Accounting in and for hybrids. Observations of the power of disentanglements

Gustaf Kastberg Weichselberger (School of Public Administration, University of Gothenburg, Gothenburg, Sweden)
Cristian Lagström (School of Public Administration, University of Gothenburg, Gothenburg, Sweden)

Accounting, Auditing & Accountability Journal

ISSN: 0951-3574

Article publication date: 18 June 2021

Issue publication date: 9 March 2022

1815

Abstract

Purpose

The authors argue that the mainstream scholarly discourse on hybridity and accounting is thus far primarily interested in the use and effects of accounting “in” hybrid organizations. Consequently, the literature has to a lesser extent explored how accounting mediates hybrid settings (while also being mediated), and the role of disentanglements in such processes. In hybrid settings, objects are difficult to define, and measures and tools difficult to agree upon. However, the literature on hybrid accounting is inconclusive and indicates that accounting can potentially both stabilize and de-stabilize relations in a hybrid setting. The authors address the research question of how accounting emerges and manifests itself in a process of entangling and disentangling in a heterogeneous emerging hybrid setting.

Design/methodology/approach

The paper is based on a longitudinal qualitative case study of the implementation of social investments, a public sector calculative framework based on the logic of measuring long term and social and economic impact of prevention. Methodologically, the study was guided by actor-network theory. In total, 18 observations and 48 interviews were conducted.

Findings

The observation the authors make in their case study is that much effort was spent on both keeping things apart and tying elements together. What the authors add to the literature is an illumination of how the interplay between entanglements and disentanglements facilitated the design idea of social investments to be enacted as multiple semi-integrated and purified hybridizations. The authors describe different translation points, each representing a specific hybridization where elements were added, recombined and disentangled. Still, the translation points were not completely compartmentalized, but rather semi-integrated where associations were facilitated through active mediation, likeness and productiveness for each other.

Research limitations/implications

One limitation is the single case approach. A second limitation arises from the ANT approach to hybridity.

Practical implications

A practical implication of this paper is that in hybrid settings, the semi-integrated character may be interpreted as a strength because it allows the mobilization of heterogenous actors. However, this may also come at the cost of governability and raises further questions of managerial practices in hybrid settings.

Social implications

The paper suggests the potentially productive role of disentanglements in allowing multiple hybridizations to evolve in hybrid accounting settings.

Originality/value

The paper suggests the potentially productive role of disentanglements in allowing multiple stabilized hybridizations to evolve in hybrid accounting settings.

Keywords

Citation

Weichselberger, G.K. and Lagström, C. (2022), "Accounting in and for hybrids. Observations of the power of disentanglements", Accounting, Auditing & Accountability Journal, Vol. 35 No. 3, pp. 706-733. https://doi.org/10.1108/AAAJ-12-2019-4318

Publisher

:

Emerald Publishing Limited

Copyright © 2021, Gustaf Kastberg Weichselberger and Cristian Lagström

License

Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/legalcode


Introduction

In this article, we address the hybridization of accounting by presenting a case study of the emergence of social investments, a public-sector governing framework based on the logic of calculating the long-term social and economic impacts of preventive social interventions. In extant accounting literature, a hybrid has been defined as a setting where two or more separate and different elements are brought together (Denis et al., 2015), which are “normally found separate” (Miller et al., 2008). The elements might be expertise, objects, actors, processes, formal structure, institutional logics, truth regimes or professional discipline (Ferry and Slack, 2021; Grossi et al., 2019; Rajala and Kokko, 2021; Vakkuri et al., 2021). We might expect hybrids to be fragile because of conflicting identities and roles (Croft et al., 2015; Grossi et al., 2017; Grossi and Thomasson, 2015; Kastberg and Lagström, 2019), because objects are difficult to define (Kastberg, 2014), or because measures and tools are difficult to agree upon (Christensen and Lægreid, 2008). Further, investigating the role of emerging accounting practices in the unstable and transitory state of hybridization remains largely unexplored (Steccolini, 2019).

We follow the urge to acknowledge the mediating role of accounting (Kurunmäki and Miller, 2011). Recognizing the mediating role of accounting means understanding it as a practice that resides “at the intersection of a variety of discursive and professional expectations”, and “suggests focusing less on the entities that populate the domain and paying greater attention to the linkages between them” (Kurunmäki and Miller, 2011, p. 237). Still, studies (Fischer and Ferlie, 2013; Kaufman and Covaleski, 2019) tend to treat the “link” of accounting as something coherent and stable. One gets the impression that organizations, management (McGivern et al., 2015) and professions (Kurunmäki, 2004) hybridize – but not accounting. As an example, in their study of how hybrid organizations deal with incompatible logics, Pache and Santos (2013, p. 986) showed how “intact elements”, including accounting, were incorporated in hybrid organizations in a process of selective coupling rather than through compromise or avoidance. However, a few researchers (Kurunmäki, 2008; Thambar et al., 2019, p. 6) have stressed the importance of recognizing how hybridization affects accounting practices. Miller et al. (2008) describe this as the “dual hybridization process” where accounting provides accounts of hybrids, while also being hybridized because of new and changing encounters. Accounting transforms and is transformed. Still, we know very little of how “hybrid control processes might manifest from hybrid organizational forms” (Thambar et al., 2019, p. 6), and how accounting mediates while being mediated (van Erp et al., 2019). This emerging theorizing of the potential dual hybridization of accounting provides the leading rationale for this study.

Hybridization is a process where “multiplication is the norm” (Miller et al., 2008, p. 961), and ever more elements are added and recombined. We argue that the main focus in extant accounting literature on hybrids is the linking and entangling of elements, that is, the expanding nature of hybridization and its effects (De Waele et al., 2021; Ferry and Slack, 2021). However, from an actor-network theoretical (ANT) perspective it is also relevant to scrutinize how things are kept a part. Not least since as elements multiply, so does the risk of destabilization (Callon and Law, 2005), increasing the risk of controversy and failure (Callon, 1998b; Fischer and Ferlie, 2013). Displacements, disassociations and disentanglements are all concepts that have been used to conceptualize measures to obstruct hybridization and keep elements apart (Callon, 1998a; Latour, 2005; Law and Singleton, 2005). In ANT founded accounting literature, such dynamic between hybridization (the linking of elements) and practice stabilization have been conceptualized as a delicate process of purification of the “impure” (see e.g. Christensen and Skærbæk, 2010, p. 525). We therefore argue the need for generating knowledge about the role of disentanglements for achieving practice stabilization in hybrid settings where the coming together of elements is a main rational, as in our case of social investments. Viewing accounting as a networked practice (Robson and Bottausci, 2018) that “emerge from processes of network formation” (van Erp et al., 2019, p. 2) our aim is therefore to explore how accounting is mediating while being mediated, with a focus on the role of disentanglements in emerging hybrid settings characterized by heterogeneity, yet a common task at hand. Hence, we address the research question of how accounting emerges and manifests itself in a process of entangling and disentangling in a heterogeneous emerging hybrid setting.

We study the emergence of accounting and the mobilization of a hybrid setting since “newness” has been identified as a characteristic of hybrid arrangements (Miller et al., 2008; Thambar et al., 2019). Also, before being “black boxed” (Latour, 2005), the political dimension of mediation and disentanglements is likely more easily observable. In general, public sector organizations have been characterized as fertile ground for the study of hybrids because of the mixture of professional competencies, legal framework and financial control (Kurunmäki and Miller, 2011).

We draw from the case of social investments, a novel governing framework around the idea of collaborative prevention, in a Swedish region (county). The design idea of social investments is that state bodies, municipalities or regions allocate resources to fund collaborative and preventive projects (the investments) with the aim to reduce future social exclusion such as long-term unemployment among citizens. This is not only beneficial for the person being helped; it will also cause less need for interventions (and hence costs) and support later on. The idea is therefore that all organizations that benefit from this should join as co-funders, contributing to the funding. However, the design idea [1] also presents further specification such as the importance of providing quantified ex ante assessments and ex-post evaluations of the social and economic impact of each investment. The actors involved need to calculate costs and benefits, and measure long-term social and economic impacts through the use of various calculative practices, and hence put management accounting at the center of framework (Hultkrantz and Vimefall, 2017).

The case of social investments provides a good opportunity to study the hybridizations involved in the establishment of an “accounting constellation,” understood as relations existing between “…institutions, economic and administrative processes, bodies of knowledge, systems of norms and measurement, and classification techniques” (Burchell et al., 1985, p. 400). In line with the concept of “accounting constellations”, we view the organizing that enables the financing and governance of inter-organizational projects and the various accounts produced of these social investments to be part of the same hybrid setting. Thus, social investments are not only a new way of carrying out preventive social projects but also a “control hybrid” aiming to incentivize inter-organizational collaboration by bringing together elements usually found separate, such as different professions and cost-benefit type calculations at street-level operations. The idea of co-funding also adds elements normally found separate, each acting at different “instances” producing interventions, for example one meeting the person as a child, the other as an adult. Thus, social investments entail a design idea that presupposes the formation of new associations and relationships as heterogenous elements and actors enter a relationship of shared resources and goals. In that sense it resembles a move from work of purification to work of translation (Latour, 1993, p. 11).

The observation we make when we follow how the design idea was put into practice is that much effort was spent on both keeping things apart, i.e. disentangling, and tying elements together, i.e. entangling. In essence, what we observed was a dual hybridization of accounting where multiplication was indeed the norm (Miller et al., 2008) and “…where objects precisely have to adapt and change shape if they are to survive” (Law and Singleton, 2005, p. 339). Translated to the vocabulary of accounting, we explicate the emergent dynamics of a semi-integrated hybrid accounting system. What we add to the literature is an illumination of how the accounting system manifested itself in an interplay between entangling and disentangling, facilitating the design idea of social investments to be enacted as, what we label, multiple translation points. Each translation point represented a specific purified hybrid since elements (“usually found separate”) of the design idea were stabilized. However, stabilizations presupposed disentanglements as each translation point also represents work of purification: at each translation point there were never “too many links to count” (Callon and Law, 2005, pp. 729–731). Still, the translation points were not completely disentangled vis-à-vis each other, but a semi-integrated system where associations were facilitated through: productiveness, likeness and active mediation by professionals.

