much like in the case of organizational self-understanding, comes to understand herself in terms of accounting and is thus not only regulated, but constituted by accounting.
As Arrington and Puxty (1991, p. 43) note, "there is [however] always an uniqueness and specificity about individual desires in private experience, and accounting is an empirical force in the construction of and responses to those needs and desires." The truthfulness that accounting representations of individuals are experienced to be imbued with is thus pivotal to such process.
To summarize, accounting's calculative practices stand in relation to three worlds, in Habermasian terms, or, according to Arrington and Puxty (1991), three types of interest, that is, the objective, the social, and the subjective. Accounting can therefore be understood in terms of three forms of action, that is, teleological, norm-conformative, and dramaturgical, and the validity claims, that is, success, rightness, and truthfulness, that correspond to them. Such multiplicity of interests, forms of action and validity claims, allows in turn for contestability, resistance, and reform.

METHOD
Based on an interpretative approach grounded in the supposition that in order to understand the social world, we need to understand its meaning and how it is communicated (Habermas, 1984(Habermas, , 1987; this study is mainly based on interviews about the changes instilled at the Swedish Public Employment Service (PES) in the name of the reform they themselves call the Renewal Journey. It reports on the beginning steps of the Renewal Journey, which is to last between 2014 and 2021 and focuses specifically on the interviewees experiences of accounting's calculative practices, their removal from the management accounting, and control system of the agency and replacement with self-leadership during the period of 2015-2017 under which we studied the agency.
The empirical material consists mainly of 54 interviews with 51 interviewees at all levels of the organization (see table 2 below). The interviewees include the director general, top, and middle managers as well as unit managers and case-workers. The interviews lasted between 30 and 120 min, with an average of 72 min, and were transcribed verbatim. They were guided conversations to obtain insights into the interviewees' lifeworld (Kvale, 2007) by allowing the interviewees to narrate their experiences and the meaning they ascribe to the renewal journey, targets, performance management and measurement, self-leadership, and their situation in view of the changes taking place at the agency. No. of second interviews with previous interviewees -9 9 No. of group interviews (two interviewees identified as a and b) 6 2 8 Total no. of interviewees 51 No. of interviewees per year 34 27 No. of interviewees on a second interview -11 Top managers 14 Middle managers 8

Unit managers 12
Case-workers 17 Average length of interview 1:09:50 1:14:34 1:12:12 As a complement to the interviews, we also made observations and participated in internal meetings, for example, top management teams' weekly meetings, seminars, and training programs and studied documents as strategic plans, annual reports, press articles, and government reports. The observations and documents were however mostly used as background material and to put the views expressed during the interviews into perspective.
The analysis is based on the three differentiated discourses presented by Habermas and adapted by Arrington and Puxty (1991), that is, the objective, the intersubjective and the subjective. The statements and discussions from interviewees were understood as actions that play a role in the constitution of the lifeworld and thus also in its potential colonization by systems and steering media of different kinds. The interpretation that follows conveys moreover an emancipatory attitude, as Arrington and Puxty (1991, p. 38) have put it, and "is less concerned to 'build' an accounting that serves a given interest and more concerned to critique how existing regimes of accounting create and sustain social pathologies that are injurious to other interests."

THE SWEDISH PUBLIC EMPLOYMENT SERVICE AND ITS RENEWAL JOURNEY
Plagued by heavy criticism, from the government, the media, its own employees, and the general public, for its inability to fulfil its missions and goals and for its detail-oriented, authoritarian, and punitive management culture, the Swedish Public Employment Service (PES) initiated a reform in 2014, the renewal journey. The overarching goal of the reform is to turn the PES into a modern agency that offers and delivers a relevant service to its customers by focusing on Electronic copy available at: https://ssrn.com/abstract=3601126 digitalization and self-leadership. In what follows, the renewal journey of the PES is described with a focus toward the previous objective management accounting and control system and its replacement with self-leadership and new management accounting and control practices.

A look in the mirror: Detail performance management and measurement
Before the Renewal Journey was initiated, detail steering permeated all levels of the PES. Management control was performed on the basis of a balanced scorecard and was focused on targets, established by management, that were broken all the way down to the individual level. The general view of the PES was that of a target-driven organization pervaded by detail performance management and measurement.
