Union and management cooperation to address psychosocial hazards at work: The role of frames of reference

Abstract

Too little is known about how the perspectives of, and relationships between, unions and management may affect the management of psychosocial hazards at work. This is despite the potential impact of these hazards on workers and calls to better integrate industrial relations and work health and safety disciplines in order to respond more effectively. Drawing on the foundational industrial relations concept of frames of reference, and informed by research in industrial relations (IR) and work health and safety (WHS), we analyse the type and extent of cooperation between a manufacturing union and management in two Australian case organisations as they sought to develop a psychosocial risk management plan over two years during the COVID pandemic. We demonstrate the utility of a frames of reference typology for analysing and explaining their attempts at cooperation. We propose that frames of reference analysis can be used more actively, including by the parties themselves; more broadly, across different issues and sectors; and more strategically, as a way for parties to shape behaviour and relationships to improve outcomes for workers and organisations.

Introduction

In industrial relations (IR), frames of reference are defined as the beliefs and assumptions through which parties at work interpret what is legitimate and important in the employment relationship (Fox, 1969: 390). IR scholars have used and extended this concept to explain how and why employees, their unions, and managers may view the employment relationship in different and potentially conflicting ways, with consequences for behaviour, relationships and outcomes at work (Budd and Bhave, 2008; Heery, 2016; Bray, Budd and Macneil, 2020). In this paper, we explore the impact of frames of reference held by unions and managers at the workplace on a specific issue, that is, psychosocial risk management. This is a topic worth exploring for four reasons.

The first reason is the growing awareness of the existence and impact of psychosocial hazards at work (Potter, O’Keeffe, Leka et al., 2019; Potter, Loh, Dollard et al., 2025). Psychosocial hazards are ‘social and psychological conditions at work which adversely affect the health (both mental and physical), safety and well-being of workers' (Quinlan, 2025: 203). Examples of hazards include unreasonable workload or job demands, poor physical environment, harassment and bullying, and poor conflict or change management (WorkSafe Victoria, 2021). The urgency of addressing the causes of psychological ill-health caused by work is evident in the growing incidence of job-related stress and other mental health injuries at work (Wang et al., 2023). In 2023–2024, for example, mental health injury claims accounted for 18% of new workers' compensation claims in Victoria (WorkSafe Victoria, 2024). This may under-report the true situation (Wang et al., 2023; Matthews et al., 2021).

The second reason to explore this topic is the substantial and complex changes in responsibility for management of psychosocial risks, reflected in legal and regulatory changes. In Australia, since 2022, changes to federal and associated state work health and safety (WHS) or occupational health and safety (OHS) regimes have made obligations to manage psychosocial hazards more explicit. Persons conducting a business or undertaking (PCBU) must pay greater attention to prevention, and face increased penalties for non-compliance (Safe Work Australia, 2023; WorkSafe Victoria, 2023; Potter et al., 2025). These changes are in addition to strengthening of obligations and penalties under employment and discrimination laws which also cover some psychosocial hazards, workplace sexual harassment, and bullying.

Third, there is little research on the development of psychosocial risk management at an organisational level. Boot et al. (2024) identified this gap in a review of 50 years of scholarship. Organisational-level studies, they claim, are inherently more difficult to undertake than those focused on the more common level of the individual because they require ‘higher levels of complexity and longer durations’ (Boot et al., 2024: 399). Moreover, as their otherwise excellent review does not make explicit, few studies focus on the topic of this paper: namely, the impact of the relationships between the workplace parties on the processes by which mental health interventions at organisational level are designed and implemented (see, for example, Boot et al., 2004: 399–400).

The first three reasons to explore this topic present an imperative to better integrate concepts in IR and WHS, especially when doing so can better explain and predict the effectiveness of approaches to managing psychosocial hazards at work. IR is one of several disciplines essential to a multi-disciplinary approach to WHS (Quinlan et al., 2010) notwithstanding that psychology has had the greater influence (Quinlan, 2025: 204). In their shared space, both IR and WHS focus on parties to the employment relationship and the processes by which they design and implement rules affecting workers (Quinlan, Bohle and Lamm, 2010; Bray, Waring, Cooper and Macneil, 2018). Both IR and WHS disciplines recognise the essential role of work, occupation and industry in shaping these processes and rules, and both also centre the role of collective voice. Practically, it is often a single department in organisations that oversees IR and OHS (and Human Resource Management (HRM)) issues. With appropriate attention to definitions and domains, relevant concepts from one discipline should be transferrable to the other, to the benefit of both. As Quinlan says, ‘growing awareness of psychosocial hazards has if anything widened the interconnections or overlap [with WHS] warranting attention by IR researchers’ (2025: 206).

Frames of reference analysis offers an opportunity to explain how perspectives on WHS can affect outcomes. For example, case studies of unions and management receiving support to move away from their traditional adversarial pluralist perspective towards a collaborative pluralist frame of reference have reported good results for both parties (Bray, Macneil and Stewart, 2017; Bray, Budd and Macneil, 2020). These included improvements in safety, although with a focus on physical safety (measured by lost-time injuries). In related research, there have been occasional calls to use frames of reference to explain differing approaches to WHS (Heino, 2013; Walters et al., 2020; Fan, Zhu, Timming et al., 2020) although this has not been done in a systematic way (for an exception, see Walters, Quinlan, Johnstone and Wadsworth, 2016).

