Navigating management control change: pathways to the future of work

Abstract

Purpose

COVID-19 has forced Big 4 firms to challenge existing management control arrangements and adapt their ways of working. Yet, we know little about how management control might be enacted in the future of the sustainable workplace. The objective of the study is to examine the patterns of management control change in the Big 4 accounting firms during the COVID-19 pandemic.

Design/methodology/approach

Adopting an exploratory qualitative research design, the authors draw on 42 interviews with directors and associates in the Big 4 professional services firms.

Findings

The findings reveal two pathways of management control change including alignment and displacement. The authors found that relatively minor adaptions to action and result controls were relied upon to respond to substantial cultural and personnel control changes.

Originality/value

The contributions are threefold: the authors take a temporal perspective to (1) unpack the changes to management control arrangements; (2) theorise the findings by developing a three-dimensional taxonomy of change pathways encompassing pace, scope and longevity of management control change and (3) contextualise management control arrangements in a hybrid work setting.

Highlights

1. COVID-19 has forced Big 4 firms to challenge existing management control arrangements.

2. Literature has focused on traditional, onsite work settings and largely ignored change pathways.

3. The authors take a temporal perspective to unpack changes to management control arrangements.

4. Big 4 firms adapted to hybrid work with substantial changes to personnel and cultural controls.

5. The authors theorise the findings by developing a three-dimensional taxonomy of change pathways.

Citation

Carr, M. and Jooss, S. (2023), "Navigating management control change: pathways to the future of work", Accounting, Auditing & Accountability Journal, Vol. 36 No. 9, pp. 390-417. https://doi.org/10.1108/AAAJ-08-2022-6005

1. Introduction

The COVID-19 pandemic has forced firms to challenge existing management control arrangements (Kober and Thambar, 2021, 2022; Passetti et al., 2021) and adapt their ways of working (Battaglia et al., 2021; Leoni et al., 2022). Many firms now have referred to developing novel future of work models, allowing for diverse combinations in how and where employees work (Gratton, 2021; Tippmann et al., 2021). To practice such a hybrid work approach, firms rely on effective management control arrangements which can assist in a swift adjustment to new circumstances by coordinating organisational activities and resources and by supporting rapid decision-making (Passetti et al., 2021). Despite embracing the need for novel working models, a recent McKinsey survey shows that 68% of firms do not yet have a detailed vision of how these new ways of working might be operationalised in practice (Alexander et al., 2021). Given the ambiguity surrounding evolving working models and the discourse around the future of the sustainable workplace, exploring how firms exercise their management controls arrangements, i.e. “influence employees” behaviours in desirable ways' to achieve their objectives (Merchant and Van der Stede, 2012, p. 6) in a COVID-19 context is of timely relevance. By management control arrangements, we refer to the processes, practices and techniques that prescribe appropriate organisational behaviour and that shape and enforce patterns of decision-making to ensure that organisational objectives are achieved (Malmi and Brown, 2008).

Prior research examining a firm's management control arrangements has been conducted in a traditional, onsite organisational setting (Bedford, 2020), with scant research considering virtual teams (e.g. Bisbe and Sivabalan, 2017) and remote working (e.g. Delfino and van der Kolk, 2021). For example research has considered support mechanisms for employees (Suff et al., 2020), the increased use of technology to monitor the workforce (Delfino and van der Kolk, 2021) and its potential dehumanising impact (Safari et al., 2022). However, research on management control changes, particularly in a hybrid setting, is embryonic. As such, hybrid work presents a novel work context and the applicability of existing management control arrangements remains to be seen (Delfino and van der Kolk, 2021; Parker, 2020).

To address this gap, we build on and expand developments in the management control domain that call for research to adopt a temporal perspective (Bedford, 2020; Ferreira and Otley, 2009) which investigates how changing organisational contexts can lead to the creation, transformation, or extinguishment of management control arrangements (Friis et al., 2015; Granlund, 2001; Malmi, 1997; Malmi and Brown, 2008; Toldbod and van der Kolk, 2022). Adopting the object-of-control framework [1] (Merchant and Van der Stede, 2012), the objective of our study is to examine the patterns of management control change in the Big 4 during the COVID-19 pandemic. The object-of-control framework has informed previous literature providing valuable empirical and theoretical insights into the collective nature of firm-wide management control arrangements (e.g. Passetti et al., 2021; Pfister and Lukka, 2019; Van der Kolk et al., 2019). In the control framework, four types of controls are distinguished depending on whether control is exercised over action, results, personnel, or culture. Utilising this framework permits a holistic perspective of the management control arrangements mobilised by the Big 4 firms during the COVID-19 pandemic to be captured, while simultaneously providing sufficient rigidity to act as a medium of analysis.

To investigate the patterns of change, in particular, we draw on insights from institutional logics (Thornton et al., 2012) and take a change pathways lens (Micelotta et al., 2017), considering the patterns of management control change in terms of pace, scope and longevity in the Big 4 accounting firms (Deloitte, EY, KPMG, PwC). Our qualitative study, encompassing 42 interviews with HR directors, audit and tax directors and associates, unpacks changes made towards existing management control arrangements, and in doing so compares pre- and midst-pandemic approaches and considers implications for the future of work. In so doing, we seek to investigate how the changed organisational context (i.e. the shift to remote and hybrid working models) influenced not only the adaption of management control arrangements but also how the changed working model may affect the evolution of management control arrangements.

Our study provides three primary contributions: First, we add to the management control arrangement literature (e.g. Bedford, 2020; Malmi, 2013; Malmi and Brown, 2008; Merchant and Otley, 2020; Van der Kolk et al., 2020), by unpacking the changes to management control arrangements in light of COVID-19. We find firms being forced to reimagining their personnel and cultural controls to reflect the new working context. To facilitate these substantial changes, some alterations were also made to action and result controls. This involved increasing reliance on existing action controls and adaptations to result controls which focused on more informal and frequent performance evaluation elements. Doing so, we provide insights into how management control elements were changed as a consequence of an external event (i.e. trigger) (Bedford et al., 2022) and how management control elements interact (Bedford, 2020; Ferreira and Otley, 2009; Toldbod and van der Kolk, 2022). We complement existing research by examining how different management control elements shaped formulating a short- and longer-term response to this exogenous shock (Bedford et al., 2022).

Second, we contribute to the management control change literature (e.g. Granlund, 2001; Friis et al., 2015; Malmi, 1997; Mouritsen et al., 2022; Pfister and Lukka, 2019; Toldbod and van der Kolk, 2022), considering the patterns of management control change. In doing so, we theorise our findings and develop a three-dimensional taxonomy of change pathways encompassing pace, scope and longevity of management control change and we map management control elements to this taxonomy. We present novel insights by indicating variance in pace and scope within management control elements and by expanding Micelotta et al.’s (2017) typology with a longevity dimension.

Third, at a practical level, we contribute to the accounting literature by contextualising management control arrangements in a hybrid work context in professional services firms. In doing so, we provide rare insights into management control in the Big 4 firms. We reveal a range of external and internal pressures which influence the longevity of management control change and the sustainability of hybrid working which the Big 4 consider their future of work model. Specifically, we identify two potentially competing perspectives; (1) a need for Big 4 firms to ensure business continuity and maintain professional standards – i.e. a client perspective and (2) a need to re-establish connectivity, enhance learning experiences, balance work-life commitments and respond to employees request for future of work models in the context of competing for talent, – i.e. an employee perspective. In so doing, we provide initial insights into the potential role that management control arrangements can play in the future of work, while also contributing to the recent literature examining the use made of management controls to navigate the pandemic (Delfino and van der Kolk, 2021; Huber et al., 2021; Kober and Thambar, 2021).

Our study proceeds as follows. First, we review the literature on management control arrangements, management control change, and our theoretical approach for analysing the patterns of change, i.e. a change pathways lens. Second, we outline our research approach, detailing the research context, data collection and research design. Third, we reveal two identified pathways of change in our findings section. Finally, in our discussion and conclusion section, we unpack our three primary contributions and discuss practical implications, limitations and future research suggestions.