We The management accounting and control practices turned in a way the organization upside-down. In effect, insofar as performance targets were established centrally and not attuned to the local labor market, case-workers found themselves forced to make strategic choices between goals dictated by targets, on which they were evaluated, and local needs, which, in function of their position as frontline employees, only they understood.
And somehow, the roles were reversed, strategic questions were discussed lower down in the hierarchy, while detail questions were handled by top management. (I37, Top manager) Along with operational matters, responsibility and accountability were referred upward because the degree of detail in management tended to drive away employees and managers alike from participation and involvement. It was, as a manager described, "kind of a downward spiral" that involved all parties and reinforced the system. The focus of the system was thus on the targets themselves. That is to say that the meaning of targets as means to an end was lost along the way. Instead, it became a matter of "satisfying the system" or "the monster" in a way that turned means into ends. The same case-worker continued: Consequently, the understanding of the agency's purpose turned into that of its targets, which were automatic responses to the political will and therefore short-sighted. Every impulse resulted in new specific indicators and goals to be followed and to be assessed on. A case-worker described that they "were completely controlled by targets" At the heart of the renewal journey lies a new leadership philosophy, self-leadership, which includes not only managers, but everyone in the organization. It is defined in terms of five dimensions (translated from Arbetsförmedlingens förnyelseresa [The PES' Renewal Journey], p. 13): • I take responsibility for the whole: I make myself acquainted with the whole and contribute with my competence, talent, and driving force.
• I cooperate with partners: Based on the whole I create strong partnerships with other parts of the organization. That way I contribute to others' success and development.
• I develop my personal leadership: By way of deepened self-knowledge, I take responsibility for my own development, motivation, and my values. I provide the conditions to succeed with my mission.
• I grasp and influence events in the surrounding world: Based on holistic perspective I can grasp and lead events in the surrounding world. By creating partnership with stakeholders, I participate and influence development in and outside my operations.
• I lead by encouraging others: I am an attractive leader that encourages co-workers to use their driving power and competence and provide the conditions for their internal and external growth.
Self-leadership is about taking on the responsibility to make the connection between the individual's inner motivation, values and goals, and an understanding of the organization's mission, collective values, and goals. It is subject-centred insofar as it is the individual's personal leadership that lies at the centre of the concept, but it Electronic copy available at: https://ssrn.com/abstract=3601126 strives toward the whole that is the organization in a surrounding world and the community of co-workers and partners.
To that effect, every employee is tasked to formulate their own "personal offer," that ii what they offer the organization and how they envision themselves working toward the success of their nearest manager. The personal offer is moreover the basis for every employee's development discussion with his or her respective manager. Thus, based on the vision, strategic areas, one's manager's goals, one's own experience, knowledge, ideas and interests, every employee is supposed to formulate (1) how they see their own situation and possibilities, challenges and needs, (2) which possibilities, challenges and needs they see themselves being the solution to, (3) a concrete proposal of such solution, (4) which are the requirements that they need in order to be able to perform and (5) which areas are no longer fruitful in one's own attitude and way of working.
The transition from the old system was however no easy feat. As a unit manager put it the first reaction was that self-leadership was not anything new because case-workers felt they had always taken care of business themselves.
But, the manager goes on, the difference lies in "thinking 'We' instead of 'I' ." Similarly, another unit manager described the following: For managers, in turn, self-leadership entails also to be able let go and rely on trust. Trust that every individual will take on the responsibility to act with good intentions and is willing to help others and work for the group and the organization. Prompted by the question about how it is done, a unit manager said the following: The instructions from our manager to all unit managers is now that we are to offer matching opportunities for job applicants, guidance and rehabilitation and spend 55% of our time on employers, the rest is up to us. A balance between the subjective world of the individual and the social world of the agency must thus be held by determining the boundaries for self-leadership. Self-leadership is in other words not free from context, but appeals to the freedom and ability of any employee within his or her employment context. As another interviewee put it, the what-question is already answered by the government and the new strategic map has been put in place in order to make such direction very clear. Self-leadership is thus not about what to do, but how to do it, which is to say what to do in order to be able to realize the agency's desired strategic states. Strategy is thus involved at every level of the organization by way of the strategic and the operational plans, but it is so in terms of the self-leadership of every individual.