The fourth reason that our topic is important is because existing research on psychosocial risk management largely ignores unions (for an exception, see Dollard and Neser, 2013), despite unions being an essential party to the employment relationship and to WHS management in many workplaces. Moreover, there is evidence that workers in unionised workplaces have better health and safety outcomes, in Australia and in other countries (Walters et al., 2016; Weissbrodt and Giauque, 2017; Potter et al., 2019). A consequence of this lack of attention to the role of unions is the lack of attention to the relationship between unions and management at the workplace, and the effect it might have on management of psychosocial hazards.

Over almost three years, we observed and explored the motivation and actions of members of one blue-collar Australian union, and managers at two manufacturing sites who volunteered to work with them. Together they sought to take a cooperative approach to psychosocial risk management at work. The frames of reference typology guiding our analysis (Bray et al., 2020) and the rich case study data collected enable us to identify or interpolate values, assumptions and behaviours of individuals and groups associated with different types of frames of reference. In doing so, our first objective was to evaluate the utility of the IR frames of reference typology to explain different approaches to a specific WHS initiative. Our second, more general, objective was to propose ways in which frames of reference analysis can be used more actively, including by the parties themselves, to identify different perspectives and their effects on psychosocial risk management; more broadly, on other issues and in other sectors; and strategically, to shape the behaviour of and relationships between them as they seek to improve outcomes for workers and the organisation.

Our research question was – How do frames of reference held by unions and management on cooperation in the employment relationship help to explain approaches to managing psychosocial risks at work?

Literature review

To apply frames of reference to the concern of psychosocial hazards, the first step is to establish the empirical and theoretical common ground between IR and WHS, specifically in relation to unionised workplaces, and to demonstrate how the two concepts can be integrated.

It is straightforward to demonstrate the connections between IR and WHS at the empirical level. Most of the 14 common psychosocial hazards identified by Safe Work Australia (2019) are either explicitly or implicitly covered in employment or discrimination law. For example, two of the more serious pyschosocial hazards - sexual harassment and bullying - have their own jurisdiction under the Fair Work Act (Cth) 2009, and sexual harassment is also addressed in the Sex Discimination Act (Cth) 1984 and most recently in the Respect at Work Amendment Act (Cth) 2021. Another hazard, poor organisational change management, is addressed through the requirement for modern awards (determining minimum wages and conditions by industry or occupation) to include a clause on ‘consultation regarding major workplace change’. Other psychosocial hazards, such as low job control, high job demands or poor organisational justice, are often implicitly covered by the same consultation requirement, where change may lead to a deterioration in psychosocial conditions.

While connecting IR and WHS on psychosocial hazards may be relatively straightforward on paper, determining in practice how to address these hazards has been complex, particularly in the period when legislative change to address psychosocial hazards was being debated but had not yet been enacted. Changes to legislation and regulation to directly address psychosocial hazards were recommended in the Review of the Commonwealth Model WHS Act (2018) (Boland, 2018). However, the Commonwealth did not introduce their revised Model WHS Act until 2023. At this point, each Australian state adopted its own legislative changes, since WHS is primarily a state responsibility. Victoria was the last state to make associated amendments to its laws,with the publication of its OHS (Psychological Health) Regulations 2025. Until then (and during the life of the project on which we report) it was the employer's general responsibility to provide a safe and healthy workplace, including both psychological and physical health and safety. In other words, the issue of what would be required of the parties in order to properly deal with psychosocial hazards at work was one on which there was a great deal of uncertainty and concern during the project.

Also at the empirical level, there is common ground between IR and WHS on the kind of roles and processes employed at the workplace to address psychosocial hazards, where unions are present and active. In Australia, consultation obligations have been imposed on employers in Australian OHS legislation since the adoption of a version of the 1974 UK Robens model (Johnstone, Quinlan and Walters, 2005: 94). This includes the expectation of worker and management representation on workplace WHS committees and the election of Health and Safety Representatives (HSRs) (often also union members or delegates, e.g., Walters, Quinlan, Johnstone and Wadsworth, 2019: 977). Unions and management have historically engaged with each other on WHS through the second avenue of traditional IR processes, such as bargaining, industrial action, enforcement of standards and rules, and claims made in and outside the workplace of safety concerns arising from the behaviour of one party or the other (Quinlan et al., 2010; Forsyth, 2023).

However, again, there is considerable ‘flexibility’ provided by the macro-regulatory regime for development of workplace policies and practices regarding physical (and psychosocial) hazards (Johnstone and Tooma, 2012: 4). Observers note the growth in an OHS systems approach which emphasises the role of management, focusing on risk, culture and climate (Walters et al., 2019: 977), but has considerably less emphasis in recent years on employee voice and decision-making (Johnstone et al., 2005; Quinlan, 2025). In summary, in terms of roles and processes, unions (where they are active at the workplace) and management have choices about how they approach WHS generally and, more specifically, on whether and how they cooperate.