2. Theory development

2.1 Management control arrangements

Research on management control has traditionally focused on single management control elements (Otley, 2016). However, more recently, scholars have increasingly emphasised the need to adopt a more holistic perspective by investigating management control arrangements, i.e. a combination of controls (Grabner and Moers, 2013; Malmi, 2013; Malmi and Brown, 2008; Sandelin, 2008; Van der Kolk et al., 2020). By management control arrangements, we refer to the processes, practices and techniques that prescribe appropriate organisational behaviour and that shape and enforce patterns of decision-making to ensure that organisational objectives are achieved (Malmi and Brown, 2008). The extant work on management control arrangements can broadly be categorised into three streams including the drivers behind the choice of management control arrangements (e.g. Davila, 2000; Gond et al., 2012; Malmi et al., 2020), the process of combining various management controls (e.g. Mouritsen et al., 2022; Pfister and Lukka, 2019; Toldbod and van der Kolk, 2022) and the performance implications of various management control arrangements (e.g. Bedford, 2015; Bedford et al., 2016; Osma et al., 2022).

Efforts to understand how firms typically utilise management control arrangements have advanced our understanding of what factors influence the combination of management controls chosen (Malmi et al., 2020), how firms formulate different management control combinations (Pfister and Lukka, 2019) and which control arrangements are effective (Bedford et al., 2016). This expanding body of research demonstrates that firms design and use a broad range of management control elements and that focusing on management control combinations as a system of interdependent tools and procedures (Grabner and Moers, 2013), a package of different tools and procedures (Malmi and Brown, 2008) or a gradual accumulation of practices that become more or less tightly coupled (Demartini and Otley, 2020), is an important area of research. In our current study, we draw on Merchant and Van der Stede's (2012) object-of-control framework to investigate the operationalisation of management control arrangements in the Big 4 accounting firms. This framework allows us to take a holistic perspective on management control, capturing the richness of control arrangements, while at the same time providing sufficient rigidity with the specific control objects of action, results, personnel and culture.

2.2 Management control arrangements and change

Traditionally, research on management control change often focuses on the changes of a specific management control element, such as the budget process (Bedford et al., 2022; Collier, 2001), the performance management system (Busco et al., 2007; Jansen, 2011; Kasurinen, 2002), or the cost accounting system (Granlund, 2001; Malmi, 1997). Recent research, however, points to the interplay among management control elements, highlighting that changes to one management control element may influence the presence of other management control elements (e.g. Toldbod and van der Kolk, 2022). Studies have begun examining how different control elements are combined in response to a particular control problem. For example, Pfister and Lukka (2019) examine how a technology firm combines stretch efficiency targets with cultural controls and personnel controls to allow managers to experience autonomous motivation and to reduce the risk of a crowding out of creativity. Similarly, Mouritsen et al. (2022) demonstrate when and why managers connect performance and risk controls in response to a supply chain issue. Adopting a process perspective, Toldbod and van der Kolk (2022) show how a change in one control practice can have a cascading effect on other controls which are subsequently amended. They examine the issue of coherence among management control elements, where coherence implies that individual controls compatibly work together and motivate employees to act in line with organisational goals.

Collectively, these single case study analyses (e.g. Pfister and Lukka, 2019; Mouritsen et al., 2022; Toldbod and van der Kolk, 2022) provide qualitative in-depth insights into how individual firms choose to combine different management control elements to respond to an internal or external shift in their organisational environment. Notably, these prior studies often only hint at, or overlook, the patterns of change in terms of the possibility that management control combination changes may be transient. Thus, the potential for management control arrangement changes to be transitory has been under researched (Bedford, 2020; Hall, 2016; Malmi, 1997; Toldbod and van der Kolk, 2022), as has been the possibility that even seemingly major variations may not accumulate to generate long-term management control changes or have lasting management control implications (Merchant and Otley, 2020). Hence, a fundamental challenge for management control studies is to theorise empirical findings and further advance our understanding of the potential for long-term variations associated with the management control combinations adopted by firms in response to a dynamic change in their internal or external organisational context. This approach matches well with the evolutionary perspective put forward by Martin (2020) and the process perspective adopted by Toldbod and van der Kolk (2022). However, we argue that management control combination changes also need to be examined from a temporal perspective (Bedford, 2020), exploring patterns of change in the different management control elements used by firms. Doing so, we suggest, can help move away from static approaches, to provide dynamic explanations about how firms choose to change their management control arrangements (Bedford, 2020; Ferreira and Otley, 2009; Malmi and Brown, 2008). To analyse these patterns of change, we introduce a change pathways lens (Micelotta et al., 2017) in the following section.

2.3 Management control outcomes, institutional logics and a pathways lens

Potential outcomes of management control change relate to the creation, transformation, or extinction of policies, structures, practices, or processes which seek to maintain organisational control (Malmi and Brown, 2008). Studies focusing on outcomes of management control change have taken three main theoretical perspectives. First, research taking an ecology perspective (e.g. Toldbod and van der Kolk, 2022) argues that firms may be unable to adapt sufficiently quickly to the modified environmental conditions, and those environmental changes trigger a cascading effect, which can involve the selection of new control elements and/or the disappearance of existing management control elements. In their empirical material, Toldbod and van der Kolk (2022) observe that where the initial management control change is structurally opaque (i.e. where a lack of knowledge or scarce resources exists), the cascade process is likely to be sequential and protracted. Conversely, they suggest that where management control change is not structurally opaque, the cascade process is more likely to be analogous and relatively swift. Second, management control change research has conceptualised firms as being capable of adaption and explores successful responses to disruptive change (e.g. Carr and Beck, 2022; Evans and Tucker, 2015; Granlund, 2001; Kober and Thambar, 2021; Malmi, 1997; Pfister and Lukka, 2019). For instance, Pfister and Lukka (2019) argue that a change in the interrelation of personnel and cultural controls with result controls supported internalisation of the latter by employees. Third, research focuses on how changes in management control arrangements can be underpinned and held together by different logics (e.g. Friis et al., 2015; Gerdin, 2020; Schäffer, Strauss and Secher, 2015). This stream of management control research highlights the need to understand the process of change and look at the potential triggers and outcomes. In other words, insights need to focus on understanding why management control arrangements are changed, the patterns of change, and how management control arrangements are interrelated (e.g. Almqvist and Skoog, 2006; Bedford, 2020; Ferreira and Otley, 2009; Toldbod and van der Kolk, 2022). In our study, we build on this third research stream, treating COVID-19 as the central exogenous trigger that forced Big 4 firms to reconsider their management control arrangements. Emphasis is placed on exploring how this identified trigger provides an impetus for change, which in our context refers to the adaption of management control arrangements.

To examine these patterns of management control change, we draw on theoretical insights from the institutional logics perspective (Thornton et al., 2012). Organisations are driven by a set of norms and beliefs, i.e. institutional logics, which shape behaviours within a particular institutional environment (Scott, 2015). Early empirical research on institutional logics focuses on temporal shifts in prevailing logics (e.g. Lounsbury, 2002; Thornton, 2004; Thornton and Ocasio, 1999). These foundational studies demonstrate the efficacy of adopting a logics perspective as a means of explanation and as a generative theoretical perspective for examining how the prevalence of changing logics gives rise to variations in behaviours. Building on these studies, Thornton et al. (2012) propose three dynamics that can drive transformational change: replacement of one dominant logic by another; blending or hybridisation of multiple salient logics; and efforts to segregate logics by time and space. In addition, they highlighted four developmental forms of institutional change: assimilation, which involves the combination of external elements into existing logics; elaboration, which refers to endogenous enhancement; expansion, by which shifts in jurisdiction drive adaptation of practices or ideas; and contraction, as decrease in a logic's scope drives similar adaptation (Thornton et al., 2012). Ultimately, the vast logic's literature highlights institutional complexity and provides a variety of mechanisms by which logic change occurs (Scott, 2015; Wooten, 2012). In reviewing the extant literature on institutional change, Micelotta et al. (2017) show how our knowledge of certain forms of institutional change remains inadequate, highlighting the need to better understand the heterogeneity of institutional change processes.

For the purpose of this paper, we utilise a typology of change pathways developed by Micelotta et al. (2017). This two-by-two typology is based on two core dimensions: the pace of change and the scope of change. The first dimension, pace of change, differentiates between “revolutionary” and “evolutionary” changes. This dimension allows simultaneous consideration to different temporalities and to capture the complexity of organisational contexts (Greenwood and Hinings, 1996). Revolutionary change, i.e. a high pace of change, is triggered by strong macro-level dynamics that disturb and challenge a firm's stability. From this perspective, when change occurs, it does so in a relatively fast manner, driven by sudden external shocks or the determined efforts of change agents (Micelotta et al., 2017). Conversely, evolutionary change, i.e. a low pace of change, is characteristically slower, triggered by less hurried forces such as slow societal changes or the deliberate introduction of humble innovations (Smets et al., 2012).