In the section following this introduction, we introduce our theoretical frame and further explain the role of disentanglements. After the methodological section, we introduce the empirical setting and describe our case chronologically. Next is a description and discussion of three translation points. In the analytical section following that, we highlight and discuss how the translation points can be understood as disentangled, yet semi-integrated. In the final section, we identify and discuss conclusions as well as avenues for further research.

On the dual hybridization of accounting

As stated, our main interest here is in the dual hybridization of accounting. However, when reviewing the literature on accounting and hybridity, most studies do not recognize this duality. Accounting is often portrayed as part of one of the elements coming together (Doherty et al., 2014; Fischer and Ferlie, 2013; Kaufman and Covaleski, 2019), often as a technique aligned with financial or market-based logic. Accounting contributes to tensions by pushing or enhancing one of several perspectives (treating other perspectives as irrelevant absences), and might cause one of the elements to dominate (Doherty et al., 2014), or an increase in conflict and eventually the splitting up of the hybrid (Fischer and Ferlie, 2013). Accounting seems to get a more coercive role if backed by resource allocation, and might bring parties to accept something that “they would prefer not to” (Kaufman and Covaleski, 2019). Accounting is, according to these studies, a coercive force, an “intact element,” in the hands of a dominant actor: neither accounting as a system, nor the objects accounted for, are transformed or unstable.

In other studies, as referred to in the introduction, accounting is rather placed “between” elements, stressing its mediating role (Kurunmäki, 2008; Kurunmäki and Miller, 2011). Accounting might change and transform elements (MacKenzie, 2004; Vosselman, 2013). These studies provide arguments for understanding accounting as resting on disentanglements vis-à-vis the different elements and their entangling with different locales. For instance, Neu et al. (2010) illustrate how a rather unchanged accounting model forces different local projects to align and disentangle their practices from local relevant approaches to AIDS prevention. Although these studies highlight how accounting can hybridize spaces by connecting domains of knowledge (Picard, 2016) or connecting and aligning interests of heterogeneous actors (Cooper et al., 2017), these studies do not explicitly recognize a dual hybridization process, that is, if and how accounting is affected. However, other studies observe how accounting is used and perceived in different ways by different actors, and hence can be regarded as fluid and changeable (Ferry and Slack, 2021). van Erp et al. (2019) show how accounting produced different accounts of the objects to different actors who used the information in different ways. They observe how the management accounting control system “developed in unexpected directions and became multiple” (van Erp et al., 2019, p. 8). In a similar manner, Themsen and Skærbæk (2018, p. 30) observed how different actors used the risk management system for different purposes, and hence how accounting allowed the “coexistence of multiple purposes” (p. 30). The cited studies contribute to our understanding of how hybrid settings proliferate and how accounting itself is hybridized. However, we still need to bring the discussion further, which we do in the following section by theorizing the dual hybridization process based on an actor network conceptual framework. This allows us to identify and problematize the tension between the need of accounting to rely on and produce stabilizations and the need to maintain hybridity and how temporal stabilizations open up for new problematizations.

An actor network approach to the dual hybridization process

In his seminal work on modernity, Bruno Latour (1993) explored the mediating link between practices of hybridization (through translation) and the purifying practices of moderns who were all too occupied with distilling the messy and impure world into neatly purified forms of the natural and the cultural. Opting for the concept of translation over network, Latour brought to the fore of analysis the association of human and non-human actors. By associations, assisted by and accomplished through a variety of techniques, entities are brought into a negotiated coexistence (Callon, 1986; Latour, 2005). Taking this as a starting point means recognizing that translation always means hybridizing and bringing different elements together. However, translations are inherently unstable which has been a recurring theme in analyses of accounting and other calculative practices. Pipan and Czarniawska (2010) show in a study of management accounting in Italian municipalities how translation processes easily fail to result in a stabilized actor network integrated enough to perform in a way that “successfully leaves an impression of unity and substance” (p. 250). The act of designing a new accounting system always runs the risk that there are “too many links to count” so that the process “… is undermined by an excess of resources that interact with and undermine one another” (Callon and Law, 2005, pp. 729–731). When elements are associated, combined and multiplied, we might expect an increase in complexity with a potential to result in paralysis or inertia (Kastberg and Lagström, 2019) because everything becomes controversial: “the identification of intermediaries and overflows, the distribution of source and target agents, the ways effects are measured…” (Callon, 1998a, p. 260). Such “impure” situations are problematic when introducing accounting systems (Christensen and Skærbæk, 2010), and might end in a situation where accounting “never became instrumental in the context of centrally orchestrated systematic decision making or control” (van Erp et al., 2019, p. 8).

However, sometimes the associations are stabilized enough to produce foreseeable effects (Latour, 2005), such as when an accounting system is in place. In order to avoid uncontrolled hybridization, stabilization has to take place described as processes of purification (Christensen and Skærbæk, 2010; Latour, 1993), with the potential of obtaining (temporarily) purified settings in the sense that the association between actors is stabilized, hybridization comes to an end and predictable effects are produced. Latour (2005, p. 40) describes these stabilized purified translations as “faithful intermediaries” because of their predictable nature. Hence, purification work have to take place in order for an accounting system to “become operable” (Christensen and Skærbæk, 2010). Purification might be about refining linkages and roles, but it might also be about reducing the elements present (de-hybridization) resulting in the creation of a “disciplined hybrid” (Kastberg and Lagström, 2019) where the associated elements are stabilized in a setting through careful linking and refining (Latour, 1993, p. 41). However, full de-hybridization may also lead to compartmentalization (failed hybrid) (Kastberg and Lagström, 2019). Whereas disciplining through purification work is a strategy for maintaining hybridity and producing effects as a result of the mobilization of heterogeneous elements (Latour, 1993), compartmentalization refers to a situation where different constellations of elements are deliberately kept apart. What unites the two developments is the importance of recognizing disentanglements, a concept Callon (1986, 1998b) puts in the center of the analysis of hybrids.

Disentanglements refer to the keeping apart of competing rationalities, systems, practices and other actors and hence signifies a process towards de-hybridization. Disentangling is an active act and signifies the need for “substantial investments” (Callon, 1998a) requiring instruments, institutions and know-how (Latour, 1993, p. 78). The effect is an ability to act more efficiently and to gain operability. Callon (1998b), when introducing the concept, used the market system as an example and illustrated disentanglements with “externalities” not accounted for in the framing of the market. For Callon, disentanglement is the answer to the question of “how is the delimiting, or framing, of relationships at a point in the network achieved?” (Callon, 1998b, p. 16). Christensen and Skærbæk (2010) illustrate in their study how this is accomplished through letting the process of introducing a management accounting system rely on “experts with a certain scientific affiliation” (p. 527). This allowed certain values and competences to dominate the process, whereas competing values and expertise were disentangled (they did however not explicitly use the term). In our analysis of the introduction of a design idea presupposing a heterogeneous actor constellation, we therefore have to pay attention to what role disentanglements play, and direct our attention to “the ambivalences and displacements that (sometimes) keep networks in place” (Law and Singleton, 2005, p. 341).

The introduction of a management accounting-based design idea presupposing the emergence of a heterogeneous setting therefore introduces what could be described as a paradox. In order to achieve operability there is a need for purification work, that is, processes aiming towards temporarily disentangled stabilizations, which however presupposes impure hybrid constellations (as this is at the core of the design idea). The dual hybridization of accounting must manifest through purifications presupposing impure settings. Our analysis is therefore directed at the operation of preliminary purification, a divided separation and a progressive reblending (Latour, 1993, p. 78). Hence, at the center of the analysis is the relationship between processes of hybridization and de-hybridization (Kastberg and Lagström, 2019), or, put differently, between too pure and too impure practices. Too pure might mean a failure to meet the rationale of the hybrid (for us the design idea) and too impure might mean a failure in operability. Although the result might be a temporal stabilization, there will always be things not accounted for and hence, there will always be destabilizing forces causing a need for further attempts to stabilize (Callon, 1998b). Still, we need to add one more layer, that of traceability, which we turn to in the next section.

The prerequisite of traceability

As described in our introduction, social investment takes the form of an idea providing a design (van Erp et al., 2019) for interaction. The design presupposes the combination of different elements resting on the assumption that it is possible to create a relationship between social operations and cost-benefit accounts. Based on our focus on how accounting is formed and transform, we draw on Latour's discussion and theorizing where the production of facts is described as translations performed in steps, illustrated by the metaphor of a chain (Latour, 1986; Robson, 1992). Whereas Latour's main interest was in the production of scientific accounts based on observations, we are interested in two different streams: one resembling the production of accounts, the other resembling the movement from the idea of the design to practices. The latter stream echoes the idea of attempts to control at distance. Since accounting is mostly a repetitive act, the process of representing leads to, or at least that is the aim, a systematization and organization of the interlinked activities (Power, 2015).

Translations are made in sequences, as in Latour's (1999) own example, of the divide between jungle and savanna: first the soil is measured and mapped, then samples dug up, then put into instruments that help carry, measure and weigh, then translated into numbers and figures, and as a last step combined with other datasets and discussed in a scientific paper. Each sequence presupposes disentanglements, that is, associations, dimensions and actors are effaced, whereas other things are associated. Each step means that new accounts are produced. That is, nothing is transported, but claims of representations are made. The production of accounts and control at distance presupposes a mobilization and association of human and non-human actors. It becomes of importance to pay attention to everything, artifacts and agents alike, that affects and forms the translations. Along this process we find actors with interests, artifacts that bend action around them and established practices. New accounting practices will meet established accounting and control practices supported by strong dense networks (Kastberg and Siverbo, 2015), resistance from strong professional identities, and artifacts important to enroll (Christensen and Skærbæk, 2007). We might expect this to certainly be the case when it comes to accounting in and for hybrids, because of the heterogeneity of elements.