The theoretical connection between IR and WHS is more abstract, and the focus of this paper. It is our contention that the frames of reference through which unions and management view workplace relations can be adapted for use in understanding approaches to WHS and, by implication, approaches to managing psychosocial hazards. The seminal explanation of frames of reference in IR comes from Fox's work (1969, 1974a, 1974b), and the three types he proposed – pluralism, unitarism and radicalism – have previously (although rarely) been used to explain different perspectives on WHS. In simple terms, pluralism is a frame of reference that can be held by unions and management, in which there is recognition of the legitimate role of the other party, where compromise in addressing shared interests is possible, but conflict is inevitable and must be negotiated. In their studies of coal mining in Australia, WHS and IR scholars point to ‘[p]articipatory mechanisms at jurisdictional, industry and workplace level’ (Johnstone et al., 2005: 94), when used and respected, as evidence of a pluralist approach to WHS (also Walters et al., 2019: 992). In contrast, unitarism is a management frame of reference, based on the assumption of managerial prerogative to decide organisational goals in the interests of all, and to determine how goals are best implemented. Unions are seen as unwanted ‘third parties’, and management seek to exclude them, wherever possible. A unitarist approach in WHS will also be evident in ‘more behaviour-based systems of direct participation in corporate safety systems, implemented and controlled by managers’ (Walters et al., 2019: 992). Fox's third frame of reference is radicalism and is only held by unions. This perspective assumes that structural conflicts of interest between labour and capital are irreconcilable. Those with this perspective may work outside agreed institutional arrangements to address their perceived interests. Holders of this perspective would reject unitarist approaches to WHS, seeing them as likely to ignore workers’ safety interests (Walters et al., 2020: 190), but also reject pluralist approaches, seeing them as incorrectly viewing safety as an area of possible convergence of interests between union and management (Heino, 2013: 154).

Applying the frames of reference typology on cooperation to WHS

As we have seen, the use of frames of reference analysis in WHS scholarship, while uncommon, is useful in characterising different and potentially contrary WHS approaches of unions and management. Development of the frames of reference concept in IR since Fox‘s early work offers extended opportunities for analysis, with expanded typologies (Budd and Bhave, 2008; Heery, 2016; Kaufman, Barry and Wilkinson, 2021). Of the newer typologies, we rely on Bray et al.'s typology of frames of reference on cooperation in the employment relationship, for three reasons. First, and practically speaking, the union representatives in our case studies were seeking to cooperate with managers at their worksites on addressing psychosocial hazards, so a typology focused on cooperation is apt. Second, this typology, and the empirical work which informed it, was based largely on case studies and analysis of union-management cooperation in energy and construction; that is, actively unionised, blue-collar industrial sectors, with a similar historical focus on physical WHS as has characterised manufacturing (Bray, Macneil and Stewart, 2017). Finally, Bray et al.'s model has been used elsewhere to investigate changes in dominant frames of reference at a workplace (Johnstone and Wilkinson, 2017), an essential concern of our case studies.

In summary, Bray et al. propose six frames of reference on cooperation in the employment relationship. What ‘cooperation’ means in five of those six frames of reference is shown in Table 1, along with the party or parties (i.e., union and/or management) that would theoretically hold them. (The sixth frame of reference proposed by Bray et al. – market perspective/egoism, is not part of our analysis, since it does not apply to collective IR.)

Table 1.

Frames of reference on cooperation between unions and management.

Frame of reference	Can be held by these parties	For those holding this frame of reference, cooperation means…
Critical/radicalist	Union only	…[reluctant] acquiescence by employees to employer-established goals and practices.
Adversarial pluralist	Union and management	…employers, employees and their representatives pursuing their separate goals while acknowledging the legitimacy of each party's interests.
Collaborative pluralist	Union and management	…employers, employees and their representatives working together on mutual goals and compromising on conflicting goals in ways that respect the legitimacy of each party's interests.
Consultative unitarist	Management only	…employers and employees working together on organizational goals, in ways that are established by management through consultation.
Autocratic unitarist	Management only	…employees following managerial directives for serving organizational goals, that in turn also are assumed to benefit employees.

Source: Derived from Bray et al. (2020: 119).

Relying on the richer descriptions of expected values and assumptions provided in Bray et al. (2020) and applying them to WHS issues in general we would expect that unions with a radicalist perspective on WHS would not work with management, rejecting the proposition that there can be any claim of ‘alignment of interests’ on safety. Any safety initiatives guided by this perspective would be developed independently, often actively contrary to perceived ‘management’ interests; and any safety initiatives developed by management would be treated with suspicion.

Of the two forms of pluralism proposed, adversarial pluralism is a common perspective in unionised Australian organisations. Those holding this perspective on WHS would recognise limited, shared interests, including on WHS, on which the parties may consult, or where the parties must consult by law (such as on OHS Committees). However, generally each party would also seek to address their WHS interests independently and have no problem making opportunistic use of the other's perceived or actual safety ‘transgressions’. For those with a collaborative pluralist frame of reference, the perceived scale and scope of shared WHS interests is enlarged, with active worker involvement, joint development and implementation of safety and related initiatives, and recourse to formal or external processes to address concerns only after internal processes are exhausted. Opportunistic wins would be avoided.

In contrast, the two forms of unitarism have in common a view that it is management's prerogative to determine a single set of goals for the organisation, including safety goals and the processes and structures to enact them. For those with a consultative unitarist frame of reference, direct communication with and consultation with employees (although not unions) is valued. Interventions on psychosocial risk management consistent with unitarism could be corporate employee assistance programmes (EAPs) (Long and Cooke, 2023); Mental Health First Aid training for managers; or corporate wellness programmes (i.e., Defehr, 2016; Joseph, Walker, Fuller-Tyszkiewicz et al., 2018; Song and Baicker, 2021). For those with an authoritarian unitarist frame of reference, communication and consultation with employees on safety would be reluctant, perhaps limited to ′compliance' training on specific issues (like bullying and sexual harassment), and risk management plans, with discussion and decisions made within management-controlled systems, structures and policies. Any ‘interference’ with managerial prerogative, including by unions, would be treated with hostility.

Frames of reference, as they have been discussed so far, are ideal types, representing values and assumptions comprising a stable world view that informs priorities and behaviour. In practice, their application is more complex (Heery, 2016). Within a particular party at a workplace, there are likely to be a broader range of views than can be captured by a single frame of reference; for that reason, we find it more useful to write of a dominant frame of reference. With sufficient data on the views of elected officials or managers, and as manifested in organisational arrangements, identification of a dominant frame of reference in action for each party, or important sub-groups of each party, should be possible.