The second dimension, scope of change, differentiates between “transformational” and “developmental” changes. This dimension captures the degree of changes in organisational logics (Micelotta et al., 2017). Transformational change, i.e. high scope of change, takes places when shared understandings, which define what is accepted and valued in the field, are overturned or significantly altered, for example, through replacement, blending, or segregation of organisational logics. Developmental change, i.e. low scope of change, occurs when alterations are relatively narrow and involves stretching rather than discarding control arrangements, for example, through reduction, integration, or expansion of organisational logics (Micelotta et al., 2017). Focusing on organisational logics, this dimension provides a higher-order framework by which to assess whether changes at lower levels of analysis (e.g. changes in management control arrangements) are transformational or developmental because when the logics governing a field change, so, too, do the values and associated beliefs that underpin the dominance of forms of controlling and its associated arrangements (Scott, 2015). Combining these two central dimensions, Micelotta et al. (2017) identify four pathways for unpacking change: (a) alignment, (b) accretion, (c) accommodation and (d) displacement.

Adopting this typology, we contend that alignment occurs where management control changes are evolutionary in pace and developmental in scope; accretion relates to management control changes that are evolutionary in pace and transformational in scope; accommodation refers to management control changes that are revolutionary in pace and developmental in scope; and displacement occurs where management control changes are revolutionary in pace and transformational in scope. These four change pathways allow us to take a nuanced approach to the study of management control, considering variance in pace and scope of changes. Our conceptual framework is outlined in Figure 1. The framework depicts (1) COVID-19 as a trigger which led to exogenous change in the organisational environment; (2) changes to existing management control arrangements including four management control elements and (3) management control outcomes, as four potential pathways of change. While we contextualise our research by highlighting a range of internal and external pressures, the focus of our study, aligned with our research objective, lies on examining the patterns of change (i.e. change pathways).

3. Research approach

3.1 Research context

Our qualitative research enabled us to explore the patterns of management control changes triggered by COVID-19 and the implications on future of work models. To do so, the paper draws its findings from the Big 4 accounting firms in Ireland. We considered this setting to be appealing for a number of reasons. First, Big 4 firms had to reimagine their traditional service delivery models as a result of the COVID-19 crisis, including formulating new ways to share and disseminate knowledge and providing inputs to problems which are key aspects for meeting client needs and for maintaining management control (Delfino and van der Kolk, 2021). Second, given that these Big 4 firms typically market their normative research to provide consultancy services, we anticipated a greater degree of sophistication in terms of management control arrangements (Alvesson and Kärreman, 2004). Finally, Ireland's restrictions in 2021 were classified by Blavatnik School of Government at Oxford University as the second toughest restrictions in the European Union and the eight most stringent restrictions globally and all firms were asked to work from home unless in-person attendance was essential (The Times, 2021).

Prior to the onset of the pandemic, Big 4 firms operated a fully onsite work model. When COVID-19 hit in Ireland in March 2020, each Big 4 firm was forced to move to a fully remote model. Akin to other sectors of the Irish economy, the Big 4 firms experienced a short period of hybrid working from July to August 2021, before being required to return to a fully remote context in September 2021. During this transition period, the Big 4 firms were actively considering their future of work model, i.e. hybrid work, combining onsite and remote work. We conducted our interviews between September and December 2021 to understand the industry's experiences to date and the initial learnings for hybrid working and management control arrangements. To negotiate access, the Managing Partner in each Big 4 firm was approached via the University's professional network. These Managing Partners acted as gatekeepers in our study, and, subsequently, helped to designate the interviewee sample.

3.2 Data collection

Our study included HR directors, audit and tax directors and associates. In total, 42 interviews were conducted, including 4 interviews with HR directors, 5 interviews with directors and 33 interviews with associates. The sample in each Big 4 firm ranged from 9 to 11 interviewees with a balance in gender (52% female; 48% male). Table 1 provides an overview of all interviewees. These qualitative insights were supplemented with secondary information provided by each firm which we utilised to understand remote working practices and policies and evolving hybrid work approaches. For confidentiality reasons, identifiers are used for all interviewees. For example, HR¹C^A – refers to HR Director 1 in company A; D¹C^B refers to Associate Director 1 in company B; and A³C^C refers to Associate 3 in company C.

Semi-structured interviews were conducted around three key elements. The first set of questions sought to gain an understanding about each firm's pre-COVID working model, and the initial control changes which had occurred during the sudden move to remote working. Interviewees were asked to speak about their firm's approach to management control prior to the pandemic. Interviewees were also asked about when COVID-19 hit, what (if any) management control changes occurred and what practices were applied over the previous 18 months. The second set of questions related to hybrid work and the emerging management control arrangements. Here, interviewees were asked about how they envisioned hybrid working to impact on standard operating procedures and monitoring systems. Interviewees were also asked to describe types of outcomes employees were held responsible for. Additionally, interviewees were asked to comment on recruitment, training and job design and to speak about the role of company values and norms and how these were being communicated during the transition to hybrid working. To conclude, interviewees were asked to reflect on the nature of management control change and what factors would impact the sustainability of hybrid working. Thus, interviewees responded intuitively about the major and minor changes to each of the four management control elements. It was subsequently in the data analysis stage that we developed and applied the notion of patterns of change including the three dimensions of pace, scope and longevity. The same interview guide was used with all interviewees to guide the conversation but was employed responsively to allow novel issues to emerge and illustrations to be provided. For example, HR, audit and tax directors, by nature of their role, provided more insights on the managerial perspectives on the management control changes made while associates provided more examples that related to their own experience of management control changes. On average, interviews lasted 40 min. All interviews were conducted via MS Teams while interviewees were working remotely and transcribed verbatim.

3.3 Research design

We adopted an abductive reasoning approach with a view to building novel theoretical insights (Grodal et al., 2021). This approach involved iterating back and forth between empirical data and pre-existing theoretical constructs (Bamberger, 2018). Allowing for insights to emerge from the data (Sætre and Van de Ven, 2021), we approached the data collection as an open-ended discovery process. We started with broad theoretical sensitivity from the literature on management control change and institutional logics. Next, we formulated a set of inter-connected second-order categories based upon the four management control elements (a) action controls, (b) result controls, (c) personnel controls and (d) cultural controls (Merchant and Van der Stede, 2012). This categorisation provided us with “structural signals” to guide our exploratory analysis. As we further engaged with the transcripts, we developed a set of first-order codes for each second-order category. First-order codes described our data using the language of the interviewees. All first-order codes were developed in an iterative way which involved merging, collapsing and dropping codes (Grodal et al., 2021). Appendix 1 presents the coding structure and evidence of our data. To ensure inter-coder reliability, both authors coded the transcripts to ensure alignment in the coding process. In addition, the authors reviewed the codebook, reflecting on and updating developed codes. While we anticipated firm-level discrepancies around management control changes, our analysis revealed strong homogeneity among the Big 4 firms, and we therefore present our data at an aggregated sectoral level. Importantly, as we engaged in theorising our data in terms of patterns of change, we identified variances among management control elements. To explore the nuances in the patterns of change, we drew on insights from institutional logics (Thornton et al., 2012). Specifically, we adopted Micelotta et al.’s (2017) two-by-two typology considering pace and scope of change, as a complementary theoretical lens. This led us to categorising management control change pathways as: (a) alignment, (b) accretion, (c) accommodation and (d) displacement. Finally, we mapped our findings to a change pathways taxonomy (Figure 2), plotting each management control element and presenting longevity as an important third dimension when examining the patterns of management control change. Changes to action controls are presented as low in pace, scope and longevity; changes to result controls are illustrated as low in pace and scope but high in longevity; changes to personnel controls are reported as high in pace and scope but low in longevity; and changes to cultural controls are described as high in pace, scope and longevity.