The discussion opens up for an analysis of entanglement and disentanglement as essential dimensions of the formation of accounting itself. Whereas the metaphor of a chain of translation indicates a process with a gap between each link (as all chains have) there is an emphasis on interconnectedness (all chains keep together). This introduces an important prerequisite when studying accounting systems. Latour (1986) stresses the importance of the ability to travel back and forth along the chain: from the figure in the report and back again to the concrete local context. It is this reversibility and traceability that is the final prerequisite for the truth claims being made. If this travel is not possible, a “lie” will be produced (Latour, 1998; Preston, 2006), we might end up “semi-confused” (Hedberg and Jönsson, 1978), or worse, being accused of “bullshitting” (Macintosh, 2009). This entangling and disentangling is then to be performed in a fragile heterogeneous network setting, as we concluded in the previous section.

Thus far, we have specified the phenomenon we are interested in from an actor-network theoretical position. We have introduced the concepts of entanglement and disentanglement, and the theoretical framing indicates the importance of using these concepts both to analyze the elements brought together (one might metaphorically think of this as a horizontal dimension) as well as the different steps in the chain of translations (one might metaphorically think of this as a vertical dimension). We described hybridization as a process where (ever) more elements are brought together and entangled. As indicated in the introduction, some sociological accounting research on hybrids has made this the main focus (Denis et al., 2015). Our theoretical discussion however also further deepens the argument that hybridization presupposes de-hybridization through disentanglements. How this comes about is at the heart of our analysis when we turn to our empirical investigation of how the design idea turned into practice.

Method

This study reports the work with social investments initiated and funded by a Swedish region (henceforth the Region). As will be described in further detail below, the idea was to fund social investment projects carried out by the Region and cooperative partners (the organizational arrangement is described in Figure 1). In Sweden, regions are self-governing local authorities in the multi-level political system, whose main responsibility is to provide health care, including health promotion. Regions also have a responsibility for other policy areas such as public transport and regional development. Around 2010, the Swedish Association of Local Authorities and Regions (SALAR) began promoting social investments, a novel public sector governing framework around the idea of collaborative prevention within the broad field of welfare issues (Hultkrantz and Vimefall, 2017). The idea is that individual municipalities or regions allocate resources to fund collaborative and preventive projects (social investments) with the aim of reducing future social exclusion among citizens. However, the framework [1] also presents further specification such as the importance of providing quantified ex ante assessments and ex-post evaluations of the social and economic impact of each investment. Based on a problematization of a general uncertainty around the efficiency and effectiveness of preventive social programs, and a lack of a governing structure of inter-organizational collaborative measures, the framework according to its protagonists would respond to both an accounting problem (lack of evidence around efficiency) and a management control problem (lack of coordination and control practices in network settings).

We view the social investment framework as an emerging hybrid calculative tool based on several problematizations put together in one package. Adding to its hybridity, social investment is also a “control hybrid” aiming to incentivize inter-organizational collaboration by bringing together elements usually found separate, such as different professions and cost-benefit type calculations at street-level operations. Thus, social investment is a hybrid accounting tool facilitating the formation of organizational hybrid objects (the investments) as actors enter a relationship of shared resources and goals.

Our study borrows from narrative approaches (Llewellyn, 1999) to management and accounting studies where the objective is to generate an “interesting story” (Law, 2008, p. 142) from the often messy and complex single-case study. While such an approach is inherently interpretive in style, our ambition has been to follow the preparers of accounts in action (Chua, 1995), and thus to refrain from giving privilege to any particular group of actors over others. We designed a longitudinal study that covered the attempts to make social investments work over a period of three years. The longitudinal case study allowed us to observe events “in action.” The approach used here means paying attention to a phenomenon under construction before it turns into a black box with a set of properties and stabilized relations (Latour, 1987). This approach urges us to investigate, and treat symmetrically, the relations and doings of human actors, techniques, knowledge, allocation of resources and information in the process of making social investments work. Thus, it is worth noting that ANT is best understood as a methodological toolbox, a guide to the question “How do things, people, and ideas become connected in larger units and remain so?” (Czarniawska, 2017, p. 146).

Data collection and analysis

The basic goal of this study was to understand what went on in the field as actions and events took place and interpretations were made by various actors about the early stages of implementing social investments in the Region. To this end, the study reported in this paper is based on a multiple data source approach, including 48 interviews, 18 observations of various meeting and events, as well as documents and other written material related to social investments in the Region. The first stage of data collection lasted from December 2014 to January 2016. We began by interviewing politicians, the working group (WG), managers from the steering group and other officials at the region head office who were involved in the work of establishing social investments. Using a themed interview guide, we asked the respondents about their professional background, how they got involved in the process, and what they perceived to be their role in relation to social investments in the Region. Questions were also asked about how different actors interacted, the measures taken so far, problems that were identified by the actors, and the solutions presented.

After this first round of interviews it became clear to us that a vast majority of the work with implementing social investments was centered around an assigned group of civil servants (the working group) and their managers (management). At this point we were given permission to follow these actors more closely, and to observe internal meetings of the working group. Complementing interview data with observations can provide more credible claims of studying accounting and control in “action” (Ahrens and Mollona, 2007). A total of 18 meetings were observed in the first stage of data collection. The meetings lasted from two hours to a full day, took place in various locations across the Region, and included various invited actors along the process. The observations of meetings were conducted in a passive manner, deliberately not engaging in conversations. As the first stage of data collection proceeded, we also conducted follow-up interviews with the members of the working group and management. During these interviews, we made small adjustments to the interview guide, for example, adding themes about current issues of the process and questions that had been raised during observations. For example, although the idea of measuring project outcomes in terms of “clinical” effects and economic impact was described by the officials as inherent in the social investment model, as we learnt more about the process in the Region, this became a topic we returned to even more during follow-up interviews. The same strategy of initial, and later follow-up interviews, was also used when the first round of social investment projects was initiated in the spring of 2015. We chose to follow three of the granted social investment projects more closely, and interviewed several involved actors from each of these social investment projects. As the projects were entering a more operative phase in the fall of 2015, follow-up interviews were conducted focusing more on measures taken in the respective project and evolving issues around the organizing of a social investment.

A third important source of data consisted of various documents and work material produced by different actors in the process. Official documents such as formal decisions and a public investigation that was conducted prior to our study helped us understand parts of the process leading up the operative phase starting November 2014. We as researchers were also granted unlimited access to work material (generally not accessible to the public) prepared by the working group and stored online. The online material consisted of time plans, templates, application forms, criteria for social investments, communication plans, educational material, organizational charts, and more. To us as researchers, these written materials provided important traces of an accounting constellation in the making, especially since much of this material was subject to prior and subsequent discussions among the actors we studied. In this sense, the material did not only provide structure for the actors in our study, it also provided narratives to us as researchers as we tried to “take up the political tasks of composition” (Latour as cited in Czarniawska, 2017, p. 149).

The first stage of data collection ended in January 2016. As the first round of social investments was granted funding for around three years, we decided to conduct a second stage of data collection in early 2018. We returned to the field and interviewed a smaller number of key (as we regarded them) respondents at the project and regional level, in order to capture reflections on how they had perceived the developments over time. In the analysis we provide a limited number of illustrative quotes by informants of three categories: politicians, WG and project representative. Politicians and the WG were regional actors while projects representatives were formally employed by either the Region (e.g. health care) or a municipality (e.g. schools) as a collaborating member of a social investment project. For the purpose of this paper, the vast majority of interview data was used in addition to documents, research notes and other sources to inform the analysis more generally. For a summary of respondents and significant documents, see Appendix.

The analysis was carried out in three steps. First, the empirical material was structured as a chronological story with a focus on key events (summarized in Figure 2). Second, we sensitized the analysis by identifying events, activities, and accounts related to the emergence and organizing of control and measurement issues. Our goal was to identify how accounting is formed and transformed in hybrid settings. Emerging findings regarding the use and perceptions of calculative practices, such as cost-benefit type calculations and process-related reporting at the level the social investment projects, were therefore analyzed more thoroughly by sorting our collected material thematically. Although our study was processual by design, we were also aware that the formation of accounting systems in practice takes place in a complex context of organizational interactions and events. Therefore, various documents including project plans, communication plans and minutes from meetings provided important complementary sources of data informing the analysis. In the final step, a re-categorization emerged in an “iterative” or “abductive” (Alvesson and Sköldberg, 1994) manner as the final analysis was carried out. We compared our findings with extant research on hybrid accounting and the concept of translations (Latour, 1999). This triggered the three narratives we conceptualized as translation points. It is important to stress that the conceptualization (as in the specific theoretical framing) and the categorization (as in the division of empirical observations) did not exist beforehand. Translation points as a concept emerged during the process of analysis and reading. Alternative labels could have been “sites” or “calculative centers”. However, we settled with translation points (as used by Callon, 1998b, p. 16) in order to stress one of our core observations: at each translation point, we observed hybridization and the entangling of heterogenous elements. It is important to recognize the processual character of the translation points. They were temporal stabilization, and although they in that sense produced irreversibility regarding time, they are in themselves not irreversible (Latour, 1993). On the contrary, we might expect the stabilizations in themselves be a source of further problematizations and hence always fragile (Callon, 1998b).

Analysis

The analysis begins with an introduction to and empirical background of the case of social investments in the Region. This is followed by an empirical analysis of the processes, events and actions taking place in what we label translation points.

Introduction to the case

The Region (a county council) initiating the work with social investments is one of the larger regions in Sweden. The main responsibility of Swedish regions is the provision of health care, including health promotion. They have also a responsibility for public transport and regional development. Several political boards are accountable for the provision of the different operations. Since, as will be addressed more thoroughly below, the work with social investments was aimed at school children, it is important to recognize that schools and social services are not regional matters, but part of the municipalities' areas of responsibility.

