Work as a government ‘scientist’: Professional status security during organisational change and downsizing

Abstract

An increasingly complex and competitive funding environment has led many public research organisations to intervene in the organisation of scientists’ work. Government scientists represent a major component of the public research workforce but little is known of how workers in this sector have fared amid institutional and organisational change. Government scientists have applied research objectives but must still meet the profession’s requirements for legitimisation through independent research and peer-reviewed publishing to be recognised professionally as ‘scientists’. This study examines the impact of organisational change and downsizing in an Australian government science agency on the professional status security of its scientists. It uses a mixed method approach involving one focus group, 22 in-depth interviews and 803 survey responses. It finds that structural change reduced scientists’ professional status security and that a context of job insecurity and limited external employment opportunities inhibits individual scientists’ capacity to resist adverse changes to their work conditions.

Keywords

government scientist job insecurity matrix structure organisational change professional status public sector employment

Context

Despite being a traditionally elite and independent profession, research scientists have had mixed success resisting reductions in their professional autonomy and degradations of their job security. Since the 1970s and 1980s, neoliberal political agendas have changed the funding of public science and the character of public research organisations (organisations whose primary funding is public monies) (Connell, 2015). Increasing selectivity in public funding has constrained scientists’ independence in research problem identification, limited the time horizon of scientific projects and increased demands for immediate and applicable research outputs (Harley and Lee, 1997; Halsey, 1992). More recently, reforms have focused on increasing the performance and productivity of publicly funded research (Meek, 2002). The associated pressure for measurable and marketable research output has further circumscribed scientists’ research activity, pushing researchers into activities and fields more productive in research metrics (Benedictus et al., 2016; Cannizzo and Osbaldiston, 2016).

The increasingly complex and competitive funding environment has led public research organisations to increase managerial control over scientists’ work (Barry et al., 2001). The introduction of principles and practice from the private sector under this new managerialism has changed the structure of many public research organisations (PROs) into more centrally controlled (Clark, 1996), enterprise-like entities (Marginson and Considine, 2000) with management systems designed to enhance stakeholder responsiveness (Etzkowitz et al., 2000). Many have interpreted these changes and the increased managerialism associated with them as commodifying scientific labour, deforming scientific organisations and distorting the principles of ‘science’ (Dearlove, 1997; Slaughter and Leslie, 1997). Others express more benign views, arguing a need for increased democratisation and accountability in the traditionally elitist scientific institution (Bäckstrand, 2003; McCormick, 2007).

These changes in public funding and expectations have increased the job insecurity of research scientists. Job tenure security reductions have resulted from increased use of flexible (precarious) employment (Kimber, 2003; Santiago and Carvalho, 2008). Job status security, relating to qualitative job characteristics (Gallie et al., 2017), has also reduced, as productivity demands and performance surveillance have increased work intensification and decreased scientists’ autonomy over task allocation (Cannizzo and Osbaldiston, 2016; Halsey, 1992; Houston et al., 2006). Both job status insecurity and job tenure insecurity are bad for employee health, well-being and performance (Witte, 1999). Despite these changes, permanently employed university researchers retain a high degree of autonomy compared to other professional groups; they are generally able to select research collaborators independently, design research projects within their designated research theme and determine their own processes of production (Barry et al., 2001; Garrett-Jones et al., 2010). This level of professional autonomy, along with the capacity to maintain professional credentialling through peer review publishing, has meant that institutional and organisational changes have had limited impact on the professional status security of permanently employed university scientists.

Little is known about how government scientists have fared in the face of institutional and organisational change. Compared with public universities, government science organisations are more politicised establishments and have more explicit, applied social, economic and political objectives. As a result, government scientists have long had a more limited scope of professional autonomy than public university scientists (Kornhauser, 1962). Government scientists’ professional activity is complicated by the need to balance public demands for solutions to major national problems with the research science profession’s requirements for legitimisation through peer-reviewed publishing (Gibbons et al., 1994). Recent studies of job insecurity have found government employees at heightened risk of both job status insecurity and job tenure insecurity, particularly under post-financial crisis austerity (Gallie et al., 2017). Studies of job insecurity among public employees use their sector of work as the primary classification without establishing associated professional classifications, such as research scientist. As a result, the impact of organisational change and austerity on public sector workers from specific professional groups is unclear. Professional status security is an important but under-studied aspect of job insecurity in public sector employment.