4. Findings

When the pandemic hit, Big 4 firms moved from an office-based work model to remote work with a short period of hybrid working in between, and, at the time the interviews were conducted, actively considered their future of work model. This evolving hybrid work approach encompassed a combination of onsite and remote work across the Big 4 firms. These new work settings required the firms to reconsider existing management control arrangements. In this section, we describe the changes made to management control arrangements. In doing so, we compare pre- and midst-pandemic arrangements, observing that relatively minor adaptations to action and result controls were relied upon to facilitate the substantial personnel and cultural control changes. Appendix 1 presents evidence of our data on management control changes which we elaborate on in this section. We identified two pathways of change: alignment of action and result controls (section 4.1) and displacement of personnel and cultural controls (section 4.2).

4.1 Alignment of action and result controls during the pandemic

Pre-pandemic, several action controls existed in an onsite setting including a set of procedures, tools and structures. During COVID, firms continued to rely on (1) standard operating procedures and tools and (2) a well-established management hierarchical structure. In addition, some amendments to action controls included (3) mirroring of traditional audit rooms in a virtual environment and (4) setting up daily check-ins with employees. The patterns of change can be characterised as low in pace and scope. The changes to action controls were largely expected to be temporal alternations (low longevity). While mirroring audit rooms was deemed necessary to operate fully flexible, the introduction of daily checks-ins was expected to be a temporal change.

First, during the pandemic and the move to remote and hybrid work, core procedures such as audit review and approval processes remained in place across the Big 4 firms. While there were limited changes to these controls, they were critical and were increasingly relied upon to deliver work in a remote and hybrid setting, allowing firms to control work despite not being in an onsite environment.

We made sure people were doing their jobs through our normal management tier structure. Managers would report to me, audit seniors report to the managers, audit juniors report to the audit senior. Within that structure, we would have our audits for the year planned in terms of who was doing what at any point in time (D¹F^D).

There is a review process. If you are an associate, documents are reviewed by the senior above you, then by the manager, then by the partner, and then upwards and upwards. They would leave review notes, and you go to the client, or if you can update it straight away, brilliant, and then it goes to the audit partner, and they review that and if they have questions again, the same process, and everything is reviewed before the sign off is given (A⁷F^A).

Second, the purpose of the virtual audit room was to replicate that in-person experience, so people were not sitting at home by themselves (HR¹F^A). This meant that oversight was maintained, and any queries could be dealt with immediately. Similarly, daily virtual check-in meetings were introduced to ensure that there was awareness and accountability of tasks, however, they were already discontinued a few months after they were introduced as they just took too much time (D²F^C).

Every manager was encouraged to have daily stand-up meetings with their associates on their jobs. So that was 15 minutes every day where you set out what are you doing today. Nobody completely fell off the radar and that was to do with the stand-up calls daily. There was accountability going on the whole time (D¹F^A).

You would have a lot more interactions with managers. I would have had a lot more calls and a lot more catchups. I would have set up an MS Teams call once a day with the whole team. The manager would not have been there before COVID (A⁷F^D).

We had all different check-in mechanisms in place. It ended up being too much. People ended up with their diaries being chock-a-block with meetings, so that sort of fell off after six to nine months. There was a realisation that we had gone too far on this and that we were forcing this connection. We had to pull back (HR²F^B).

Pre-pandemic, result controls existed at individual and organisational levels, and were accompanied with a performance appraisal system and process. In an onsite pre-COVID setting, organisational KPIs encompassed revenue, probability, risk, quality metrics and utilisation rates. At an individual level, employees were assessed based on the quality of their work papers while the performance appraisal system and process consisted of biannual or triannual performance reviews and wider feedback on projects. During the pandemic and the move to remote and hybrid work, (1) organisational-level KPIs and (2) individual-level metrics remained unchanged, while some minor amendments were made to (3) performance systems and processes. The patterns of change can be characterised as low in pace and scope. The changes to result controls were expected to be perpetual alternations (high longevity).

First, although the key performance metrics were identified as having remained stable during the various transition points, the practices around evaluating individuals were thought to have been altered. For example, a more holistic perspective, beyond hard metrics, was adopted.

The tools themselves did not change. What did change or what we were looking out for was what we were seeing in terms of activity amongst clients, what were the revenues telling us, what were the debtors days telling us? What were the utilisations saying in terms of how busy people were? (D¹F^D).

One of our lessons with COVID has been that human interaction, sitting back, and that fireside shot rather than just the cold metric of met deadline, yes/no, met target, yes/no. We have realised that social interaction will help to explain targets being met or not met (HR¹F^D).

Second, it was perceived that, pre-pandemic, firms had been discussing the need for the appraisal process to become less formal and more flexible, but that progress had been slow. The pandemic was perceived to have accelerated these developments. Directors were also trained and encouraged to be more cognisant and to formally recognise and reward individuals who had made specific efforts to foster collaboration and connectivity, and to have more open and meaningful conservations (HR¹F^C) with individuals about their performance. Therefore, during the pandemic, we observed adaptations to result controls which focused on more informal and frequent evaluations.

Before COVID, it would be very much, you are doing well, or you are not doing well. Whereas now, if somebody is not going too great, we are more into chatting about the person now and seeing what is going on with them, which I think is good because it needed to happen. There is still the work element of it, but there is more chat about what is going on behind the scenes and what is going on with the person (D¹F^C).

The performance evaluation idea or system, that process has not changed. But behind the scenes it is has become about finding out how someone is getting on or how they are doing. So, there is that bit more effort being placed on getting the whole picture about someone and their performance (HR¹F^B).

I think what the firm changed is the smaller thank you rewards, just saying a quick thank you. There are theses lower amounts that you can now award to people. Just to say thank you for the job they did (D¹F^B).

While these changes were perceived favourably, the realities of remote working remained difficult to consolidate into an appraisal process as indicated in the below quote. Nevertheless, most interviewees perceived the identified amendments to result controls positively, particularly the expansion of real-time, informal feedback, combined with empowerment through revised rewards practices. It was believed that these amendments would continue in the hybrid work approach.

If you were in the office, they would see you work until seven or eight, whereas now they will not see you working. They do not physically see the tiredness when you are working late. When you are in the office, they would see, he is wrecked whereas now it is just, he is at home, I do not know if he is tired or not (A⁷F^A).

Our research reveals that action and result controls exhibited limited changes in both remote and hybrid work environments compared to the pre-pandemic onsite setup. This can be attributed to the persistence of significant standard operating procedures and a rigid management hierarchy during the transition to new work arrangements in the Big 4.

Moreover, we observe a growing significance placed on human interaction and communication within both action controls, evident through heightened monitoring and increased check-ins, as well as within result controls, reflected in the adoption of more informal feedback mechanisms. Specifically, action controls were reinforced through frequent check-ins and heightened managerial supervision in remote work scenarios. Meanwhile, result controls reflected a more informal approach, incorporating real-time feedback and empowering reward systems. Despite these relatively minor adjustments, the Big 4 emphasised the significance of action and result controls as vital mechanisms for navigating the evolving work landscape. In the next subsection, we demonstrate how these relatively minor adaptions to action and result controls were relied upon to respond to substantial cultural and personnel control changes.

4.2 Displacement of personnel and cultural controls during the pandemic

Pre-pandemic, personnel controls existed in terms of recruitment and selection practices, training and job design. When the pandemic forced firms to move to a remote and hybrid work setting, these controls were significantly altered and the pattern of change can be characterised as high in pace and scope. Major changes related to (1) onboarding and interview formats, (2) re-assessment of learning and (3) the setup of onsite and virtual workspaces. While some of these changes were expected to be lasting (e.g. workspace design), the majority of changes to personnel controls were perceived as being temporal alternations (low longevity).

First, alternations to recruitment and selection practices were demanded, with efforts directed to developing novel virtual practices. While acknowledging the necessity to move on-line, a return to onsite practices was expected. Weaknesses associated with onboarding new recruits and conducting exist interviews were suggested to be particularly problematic.

Onboarding of new people is exceedingly difficult. It is such a critical stage to bring somebody in and really connect them. It is the individual learning. It is the building of relationships. And it is the informal stuff. We can have a lovely onboarding pack and all the lovely stuff. At the end of the day, the person is sitting at home in their kitchen and when they are stuck with something they must go, oh no, who do I ask? (HR²F^B).

Exit interviews. Having that connection and meeting with people to understand and share their experience and have an honest conversation. It is important that we connect and still have that. We were trying to be very deliberate on what are those times in a hybrid working model (HR¹F^A).