While the very first discussions about social investments among officials with public health competence were held in 2010, more intensified work did not start until 2012. A political board then decided that a more thorough investigation should be conducted. At that time, as the respondents say, little was known about social investments or how to work with them, hence the need for an investigation. This resulted in a commission of enquiry aimed at describing what social investments was about in a broad sense, and identifying important aspects to consider when organizing the work. This resulted in a formal political decision describing overall direction regarding the definition of social investments. Some things concluded as a result of the report were that the work should be narrowed down to targeting young children who risked leaving school without high school eligibility and how the work should be organized and rest on cooperation between different parts of the Region.

Since there was a lack of experience of this kind of work, both within the Region and within the public sector at large, it was decided that the first year (2014) would be a trial with limited resources amounting to SEK 5 million (approx. US$0.5 million). In the following year, the project would scale up and invest SEK 30 million (approx. US$3 million). Initially, the idea was to invest in a few pilot social investments projects, and to learn from them. Still, when 2014 had passed no social investments had been made. A few potential projects had been identified by administrators at the Region, but the projects were not regarded as ready to be invested in. The projects did not meet the criteria of what was regarded as a social investment, while the process of identifying projects resulted in increasingly specific criteria. As a respondent representing one of the projects stated: “they keep coming back with more and more demands.” What we observed was a lot of activities related to organizing objects and relationships.

In 2015, the first five investments in projects were initiated. This was a result of a general call for projects. While more than 20 applied for grants, only the five successful applicants were regarded as being close to meeting the requirements. Parallel to the start-up of the projects, control initiatives were introduced in order to safeguard that the projects would live up to the agreements (Figure 2).

Observations of translation points

What we observed was the development of a calculative framework through which associations and disentanglements were asymmetrically negotiated across different sites of the network. We call them translation points since each of them represents sites where both the representations of the design idea and the evolving construction of relations were most visibly negotiated and temporally stabilized. At each translation point, elements were disentangled and re-combined, triggering responsive action (at distance) from other translation points. Accounting was mediated as well and manifested in different ways at each translation point. However, the translation points were not limited in time making up sequences. Instead, they represented stabilizations and co-existed. We will return to these observations, but start the analysis by presenting what we label translation point 1 – Strategizing.

Translation point 1 – strategizing

We describe the first translation point as “strategizing” since the activities observed and the rationale of introducing social investments were highly related to other overarching issues in the Region and about making general decisions regarding the overall direction and organization of the social investments. Here, the design idea, in itself a hybrid because of its emphasis on relating costs to benefits (financial and non-financial values), was translated into concrete associations. Hybridization did not stop but was transformed and proliferated into a new constellation of actors and practices. Decisions were made about how to allocate responsibilities and accountability structures within the large Region's organization and elaborate discussions on what a social investment is took place.

Core actors enrolled in strategizing were politicians, officials and actors on a national level acting as experts. At this translation point, social investment was regarded as strategically important, as it promised an efficient use of resources due to the strong relationship between costs and benefits. For the politicians it was tempting to introduce a model which applied to an economic logic, and that could improve people's quality of life:

The novel aspect of social investments, I guess is that one looks at investments in children and youths as investments, just like an investment in infrastructure or in buildings. Just as we create opportunities for operations, we also create opportunities for a foundation for children and youth to grow up. Because if we do not…the costs will increase, like for health care and therefore we have to take our share and invest. (Politician (#9), 2015)

One important initial activity when the Region started to investigate social investment and how it could be operationalized was to define it in a broad sense. Politicians decided that for the Region, social investment should be about improving the conditions of children and young people. This would be done through proactive, preventive efforts, with a narrower formulation that social investment projects should improve the number of children who leave high school with completed grades. While attempts were made to specify the object, we observe how it still was a fairly broad definition.

Since the argumentation was that the investments would pay off manifold in the long run, it was argued that social investment should be kept separate from other issues related to the conventional annual budgeting system and prioritizations. The allocated resources were sometimes referred to as the social investment fund. Yet, while singling out social investments in relation to ordinary practices of resource allocation was described as important, so was associating the new model to existing political goals. One official described the background of social investments as a result of a growing interest in putting results in relation to resources invested: “… that is, to measure the outcome in fiscal terms” (WG representative). Still, the idea of social investments could also be aligned with other issues regarded as important such as social sustainability, a concept growing in popularity as well. During interviews, references were made to the strategic policy document “Det goda livet” (The Good Life), which contained ambitions on social improvement and social sustainability:

… we produced this plan for social sustainability, so social investments were a way to start that work, in order to create a more equal health, it is a number of factors. (Politician #9)

At the site of the emerging network that we label “strategizing,” core activities also included allocating assignments, resources, and responsibilities regarding the operationalization of the work with social investments within the organization. More specifically this included decisions about the formal structure of governance such as the issue of which departments would be involved in the development and organizing of social investments.

Organizationally the funds came to be sorted under the Regional development board (RDB). However, this was not a clear-cut decision in everyone's eyes. Prior to the assignment of the RDB, there had been a discussion about which department would be suitable for running the program, and other alternatives were discussed. For example, officials representing the Board for Public Health (BPH) argued that social investments belonged to their operational competence and that it would have been reasonable to place the program under their responsibility. However, the political decision made it clear that the overall responsibility of the social investments program was to be sorted under the RDB, but also, as part of strategy to escape the confinement of conventional budgeting and organizing, that the program should be implemented in cooperation with representatives from other boards, for instance the BPH. One politician explained during an interview that this decision was made because the RDB had experience in implementing projects. Further, the RDB was perceived as action oriented in contrast to BPH, who according to the politician tended to be oriented less toward concrete operations and more towards structural questions.

The RDB was more used to setting up project organizations running over a limited time and really getting things done while the BPH has been more overarching and maybe not so concrete. (Politician #8)

For the politicians, social investments were clearly about accomplishing specific and concrete social interventions based on calculations of costs and benefits. However, we also observed a deviation from the integral idea of social investments: to set up a funding model making it possible to reallocate resources to the program based on actual savings made as a result of interventions. Instead, a “budget model” was chosen, meaning that the Region would allocate resources to the social investment program annually, without affecting the budgetary frames of specific departments. This choice of financing model allowed the work to continue without internal negotiations about the size and timing of realized savings across departments within the Region.

Within the Regional Development Secretariat (RDS), a steering group made up of representatives from the RDS and other departments was appointed. A more operative group, the three-member Working Group (WG) managed the day-to-day activities. One of the first tasks of the WG was to investigate and propose a plan for politicians, concerning how social investments were to be realized in practice. This however turned out to be a challenge and for some time two alternatives were weighted against each other, each with advocates in the WG and the wider network of actors. One idea was to proceed with a project-based approach where the WG acted as “investors” setting up criteria and demands. The other idea was to let the WG actively act as a co-producer who negotiated appropriate measures with concerned stakeholders (other regional units, municipalities and other relevant actors) regarding co-financing and joint operational responsibility. Not without struggle and conflict, the first model was chosen – not least since it became important to show some results in the form of actual investments. For instance, politicians were eager to show progress. The first model seemed an easier way to proceed, as fewer had to get involved at a strategic level. The WG would therefore, according to the selected model, not carry out projects but channel resources to applicants describing promising social investment projects (see Figure 1). This resulted in a proposal of a number of criteria for what was to be regarded as a social investment. The criteria included structural organizational requirements that reflected the strategic aim of politicians and administrators within the newly appointed steering group of managers. For example, social investment projects were required to be novel (in relation to existing practices at the level of operations), and had to be carried out in a collaboration between regionally governed operations and municipally governed operations.

Thus far we have described the role of the strategizing translation point in the move from idea to practice. Following the stream in the other direction, we observed the opposite move. The multidimensional world was translated into accounts of the investments. Investment applications, now ranked in a document, provided a basis for decision-making by politicians. We again observed a disentanglement from the design idea. The ranking, the construction of which will be elaborated in the next section, was not based solely on cost benefit calculations, but a summing up of several dimensions was taken into consideration. This move made possible political decision-making on social investment projects, and thus, reconnected to political ambitions and strategy.

In summary, strategizing represents a translation point where the abstract design idea had to be concretized and actors enrolled (Latour, 1986b). Initially, actors with different competencies and from different departments were enrolled, temporarily resulting in dispute within the WG over the course of action and the project-based approach. Here, active engagement of politicians resolved the dispute and stabilized the hybrid setting temporarily. Following the entanglements of hybridization we observed how a hybrid setting emerged as politicians and representatives from different units and boards of the Region got involved, and that ideas of interventions were associated with financial considerations, as one “looks at investments in children and youths as investments.” Following the disentanglements, we observe how practices and actors were distanced. Strategizing involved disentangling from traditional budgeting and single-purpose organizing. The definition of social investments as concerning schoolchildren at risk of not obtaining sufficient grades allowed the work with social investments to gain focus and associate relevant actors. It, however, also meant disentangling the work from other social concerns such as unemployment or integration. The models of re-financing the program through decreased future departmental budgetary frames and of co-production was disentangled in favor of conventional budgeting and a project-based model. Both choices meant more issues could be settled within the mandate of the Region, causing less of a need to deeply involve (on this strategic level) other actors such as municipalities, and hence, a gaining of operability. Following the stream in the opposite direction, we observed further transformations. The idea of judging investments based on cost-benefit based calculations was partly abandoned and replaced with decisions on ranked projects, where the ranking, as we will describe in the next section, was a hybrid construct, shaped and formed in relation to the translation point we name administrating. Altogether, we observe purification work resulting in a stabilized set of associated actors and, in relation to the design idea, redefined accounts representing investments.

Translation point 2 – administrating

We describe the second translation point as administrating. We observe the emergence of new associations as well as how elements of the design idea were disentangled from the strategizing arena, making it relevant to be represented as a translation point on its own terms. At this translation point, the work was operationalized into control models, more specified objectives and an expansion of the network to include expertise, administrators and yet again politicians (but with a new role). Most engaged was the WG, which led the operative work. The people involved had different backgrounds and competencies, but shared the interest of addressing social problems in a proactive manner.