Professional status and organisational control

Professional status is the professional membership and positioning of an individual in the status hierarchy of their profession (Abbott, 1981). It is intertwined with job status and depends on qualitative job characteristics that allow a professional to maintain their professional membership and advance their professional positioning. Professional status security has implications beyond an individual’s immediate employment, extending to their capacity to maintain and advance their career in the profession for which they have trained. Membership of the research science profession is granted through demonstration of capability as an independent researcher, typically through completion of a doctorate of philosophy. Professional membership is maintained through ongoing, peer-reviewed scientific publishing, which also serves as the basis for contestation of positioning in the status hierarchy (Cole and Cole, 1973; Merton, 1973). For scientists, this status is necessary to ensure continued opportunity to advance their reputation, salary and rank within the professional hierarchy as producers of scientific knowledge, for those who only apply existing knowledge are unable to compete as ‘scientists’ (Gibbons et al., 1994).

Elite professions tend to be reasonably well protected from status loss in broader social hierarchies as they have maintained demand for their professional knowledge, an entrenched monopoly over it, and a system of internal self-regulation. The relative absence of external status competition for elite professions is countered by high levels of internal competition between different specialisations and the professionals within them (Abbott, 1981). Compared with other elite professions such as medicine and law (Rothman, 1984), professional credentialling in science is more frequent because the ‘publish or perish’ dictum has long dominated status determination (Merton, 1973). The need for high-frequency credentialling has intensified with the growing centrality of research metrics to performance measurement and reward at institutional, organisational and individual levels (Benedictus et al., 2016).

As workers, organisational systems of work design, work allocation and internal authority are important determinants of individuals’ professional status and status security (Macdonald, 1995). Professional status security in employment is most important for elite professions such as science with high costs of entry (e.g. extended education) that require ongoing credentialling to maintain professional membership and have strong internal hierarchies. Given the high cost of entry, there are significant opportunity costs for members of these professions who stagnate in low-status positions, and high real costs for those who lose their professional membership as a result of failure to maintain the requisite credentials (Abbott, 1981).

Conflict between professionals and their employing organisations has long arisen as a result of differences in priorities and values (Kornhauser, 1962). Clark (1996) described research specialisations and research organisations as the ‘grand matrix’ within which competing priorities for scientific work must be managed (Clark, 1996: 419). Many of the organisational changes enacted by PROs over recent decades are designed to better align the activities of their researchers with organisational and stakeholder objectives. Some PROs have established matrix-like units in their organisational structures to better enable external engagement and enhance collaboration across disciplinary groupings (Etzkowitz et al., 2000). These units sit outside traditional discipline-based departmental structures in the form of research centres and technology or knowledge transfer offices. They draw personnel from across departmental units for temporary or secondary roles (Boardman and Bozeman, 2007; Debackere, 2000).

Scientists working in these types of cross-disciplinary research centres or units in addition to their home department can suffer work intensification and role strain as their department’s objectives and priorities often differ from those of research centres (Boardman and Bozeman, 2007). However, these represent single units that cut across the traditional, departmental format rather than matrix organisational structures. In these cross-departmental units, scientists largely remain ‘free agents’, able to withdraw from participation in the secondary research role if their expectations are not met or expertise development inhibited (Garrett-Jones et al., 2010: 452).

Little is known of how the implementation of a matrix structure across an entire PRO would impact the professional status security of research scientists. The structures involve two permanent, cross-departmental group structures: one based on functional or disciplinary groupings and one based on output groupings, such as projects, products or customers. They are a method of coordinating cross-departmental work by drawing workers from a range of specialisations around a particular product line or customer group (Ford and Randolph, 1992).