What can be anticipated is that organisational arrangements will be consistent and well-understood if the two parties essentially share the same frame of reference (Kaufman et al., 2021: 211; see also Johnstone and Wilkinson, 2017). In all other combinations, frames of reference will be in conflict, and a joint approach to the creation and implementation of organisational policies and practices will be difficult (Budd, Pohler and Huang, 2022). As a result, the balance of power within the workplace will be especially important in determining employment and WHS arrangements.

Also, in acknowledging that the dominant frame of reference in action for each party may not be shared by all within that party, it is also possible to see how a dominant frame of reference might change. A senior manager with an autocratic unitarist perspective, newly appointed to lead the organisation, could alter the consultative unitarist status quo amongst the whole management group. Similarly, a reformist union ticket winning an election on a promise to replace high-level conflict with managers with active negotiation could alter the dominant frame of reference in action for the union from radicalist to adversarial pluralist.

Turning to the WHS literature on the role of unions, there is widespread evidence that union participation in WHS matters delivers good outcomes. A US study, for example, found that unionisation is associated with improved health and safety, with the effect most pronounced in manufacturing (Li, Rohlin and Singletone, 2022: 418), the sector in which our study is conducted. Other studies confirm the benefits of union participation, in Australia (Bray et al., 2017) and overseas (e.g., Seghezzi, 2023). In circumstances in which unions are active at the workplace, and acknowledging their beneficial influence, what frames of reference might we expect to see associated with the best WHS management, and psychosocial risk management in particular?

Where the frames of reference of union and management are not shared, a union approach ‘marked by resistance’ (possibly representing a ‘radicalist’ perspective) can be effective, particularly in the face of unitarist prioritisation of production by managers (Walters et al., 2016: 392). Where unions are active at the workplace but the dominant management frame of reference is unitarist – particularly of the authoritarian type – dissonance between the parties will create challenges in WHS management and interfere with worker voice, generally recognised as essential to the WHS risk management process (Quinlan, Bohle and Lamm, 2010; Walters et al., 2016). A better understanding of the characteristics and consequences of these conflicts would be very helpful.

If the best WHS outcomes are supported by a shared frame of reference between union and management, then this can be either where the parties share an adversarial pluralist perspective or a consultative pluralist perspective. The existing evidence is not sufficiently nuanced to suggest which shared perspective might be more beneficial, nor what challenges arise if the parties are pluralists of different types, but it does anticipate better outcomes when unions and management cooperate. Walters et al. argued that ‘research on worker representation suggests that [an] essentially pluralist approach, when implemented with adequate support, is likely to result in improvement in OHS outcomes’ (2016: 379); and others claim that a trusting or cooperative relationship between management and unions improves health and safety outcomes (e.g., Gunningham and Sinclair, 2014). While psychosocial safety benefits of union-management cooperation have not been assessed systematically, an argument has been made that psychosocial safety programmes designed with and trusted by employees deliver better outcomes (Harvey, 2019). Generally, ‘mutual gains’ for workers and management associated with union-management cooperation commonly include greater job satisfaction, decreased stress and a happier work environment in which managers and workers enjoyed (rather than hating) coming to work (Kochan, Eaton, McKersie and Adler, 2009), all factors associated with increased psychosocial safety (see Bray et al., 2017; Bray and Macneil, 2024). Moreover, unions in a cooperative relationship with management may provide support on issues that would otherwise thwart effective psychosocial risk assessment and management, such as lack of awareness, resources, and expert advice and guidance (Metzler et al., 2019; see also Fan et al., 2020; Potter et al., 2019; Walters et al., 2016; Weissbrodt and Giauque, 2017; Valizade et al., 2016).

In summary, we propose that this integration of IR and WHS literatures through the analysis of frames of reference held by unions and management on cooperation in the employment relationship will help to explain approaches to managing psychosocial risks at work.

Method

The project and its participants

With increasing recognition that organisers and delegates were neither trained in, nor comfortable with, tackling psychosocial hazards at work, the AMWU Victorian state conference resolved in 2019 to ‘develop a preventative project to assist members, delegates, health and safety rep[resentative]s, and organisers to secure a workplace that is safe from psychosocial hazards, including bullying and harassment’ (AMWU official). The project was managed by AMWU officials, especially the WHS Officer, the State Secretary and a Project Officer.

The union received funding from WorkSafe Victoria for a project to collaborate with companies to prevent psychosocial safety injury in manufacturing work. WorkSafe also provided timelines and templates to help keep the project on track. Project funding paid for external consultants with psychosocial safety expertise to conduct training. Using results from the 12-item Psychosocial Safety Climate Questionnaire, or PSC-12 (Hall, Dollard and Coward, 2010), open to the whole site workforce, the consultants helped to create psychosocial safety action plans (which were called Mental Health Action Plans (or MHAPs).

The project was overseen by an active Steering Committee comprising project teams from each of four participating sites and AMWU representatives. The Chair of the Committee was a retired Fair Work Commissioner, with extraordinary expertise in facilitating union and management relationships shaped by all frames of reference. Prior to joining the Commission, he had worked in the manufacturing sector. Also in attendance were the external consultants, and the academic team engaged by the AMWU to conduct the project evaluation from start to end. The latter team had expertise in in IR, organisational psychology, mental health at work and manufacturing work. Their final report was delivered to the AMWU in December 2022.