One central category of activities concerned enrolling operative resources such as core personnel with the right competence, and personnel representing organizations defined as important to enroll. One expression of the former was the staffing of the expert committee with the assignment to suggest what projects would be granted resources for making social investments. One expression of the latter was the establishment of a social investment political advisory board acting as link between the administrative and political levels of the Region. (This was a new set of politicians who were not members of the board that made investment decisions.) The advisory board occasionally met with the WG, to get information on the development of the work. One central theme at these meetings was how to operationalize the work of social investments in more concrete terms. While it was easy to gain support for the proactive idea of social investments, it was more challenging to communicate what social investments were more specifically. In the work process the object was specified further. The WG came up with an extensive list:

  1. The social investments have a clearly defined target group. The potential clients must be specified.

  2. Needs of clients should be well defined.

  3. Well-defined and valid method/model, or idea of how a method should be tested.

  4. Well-defined effect; what to expect regarding outcome and causal relations.

  5. Must be novel and different and not be part of other ongoing operations.

  6. Clear time frame/plan, while aiming at long-ranging effects.

  7. The project must be well planned regarding execution (beginning, ending, sub-goals).

  8. The social investments must consist of inter-organizational cooperation.

  9. The social investments must be measurable and followed up.

  10. A cost-benefit type estimation of the social investments must be included.

  11. A plan for continuation after the project is ended must be presented.

  12. There must be a structured and prepared organization for the project (project owner).

  13. The project must be managed by the applicant organization (and not by a hired consultant).

  14. There must be some co-financing from the cooperating organization.

The linking of actors to the social investments projects was also about safeguarding commitment and a stable relationship to other ongoing operations. Since one of the aims of social investments was that projects should be integrated with ongoing operations after the projects had been finished, it was important that the investments made by the Region were mirrored by efforts made by the cooperating organizations. Work hours, physical space, and other facilitating arrangements should be provided by the applicant organizations in order to signal that they are also making an effort. This proved to be challenging for applicants that represented small organizations, since it was difficult to find replacements for the persons that were to engage in the projects.

We can notice how processes had to be linked, decisions had to be made in different arenas and budget processes of different interacting parties had to be taken into account. The latter proved to be an obstacle also for the WG who, despite the definition of social investments as long-ranging and aiming at overcoming the fiscal year, also had to plan in order to avoid negative aspects of the Region's budget process; resources not spent on social investments during a fiscal year would not be available the following year.

While we observed a lot of activities aimed at stabilizing a supportive network of actors, there were many other organizational activities going on as well. While some of them concerned creating coordination within the WG and between the WG and the social investments projects, this was also very much about establishing set identities and roles. Regarding the coordinative activities and the work processes around social investments, a central issue was to establish a process for evaluating potential investments, and later on to establish a way to work in relation to the projects that had been granted resources.

The idea of cost-benefit analysis and socioeconomic motivations were described at the center of the argumentation in translation point 1. Here at translation point 2 the focus on calculations, guidelines regarding tradeoffs or discussions about priorities between different social investments projects based on costs and benefits were toned down. Instead, the argument was that this could be refined later, and instead the focus was shifted towards the budget and financial planning of the different projects. The motivation of not being too engaged in cost-benefit calculations was motivated partly by a recognized lack of competence at the level of project organizations.

During the observed period, an important change of direction was the decision to work according to a model where different projects could apply for grants. It was arranged through an announcement in the spring 2015. The decision to make an announcement led to an intensive period with a range of activities related to the establishment of a sufficient process both regarding the announcement and the evaluation of social investment applications. One member of the WG described how this intensified the work with specifying the process:

… that we actually decided to make an announcement with a set date. We had to stop discussing and had to settle on criteria, application form and frame social investment and what kind of investments we should make. We had to set an assessment committee with experts and all of that. (WG representative #1)

A vital activity was the communication to potential applicants concerning what was needed in order to be eligible for grants. Another activity regarded producing accounts or representations of the projects. The collaborative efforts that the projects consisted of (more thoroughly described in the next section) were here turned into spreadsheets containing numbers describing qualitative aspects, costs and also probability of success.

… we want to assess them based on numbers, and yes, we want a scale and that scale should illustrate to what extent they live up to something … we divided the assessment into different main areas … one was analysis of problem and attainment of objectives, executability … is the budget reasonable? (WG representative #1)

In the process of turning the projects into numbers, the expert committee set-up played a pivotal role. It made judgments based on local knowledge about the projects and their context, combined with abstract knowledge about methods, causes and effects. The final result was a ranking of the projects. All in all, 21 projects applied for grants, and five were successful.

To sum up the observations at translation point 2, we observe hybridization as new associations emerged. The set of actors and practices were a different constellation compared to translation point 1. The design idea was again transformed since the strategic cost-benefit focus recognized at translation point 1 was downplayed here. Instead, we observed how the idea was turned into a focus on credible output control, administrative routine and a wide set of criteria. The definition of an investment was increasingly taking the form of a more specified object equipped with criteria and requirements. Social investments were hybridized and related to assessable methods, organization and output definitions materialized in the application form. New competences and practices were introduced as well. In addition to the officials of the WG, experts and politicians in an advisory committee got involved in the work. Both the politicians and the experts were mediating by facilitating stabilization, and by lending the processes trustworthiness. For example, the experts had a focal role in the reductionist move where the projects were turned into numbers, which allowed ranking. This observation also signifies the processes of purification the design idea underwent. As described above, the relationship between costs and benefits was downplayed and instead manifested in accounts of output control, materializing in for example an application form (following one stream) and a ranking (following the other stream).

Whereas actors and practices were associated, others were disentangled. At translation point 2, the involvement of the politicians (of the board) and other higher-ranking officials was less visible. The disentanglement of cost-benefit analysis also meant that the involved parties could reside within their own circle of competence and not involve experts to, first, conduct the analyses and, second, to create relationships with parties benefiting from the social investments. However, perhaps the most important purifying act was the decision to choose a model where projects applied in competition for grants. As described, this changed the character of the work, so it became about narrowing down and specifying to such a degree that the idea could be materialized on an application form. Indeed, this move gave rise to new hybridizations and purifications in what we label “aligning conduct.”

Translation point 3 – aligning conduct

We label a third translation point “Aligning conduct.” The activities here concerned the conduct of carrying out operations that could be regarded as social investment projects. Since the investments were not physical or tangible, but instead cooperative coordinated activities, much effort was devoted to aligning activities in relation to the demands put forth at translation point 2 and materialized in the application form. This is also why the label “aligning conduct” was chosen; it was not just about counting projects out there, but about making projects count. While all of the projects that were allowed grants in the first announcement were planned before the announcement, they had to be altered in order to correspond to the templates of social investments. As will be described, the projects had to be made controllable as social investments, and had to be able to be represented as social investments. Once this trick had been carried out, well-run projects could be regarded as social investments. The design idea shifted form into new representations as new associations emerged.

The most active actors here were social workers, teachers, researchers, psychologists, nurses and other occupations working actively within the social investment projects. As a consequence of the requirement of inter-organizational operations, they represented different parts of the public sector. The constellations differed between the projects depending on aim and focus. For actors at the level of the individual project, translating their ideas of important social work projects according to the structure, vocabulary, and requirements of the social investments was key to secure funding. In relation to the projects, the WG took on an active role as communicating the requirements and the benefits of organizing the social investments in a way that facilitated measurable outcomes. It was for example important to communicate the importance of a having well-defined target groups, reliable organizational structures and reasonable output measures. Further, the WG stressed novelty, that is, the projects had to be clearly demarcated in relation to other ongoing operations.

Several activities and objects aimed at generating descriptions of the projects. Templates produced by the WG provided a specified structure for prospective projects applying for funds: target groups, defined cooperating parties and expected outcomes were examples of what had to be specified. The applications also had to include descriptions regarding staffing, formal support of the cooperating organizations, and social work methods that could be developed or applied. Yet another dimension was that of benefit estimations. As we observed meetings between prospective projects and the WG, we observed how this calculative requirement was downplayed in relation to project applications. This was also mirrored in the subsequent completed applications received by the Region. As the WG received applications for a first assessment during spring 2015, a group of experts from different fields was assembled (business management, social work and medicine). Their task was to discuss and rank the applications. Although calculation was one of many factors included in their assessment, we observed how the experts perceived the calculations as vague, schematic and generic. This resulted in a rather permissive attitude regarding the financial dimension from the expert committee when assessing the projects in relation to the cost-benefit dimension. The project organizations, being the most important new group of actors enrolled at this translation point also meant new controversies as well as a need to purify the work. In interviews with project representatives, we were told that describing and calculating on social costs and benefits was difficult, not least since this was a new activity for them and they lacked relevant knowledge.

While the importance of calculations was downplayed in practice, what really mattered was that the project could meet other requirements. As reflected in the quote below, where a WG representative talks about a discussion with an applicant, it was not easy for all projects to meet the requirements:

But they too did not perceive themselves as ready for investments as we started to ask questions. What are your target groups, who are the children you are addressing, why are you doing this? And then they thought that, no, we have to do more work first and what do our forms and structures for cooperation look like. (WG representative #1)

All of these demands illustrate calls to align conduct with the design idea of social investments. The projects had to be restructured in order to meet the criteria. Still, as for the criteria of calculating benefits, criteria about novelty and innovation turned out to be similarly difficult to realize in practice. None of the five projects receiving grants were made up just as a consequence of the Region's initiation of social investments. Instead, they were all planned and an organization existed for cooperation, although there were considerable differences regarding how far they had proceeded in planning and concretizing the projects. Still, for all of them it meant a lot of additional planning and organizing in relation to the specific social investments demands. As social investments should consist of cooperation between different actors answering to different mandators, like municipal and regional boards, a lot of effort therefore had to be invested in arranging a convincing organization with one appointed mandating organization and one or more cooperating organizations. This caused a need for the interacting actors to produce letters of intent, time reporting (of the actors' involvement), and formal decisions by, e.g. political boards and higher-level officials.

