کنترل و تغییر؛ تجزیه و تحلیل فرایند نهادینه سازی

This paper studies the process by which a change in the institutional logic of an organisational field diffuses through the management control system of a firm. The theoretical framework proposed by Hasselbladh and Kallinikos (2000. The project of rationalization: a critique and reappraisal of neo-institutionalism in organization studies. Organization Studies, 21, 697–720) enables us to describe the institutionalisation process of management control systems in more detail by observing how ideals are translated into discourses and control techniques. We argue that both the process by which institutional changes are implemented inside organisations and the process of decoupling are two aspects of the same issue. Revisiting core notions of new-institutional theory such as internalisation and decoupling, our findings question the systematic as well as the linear nature of the institutionalisation process. Empirical findings, based on a field study conducted in the French subsidiary of a pharmaceutical laboratory, highlight how the discourses of organisational actors contradict new ideals and control techniques. More particularly, it appears that, when a discourse cannot be heard, it can be partly bypassed using techniques. It is also argued that intra-organisational change builds on stable discourses and ambivalent technology, which foster insidious dissemination of the new institutional logics when “what can be done cannot be said”.

Neo-institutional research has demonstrated the importance of studying institutional factors, as well as traditional economic factors, in order to understand why Management Control Systems (MCS) change (Covaleski and Dirsmith, 1988, Abernethy and Chua, 1996, Granlund and Lukka, 1998, Brignall and Modell, 2000, Burns and Scapens, 2000, Collier, 2001 and Modell, 2001). In this paper, we draw on New Institutional Sociology (NIS) to address the issue of change in management control. We thus rely on a sociological and organisational approach to change (rather than on more functional economic views) and account for change through the institutionalisation process of new MCS. For us, NIS is a relevant framework because it enables us to show that MCS changes are driven mainly by institutional factors and not only by efficiency motives (Granlund, 2001). It underlines that organisations adopt MCS to appear well managed and to be legitimated in their organisational field (Covaleski et al., 1996). Therefore, NIS invites us to consider MCS as vectors as well as targets of change. Moreover, this perspective does not imply any moral judgement regarding change, being associated with neither progress nor regression (Burns and Vaivio, 2001). However, NIS does not provide a perfect and absolutely critique-proof framework for studying MCS change. NIS focuses on dissemination models, remaining at a macro level (Hasselbladh and Kallinikos, 2000) and does not really explain the process by which institutional changes are implemented within organisations (Siti-Nabiha and Scapens, 2005). The effects of change through the MCS are also widely debated in this perspective. NIS theorists originally considered that institutional constraints were satisfied by decoupling the activity, as it was presented externally from the activity as it really occurred (Meyer and Rowan, 1977), and conceived decoupling as a largely inherent attribute of institutionalised practices and organisations (Modell, 2003). In this perspective, the MCS are implemented to satisfy institutional demands and should not interfere with day-to-day activity in the organisation (Carruthers, 1995). On the one hand, some studies identify various decoupling phenomena (Ansari and Euske, 1987, Fernandez-Revuelta Perez and Robson, 1999 and Siti-Nabiha and Scapens, 2005). On the other, Covaleski and Dirsmith (1988) deny any ability to decouple the MCS from the day-to-day activity because of their structuring impact. The question of decoupling has to be tackled to improve our understanding of change in a neo-institutional perspective. In this paper, we consider that both the process by which institutional changes are implemented inside organisations and the process of decoupling are two aspects of the same issue. Both require an in-depth analysis of the problem of institutionalisation of MCS inside the organisation (Collier, 2001 and Siti-Nabiha and Scapens, 2005). Indeed, the institutionalisation refers to both the implementation and the internalisation of new practices (Kostova and Roth, 2002), and institutionalisation cannot be completed in the event of decoupling (Meyer and Rowan, 1977). By analysing the process of institutionalisation carefully, we try to explain in depth what change is (Quattrone and Hopper, 2001). More precisely, this research endeavours to answer the following question: How is change in a given organisational field institutionalised in a management control system of a given firm? To respond to our research question, we observe the institutionalisation process of a new institutional logic inside a firm in the pharmaceutical sector. Since the end of the 1990s, companies in this sector seem to be changing their development strategies and are turning to modes of selling that are similar to those of the fast moving consumer goods industry (Panigyrakis and Veloutsou, 1999 and Moss and Schuiling, 2003). This change characterises a new institutional logic in the sector, which may be driven not only by efficiency expectations but also by institutional factors (Scott et al., 2000). In order to provide an in-depth understanding of the institutionalisation process at a micro level, our research deals with one firm: Antalgyx. More precisely, we analyse how this laboratory has imported and incorporated MCS from the consumer goods industry. The case study is analysed from the angle of NIS. The theoretical framework of Hasselbladh and Kallinikos (2000) is used to understand changes resulting from the diffusion of the “consumer goods” logic in Antalgyx. As suggested by these authors, we carefully study the ideals, discourses and techniques that deal with the diffusion of the institutional logic inside the organisation. Responding to criticism that a more comprehensive MCS should be studied (Otley, 1980), this study investigates a combination of controls including cultural, behavioural and output controls (Merchant, 1982 and Merchant and Van der Stede, 2003). Three elements are specifically analysed: the management culture, the division of powers amongst the departments and the principles of accountability (Abernethy and Chua, 1996). Our empirical study shows that the process of institutionalisation is neither as systematic nor as linear as the theory suggests. This research highlights decoupling phenomena, meaning that change does not always substantially modify the daily activity of organisational actors. In our case study, change is either slowly implemented or rejected by actors, or ceremonially accepted, depending on the element of the management control system under consideration. The decoupling of MCS can be not only an organisational response to institutional demand but also an attempt developed by internal coalitions to resist the new logic (Townley, 1997, Basu et al., 1999, Burns and Scapens, 2000 and Major and Hopper, 2005). Whatever its causes (if, in fact, there are any), the decoupling of MCS reminds us that organisational change stems from stability (Feldman, 2000 and Feldman and Pentland, 2003). Indeed, in the pharmaceutical sector, stable practices conflicting with new ideals and techniques enable actors to cope with the evolution of their organisational field. Indeed, according to Hasselbladh and Kallinikos (2000), the institutionalisation process is supposed to build on a new ideal that develops into discourses and then techniques. However, our findings show that the process of change is not as linear as presented. Ambiguous coexisting ideals can give birth to new discourses and techniques (all the more so when new ideals are sufficiently ambiguous to appear as non-conflicting with the former ideals). Moreover, when a discourse dealing with the new logic cannot be heard by organisational actors, this does not prevent control techniques that are coherent with the new ideal from developing, meaning that organisational taboos can be bypassed by techniques. However, our empirical findings suggest that the institutionalisation is not completed whilst ever ideals, discourses and techniques are not coherent. In this respect, we give an empirical echo to a core NIS notion: the need for internalisation1 to enable intra-organisational change to develop fully. When ideals, discourses and techniques do not form a coherent and linear whole, as stated above, objectification, i.e. the visible side of the institutionalisation process, can occur. However, in this case, subjectification, i.e. the invisible side of the institutionalisation process, is hindered. The remainder of the paper is structured as follows. In the next section, we highlight the relevance of NIS to analyse organisational change and the role of management control systems in this respect. Then, we describe our research and data-gathering method. Sections 4 and 5 expose the case study background and our findings regarding the institutionalisation of a new logic in MCS in one pharmaceutical firm. Finally, the last section will be devoted to the discussion.