As the project was getting started, so was the COVID-19 pandemic. Substantial efforts were made to keep the project on track. Project-wide activities organised by the AMWU in 2020–2022 included project management and governance activities; two full-day union training forums (in 2021 and 2022) to educate participants on managing psychosocial hazards at work; use of project collateral to launch a union campaign on recognising psychosocial safety hazards at work. The AMWU and the Steering Committee Chair also arranged for members of the Fair Work Commission (FWC) with expertise in their ‘Collaborative Approaches’ program, to offer training (FWC, 2020).

The research reported in this paper focuses on the experience of the union and management at two of the original four participating manufacturing sites.

Research procedure

The research reported here comprises two case studies on attempts by managers and union representatives to jointly develop plans (MHAPS) to address psychosocial risks and prevent mental health injury at their workplaces. The cases are treated as ‘critical’ cases (Flyvberg, 2006; Ruddin, 2006), with the change in circumstances generated by the COVID pandemic offering an unusual opportunity to explore the effect of strong forces, first for and later against change.

In 2020, these two workplace projects could be characterised as most likely cases to succeed; that is, if cooperative relationships between union and management could not be developed in these workplaces, it would be difficult to do so anywhere. Union organisation was strong and interest high with significant resources allocated. Management participation was voluntary; but one incentive to participate was that management responsibility would escalate with the proposed regulatory changes. Cooperation was bounded by both topic (psychosocial safety action) and time (initially a 2-year project); making participation less risky than a broad, open-ended commitment.

However, through 2021 and into 2022, the likelihood of success in achieving a cooperative approach to managing psychosocial hazards was made least likely by contextual factors. The COVID-19 crisis delayed the project and adversely affected participants. It divided the workforce between those who were required to be present at work, and those who were not. In a city (i.e., Melbourne) with some of the most severe and lengthy lockdowns in the world, at the two case study sites new and unexpected challenges arose as a counterforce to initial support (Manisaligil et al., 2023). With this complexity, an unusual opportunity arose to examine how the plans of union and management to cooperate might change over time, and under pressure. The case studies are therefore ‘critical cases’, with rich data ‘facilitat[ing] and encourag[ing[ critical reflection to enhance our ability to challenge, refute, refine and illustrate theory’ (Alvesson and Karreman, 2011: 13–14).

Thirteen interviews were conducted for the Company 1 case study. Of six participating managers and union delegates, three were interviewed at both the beginning and end of the project. Four other project participants were also interviewed. Fifteen interviews were conducted for the Company 2 case study. Four actors – managers and union delegates – were interviewed at the beginning and the end of the project, with seven other interviews conducted. Interviews, of 40–70 minute duration, were recorded, transcribed, and approved by interviewees. Participants are identified only by their role. The researchers also had access to project presentations and documents, and notes and observations from their participation in project events, all providing data on the process of building cooperation to improve workplace psychosocial safety.

The case studies were written consistent with the IR tradition of ‘thick description’. A broad issue was identified, in this case, how the parties’ frames of reference influenced their approach to cooperation on improving workplace psychosocial safety. The method was inquiry-based, ‘to learn enough about the case to encapsulate complex meanings into a finite report but to describe the case in sufficient descriptive narrative so that readers can experience these happenings vicariously and draw their own conclusions’ (Stake, 2023: 128). The challenge, as always, was to convey enough of the sequence, uncertainty, contradiction and context to allow readers to do so. Fortunately, this approach to case research is unusually flexible, since the research was framed in one way, and then fundamentally altered by the catastrophic global pandemic. The case narratives are initially chronological, providing context and descriptions of relationships before the project, of the project activities unfolding on site; and of specific attempts to build cooperation and their fate. The case studies do not seek to report the outcomes of the MHAPs. The focus is on the approach taken by each party, and its affect on the relationship between them. At the end of each case, a short, explicit analysis is provided of how the researchers perceived the frames of reference of the parties, and evidence of any changes in the dominant frame or frames in action.

Findings

Case study 1

‘Company 1’ was a single work site in Victoria, part of a large multinational manufacturing enterprise, with around 100 production workers and 40 office workers and managers. Almost all production employees (workers and management) were men, of long tenure with the company. AMWU membership amongst production workers was high, with a strong, active workplace delegate structure. A Memorandum of Understanding for the project between Company 1 and the AMWU was signed in December 2020 and the project ran from early 2021 to late 2022.

History and context for the project at Company 1

Workplace relations at the site were historically adversarial, even described by interviewees as ‘hostile’ and ‘toxic’. Poor behaviour was acknowledged on both sides over many years, ranging from spiteful language and disrespectful actions by individuals to unpredictable and obstructive collective relationships, sometimes leading to bullying, harassment and favouritism. As one union delegate summarised, ‘The company thinks that employees at [the site] are just assholes, militant assholes, and we [i.e., the union] have the view that managers … are incompetent fools’.

One manager saw at least some of the poor behaviour as related to mental illness. It was, he said, bad for people and created low morale and high absenteeism. Changing behaviour and minimising its consequences was one of his motivations to participate. A union official reported that the site was selected in the hope of improving union-management relationships as well as promoting better psychosocial safety. Coming into the project, one union delegate was unconvinced about management commitment; another saw potential benefit if management ‘followed through’.

Managers and union representatives agreed that pre-existing company policies and practices were quite reactive, largely limited to the company's EAP. Union WHS focus was primarily on physical health and safety.

The cooperative project on site

In the first full year (2021), there was project engagement from managers and union representatives; helpful support from the project Steering Committee; and resources, expertise and assistance available from experts, directly and indirectly through the project. A major session on union-management cooperation was presented in the first Training Forum in April 2021. Management and union interviewees thought the first year of the project went well, despite challenges presented by the COVID-19 pandemic ramping up.