As for the actual work methods carried out in the social investment projects, much effort was spent on establishing causal relationships. This was a recurring topic during conversations between the WG and the projects. Not only were project performance and outcome measures regarded as key features of social investments as such, they were perceived as important information for subsequent decisions to scale up and implement successful projects as part of ordinary operations. As illustrated in the conversation between a WG representative (#2) and a project representative (#20) below, specified outcomes of the projects had to be measured and indicators identified and systematically followed up:

WG representative: One must be able to see the effects, the outcome of the efforts. We must have indicators. Absence (from school) is easy to measure.

Project representative: Do we have to specify the exact methods we use to measure? We would like to allow some trial and error.

WG representative: You have to be specific. We must measure and capture the difference between today and the effects after some years. We want to see more indicators. Of course, you cane fine-tune it after the application has been sent in. You can have a frame within which adjustments are made.

Project representative: The target group is 3–10 years. Can grades in level 6 pass as an indicator?

WG representative: Yes, but we do not make such long-reaching investments. You should implement a 2–3 year long investment in your operations. You must have an idea about how to implement it in your ordinary operations afterwards.

When summing up translation point 3, we can observe, as indicated in the last quote, that the time frame had narrowed further and the request for well specified indicators raised questions. While the long-term strategic ambitions we observed at translation point 1 were still discursively present in conversations and applications, in practice, it was about two to three years, with identified short time indicators. At translation point 3, the social investment network of actors had expanded further and now included the projects, each hybridizing the network to include the different competences needed for conducting the projects as well as to respond to organizational and calculative demands from other translation points. In practice, such responses also meant further disentangling the idea of cost-benefit calculations and the criteria of novelty. Since the projects did not need to prove that they concerned entirely new approaches or methods, new sets of actors did not have to be involved. Existing constellations of actors could be brought in and, after adjusting the projects a bit, count as investments.

The design idea was represented in a new way, again; it was now about division of labor and coordination. We observed disentanglements regarding output control as well as added associations materialized as criteria of coordinative formalization of projects, and application of methods likely to have effects. Costs, when discussed, were about staffing and allocation, and less so about the relationship between costs and benefits in the long term. Still, the underlying idea of output orientation was present, but here manifested in the form of credible (likely to produce desired results) methods to develop or apply. Arguments based on cost-benefit calculation had largely been replaced by arguments about causal relations between interventions and effects.

Discussion: hybridity multiplied and de-hybridization

Our empirical case is not about how an accounting system is introduced in a hybrid setting (Kurunmäki, 2008), or how accounting forms part of one element (Kaufman and Covaleski, 2019). Rather, it is a case of how accounting as a design idea was essential in mobilizing a hybrid setting (Thambar et al., 2019). Rather than an integrated design idea put in practice, we observed the emergence of a semi-integrated system. The accounting system, as expressed in the design idea was constitutive of a “dual hybridization process” (Miller et al., 2008) where accounting mediates while being mediated (van Erp et al., 2019).

What we observed was different temporal stabilizations that we suggest to conceptualize as translation points. At each translation point we observed purification work, where disentanglements allowed heterogeneous elements to stabilize. Each translation point manifested its own negotiated order: social investment was about relating cost to benefits, it was about output control, and about trustworthy application of methods (as illustrated in Figure 3). Yet, it was not a full compartmentalization (Kastberg and Lagström, 2019), where the translation points operated without regard to each other. Traceability was not lost. Instead, what we observed was how the translation points acted as “faithful intermediaries” (Latour, 2005, p. 39), producing predictable effects for each other. They represent “co-centers of action” (Yu and Mouritsen, 2020, p. 15). The design idea mediated while being mediated.

For each translation point, the disentanglements facilitated purification. All the elements and their associations represented at the other translation points had not to be considered. It was the “displacements” (Law and Singleton, 2005, p. 341) that kept the network in place. Still, and this is an important observation, whereas each translation point produced its own representation, as we have described, their interrelatedness makes up parts of what an observer at distance (like the researcher) would describe as a hybrid. This observation allows us to elaborate on the relationship between the translation points, extending the analysis from observations also made by others that interpretations and enactments might vary across a wide and heterogenous network (Themsen and Skærbæ; van Erp et al., 2019). From an ANT perspective, it is the stabilizations that need explanation (Latour, 2005, p. 39) and we have provided insights about the negotiated order of disentanglements. We discern three mediating factors facilitating the stabilization of the translation points with regard to each other: productiveness, likeness and active mediation by experts.

While being disentangled vis-à-vis each other, the translation points proved to be productive for each other as well. They produced predictable effects. At each translation point, we observed a double move; we could view this as a demand side and a supply side. What is intriguing about this is that the two movements also represent transformations, but of slightly different kinds along the chain of translations. On the demand side, complexity was increasing, e.g. politicians' demanded solutions to problems of social exclusion, administrators demanded calculable projects, and projects demanded resources. On the supply side however, projects supplied partial and incomplete translations of their respective projects along the chain. At administrating, these objects were transformed once more, and turned into a ranking in a spreadsheet in order to facilitate political decision-making and resource allocation (as summarized in Figure 3).

Although disentangled and producing different representations, the translation points were not fully disintegrated, which is mirrored in the representations signaling likeness, while not resembling sameness. All the actors agreed on the underlying problematizations, as conceptualized by Callon (1986): the importance of enhancing output orientation. At the translation points the focus varied, different problems were centered and different solutions put forth. Still, there were common themes, although these were addressed in different ways. For example, the idea of focusing on outcomes meant strategizing with cost-benefit as the strong argument, organizing around output control and investing objects coordinating and providing performance evidence. Outcome orientation turned into different things, but there was also likeness in the sense of resemblance.

Active mediators between the different sites were not systems, as observed by Themsen and Skærbæk (2018), but rather experts. As social investment applications were received by the WG, mediation between the multifaceted character of social investments projects and the strategic ambitions of social investments was needed. We observed how the expert group played a pivotal role in this translation. The different backgrounds and fields of expertise provided credibility to the process of ranking applications. Thus, in one direction, administrating meant translating strategy into output control and accountability, and in the other direction, it meant translating the ambitions and plans of projects into calculable (or at least commensurable) and strategically motivated investments.

When we as observers move our attention between the different sites, we are able to scrutinize the translation points as well as the relationships between them. What we have observed is hybridization when different elements at each translation points were entangled, as well as de-hybridization when different translation points were disentangled. Rather than observing different elements being brought together, making up one entangled translation, we observed multiple sites combining elements, hence being able to be described as separate negotiated translation points resting on work of purification. The hybrid design idea multiplied when enacted at the different translation points. The design idea was divided and separated and then “reblended” (Latour, 1993, p. 78) and each translation point represented its own negotiated order where a blend of elements were stabilized. Each translation point could be characterized as disciplined (Kastberg and Lagström, 2019) in the sense that elements were carefully (Latour, 1993) brought together and this was, as described, facilitated through disentangling.

Conclusion and contribution

The theorization of hybridity and accounting is thus far much inspired by new institutionalism (De Waele et al., 2021) resulting in important contributions regarding the use and role of accounting, accountability and performance measurement “in hybrid organizations” (Ferry and Slack, 2021; Rajala and Kokko, 2021; Vakkuri et al., 2021; Zawawi and Hoque, 2020). However, we argued in the introduction that this mainstream literature only to a limited extent recognizes or discusses disentanglements. The ANT literature on the other hand has put disentanglements at center of our understanding and instead problematizes entanglements, or rather, how they come about. If we in relation to the first strand of literature add by conceptualizing and illuminating disentanglements and how this allows multiple translation points to co-exist, in relation to the ANT literature we add by illuminating the relationship between the translation points and how traceability is maintained.

In order to elaborate on our contribution regarding disentanglements, we here firstly outline how accounting manifested as the design idea of social investments was realized. The design idea mediated and an “accounting constellation” was established where elements did come together and got entangled as often illustrated in the accounting literature on hybrids. Not only did different formal organizations get involved, but a variety of actors, such as politicians, administrators and experts (representing different professions), came together. Still, the design idea was mediated as well. It could not be described as coherent and stable (Fischer and Ferlie, 2013; Kaufman and Covaleski, 2019). This resembles the notion of dual hybridization (Miller et al., 2008) indicated by the combining of metrological activities and expert judgment, and how qualitative data were turned into quantitative. Still, these different elements were not merged or present at one translation point. What we observed was different disentangled translation points resting on purification work. Actors, perspectives and time frames were added, while the disentangling prevented the situation from turning into too complex a situation (Callon, 1998b). The hybridization as it was enacted allowed parallel representations and accounts, which is in line with other observations (Themsen and Skærbæ, 2018; van Erp et al., 2019; Yu and Mouritsen, 2020), while disentanglements of translation points vis-à-vis each other in turn made it possible to enroll and distance different actor groups. Disentanglements allowed a hybrid design idea to be enacted as multiple disciplined hybrid translation points. Accounting manifested itself as a semi-integrated system. Hybridization and entanglement was the norm (Miller et al., 2008), but it presupposed disentanglements; at each translation point there were never “too many links to count” (Callon and Law, 2005, pp. 729–731) and provided a temporal stabilized achievement.

Yet, and this is an important dimension of our contribution, the disentangling did not result in fully disintegrated translation points. Our empirical study indicates that it is not only important to recognize the disentanglements that emerge between translation points, but also that they kept being interrelated or as we described it semi-integrated. The prerequisite of traceability as we introduced it in the theoretical section was met. Our conceptualization of separate translation points allowed us to also describe how they were related and this is how we contribute the ANT based accounting literature. In the discussion, we illuminated how this came about through productiveness of the translation points in relation to each other, through likeness and shared problematizations and through active mediation by professionals. All of these mediating factors allowed the different translation points to function as “faithful intermediaries” (Latour, 2005, p. 40) in relation to each other, making them gravitate. Still, these mediating factors also indicates the paradox indicated by Latour (1993). The work of purification and stabilization making it possible to talk about disentangled translation points rests on mediating activities. The work of purification is re-embedded in the work of translation and always open for entanglements.