Second, pre-COVID, all firms placed a strong emphasis on learning by working closely with colleagues in a face-to face setting where project management, the application of accounting techniques, conflict resolution and networking skills were all developed via direct observation. The sudden removal of this face-to-face personal control posed a significant challenge to each of the firms, who were required to fundamentally reconsider their training and learning practices. While there was an attempt to reconcile the removal of these personal controls through newly developed action controls, it was evident that the amendments to personnel controls had negatively impacted control efforts. This view was shared across the firms.

Learning is a big gap. It is that osmosis, sitting in the office, hearing a conversation that you may or may not get involved in. But you hear about how somebody is dealing with the problem. So that stuff is missing 100 per cent. That is something that I do not think we have figured out how to compensate for yet, that informal learning. We can put in all the training courses and do all that good stuff, but it is the osmosis, it is stuff you hear, or it is where you hear a conversation and go, oh, that is interesting (HR²F^B).

We had an open floor, there were always people asking questions, and someone had always come across an issue before, so you get your work done a lot faster whereas at home you tend to spend longer at the work yourself before going to Teams to ask a question. Or you might only be asking your friends, so you might not get the full benefit of all the other associates (A³F^C).

Third, the need to redesign workspaces in terms of the practicalities of onsite and remote working was referenced by several participants, who considered firms had been responsive. This included the setup of virtual workspaces and ergonomic assessments as well as reimagining the use of the office. Both were deemed as important aspects of the future of work model and were perceived as lasting changes.

There have been a lot of changes. In the beginning, I did not like working from home. I wanted to get back into the office. Now I am used to it. I am set up. The firm did the ergonomic assessment for everyone, and they sent tables, chairs, screens, and laptop stands. That helped a lot to set up an office for yourself at home (D¹F^B).

We have reimagined all our office spaces, we have taken away the assigned desk for everyone. We have moved more into neighbourhoods. We see our office as a place where they come together to collaborate and collide and do something specific. We will be making our offices more welcoming and more places of collaboration (HR¹F^A).

Pre-pandemic, cultural controls focused on group norms which were established based on onsite work with a high degree of visibility. This was reinforced by a tone from the top which gave recognition to being present and role modelled by senior management. COVID triggered a substantial change including (1) the tone from the top and (2) the establishment of new norms. The patterns of change can be characterised as high in pace and scope. The changes to cultural controls were expected to be perpetual alternations (high longevity).

First, the tone from the top changed substantially during the pandemic. While flexible working had been explored, little progress had been made in its implementation pre-COVID. The notion of a substantial proportion of staff working remotely or in a hybrid fashion was described as unimaginable (HR¹F^C) and a hard sell before COVID hit (D¹F^D). The focus continued to be on being a high performing professional firm but with great flexibility (D²F^C) and the Big 4 firms realised that a change in ways of working was necessary to ensure their culture was fit for purpose and future proof (HR¹F^C). It was evident from our findings that this new way of working had been embedded, particularly by senior management who were found to have championed this change which highlighted the importance of tone from the top as a form of control in this context. In terms of employees adapting the new ways of work, role modelling of this flexible approach was perceived as critical to ensuring a long-term cultural shift.

People have now begun to take flexibility if they wanted to do something; it did not matter if it was something they had to do or if they wanted this afternoon off. For example, if they were meeting up with a friend, or they were taking a longer weekend. It does not have to be a medical appointment. Which is what flexibility would have been used for in the past (A⁴F^D).

Our Head of Audit is really focused on making this work and he has been clear with the leadership group. Because there are mixed opinions, but he has been clear with them ensuring that tone is sent from the top and he expects them to be role modelling it and he is role modelling it himself without question (HR¹F^A).

Second, the establishment of new norms associated with hybrid working were highlighted. This included practical considerations such as unconscious bias, ambiguity around work time and place expectations and even dress code. The issue of proximity bias was referred to, but as of yet has not been materialised. Participants stated that this issue would particularly arise when promotional decisions were being met in a hybrid context.

The real task is how do we avoid proximity bias. Because whether it be assigning work, whether it be giving promotions. We are going to be biased consciously or unconsciously, and how do we put in the protective measurements needed as an organisation or as a world to make sure that is reduced (HR¹F^C).

The old thing, if someone is in the office until 9 pm, they are working harder than everyone else, there will be a bit of that. Some people might view someone who is coming into the office more often as working harder which is not the case. Also from the associates' perspective, if they see a manager in the office more, are they going to think that he/she is more involved, or view him/her differently to someone who is at home? (D¹F^A).

A related concern in terms of new norms emerged around the ambiguity of the ways of working as the hybrid model continued to evolve. Participants shared mixed views on the degree of structure and formality in terms of the number of days per week in the office versus at home, the working hours on a given day and the dress code expectations. While some participants were hoping for maximum flexibility, others preferred some guiding structure.

You are an adult, so we are not going to monitor any person. We are not going to give someone a tap on the shoulder and say, where were you on Tuesday? We are not asking people to say on their time sheet, I was in the office or at home. It is not envisaged that we would have a slide deck at the end of the year showing who was in the office. We are monitoring the numbers holistically (D¹F^D).

I do not mind being told to come in Mondays, Tuesdays. Being told that you are in on these days, I could plan my October. If they left it up to me, I might not bother going in there. It is too easy to stay at home. There was leniency if you could not make it, but they gave us a bit of direction. There was no forced pressure (A²F^C).

They are more flexible, they are open to having more flexible work hours, and they are also open to ‘dress for your day.’ If you know you are going in the office for a meeting with the client or with the partner, it is ‘dress appropriately.’ If you know you are just working away yourself, you will be free to come in wearing jeans and a jumper. It is not as formal as it was pre-COVID, which is a good thing (A⁷F^C).

In sum, we observed that the transition to remote and hybrid work necessitated a reimagining of personnel controls and cultural controls. In this regard, simply strengthening or weakening controls, or making them more formal or informal, was inadequate. Instead, the Big 4 firms had to make substantial modifications to their personnel and cultural controls. This encompassed changes in leadership messaging, the establishment of new norms, workspace redesign and the implementation of innovative learning approaches, among other adaptations. Specifically, our findings indicate that human interaction and communication were recognised as crucial elements in managing and enhancing organisational learning. While changes to cultural controls were expected to endure, most of the alterations to personnel controls were regarded as temporary adjustments with limited longevity.

4.3 Contextual pressures influencing longevity of management control change

Participants referred to a set of contextual pressures that have influenced considerations around management control changes during the pandemic (see Appendix 2). Overall, the Big 4 firms viewed hybrid working as the future of work model and were committed to its roll out. For example, it was emphasised, we are not going back (HR¹F^A) and that not proceeding with hybrid work would be a missed opportunity (HR²F^B). At the same time, participants acknowledged that adjustments to management control arrangements might be needed as the model evolves further. Participants highlighted that firms are in a learning phase which contains a lot of trial and error; it is a big experiment for now (D²F^C). Most associates viewed hybrid working as the future of work while few associates noted that they feared management would return to old ways of working. Directors acknowledged that some adjustments might be required but emphasised that the long-term plan was hybrid working.

Managers are telling us it is going to remain hybrid going forward, and they themselves have enjoyed the option to work from home, but a part of me does feel that, if they can get us in, they will get us back in. I would love if they did not, but I just have that kind of feeling; the nature of the Big 4 and the busyness of it (A⁷F^C).

Overall, hybrid has worked, it has been fine. But adapting it to becoming a permanent solution is a different kettle of fish. Now we are in this flux period where we are just trying to bed things down (D¹F^D).

Now everyone is doing their own thing and rowing their own boat, and we are trying to see how that works before we see is there more guidance or principals required to try and make it work longer (HR¹F^D).

Two core external pressures were identified: (1) ensuring business continuity and (2) maintaining professional standards. For example, regulatory and client demands as well as quality of the professional service provided were concerns highlighted by participants. A core question asked was whether changes will impact the audit quality which would be an important thing for us and in all the Big 4 (A⁴F^C).

The regulator will not say, oh it is fine, you did less work because people were at home, and it was hard. The regulator does not care. They are like, you signed an audit opinion, so you need to get the evidence to support it (D¹F^D).

I suppose the Big 4 is a tough environment and there are certain pressures, and you are trying to get the best to your clients and deliver things quickly and rapidly and at a high quality, so I think trying to deliver that when you are remote and to build that culture and that ethos is difficult (D²F^C).