There were increasing problems in 2022. At one point early in 2022, 31 of 100 production workers were absent from site. Simultaneously, the closure of a nearby site increased workloads and customers faced extended delivery times. One union representative said that resulting production strains meant the project was not a high priority for management.

There was a change of site manager in early 2022. The new manager was still formally committed to the project, but his role was limited and participation by other managers diminished. After attending the FWC training in March, one manager said he thought the training was too pro-union. Another said he would have found it helpful if the initiative was company-wide rather than confined to the site, and if it were associated with some new company policies and procedures. The full-time union organiser and head union delegate who initially took project leadership were absent from site due to COVID pressures. Other delegates stepped in but became increasingly disengaged. The company's new collective agreement was being renegotiated in an adversarial process.

Managers and union representatives participated in a session in May 2022 on improving psychological safety at work and agreed on a MHAP. It was relatively simple – identifying the hazards to be addressed (both related to communication, on psychosocial safety and wellbeing), evaluating the level of risk, relevant policies, practices and procedures (of which the group could identify none); proposed interventions (focused on information, training, including Mental Health Champions, and opportunities for discussion and feedback). Responsibilities and timelines were also noted. However, the site manager felt that training could not go ahead for ‘production reasons’. Without training, commitment on both sides lapsed, participation in the Steering Committee ended, and the MHAP was suspended.

Attempts to build cooperation

Managers acknowledged in principle that cooperative union input into decisions could be valuable, especially on psychosocial risk management and in highly unionised workplaces, because many workers trusted their union more than their managers. Union representatives also believed management should work with union delegates. One explained it could protect workers by ensuring workplace changes were ‘practical’. Another said that collaboration on psychosocial safety could bring ‘more eyes and voices’ to an important issue.

Despite these sentiments, managers made cooperation and psychosocial safety low priorities. One union representative recognised that collaboration was much more difficult than the relationships to which they were accustomed:

It's really easy for me to galvanise my members against a common enemy, be it managers, the company. That's the easy part of my job. But to bring them along on a [cooperative] journey where both parties win, where we all get treated with some dignity and respect and get listened to – that's hard.

As relations on site slipped back in 2022 to a more adversarial tone, union representatives largely blamed management. One union delegate was slightly more sympathetic to management circumstances, recognising that it was at least partly explained by management being under-staffed and time poor. One pointed to poor communication with the workforce, ‘always … an issue on this site’, also arguing that enterprise bargaining was a distraction which ‘tended to encourage adversarialism, rather than [the] collaboration’ required of the project. A third union representative attributed the project's suspension to management's ‘old-school adversarial’ approach towards unions and their deeply held belief in managerial prerogative.

Correspondingly, there was blame attributed to the union by managers. One manager argued that collaboration was more difficult because the union was excessively aggressive and failed to recognise the significant challenges of the site's situation. Another manager said that union representatives did not accept that participation in the project included a need to regard ‘managers as people’. Both managers also wanted to ensure that non-union employees were included and criticised the union for not recognising the need to ‘engage everyone’.

Taking action to prevent psychosocial safety injury was new – no one was confident with their knowledge and skills on the topic. One union delegate advocated the appointment of a specialist manager to provide support on-site. The opportunity to seek advice and support from ‘experts’ involved in the project was well-received by both sides, but it was not sufficient. Despite all this, one union representative saw a bright side, saying ‘At least the conversation has started. It can’t be ignored now’.

Frames of reference and any changes

Before the project, the perspectives of both union and management on site were, overall, consistent with ‘adversarial pluralism’, although within the parties, there was some variation. Some union representatives were more inclined to a ‘radicalist’ perspective, and some managers more inclined to ‘autocratic unitarism’. Most managers adopted an ‘adversarial pluralist’ perspective in action, in pragmatic recognition of the deeply embedded union role on site. Interviewees were not complimentary about the state of their relationships but accepted them.

During the first year of the project, a window briefly opened on the possibility of a shared collaborative pluralism approach at least on the issue of psychosocial safety. On the union side, it was supported by the officials initiating the project; and the long-standing example of cooperation at another unionised site in the same industry. Union delegates also hoped for less confrontation. In the second year, this interest was infected by increasing pessimism as trust declined, and ‘adversarial pluralist’ relations returned.

On the management side, there was some initial commitment to change and in principle recognition that a more ‘collaborative pluralist’ approach could be beneficial, but little action was taken, especially in the second year. The pandemic and production pressures were the stated cause, but preferences more consistent with ‘autocratic unitarism’ seem to have reinforced this choice. In the end, with little new action on psychosocial safety initiatives, ‘adversarial pluralism’ was reinforced as the dominant frame of reference.

Case study 2

Company 2 was a single Victorian manufacturing site, part of a larger international organisation. Established around 1970, the site was scheduled to cease production in late 2022, an announcement made prior to project commencement. There were about 180 production workers on site, with high AMWU membership, and around 120 office workers and managers.

History and context for the project

Union and management interviewees agreed that, before a change in site management around 2017–2018, workplace relations were decidedly adversarial. For a long-term union representative, facing regular turnover in management, ‘everything was a battle’. With the appointment of new management, however, the situation changed:

The relationship with the company and the union's improved tenfold and they’ve learned to be able to consult and listen. Now that they are, we are … cooperating, we’re working together and getting things done.

Psychosocial risk management initiatives at Company 2 before the project were, as one manager acknowledged, developed neither locally nor cooperatively. There was little uptake of the company EAP, the primary corporate resource to support workers’ mental health. Explanations differed. A union representative was cynical, arguing that managers were simply ‘ticking the box’. A manager expressed bewilderment about low uptake, after active encouragement for employees to use the programme.