Our interpretation is that this semi-integrated system is at the same time a strength and a possible challenge in a hybrid setting. Let us begin with the strength: we can only speculate if a different translation, with less demarcated translation points, would have caused failure due to lack of support (Qu and Cooper, 2011). Studies have shown that integrated accounting systems have had strong coercive effects (Covaleski et al., 2013) and also how this has caused a breakdown of the hybrid setting (Fischer and Ferlie, 2013). An integrated accounting model put into practice might therefore not only mediate, but also turn the setting into a fully de-hybridized practice, which in fact is a failed hybrid (Kastberg and Lagström, 2019). The strength in our observations was not the stable and integrated nature of accounting, but the potency when it came to mobilizing meditators. The semi-integrated accounting system allowed different elements to be added and to coexist. Actors conducting the projects could go ahead without having to be bothered by proving cost-benefit gains, whereas decision-making politicians could align with the idea of a cost-benefit rational.

However, from a managerial perspective, and now we are touching upon the potential challenge, the multifaceted character of social investment could also cause uncertainty regarding what social investment actually is, and how to manage it. As we have shown, social investment is best understood as more “fluid” (see Law and Singleton, 2005, p. 337) and flexible concept providing facilitating gravitation rather than hard facts or a clear path of action. This might indicate that disentanglements like those that we observed facilitate flexibility at the cost of governability from a managerial perspective (at a distance). This observation suggests that the emergence of hybrid settings, as a result of implementing calculative tools and practices to achieve complex goals that reach beyond the control of a single organization, may require managerial practices that increasingly mediate between multiple perspectives across the network of actors and that acknowledge the inherent complexity of the task.

What do we learn about the phenomenon of social investments? While the social investment project cannot be described as an investment in the more traditional sense, the results in this study allow us to argue that what we observed was a movement towards a formalization of a calculable long-term impact of social work. The convincing idea of investments became a tool in a transformation of the conduct of social work. Still, recognizing the two “streams” referred to in the theoretical section, we can observe how this performative movement was realized in several steps. The idea of cost-benefit based management was turned into output-control oriented organizing, which in turn was turned into an enhanced focus on operative project with a reliance on evidence and strong coordination. For the projects this meant a tilt towards a new way of thinking about the way of organizing and conducting social work (see also Cooper et al., 2016), although the new way of doing things by no means was a total break with earlier ways of doing things. Following the stream in the other direction, we can observe how the messy local reality was turned into numbers and figures, and finally could be presented in a spreadsheet, and as a consequence, how it aligns with the more general idea of social investments. Following these observations, one might argue that what we observe is how the chain of translations made it possible to connect previously rather distanced elements, cost-benefit management calculations and traditional social work, in a way that meant that both had to be adjusted in order to realize that connection.

The discussion and conclusions indicate that there is a need for further research. The paper suggests the potentially productive role of disentanglements in allowing multiple hybridizations to evolve in hybrid accounting settings, and a practical implication of our paper would be that semi-integrated accounting may be beneficial in hybrid contexts, however, potentially at the cost of governability. Yet, it also indicates some limitations of the paper. First, a possible limitation arises from the ANT approach of the paper as it presupposes a proliferation of hybrids in society. Thus, the approach differs from the more macro- and meso-level approaches to hybridity that primarily concentrates on formal structure and governance (Denis et al., 2015). Second, since our conclusions are founded in a specific case setting, more research is needed to further our insights about how we can understand when disentanglements stop being productive (when they rather cause a hybrid failure and/or lack of manageability) and what mediating factors in different settings contribute to the survival of calculative networks.

Figures

Organizational model of social investments

Figure 1

Organizational model of social investments

Timeline of organizational events

Figure 2

Timeline of organizational events

Different translation points

Figure 3

Different translation points

List and summary of respondents

ID No. Int.Year(s)Note
Role at region office
1Adm. Working group22014, 2016WG member RDS
2Adm. Working group42014, 2015, 2016, 2018WG member BPH
3Adm. Advisory board12015
4Adm. Advisory board22015, 2017, 2018Worked close to WG
5Adm. Advisory board12015
6Adm. Working group12015WG member health care. Terminated position 2016
7Adm. Advisory board12015
8Politician RDB12015The Green Party
9Politician PHB12015The left party
10Adm. Steering group2Jan 2015, Oct 2015Head of business dev. RDS
11Adm. Advisory board12015
12Adm. Steering group12015
13Adm. Steering group12015
14Adm. Steering group12015
15Adm. Working group12017Appointed WG 2017
Role SI project level
17SI project 1, member22015, 2018
18SI project 1, member12015, 2018
19SI project 1, member12015
20SI project 1, member22015, 2018Project leader SI 1
21SI project 1, member12015, 2018
22SI project 1, member12015
23SI project 1, member12015
24SI project 1, member12015
25SI project 2, member22015, 2018
26SI project 2, member12015
27SI project 2, member22015, 2018
28SI project 2, member22015, 2018
29SI project 3, member12015
30SI project 3, member12015
31SI project 3, member22015, 2018
32SI project 3, member22015, 2018
33SI project 3, member12015
34SI project 4, member12015

Note(s): Adm = administrative; SI = social investment; WG = working group

Summary of interviews and observations

ActivityInformant (no.)No. of interviewsPeriod
InterviewsRegional office administrators and politicians (18)222014–2018
InterviewsProject staff (18)262015–2018
ObservationsRegional office meetings112014–2016
ObservationsProject meetings72015–2016

List of significant documents studied (registration number)

1. Investigation of social investment fund, VGR (RUN 610-0488-13)2013
2. Official statement, social investment fund, VGR (RUN 610-0488-13)2013
3. Compilation of applications to SIF2015
4. Application form, SIF2015
5. PowerPoint SIF, working group to inform future applicants2015
6. Completed applications by projects2015
7. Project plans from all granted projects2015
8. Notes from various meetings2014–2016
9. PowerPoint project management, educational day for project members2015
10. Communication plan2015
11. Internal procedural documents: “to-do lists,” “time plan,” “implementation plan” and “organizational charts” within the working group2014–2016
12. Assessment template for project applications2015
13. List of criteria and procedural guidelines for assessing applications2015
14. Political motion (RSK 792-2010)2010
15. Political decisions for the establishment of SIF (RS 2149-2012)2014
16. Public announcement of SIF, informing potential future applicants2015
17. Press material related to the announcement of funds2015

Note

1.

As promoted in the SALAR guidelines and other material circulated to public authorities.

Appendix

References

Ahrens, T. and Mollona, M. (2007), “Organisational control as cultural practice. A shop floor ethnography of a Sheffield steel mill”, Accounting, Organizations and Society, Vol. 32 Nos 4-5, pp. 305-331.

Alvesson, M. and Sköldberg, K. (1994), Tolkning och reflektion: vetenskapsfilosofi och kvalitativ metod, Studentlitteratur, Lund.

Burchell, S., Clubb, C. and Hopwood, A.G. (1985), “Accounting in its social context: towards a history of value added in the United Kingdom”, Accounting, Organizations and Society, Vol. 10 No. 4, pp. 381-413.

Callon, M. (1986), “Some elements of a sociology of translation: domestication of the scallops and fishermen of St Brieuc Bay”, in Law, J. (Ed.), Power, Action and Belief, Kegan Paul, London.

Callon, M. (1998a), “An essay on framing and overflowing: economic externalities revisited by sociology”, in Callon, M. (Ed.), The Laws of the Market, Blackwell Publishers, Oxford.

Callon, M. (1998b), “Introduction: the embeddedness of economic action”, in Callon, M. (Ed.), The Laws of the Market, Blackwell Publishers, Oxford.

Callon, M. and Law, J. (2005), “On qualculation, agency, and otherness”, Environment and Planning D: Society and Space, Vol. 23 No. 5, pp. 717-733.

Christensen, T. and Lægreid, P. (2008), “The challenge of coordination in central government organizations: the Norwegian case”, Public Organization Review, Vol. 8 No. 2, pp. 97-16.

Christensen, M. and Skærbæk, P. (2007), “Framing and overflowing of public sector accountability innovations: a comparative study of reporting practices”, Accounting, Auditing and Accountability Journal, Vol. 20 No. 1, pp. 101-125.

Christensen, M. and Skærbæk, P. (2010), “Consultancy outputs and the purification of accounting technologies”, Accounting, Organizations and Society, Vol. 35 No. 5, pp. 524-545.

Chua, W.F. (1995), “Experts, networks and inscriptions in the fabrication of accounting images – a story of the representation of 3 public hospitals”, Accounting, Organizations and Society, Vol. 20 Nos 2-3, pp. 111-145.

Cooper, C., Graham, C. and Himick, D. (2016), “Social impact bonds: the securitization of the homeless”, Accounting, Organizations and Society, Vol. 55 No. 1, pp. 63-82.

Cooper, D.J., Ezzamel, M. and Qu, S. (2017), “Popularizing a management accounting idea: the case of the balanced scorecard”, Contemporary Accounting Research, Vol. 34 No. 2, pp. 991-1025.

Covaleski, M.A., Dirsmith, M.W. and Weiss, J.M. (2013), “The social construction, challenge and transformation of a budgetary regime: the endogenization of welfare regulation by institutional entrepreneurs”, Accounting, Organizations and Society, Vol. 38 No. 5, pp. 333-364.

Croft, C., Currie, G. and Lockett, A. (2015), “Broken ‘two way windows’? An exploration of professional hybrids”, Public Administration, Vol. 93 No. 2, pp. 380-394.

Czarniawska, B. (2017), “Bruno Latour and Niklas Luhmann as organization theorists”, European Management Journal, Vol. 35 No. 2, pp. 145-150.