And more renewal: A new strategic management control tool
Along the new leadership philosophy, a new strategic management control system was also introduced. It is based on a strategic map that consists of five perspectives (1) customer/societal value, (2) offer and services, (3) self-leadership and culture, (4) good administration, and (5) management control. In turn, the targets and indicators of the balanced scorecard were removed along with the scorecard and replaced by desired states on the strategic map. This was a move designed to promote change.
We have the strategic map on the strategic level and then every manager gets to take the strategic actions they The focus of management accounting and control is thus on how activities put in place to achieve desired states are coming along by discussing how a section, a group, and every individual, respectively, contribute and may contribute to the achievement of the agency's desired strategic states. Along the way, however, critical voices have been raised. They remark that performance measures are important because they nonetheless provide information about the relative situation to the desired outcomes. Measurement provides accordingly knowledge that is presently largely absent, the aggregate level notwithstanding.

It's damn scary. I mean, had I felt completely confident in the competence and will of my co-workers, then I would put my trust in them, and I do trust some of them completely […] but to let go […] makes me feel insecure and [as it is now] I don't have the means. (I22, Unit manager)
In the middle of 2016, moreover, a renewed demand for performance measurement began to surface at the agency.
In effect, after more or less 2 years of reform, the government was eager to know what the Renewal Journey has led to and expected consequently to see results. The need to produce results became thus clearer and clearer to everyone at the agency. Not least after the Director General's speech at the 2016 ′ s late summer kick-off conference, where he clearly stated that the first stage of the Renewal Journey had been more or less completed and that it was time to show the fruits of such endeavor.
In turn, employees at the lower levels of the hierarchy began also to wonder and ask how, given the work they have put into the renewal journey, they were doing. In that vein, performance measurements were also highlighted as an important element of the learning process of any group, and in that sense an important component of the ability of individuals and groups to perform their mission through self-leadership.
We are trying to create a learning organization. To improve constantly is an integral part of the renewal journey and then one needs to know how things are going. (I42, Top manager) Whether performance measurement is performed should thus, from a self-leadership perspective, be a question that is up for groups at all levels of the organization to decide for themselves. In effect, if it is in the best interest of the contribution of a group to the organization, performance measurement should be performed and available to the group.
Hence, there was a longing for concretization among employees and a perceived need to take matters in their own hands, as per the self-leadership philosophy, that opened up the gate for performance measurement anew, if applied in a good way. But, the spectre of the old system still loomed large at the agency. And the balance between the new and the old, which is very much what is at stake whenever the inception or reinception of performance measurement is entertained, was perceived as a difficult task. The problem, as a manager put it, was that the agency still was very sensitive to any input that reminded it of the system of old and what a case-worker referred to as "a monster that demands to be fed." But, although the demand for performance measurement was growing at the agency, there was also fear that the system of old was surfacing again, sometimes secretly. In conclusion, the absence of the system of old was thus not the absence of a desire for objective evaluation, which was felt both from above, all the way from the government to strategic level management, and from below. From that perspective, some talked about a balance to be held in which both self-leadership and performance management can be combined. Perhaps, some said, by measuring without being so specific about targets to be met. Others described such balance in a different fashion, namely as a matter of letting the desire for objective evaluation come from the employees themselves in order to suppress the darker connotations of the performance management system of old.

ACCOUNTINGIZATION, DE-ACCOUNTINGIZATION, AND SOMETHINGIZATION
Arguably, the example of the PES presents an attempt to counter the colonizing tendencies of accounting (Campanale & Cinquini, 2016;Oakes & Berry, 2009;Power et al., 2003) by partially de-accountingisizing the agency, as it were, and limiting the reach of the objective system of targets, performance management, and measurement. In effect, targets had become goals in themselves and developed into a vicious circle of action that reinforced itself because success at every level was determined teleologically, that is, whether or not employees, units and ultimately the agency met the performance targets and failure to meet them only produced more targets. Through its regulative role, the system came to define and constitute the world of the agency. Targets were set systematically and, mediated by the power of management, turned into incontestable ends. In terms of the adaptation of Habermas work by Arrington and Puxty (1991), this involved in consequence the setting aside of both social and individual interests insofar as the Electronic copy available at: https://ssrn.com/abstract=3601126 involvement with such practices became mainly mechanistic and geared toward objective interests. As Miller (2001) writes, this could be understood as a way to engage the individual by promoting individualization and responsibility, but the proliferation of specific targets also ensued contradictions between targets and forced individuals to choose between targets. At that stage, the only certainty about the system was that the individual could only do wrong by choosing one target over the other.