The cooperative project on site

Management enthusiasm for the project initially came from senior site managers after meeting with the union. A member of the site leadership team was appointed project ‘champion’ and formed the project team, with a specialist WHS manager, an office worker (not a union member) and the senior site union delegate.

This team led the project on-site. Members attended steering committee meetings and participated in the FWC's interest-based problem-solving workshop (for details, see FWC, 2020). Both activities were designed to build understanding and skill in how to frame shared problems, explore and acknowledge all interests related to that problem, including conflicting ones, before seeking a consensus solution. Both activities were praised by participants. Another initiative, funded by the project and designed to engage a broader spectrum of the workforce, was the psychosocial safety training: small-group sessions were held on site and directly involved most of the workforce. Two union representatives were complimentary about the content, and the degree of workforce involvement, as were participants. Managers saw these training sessions as one of the most successful aspects of the project.

Project team members contributed to planning and promotion of the project. They participated in a panel discussion at the project's second training forum in November 2021, talking about their experiences, challenges and successes, subsequently working with the AMWU videographer to create a short educational video. Management members of the project team attended a global corporate WHS meeting, presenting the project and receiving encouragement to continue the project and share its outcomes.

Cooperation on psychosocial risk management

The project improved collaboration between management and the union, especially evident within the project team whose members observed that ‘there is more trust and transparency than there used to be’ and that the project had been ‘…far more transparent than I thought it would have been’. An external union representative, who closely followed developments at Company 2, observed:

[T]hey talked to one another, they discussed things, they came up with their own ideas, their own initiatives that they could implement. And, really, I feel that that's probably the most critical factor: that they worked together well as a team of their own volition, and that really seemed to improve outcomes within their organisation.

A key example was team members working together to develop the MHAP. Managers led, but the development of the MHAP was done collaboratively. The MHAP comprised 13 separate actions, and ownership was delegated across various groups, depending on the item – the senior leadership team, the operations manager, the EHS team, the safety committee, department leads and front-line leaders, and workers. The action items were specific and focused heavily on psychosocial risk awareness and making assistance available. Communication about the plan and its actions was widespread; psychosocial safety issues were added to meeting agendas and toolbox meetings; information was shared via pamphlets and plasma screens; and one-on-one discussions. Consistent with the action plan, message content and communication mechanisms were jointly determined and delivered, mostly by the project team.

Managers interviewed did not seem to be great supporters of unions generally, emphasising instead the importance of ‘leadership’. They were committed to ensuring that non-union employees were included in all discussions. However, they came to see value in union involvement in the psychosocial safety project. A senior manager, for example, said that the

[senior union delegate] walk[ing] around the floor and talk[ing] about the program [was helpful] because he represents the people. They respect him, and they trust him. I'm not saying they don't respect and trust me. I probably don't know everyone as well as he does. He works with them all the time.

Another manager on the project team agreed that union engagement was an essential component of the process. A union representative on the project team went further:

Collaboration has to be the way. People won’t do it if I’m [as the union representative] not on board with it… Can’t have the company run it, otherwise it's not going to work.

Frames of reference and any changes

Pre-project, Company 2 union representatives were inclined to ‘adversarial pluralism’, although not of a particularly combative kind. Managers had a unitarist perspective, leaning towards the consultative rather than autocratic, influenced by their own values and commitment to workforce welfare. Activities on site related to psychosocial safety were consistent with this characterisation, i.e., management EAPs, and union activities on physical safety, with neither engaged in the other's activities. The two contrasting frames of reference in action co-existed, but not to the obvious satisfaction of either party.

During the project, union and management both shifted, especially within the project team, working towards a shared, ‘collaborative pluralist’ approach. Project team members were highly regarded on-site and supported by their respective leaders. Collaboration gained traction. Management's company-wide initiatives were supplemented with jointly designed site-specific initiatives. Management joined the union in creating educational material for workers. Concurrent activities to support the site's shutdown reinforced the importance of voice, communication, and a senior management focuses on workforce welfare.

Discussion

This paper addresses the question of whether and how frames of reference held by unions and management on cooperation in the employment relationship explain approaches to managing psychosocial hazards at work. Drawing on data from a union-led project to work cooperatively with management to improve psychosocial safety at work, a project overview and two case studies have been presented, using the frames of reference on cooperation to guide analysis.

Our first objective was to evaluate the utility of the IR frames of reference typology to explain different approaches to a specific WHS initiative. In summary, at the start of the project, both union and management in Company 1 acted consistently with an adversarial pluralist frame. From this shared base, and an early willingness from both parties, progress was made in 2021 towards active collaboration on the specific issue of psychosocial risk management. However, as the effects of the pandemic were felt more deeply and ‘business concerns’ become an overriding focus, management action at Company 1 became more consistent with autocratic unitarism. A jointly developed plan to address psychosocial hazards at work, light on detail beyond ‘improving management communication’, was not implemented. Union and management on-site slipped back to their pre-COVID adversarial pluralist status quo. In contrast, at Company 2, at the start of the project, union and management action was characterised by different frames of references. Union action was consistent with an adversarial pluralist frame, and management action with a consultative unitarist frame. Progress was made in 2021 towards active collaboration on workplace psychosocial safety. The effects of the pandemic were also felt at Company 2, but joint activities, a more robust plan, and growing mutual respect cascaded across the site, increasingly consistent with a collaborative pluralist frame of reference.