Denis, J.-L., Ferlie, E. and Van Gestel, N. (2015), “Understanding hybridity in public sector organizations”, Public Administration, Vol. 93 No. 2, pp. 273-289.

De Waele, L., Polzer, T., van Witteloostuijn, A. and Berghman, L. (2021), “‘A little bit of everything?’ Conceptualising performance measurement in hybrid public sector organisations through a literature review”, Journal of Public Budgeting, Accounting and Financial Management, Vol. 33 No. 3, pp. 343-363.

Doherty, B., Haugh, H. and Lyon, F. (2014), “Social enterprises as hybrid organizations: a review and research agenda”, International Journal of Management Reviews, Vol. 16 No. 4, pp. 417-436.

Ferry, L. and Slack, R. (2021), “(Counter) accounting for hybrid organizing: a case of the great exhibition of the North”, Accounting, Auditing and Accountability Journal, Vol. ahead-of-print No. ahead-of-print, pp. ahead-of-print.

Fischer, M. and Ferlie, E. (2013), “Resisting hybridisation between modes of clinical risk management: contradiction, contest, and the production of intractable conflict”, Accounting, Organizations and Society, Vol. 38 No. 1, pp. 30-49.

Grossi, G., Kallio, K.-M., Sargiacomo, M. and Skoog, M. (2019), “Accounting, performance management systems and accountability changes in knowledge-intensive public organizations: a literature review and research agenda”, Accounting, Auditing and Accountability Journal, Vol. 33 No. 1, pp. 256-280.

Grossi, G., Reichard, C., Thomasson, A. and Vakkuri, J. (2017), “Theme: performance measurement of hybrid organizations: emerging issues and future research perspectives”, Public Money and Management, Vol. 37 No. 6, pp. 379-386.

Grossi, G. and Thomasson, A. (2015), “Bridging the accountability gap in hybrid organizations: the case of Copenhagen Malmo Port”, International Review of Administrative Sciences, Vol. 81 No. 3, pp. 604-620.

Hedberg, B. and Jönsson, S. (1978), “Designing semi-confusing information systems for organizations in changing environments”, in Otley, D. and Merchant, K. (Eds), Readings in Accounting for Management Control, Springer, Boston, pp. 149-173.

Hultkrantz, L. and Vimefall, E. (2017), “Social investment funds in Sweden: status and design issues”, Scandinavian Journal of Public Administration, Vol. 21 No. 3, pp. 85-108.

Kastberg, G. (2014), “Framing shared services: accounting, control and overflows”, Critical Perspectives on Accounting, Vol. 25 No. 8, pp. 743-763.

Kastberg, G. and Lagström, C. (2019), “Processes of hybridization and de-hybridization: organizing and the task at hand”, Accounting, Auditing and Accountability Journal, Vol. 32 No. 3, pp. 710-725.

Kastberg, G. and Siverbo, S. (2015), “The role of management accounting and control in making professional organizations horizontal”, Accounting, Auditing and Accountability Journal, Vol. 29 No. 3, pp. 428-234.

Kaufman, M. and Covaleski, M.A. (2019), “Budget formality and informality as a tool for organizing and governance amidst divergent institutional logics”, Accounting, Organizations and Society, Vol. 75, pp. 40-58.

Kurunmäki, L. (2004), “A hybrid profession-the acquisition of management accounting expertise by medical professionals”, Accounting, Organizations and Society, Vol. 29 Nos 3-4, pp. 237-250.

Kurunmäki, L. (2008), “Management accounting, economic reasoning and the new public management reforms”, in Chapman, C., Hopwood, A. and Shields, M. (Eds.), Handbooks of Management Accounting Research, Elsevier, Vol. 3, pp. 1371-1383.

Kurunmäki, L. and Miller, P. (2011), “Regulatory hybrids: partnerships, budgeting and modernising government”, Management Accounting Research, Vol. 22 No. 4, pp. 220-241.

Latour, B. (1986), “Visualisation and cognition: thinking with eyes and hands”, in Kucklick, H. (Ed.), Knowlede and Society Studies in the Sociology of Culture Past and Present, Jai Press, pp. 1-40.

Latour, B. (1987), Science in Action: How to Follow Scientists and Engineers through Society, Harvard University Press, Cambridge, Massachusetts.

Latour, B. (1993), We Have Never Been Modern, Harvard University Press, Cambridge, Massachusetts.

Latour, B. (1998), Artefaktens återkomst: ett möte mellan organisationsteori och tingens sociologi, Nerenius and Santérus, Stockholm.

Latour, B. (1999), Pandora's Hope: Essays on the Reality of Science Studies, Harvard University Press, Cambridge, Massachusetts.

Latour, B. (2005), Reassembling the Social. An Introduction to Actor-Network Theory, Oxford University Press, Oxford.

Law, J. (2008), “Actor-network theory and material semiotics”, in Bryan, T. (Ed.), The New Blackwell Companion to Social Theory, Blakwell, Oxford, pp. 141-158.

Law, J. and Singleton, V. (2005), “Object lessons”, Organization, Vol. 12 No. 3, pp. 331-355.

Llewellyn, S. (1999), “Narratives in accounting and management research”, Accounting, Auditing and Accountability Journal, Vol. 12 No. 2, pp. 220-237.

Macintosh, N.B. (2009), “Accounting and the truth of earnings reports: philosophical considerations”, European Accounting Review, Vol. 18 No. 1, pp. 141-175.

MacKenzie, D. (2004), “The big, bad wolf and the rational market: portfolio insurance, the 1987 crash and the performativity of economics”, Economy and Society, Vol. 33, pp. 303-321.

McGivern, G., Currie, G., Ferlie, E., Fitzgerald, L. and Waring, J. (2015), “Hybrid manager-professionals' identity work: the maintenance and hybridization of medical professionalism in managerial contexts”, Public Administration, Vol. 93 No. 2, pp. 412-432.

Miller, P., Kurunmäki, L. and O'Leary, T. (2008), “Accounting, hybrids and the management of risk”, Accounting, Organizations and Society, Vol. 33 No. 7, pp. 942-967.

Neu, D., Rahaman, A.S. and Everett, J. (2010), “Accounting for social purpose alliances: confronting the HIV/AIDS pandemic in Africa”, Contemporary Accounting Research, Vol. 27 No. 4, pp. 1093-1110.

Pache, A.C. and Santos, F. (2013), “Inside the hybrid organization: selective coupling as a response to competing institutional logics”, Academy of Management Journal, Vol. 56 No. 4, pp. 972-1001.

Picard, C.-F. (2016), “The marketization of accountancy”, Critical Perspectives on Accounting, Vol. 34, pp. 79-97.

Pipan, T. and Czarniawska, B. (2010), “How to construct an actor-network: management accounting from idea to practice”, Critical Perspectives on Accounting, Vol. 21 No. 3, pp. 243-251.

Power, M. (2015), “How accounting begins: object formation and the accretion of infrastructure”, Accounting, Organizations and Society, Vol. 47 No. 8, pp. 43-55.

Preston, A.M. (2006), “Enabling, enacting and maintaining action at a distance: an historical case study of the role of accounts in the reduction of the Navajo herds”, Accounting, Organizations and Society, Vol. 31 No. 6, pp. 559-578.

Qu, S.Q. and Cooper, D.J. (2011), “The role of inscriptions in producing a balanced scorecard”, Accounting, Organizations and Society, Vol. 36 No. 6, pp. 344-362.

Rajala, T. and Kokko, P. (2021), “Biased by design – the case of horizontal accountability in a hybrid organization”, Accounting, Auditing and Accountability Journal, Vol. ahead-of-print No. ahead-of-print, pp. ahead-of-print.

Robson, K. (1992), “Accounting numbers as ‘inscription’: action at a distance and the development of accounting”, Accounting, Organizations and Society, Vol. 17 No. 7, pp. 685-708.

Robson, K. and Bottausci, C. (2018), “The sociology of translation and accounting inscriptions: reflections on latour and accounting research”, Critical Perspectives on Accounting, Vol. 54, pp. 60-75.

Steccolini, I. (2019), “Accounting and the post-new public management”, Accounting, Auditing and Accountability Journal, Vol. 32 No. 1, p. 24.

Thambar, P.J., Brown, D.A. and Sivabalan, P. (2019), “Managing systemic uncertainty: the role of industry-level management controls and hybrids”, Accounting, Organizations and Society, Vol. 77 No. 6, pp. 45-87.

Themsen, T.N. and Skærbæk, P. (2018), “The performativity of risk management frameworks and technologies: the translation of uncertainties into pure and impure risks”, Accounting, Organizations and Society, Vol. 67 No. 4, pp. 20-33.

Vakkuri, J., Johanson, J.-E., Feng, N.C. and Giordano, F. (2021), “Governance and accountability in hybrid organizations – past, present and future”, Journal of Public Budgeting, Accounting and Financial Management, Vol. 33 No. 3, pp. 245-260.

van Erp, W., Roozen, F. and Vosselman, E. (2019), “The performativity of a management accounting and control system: exploring the dynamic relational consequences of a design”, Scandinavian Journal of Management, Vol. 35 No. 4, pp. 1-11.

Vosselman, E. (2013), “The ‘performativity thesis’ and its critics: towards a relational ontology of management accounting”, Accounting and Business Research, Vol. 44 No. 2, pp. 181-203.

Yu, L. and Mouritsen, J. (2020), “Accounting, simultaneity and relative completeness: the sales and operations planning forecast and the enactment of the ‘demand chain’”, Accounting, Organizations and Society, Vol. 84, pp. 1-16.

Zawawi, N.H.M. and Hoque, Z. (2020), “Network control and balanced scorecard as inscriptions in purchaser-provider arrangements: insights from a hybrid government agency”, Accounting, Auditing and Accountability Journal, Vol. ahead-of-print No. ahead-of-print, pp. ahead-of-print.

Corresponding author

Gustaf Kastberg Weichselberger can be contacted at: gustaf.kastberg@spa.gu.se

Related articles