By way of the Renewal Journey, the PES introduced instead a new system that relies on self-leadership and the responsibility of every employee and unit at the agency to translate the desired strategic states into their own. It relies therefore also on the willingness and competence of individuals to translate the agency's strategic map into their own situation and to take responsibility by starting with themselves, with the "I" of individual. Through self-leadership and techniques such as result and development discussions, the "I" was however not completely left to its own devices. It was put into the context of cooperation with partners in a "surrounding world" and "responsibility for the whole" in order to unite dramaturgical and the norm-conformative forms of action and induce the individual to take charge of his or her story within the context of the agency. Self-leadership attempts thus to enter the realm of individual preferences, beliefs, imagination, wills, desires, and motives in order to attach them to the agency's strategic map and harness individuality through the collective. In effect, although employees are to take charge of their own story at the agency, self-leadership is not up for debate. It is the leadership philosophy of the agency and the way employees are expected to go about their work. In other words, the individual at the agency has been freed to be free to take initiative within the frame of the agency's strategy and leadership philosophy.
In this sense, there is not much difference from the previous system. It is still a system that promotes individualization and responsibility, but it does so without the calculative practices of target setting, performance management and measurement, by expecting norm-conformative instead of teleological action from individuals. The individual is thus expected to appropriate the desired states of the agency's strategic map and realize them by making them into his or her own. In this sense, the appropriation works also the other way around, that is, it is also an attempt to colonize the truthfulness of the subjective realm of the individual by way of the rightness of the norms imparted by the new system of self-leadership. It is important to state here that such norms are not necessarily society's shared knowledge and pre-understanding, but the agency's own mediated by the coercive and financial power of organization, in this case the legitimacy of the agency as a state controlled organization and a source of remunerative employment.
The inception of self-leadership can thus be understood as an attempt to create a new intersubjectivity born out the involvement of the individual, a new understanding of the goals at hand that is both collective and individual at the same time, in the sense of a generalized subjectivity. In this context, self-leadership, result, and development discussions frame thus the individual into a system that rests ultimately on organizational coercive and financial power based the employment contract and presumed choice of the individual. Accounting becomes then something else, namely the noncalculative ability to account for one's own performance in dramaturgical, but within the norm-conformative frame of the system of self-leadership. It is about the presentation of the self, not in calculative and objective terms, but in the generalized subjective terms of an individual engaged in the system of self-leadership. In a sense, de-accountingization can be said to have taken place as accounting's calculative practices were removed from the operational level of the agency, but only to be replaced by self-leadershipization and a new form of accounting that harks back to accountability as endemic to the social (Carmona & Ezzamel, 2007) and its religious and moral roots (Aho, 2005) that bear more resemblance to recounting than counting.
In effect, the system of self-leadership and the techniques of result and development discussions seek to remodel the moral fibre of the agency to bind the subjective realm of private interest to the generalized norm-conformative organizational system by expecting validity claims of rightness to come from the individual's own dramaturgical presentation of the self. The question "what can I do to ensure the success of my superior," or "the success of others" illustrates perhaps best the inclusion of the generalized subjective morality that the system imparts to the subjective realm. The renewal journey is then also a change in the kind of validity claims to be accepted, from those of NPM (Diefenbach, 2009), where accounting provides objective information and tells a truth modelled after degree of target attainment, to that of the colonization of the subject by the organizationally mediated subjective realm of rightness, which is to follow the management philosophy, take responsibility, and account for his or her own actions accordingly.
The introduction of a generalizable self-leadership and the removal of accounting's objective calculative practices create however a void also. The realization of "infinite freedom," as a manager called it, is not only positive and empowering. It awakes also an existential anxiety insofar as the individual needs to rely solely on his/her subjectivity and take responsibility for the translation of the norm-conformative strategic map into a generalizable private dramaturgical logic of action. In effect, although the system of self-leadership seeks to overtake the subjective realm of the individual, it does not so to provide guidelines or solutions, but to foment responsibility and individualization, as do accounting's calculative practices (Miller, 1992(Miller, , 2001. It does so, however, without an objective system of calculation and the apparent simplicity of numbers, which entails the inability to make validity claims to truth and the state of affairs of things (Arrington & Puxty, 1991). The individual may thus feel as though he or she is doing the right thing, self-leadership, without knowing if he or she is doing it right because there are no results to validate his or her actions. In sum, although the spectre of the objective system of old is not easily erased and produces anxiety among employees, its absence also awakens its presence in terms of uncertainty and insecurity because subjective and intersubjective validity claims cannot fully fill the void created by the absence of and longing for objective validity.