The case studies and associated analysis therefore support the initial classification of union and management action to a particular frame of reference. They also support a change in classification, at least on this issue, as the project progressed and new actions were taken. The cases provide sufficient data to classify actions to a frame with considerable confidence. However, there are clearly different ‘camps’ or points of view within each group, for example, within the union at Company 1; on-site managers vs. ‘international’ managers at Company 2. This is a timely reminder that a change in a key person with sufficient power – for example, a union official, senior delegate, or site manager – who holds a different frame of reference and acts in accordance with it over time can sway the frame in action of the whole party.

Overall, the decision to designate these two case studies as critical cases is supported by the findings. The cases provide evidence on what can bring parties together to consider a more cooperative approach. The unprecedented change in external circumstances which occurred after the project started – brought on by the COVID pandemic – also confirms the value in understanding the dynamic nature of frames of reference in practice (a contribution of the Bray et al. framework previously noted by Johnstone and Wilkinson, 2017).

Our second, more general, objective was to propose ways in which frames of reference analysis can be used more actively, more broadly, and more strategically. The active use by the AMWU of the Bray et al. typology on frames of reference on cooperation to frame the project and invite the companies and other participants into it, suggests that frames of reference are a concept that can be used by practitioners, to identify where they are, and where they’d like to be. The two case studies provide nascent support to the union's initial proposal that ‘collaborative pluralism’ has potential to improve WHS in unionised workplaces, specifically how cooperation between unions and management could promote good psychosocial safety. The fact that the context for the project changed so unexpectedly and radically does not diminish what was achieved, particularly in Company 2. Union and employee voice was said to have improved the quality, practicality, and acceptance of the MHAP. Mutual support by union and management helped to overcome historical employee and union mistrust of management initiatives (e.g., EAPs) and, similarly, management mistrust of union safety campaigns run on-site. Union officials from off-site, the consultants, the academics and the Steering Committee Chair brought in new knowledge and skills, supporting site delegates as well as managers in communication. In particular, union officials supported delegates in encouraging workers to accept and participate in plans. While some of these benefits could have been realised under ‘adversarial pluralism’, the reciprocity of ‘collaborative pluralism’ provided an integrated approach, making better use of existing resources and building on them. An optimist would suggest that a positive experience of cooperation on one issue may predispose parties in the expansion of their joint agenda for the future.

Through observation and discussion, unions or management may find the frames of reference typology useful to identify the dominant perspective characterising the relevant parties’ actions on psychosocial risk management. In this, future research may confirm with more confidence the observations we make – that some types of initiative are more likely to be associated with some dominant frames of reference in action. With these insights comes a more realistic assessment of whether cooperation is a viable alternative. If the dominant frame of reference in action for management is autocratic unitarism, it is unlikely that cooperation with unions on this (or any) issue will be on the table. Only management turnover or a change in regulation may change this. Similarly, if union action is strongly consistent with radicalism, only a new union ticket may open the way to cooperation. Second, the observation and discussion process may reveal some diversity of perspective within union or management. If this diversity includes adversarial pluralism (for unions or management) or consultative unitarism (for management), the opportunity may exist to build a coalition within or across the parties, with those more willing to share information, build trust and skills, and create plans on a common issue like improving psychosocial safety. As noted earlier, the boundaries are not always clear and can change over time or on different issues. There are risks in this approach to trying to shift a dominant frame of reference in action, but also opportunities.

In seeking to draw broader lessons, there were case factors that acted as a support, and later became more of a burden, in seeking to build a more cooperative relationship between union and management. For example, attempts to frame psychosocial risk management as a common interest gained some traction, but the continued uncertainty about the regulatory environment, and the physical and psychological effects of COVID-19 – at work and at home-started to wear away at commitment, particularly in the face of production challenges. The reporting requirements attached to the government funding initially provided structure and process discipline and were certainly consistent with the good operation of any effective WHS initiative or broader management system (see Metzler et al., 2019), but these responsibilities took a toll as COVID slowed everything down and made site engagement more difficult. This experience is highly consistent with earlier cited literature on union-management cooperation; that is, attempting cooperation is rarely done; and even with many contextual factors working in favour of the agents, it is difficult to sustain.

At a strategic level, frames of reference analysis has potential to improve understanding IR and WHS issues, separately and together. It provides a descriptive (rather than prescriptive) way to understand the nature and practical consequences of different values and assumptions, that is, frames or reference are ‘about how you see things, not how you’d like others to see them’ (Budd and Bhave, 2008). It helps us explain to ourselves and others why union or management in one workplace or context acts in a different way or seeks different outcomes to another. Illustrative examples can be developed for other common issues; and would be particularly useful in circumstances with other influential stakeholders, including regulators; and where there is crossover in jurisdictions. One specific context could be sexual harassment (itself a serious psychosocial hazard), where unions and management actions are influenced by different regulatory regimes and bodies at the state and federal level; where there are many disciplinary perspectives; and powerful advocates for very different ways to address manage the risk. The typology could also be refined further to reflect industry differences; for example, in hospitality or mining. Finally, various countries, with their institutional and legal differences, may have different parameters or examples to be reflected in the frames of reference typology. Frames of reference analysis has the potential to generate additional theoretical propositions to be tested in subsequent studies of wider scope.

Conclusion

The objective of this paper was to explore how the frames of reference held by unions and management on cooperation in the employment relationship, and the relationship between the parties, could explain approaches to psychosocial safety at work. As demonstrated in two critical cases, a typology of frames of reference can provide a valuable lens through which to interpret actions, how they change over time, through agency or context, and the implications for cooperative efforts on this and other WHS and IR issues.