Matrix structures are also a divide-and-conquer method for debasing professional power by requiring professional priorities be subordinated to output objectives (Kuprenas, 2003). The implementation of these structures is an act of centralisation designed to curtail the independence of functional/professional groupings. Research in engineering firms has found that matrix structures can reduce the autonomy for professional employees (Denis, 1986), but the impact on the professional status security of individuals in traditionally elite professions such as science has not been considered. Nor has the role of fiscal austerity on a PRO’s implementation of a matrix. Matrix structures introduce a layer of organisational control over research scientists beyond that achievable under traditional, discipline-based structures. Can membership of a traditionally elite profession protect against professional status insecurity or deprofessionalisation under a matrix organisational structure? How does the context of funding cuts influence scientists’ capacity to resist threats to their professional status security?

This study addresses this gap in existing knowledge. It presents a mixed method study of how the implementation of a centralised work allocation system under a matrix organisational structure in a large Australian government science organisation (henceforth GSO) impacted the professional status security of scientists. GSO reorganised its traditional, functional structure into a matrix structure as part of a strategy to coordinate cross-disciplinary research and whole-of-organisation solutions to problems of significance to its external stakeholders. Output groupings based on these major problems were created while the existing discipline-based divisions remained as the input groupings.

The structure was implemented incrementally and was first introduced during a period of stable public funding to the organisation. However, the period during which some of the most substantial reorganisation of scientists’ work was being implemented coincided with a prolonged period of reductions in real funding that led to a substantial reduction in staff numbers. This research was conducted in 2010–11 at the beginning of this period of funding decline. Between 2010 and 2016 there was a reduction of more than 20% in research scientist employees, with similar reductions in research support staff (GSO, 2010, 2016). Within this context, this study addressed the research question: how did the centralised work allocation system under GSO’s matrix impact the professional status security of its scientists? The following section will present the methodology for this study.

Methodology

This study used a mixed method research design involving one focus group of four participants, 22 in-depth interviews and 803 survey responses. Integrating qualitative and quantitative methods is valuable when investigating multi-level phenomenon (Currall and Towler, 2003) such as job insecurity, which involves an interplay between individual and organisational factors. This research uses a QUAL > quan approach, meaning that the dominant data are qualitative and that the phases were separate and sequential (Morse, 2003). Method combinations are valuable in developing complementarity between methods and convergence of research findings (Morgan, 1998). The survey research was designed to help evaluate qualitative research findings; it allowed attitudes and experiences addressed qualitatively to be assessed for broader applicability among survey respondents. This supports identification of the scale of impact across the population and the relative experiences of different groups. Survey data were also collected on factors more easily reported quantitatively, such as working hours, or factors that can benefit from scale ratings such as work attitudes and level of work overload.

The study was designed to develop a detailed understanding of scientists’ experiences of work under a matrix structure. Primary data was collected with the support of the GSO staff association and secondary, administrative data about organisational structure and strategy was collected from annual reports from 2000 to 2016. The first stage of data collection was an exploratory focus group. Focus groups offer an opportunity to gain an insight into the nature, language and key issues in the context of study (Finch and Lewis, 2003). In this study, the focus group sought to develop a general understanding of employees’ perspectives of the structure, terminology and key employment issues being experienced by staff under a matrix.

The interview component of this study was designed to achieve contextual explanatory data (Ritchie et al., 2013). To achieve this objective, semi-structured interviews based on open questions were conducted. This allowed the researcher to guide the topic of discussion while maintaining flexibility in discussion to allow the research participant to introduce relevant or expanded interpretations of the subject matter. Open questions put the onus on research participants to supply the content of the answer, hence allowing for a stronger participant voice (Patton, 2005). Interviews addressed the nature and organisation of work at GSO, with themes including employment conditions, worker control and autonomy, authority structures, the reward system and professional development. As university researchers, the interviewers had a degree of familiarity with key issues affecting GSO scientists’ work and working environment, but were not insiders from the perspective of shared discipline, sector or organisational membership.

Purposive sampling was employed with the aim of recruiting participants from across the salary structure to identify any variation in experiences of organisational change within the internal hierarchy (e.g. Were mid-career scientists more effected than junior scientists? What affect did existing authority have on response to change?). The GSO salary structure is similar to that in Australian universities, with scientists (except the most senior leaders) ranked on a five-level structure from entry-level/post-doctoral researcher to senior research scientist. All qualitative data were transcribed verbatim and data analysis was conducted using NVivo software and thematic analysis. This method of analysis allows for the identification of recurrent and emerging themes from research participants through a data coding process.