At this juncture, it is important to relate the renewal journey to the cultural context within which it is being introduced and established because any attempt to reform, especially in the case of public agencies, is concomitantly reliant upon the reception of internal efforts both internally and externally. An agency may attempt to reform itself but that does not necessarily reform its wider context also. In the case of the PES, performance is still being monitored at the aggregate level in order to report the agency's level of success to the objective interest of the political establishment, whose demand for results are increasingly being felt as the Renewal Journey progresses. This can only exacerbate anxiety because failure to produce objective results at the aggregate level increases the risk of detail steering from the government, and subsequently the risk of a return to the objective system of old.
Although objective accountability may be absent at the operational level of the agency, it is still present in ways that are important to highlight. In effect, the objective interest for validity claims of success is very much present insofar as the government, albeit at an aggregate level, requires to see the results of the Renewal Journey. A similar desire comes also from below as employees, who have taken responsibility within the Renewal Journey, yearn to verify success in ways that validity claims of rightness and truthfulness cannot. De-accountingization, in this case the removal of performance management and measurement, means thus not the absence of objective interest per se, but a different presence that can be related to the wider cultural system whence it once was spawn, the lifeworld, which albeit in more and more differentiated fashion, consists of an objective, an intersubjective and a subjective world, and the systems we create to operationalize it: in this case, a system from the past in the form of the haunting spectre of the old system, the present system and the ambiguity employees now face in leading themselves as per self-leadership, and the uncertainty that the future brings in association with the need of the agency to prove itself before the overarching system and its political will. As Kurunmäki, Mennicken, and Miller (2016), p. 11) remark, "even the most utopian and apparently benign projects can become dystopias," but this is true of all systems that focus excessively on any one interest, be it objective and realized through calculation as Kurunmäki et al. (2016) warn us, generalizable as in this case, or private.

CONCLUSION
Through the Renewal Journey of the PES we have attempted to show an example of what we have called deaccountingization and shown in essence that de-accountingization, in the sense of the removal accounting's calculative practices, does not imply the elimination of accounting, only its conversion, in this case, to accountability as per the self-leadership philosophy. Furthermore, it does not imply the elimination of objective interest and validity claims of success either. In effect, this study shows that in their absence the objectivity of accounting's calculative practices fulfils an existential function, a longing to reduce uncertainty and ambiguity. As Miller (2001) remarks, the simplicity of Electronic copy available at: https://ssrn.com/abstract=3601126 the single accounting figure should not be underestimated and, good or bad, the single figure that accounting's calculative practices can produce yields not only comparability and management at a distance (Miller, 2001), but also certitude and stability and, as Vollmer (2003) notes, the maintenance of social order.
From this vantage point, reforms can be understood as attempts to reshape the balance between objective, generalizable, and private interests within a system by way of a particular steering media. In some instances, as in the dominance of NPM (Hood, 2000;Lapsley, 2009;Pollitt, 2007) and what is referred to accountingization (Campanale & Cinquini, 2016;Lapsley, 2007;Power et al., 2003), systems cultivate objective interests and differentiate themselves from the generalizable and private interests that also make up the lifeworld. In other instances, as we have attempted to show here, systems may cultivate other interests. This is however not surprising because organizations are essentially palpable concretizations of the lifeworld and as such can only be simplifications and always already reduced to be coordinated through the steering media of power and or money, as opposed to the ideal speech situation that Habermas champions and where consensus is the goal. In that sense, systems are in a way always already "colonized" and the important question is rather how we can learn to take other interests into consideration within a system. This is not to say that systems are simple, but that the thesis of internal colonization in the field of accounting (Broadbent et al., 1991;Campanale & Cinquini, 2016;Oakes & Berry, 2009;Power & Laughlin, 1996;Power et al., 2003) needs perhaps to be revisited from such vantage point. Indeed, as Habermas (1984Habermas ( , 1987 and Arrington and Puxty (1991) remark, contestability is of great importance. It marks the possibility of change and, to that purpose, it is perhaps more important to be aware of somethingization, as there is always something with hegemonic tendencies in the systems we create.