In addition, four internal pressures were identified: (1) re-establishing connectivity, (2) enhancing learning experiences, (3) balancing work-life commitments and (4) responding to evolving future of work preferences. Participants emphasised that working remotely reduced the onsite interactions which put pressure on the Big 4 firms to consider initiatives to re-connect with employees and rethink the purpose and role of the office in a hybrid setting. Concerns were also raised around the wider informal learnings which were seen as critical, but which were more difficult to realise in a remote and hybrid work environment.

We are human beings; we need to connect to thrive. That is not about forcing people to come in, it is using that time and we are doing a bit of work on the purpose and the role of the office. Why you would choose to come into the office and the best use of that. It is absolutely relationships, connection (HR¹F^C).

Working from home you are not going to be as connected. Even the values and stuff are on the office wall. When you walk past, you are going to read them. You have a quick glance, and again, you see them in your head every day. At home, you just do not see those things. I could have been working for firm x or y, or anyone (A²F^C).

We obviously regard a big part of our offering that we have formalised training supports in place but also informal training in the office and sometimes it is just being in a room with more people on an audit, and your work is going fine but there is a commotion in the corner or something else is going on and the client comes in and there is a bit of a standoff, and we learn by observing others. We cannot replicate that piece (HR¹F^D).

Moreover, participants flagged the significant challenge of attracting and retaining talent in a dynamic market, stating that the big resignation is real (HR¹F^A). They also pointed out to changing expectations of employees which required the firms to closely monitor their flexibility and compensation offerings, as the firms did not really want to lose good people (D¹F^D). At the same time, employees were conscious of what competing firms offered and flagged the preference for hybrid work.

I think around recruitment in the Big 4 there has been a challenge, because a lot of people can now move around. A lot of people took a chance and I saw that across friends telling each other what is happening in other Big 4 firms. A lot of people go out of the Big 4, taking a chance on doing something in the industry (D¹F^B).

Our vision of the future is flexible. And I am always about ‘why’ we are doing this. And the purpose is we need to remain the leading professional services firm and we need to remain extremely attractive to talent and to retain our people. They are why we do everything. For that reason, we absolutely need to listen and act upon what we were hearing (HR¹F^C).

5. Discussion and conclusions

The purpose of this study was to investigate the operationalisation of management control arrangements in the Big 4 firms and specifically to examine the patterns of management control change during the COVID-19 pandemic. In the following, we discuss our key findings, structured around our three primary contributions. Our first contribution lies in unpacking the changes to management control arrangements during the COVID-19 pandemic by comparing pre- and midst-pandemic management controls. In doing so, we build on the extant management control literature, by taking a temporal perspective (Bedford, 2020) to understand how firms make changes to their combination of different management control elements (Malmi, 2013; Malmi and Brown, 2008; Merchant and Otley, 2020; Van der Kolk et al., 2020). We illustrate that personnel controls and cultural controls were reimagined as a result of moving to remote and hybrid work. Here, it was not sufficient to strengthen or weaken controls or making controls more formal or informal, but instead, the Big 4 had to substantially alter their personnel and cultural controls. This included, for example, a changed tone from the top, the introduction of new norms, redesign of workspaces and creation of innovative approaches to learning. In contrast, we reveal that action and result controls remained relatively unchanged in both remote and hybrid work contexts in comparison to the pre-pandemic onsite setting. This could be explained by the significant standard operating procedures and a rigid management hierarchical structure which were maintained when the Big 4 transitioned to a new way of working. More specifically, action controls were strengthened through more frequent check-ins and increasing managerial oversight while working remotely. Result controls became more informal with more real-time feedback and empowering reward mechanisms being added. Despite these changes being relatively minor, firms emphasised the importance of action and result controls as a mechanism to navigate the new way of working. Therefore, our findings align with recent research which emphasises the interplay among management control elements (Pfister and Lukka, 2019; Toldbod and van der Kolk, 2022). We find that relatively minor adaptions to action and result controls were relied upon to respond to the substantial cultural and personnel control changes. These observations further our understanding of the changes firms made to their management controls in response to the COVID-19 crisis (Bedford et al., 2022; Huber et al., 2021).

Our second contribution stems from theorising the patterns of management control change by revealing a set of change pathways. In so doing, we contribute to the management control change literature (e.g. Granlund, 2001; Friis et al., 2015; Malmi, 1997; Mouritsen et al., 2022; Pfister and Lukka, 2019), considering the outcomes of such change (Bedford, 2020; Ferreira and Otley, 2009; Toldbod and van der Kolk, 2022). Specifically, we argue that the change pathway of action controls and result controls across the Big 4 firms can be categorised as alignment, i.e. variations that are evolutionary (low) in pace and developmental (low) in scope (Micelotta et al., 2017). The changes made to action and result controls did not disturb the stability of the firm and were only somewhat accelerated by the COVID-19 pandemic. For example, more informal frequent feedback was considered by firms pre-COVID-19 but did not fully materialise until the pandemic. Similarly, these changes were developmental in scope in that they were mainly expansions or diffusions of previous action and result controls. For example, action controls were expanded by adding daily check-ins. These observations also allow us to suggest that, during a crisis, management control arrangements can be mobilised to accelerate and embrace opportunities (Kober and Thambar, 2022). For example, we illustrate that COVID-19 provided an impetus for the practices around evaluating individuals to be altered, with a more holistic perspective, beyond hard metrics, being adopted. It was perceived that, pre-pandemic, firms had been discussing the need for the appraisal process to become less formal and more flexible, but that progress had been slow. The pandemic was perceived to have accelerated these developments, which was viewed as being a positive outcome.

In contrast, changes to personnel controls and cultural controls across the Big 4 firms can be categorised as displacement, i.e. changes that are revolutionary (high) in pace and transformational (high) in scope. Changes to personnel and cultural controls were fast in pace and disturbed the stability of the firms' logics. These changes were transformational in scope in that they altered shared understanding (Micelotta et al., 2017). For example, developing hybrid working models and reassessing existing learning approaches required the Big 4 firms to reconsider long-established personnel and cultural controls. We assert that, on the one hand, Big 4 firms operated within a set action and result controls, based on well-established norms which firms did not want to and could not overrule. On the other hand, imposed pressures required the Big 4 firms to radically alter their institutional logics and to adapt their personnel and cultural controls in remote and hybrid work settings. In doing so, we empirically support Toldbod and van der Kolk's (2022) argument that management control change is a dynamic process with heterogeneity among control elements in terms of pace and scope. Our observations also suggest that personnel controls play a pivotal role in management control efforts. In particular, the re-assessment of employee onboarding as well as employee learning is viewed as being critical. That is, the personnel control changes driven by the pandemic were partially viewed negatively and expected to be temporary alternations, which would need to be reconsidered in the future of work model. This is consistent with prior research that has shown that personnel controls are essential in volatile circumstances (Abernethy et al., 2015; Kober and Thambar, 2021).

In addition to pace and scope (Micelotta et al., 2017), we assert that longevity is an important dimension of change. Across the Big 4 firms, changes made to results and cultural controls were mostly expected to be long-lasting changes, necessary to meet expectations of talents in terms of the future of work (e.g. offering greater flexibility). In contrast, we found evidence that some of the changes made to action and personnel controls were expected to have a shorter lifetime or were already overhauled since their introduction, either because they related to the immediate pandemic response (e.g. daily check-ins) or because they proofed to be challenging over time (e.g. development of soft skills in an online environment). We therefore expand Micelotta et al.’s (2017) typology of pace and scope dimensions of change pathways by introducing longevity as a third dimension. Longevity can be conceptualised by considering temporary changes (low longevity) and perpetual changes (high longevity). We assert that the addition of longevity provides a more holistic perspective on the pathways of change and is a vital dimension when considering management control change pathways over time.