To conclude, our intention was to analyse one example of intra-organisational change by detailing the institutionalisation process of new MCS within one firm. With this in mind, we have studied the impact of the diffusion of the consumer goods industry logic on the MCS of an organisation in the ethical pharmacy sector. To go beyond criticism levelled at the neo-institutional school of thought, an in-depth analysis of the process of institutionalisation was conducted based on the theoretical framework proposed by Hasselbladh and Kallinikos (2000). The case study presented here is used as additional empirical evidence to generate theory about institutionalisation and decoupling. This paper not only explores how the process of institutionalisation takes place, but also whether it actually takes place and, if so, to what extent. This research has identified signs of change in management control when confronted with a new institutional model. The use of the theoretical framework proposed by Hasselbladh and Kallinikos (2000) offers new insights into understanding the decoupling phenomenon by studying the institutionalisation process through ideals, discourses and techniques. Decoupling is usually considered as the result (maybe unconscious) of an organisational choice not to entirely implement the new technique inside the organisation, as traditional NIS scholars argue (Meyer and Rowan, 1977), or as the result of resistance to change (Siti-Nabiha and Scapens, 2005). Our empirical findings call into question the systematic as well as the linear nature of this process. Ideals, discourses and techniques do not naturally translate from one to another: decoupling appears when there is no coherence between them. Moreover, we argue that ambiguous ideals and ambivalent techniques may facilitate the implementation and the internalisation of a new institutional logic in a given organisation. These findings have several implications for future research. They invite us to analyse change in MCS in relation to the stability of practices and to the ambivalence of control devices. This paper revisits the centrality of discourse underlined in postmodernism and organisational research (Alvesson and Deetz, 1996). Building on a Foucauldian approach to discourses (Foucault, 1975 and Foucault, 1980), we have shown how discourses contribute to producing particular forms of subjectivity according to, in resistance to, or independently of, changing power techniques and evolving ideals. We have confirmed that discourse is central to the institutionalisation process. It is central to change because it not only “ties together ideals and systems of measurement and control” (Hasselbladh and Kallinikos, 2000, p. 706), but also structures the individual's subjectivity, “providing him/her with a particular social identity and way of being in the world” (Alvesson and Deetz, 1996). We have actually highlighted empirically how discourses that conflict with new ideals and techniques enable actors to cope with the evolution of their organisational field and find their way in their organisation. Therefore, change seems able to develop only if some practices (such as the roles envisaged by organisational actors) remain stable. Furthermore, this research concludes that change can build on the ambivalence of ideals and techniques. As stated before, the ambivalence of ideals and techniques enables change to diffuse gradually and disseminate into organisational actors’ daily practices. Internalising norms, the ultimate stage in institutionalisation, is thus facilitated. For these reasons, it seems fruitful to import perspectives offered by research on properties and on the impact of technology to understand the institutionalisation of norms. This research suggests that change can be better analysed when crossed with research about accounting as a socio-institutional practice (Hopwood and Miller, 1994, Townley, 1995 and Jeacle, 2003). In order to further analyse changes in MCS, we propose to explore the socio-institutional processes through which MCS is internalised and actorhood is shaped.