Our third contribution stems from contextualising management control change in a hybrid work setting in the professional services sector. In doing so, we identify two potentially competing perspectives which are likely to influence the longevity of management control changes and the sustainability of hybrid working in a Big 4 context. On the one hand, we identify a client perspective premised upon the needs of the clients. Frequently cited pressures by our participants related to ensuring business continuity and maintaining professional standards which was likely to determine how employees could carry out their work. A central concern for interviewees was clients' preferences for onsite, remote or hybrid interactions, which confirms the long-standing service ethic in Big 4 firms, seeing many decisions being made “in the name of the client” (Anderson-Gough et al., 2000: 1151). On the other hand, we also identify a second perspective which evolves around employees' request to firms to respond to future of work models. Here, pressures that related to competing for talent were strongly present in the Big 4 firms as they analysed emerging workforce trends. Closely monitoring competitors changing approach to management control and hybrid working was particularly important in the context of the ongoing “war for talent”, with firms viewing hybrid working as an essential talent attraction and retention mechanism. These views align closely with McKinsey research indicating that hybrid work could be a crucial factor in turning the “great attrition” to the “great attraction” (De Smet et al., 2021). For professional services firms, such as the Big 4, learning and development and knowledge sharing will remain key priorities and will be strong factors to consider when changing any management control arrangements or hybrid work approaches. Our research also indicates that hybrid work is fluid in that its formulation will change and develop over time as the way of working will depend on the strategic requirements of the employees. For example, professional services firms might have certain phases throughout the year where strong internal and external collaboration requires onsite presence while during other phases, more individually driven work can be completed remotely. We find that firms took a more critical look at what activities required onsite presence in a hybrid work model, confirming Delfino and van der Kolk's (2021) findings that connectivity is a critical aspect of management control, particularly in a remote setting.

Our study offers several practical implications for firms. First, considering the pace, scope and longevity of management control change can help firms to get a more holistic perspective of the patterns of change and increase the ability of firms to respond effectively to crisis situations. We posit that, if firms wish to successfully respond to a crisis, they should be conscious that involuntary changes imposed by the crisis to one type of management control can be managed by relying upon or making alternations to other management control types. Consequently, when a crisis occurs firms should closely consider what changes are required to action, results, personnel and cultural controls. Our observations also suggest that, as the move from a traditional, onsite model to a hybrid model presents a new way of working, reimagining cultural controls (Kober and Thambar, 2022) will be essential and can only be accomplished with an aligned tone from the top and the establishment of new norms that support hybrid working.

We acknowledge that our study, like all studies, has some limitations which we see as opportunities for future research. Our study was exploratory and qualitative in nature and examined management control changes in a particular context – COVID-19 and Big 4 firms. Hybrid working models were evolving in the Big 4 firms at the time our interviews were conducted, and amendments to these models will likely continue post pandemic. It was also beyond the scope of this study to establish any quantifiable relationship between individual control types or between management control arrangements and hybrid work. To address this, future studies could deploy a large-scale survey to examine the interdependence of controls (Speklé and Widener, 2020) in a hybrid work setting and the role of hybrid work in management control. Finally, at a practical level, we have identified a set of pressures to contextualise our findings. These pressures aid firms in understanding how two potentially competing perspectives may influence patterns of management control change; however, the theorisation of these potentially competing perspectives was not the focal aspect of our study and would merit future research.

Furthermore, we recognise that Merchant and Van der Stede's (2012) object of control framework is built on the assumption that each object – action, results, personnel and culture – needs to be controlled due to limitations in personnel, motivational issues and a lack of direction. Therefore, we chose to use their control framework based on the expectation that the Big 4 firms wished to manage and control human behaviour to ensure continued alignment with pre-COVID objectives of the Big 4. That is, we anticipated that many firms, but particularly the Big 4, would continue to adopt a traditional command-and-control management approach (Anderson-Gough et al., 2000; Spence and Carter, 2014) in response to COVID-19. This understanding of command-and-control is similar to Anthony and Govindarajan’s (2007) perspective of management control, but Merchant and Van der Stede's (2012) study also encompasses informal control mechanisms such as personnel and culture. Our findings support the expectation of a command-and-control approach, with the Big 4 firms making relatively minor adjustments to action and result controls while responding to significant changes in cultural and personnel controls. Additionally, our research highlights the growing importance of human interactions and communication across all four control objects which merit further research. To delve deeper into these critical aspects, we suggest that future studies in the context of the Big 4 could employ more interactive approaches like pragmatic constructivism (e.g. Mitchell et al., 2021), which has been found to be especially useful in exploring the role of trust (Liboriussen et al., 2021) and values (Mitchell et al., 2021). In particular, we suggest that future studies in Big 4 settings should pay particular attention to the issue of working environment and the probability that job performance will be achieved (Merchant and Van der Stede, 2012).

We are also conscious that our central point of interest lied on management control arrangement changes at a sectoral level. Future studies should accompany our findings by considering individual- and firm-level analyses. We contend that a focus on the role of situated rationalities of key actors will likely influence changes to management control arrangements and hybrid work considerations (Berts and Quinn, 2022; Ter Bogt and Scapens, 2019). Furthermore, research on how individuals' needs for autonomy, competence and relatedness are affected by changes to management control arrangements in a hybrid work setting would further aid organisations in drawing conclusions on the effectiveness of management control changes (Hall, 2016). Similarly, future studies might find that approaches to hybrid work differ across firms depending on their strategic directions and operational needs (Bedford et al., 2016; Gratton, 2021; Tippmann et al., 2021). To date, the future of work model research is at an embryonic stage, and empirical firm-level case studies are enquired to unearth differences across firms. Considerations in such studies might include the future of work strategies and the characteristics of the work model. For example, the future of work models might differ in their flexibility, i.e. constrained or unconstrained, around place and time (Gratton, 2021); these models might also differ depending on the extent of internationalisation, which links to the number of employees in a location, and the task context which might be more relational (onsite) or more transactional (remote) type of work (Tippmann et al., 2021). We hope that our paper provides a useful starting point for development of research on patterns of management control change in a hybrid work setting which the Big 4 consider the future of work.

Figures

Figure 1

Management control arrangement outcomes: a pathway lens

Figure 2

Three-dimensional taxonomy of management control pathways of change

Table 1

Interviewee sample

Firm	Position	Gender	Age	Tenure (years)	Identifier
A	Associate	M	20–29	4	A1FA
A	Associate	F	20–29	4	A2FA
A	Associate	M	20–29	4	A3FA
A	Associate	F	20–29	4	A4FA
A	Associate	F	20–29	4	A5FA
A	Associate	M	20–29	4	A6FA
A	Associate	M	20–29	4	A7FA
A	Associate	M	20–29	4	A8FA
A	Associate	F	20–29	3	A9FA
A	Audit Director	M	20–29	6	D1FA
A	HR Director	F	40–49	2	HR1FA
B	Associate	F	20–29	4	A1FB
B	Associate	F	20–29	4	A2FB
B	Associate	F	20–29	3	A3FB
B	Associate	F	20–29	4	A4FB
B	Associate	M	20–29	4	A5FB
B	Associate	M	20–29	2	A6FB
B	Audit Director	M	30–39	4	D1FB
B	HR Director	F	30–39	6	HR1FB
B	HR Director	M	40–49	4	HR2FB
C	Associate	F	20–29	4	A1FC
C	Associate	M	20–29	4	A2FC
C	Associate	F	20–29	4	A3FC
C	Associate	F	20–29	4	A4FC
C	Associate	F	20–29	4	A5FC
C	Associate	M	20–29	4	A6FC
C	Associate	F	20–29	4	A7FC
C	Associate	M	20–29	4	A8FC
C	Audit Director	M	30–39	7	D1FC
C	Tax Director	M	30–39	11	D2FC
C	HR Director	F	40–49	15	HR1FC
D	Associate	F	20–29	3	A1FD
D	Associate	F	20–29	3	A2FD
D	Associate	F	20–29	3	A3FD
D	Associate	F	20–29	3	A4FD
D	Associate	M	20–29	3	A5FD
D	Associate	M	20–29	3	A6FD
D	Associate	F	20–29	3	A7FD
D	Associate	F	20–29	3	A8FD
D	Associate	M	20–29	3	A9FD
D	Audit Director	M	40–49	21	D1FD
D	HR Director	M	30–39	11	HR1FD

Source(s): Author's own creation

Table A1

Changes to management control arrangements

Second-order categories	First-order codes	Empirical evidence
Action controls	Standard operating procedures and tools Management hierarchical structure Virtual audit rooms Daily check-ins	A function of the way we operate is, most of the firm's managers will still be reviewing all our associates work either in an in-person scenario, pre-Covid, or in a hybrid working model, so I think we will still on a day-to-day basis be able to track someone's progress on several activities (HR1FD) There is nothing wrong with those tools themselves. They are the sort of thing that will monitor our business and morale, etc. regardless of where people work. So, I think they will all continue as they are (D1FD) Pre-pandemic, there was less oversight. You are given your time; you do your work and then two weeks after there will be just a review on the whole file. Whereas when we moved to remote working, I felt there was a lot more communication between us and he will probably go in the file more often to review sections (A5FC) At the start of Covid, management very much expected that it would just be a complete disaster, that none of us would be able to work from home. They were afraid they could not keep an eye on and that everything just would not work out. Obviously, the opposite happened. That was a surprise to them. (A7FC) In the office, if I had a question, I would swing around my chair and I would ask the person behind me, whereas now, at the beginning of Covid, I would have to see if I got get them on a Teams call. It is hard trying to contact people. That is why we now do the virtual audit room. So, it is in the background if anyone has any questions, they just shout away straightaway, just like the office basically. (A7FA) We asked coaches and managers to check in with people more regularly. A lot of the teams put in daily stand ups. So, everyone came together at the start of the day to talk about what they were working on and where they had blockers or where the needed support (HR1FA) We might have more regular catchups throughout the audits with managers just to make sure that the audit is moving (A4FC)
Result controls	Organizational-level KPIs Individual-level metrics Performance systems and processes	Obviously, revenue and profitability are key like any business. Outside of that then, within assurance we have a risk and quality score card. So that is a set of criteria that I am assessed against making sure I am doing my day-to-day job right. Then we do our global people survey which is a global initiative. I have not witnessed any major change in these things (D1FD) To be fair, I think you were still judging off the same categories, the same skills. That has not changed. You are always going to look at, did they get the job done, how well did they get the job done there? Their actual accounting knowledge, their ability, etc. (D1FA) Once a week they help bring the team together so that then would have played a part in terms of performance appraisals. You would have recognised a team member's effort to help bring connectivity or connection within their own teams so that would have been something that would have been new in the context of Covid that we would not have had before (HR1FD) I do not think there would be any issues with promotions if you are virtual because you are still in contact with everyone anyway. But maybe there might be a bit of an element of out of sight, out of mind, but I do not think so, but I guess it remains to be seen (A6FD)
Personnel controls	Onboarding and interview formats Re-assessment of learning Set up of onsite and virtual workspaces	I think connectivity was a big buzz word as well in terms of the first-year associates now who have onboarded completely remotely. So that's a bit part of the hybrid working as well, to make sure that people are conscious that you do want to get people to meet up and you do want to be able to connect appropriately and kind of get to know these people, make sure they have a smooth onboarding process (D1FA) Obviously, there are other types of learning where it's much better in person. Onboarding type activity and connecting people with teams, that's another one. There's a whole framework that we've set out (HR1FA) A mix is good, especially for the kind of work we do, because it is a training firm, those two days in the office will be very beneficial to the training. Bouncing questions off each other, being able to go up to them with the laptop, show them in person. The work will remain the same but in terms of training members, I think there will be an improvement by even just getting those two days in the office. Just to train them properly because it is tough trying to do it over Zoom, and training people you never met in person, it is very tough (A7FC) Everything mobilised very quickly and the supports department even from an IT infrastructure perspective, just enabled people to be able to continue as normal. We provided people with more resources and equipment (HR1FC) Home equipment was a big one, employees got ergonomics assessment completed and to make sure that their home set up was appropriate. Other than that, it is just working with our teams so allow them the flexibility to do what is right for them (HR1FB)
Cultural controls	Tone from the top Establishment of new norms	The message we are getting now is that we will never go back to five days a week in the office. There has been a big investment in technology, there seems to be a strong commitment to it from the firm (D1FC) People attribute success to what got them there. So, if I think of the partnership, they were in the office five days, they worked x number of hours, and they were successful, and the firm has been successful. And therefore, we should apply that same, that belief carries with us. And we are a training firm and there is a huge amount of on-the-job learning, 70/20/10. Also think nobody really asked for it, because it had not been done, so people did not think it was an option they just assumed it would not be an option (HR1FC) It is that flux of being in the office and not being in the office. There will be an unconscious bias. Hybrid will not be a huge issue for us in terms of someone who is performing. I think the challenge will be for someone who is not performing, and they are primarily at home. I just wonder if that will create an unconscious bias, oh they checked out and are working from home. It will probably be a little less controversial on account of someone who is performing well and meets expectations (HR1FD) The director was saying that it would be good to bring everyone in for a day. They thought it would be easier if we were altogether and I said, I do not mind, I can travel up if you want, but I was reluctant. Then the manager told me the next day, there was no need and I just dialled into the call while they were all in the same room in the office (A3FA)

Source(s): Author's own creation

Table A2

Contextual pressures

Second-order categories	First-order codes	Empirical evidence
External pressures (a client perspective)	Ensuring business continuity Maintaining professional standards	There are a lot of concerns, you know, this sounds great, we all want to do it but what are we going to do when clients demand that we x, y, and z? In the past your client wants you there, ok, we are all there. Again, we would say a lot of our clients are probably in the same situation that they are reimagining everything, but then there are also others where it not something you can debate with (HR1FA) We are known for going above and beyond for our clients. That is just embedded all the way through, and it is the way people act from the top down. I think that is a big factor, irrespective of what controls are in place (HR1FC) The client will probably have a degree of precedence. If the client is saying, I need people onsite, we will do that (HR2FB) What they are always going towards is audit quality. They do mention that the employee matters as well and how we feel about it. The audit quality is the main thing, but they do try to take onboard how the employees feel about it (A7FC)
Internal pressures (an employee perspective)	Re-establishing connectivity Enhancing learning experiences Balancing work-life commitments Responding to evolving future of work preferences	Biggest struggles, one of them was staying connected and I think we over indexed on you know team meetings, one-to-ones, coffee catch-ups, bingo games. We were really going hard on making sure that people had that connection (HR2FB) Culture and people are such an important thing in the Big 4. You go in there with your year group. That is the biggest thing they are missing (D1FA) A lot of people miss that informal learning. You pick up so much in the office just listening to what people are saying. You hear about stuff that is going on. So, for me, you can continue to structure, you can continue to have processes and everything in the diary, it is all great, but we need the informal learning (HR2FB) Because we are a training firm, you bring people up through the ranks and I do not think it is the same training experience virtually as it is when people are together. And I think we have lost that in the past year or so and we have to kind of recapture it (M1FD) The market is hot; work-life balance has always been an issue in professional services. Our number one negative tenant on every survey and everything that we look at is work-life balance. So, this is a meaningful way for us to get after that as a business problem (HR1FA) Some of our response will be dictated by the market. Because you have people all over the country living at home, so, it will be hard to have a one-size-fits-all answer; we are in a funny phase now because we are just trying to roll out something new and different in the same way everyone else is (D1FD) If I was going for a job now and it was five days in the office, I would be a small bit iffy, and if it was 5 day at home, I would be a small bit iffy. I think that kind of split role, we need to get it right. I do think it is massive (D1FC) I suppose our thinking was to kind of make sure we were progressive in that hybrid area and that we were keeping up with what is going on in the rest of the world (HR1FB) They do not want to risk losing staff who feel the pressure to be in 9 to 5 if it does not suit them. I think if people can see it in other firms and if they are not getting it themselves, they will move once they hit the finishing contracts. It is such a bonus, and you are getting time for yourself when it works for you. I think it is something that people are wondering about (A4FD)

Source(s): Author's own creation

Note

1. The object-of-control framework distinguishes four categories of control. Action controls are controls that ensure employees perform actions that are beneficial to the firm and limit detrimental actions; Result controls refer to controls that involve rewarding certain outcomes or holding employees accountable for not achieving outcomes set; Personnel controls relate to appropriate recruitment (i.e. recruiting the “right” employees), training (i.e. providing employees with knowledge, skills, and abilities), and job design (i.e. considering job tasks, and how they are performed); Cultural controls often link back to the “tone from the top” and encourage individuals to pursue group norms (Merchant and Van der Stede, 2012).

Funding statement: This research has been kindly supported by funding received from the Institute of Chartered Accountants Ireland Educational Trust.

Conflict of interest: The authors do not have any competing interests to disclose.

Data availability statement: The data that support the findings of this study are available upon request. The data are not publicly available due to containing information that could compromise the privacy of research participants.