The aim of the survey was to achieve a representative sample to assess the generalisability of qualitative findings within the organisation and to generate data about the working conditions of research scientists at GSO such as working hours, workload and work distribution. It was also designed to validate qualitative findings and support the methodological convergence within the research project. Attitudinal items were rated on a five-point Likert scale. Demographics were also sought, along with salary level and management responsibility. Space was also provided for open-ended comment at the end of the survey. The survey was handed out in person by staff association representatives. The responses of 803 research scientists were received, including 274 comments. The response rate was 49%. The survey population was biased towards union members with 63% of respondents being members compared with 47% in the overall staff population. Analysis of survey responses did not, however, find significant differences between members’ and non-members’ responses.

Findings

The matrix structure reduced scientists’ job status security generally, and professional status security particularly. Research participants were generally supportive of the organisational strategy to enhance collaboration across disciplinary research departments but critical of the use of a matrix structure to achieve it due to its negative impact on their professional autonomy and professional relationships. Only 14% of survey respondents reported that the matrix enhanced the organisation’s ability to produce quality science and scientific solutions. Rather, it was seen as creating a consultancy style business for organisational stakeholders in industry and government departments. A particular criticism was the reduction of scientists’ autonomy to develop their own research rather than follow the agenda set by output units and external stakeholders. Only 20% reported that the matrix provided opportunities for them to do work they are passionate about.

The employment context at GSO was impacting scientists’ response to organisational changes. In addition to substantial reductions in scientific staff numbers, changes in research priorities generated staff churn through redundancies in discontinued fields and appointments in new ones. This context created a strong sense of job tenure insecurity among GSO’s research scientists. Only 34% of survey respondents (38% of permanent employees) reported that their job was secure. Poor external employment alternatives were also reported, with only 52% of survey respondents reporting that they could likely find alternative employment if their job at GSO was discontinued. The perceived limitations in the external labour market constrained scientists’ opportunities for resistance to GSO’s unfavourable organisational conditions or exit from the organisation.

Exit was also subdued by high levels of organisational commitment; 86% of survey respondents reported that they really cared about the future of GSO. Morale was being degraded by the negative impact of the matrix on research scientists. Many felt their commitment was exploited when poor working conditions were imposed under the matrix. Only 37% of survey respondents reported feeling personally valued by GSO. An additional factor influencing scientists’ responses to the matrix was a widespread perspective that the structure was an ineffective ‘management fad’ which would eventually be unwound, as many previous organisational structures had been. From this perspective, poor working conditions were a temporary hardship to endure. The following section will examine how work was organised under the matrix at GSO.

Organising work under the matrix

Under the matrix, scientists’ work was organised using an organisation-wide work allocation system in which output departments sourced staff from input departments. Project funding was controlled by output line management, which purchased units of staff labour from input line management. Input line managers retained responsibility for scientists’ career development and were provided with some specific, though limited, funding for this purpose. The impact of the matrix on individual scientists’ work varied widely. There was substantial overlap between the research profiles of some input departments with output departments. Individuals in these input departments were largely excluded from cross-departmental allocations. Individuals working on a single, long-term research project also experienced little effect. For many, however, the introduction of the matrix marked a significant shift towards more direct, centralised control of scientists’ work.

Under GSO’s work allocation system, employee time was allocated on a proportional full-time equivalent basis. For example, a scientist could be allocated 40% to one project and 60% to another, and may or may not be co-located with project peers. Survey respondents were asked how many projects they were allocated to; the numbers reported ranged from 1 to 11. In format, the allocation system at GSO held similarities to those utilised in universities as tasks and workers’ time were unitised to measure and distribute workload (Hull, 2006; Houston et al., 2006). In content, however, it differs significantly as allocates time to specific research projects rather than broad categories of work (research, teaching, administration, etc.).

Analysis of work allocation across employment types illustrates significant differences across categories. The survey population consisted of 80% of permanent employees and 20% temporary (casual or fixed-term) employees. Figure 1 presents the project allocation of survey respondents by employment type. It shows that the majority (57%) of temporary employees were working on a single project. Of those who were not, the median project spread remained significantly lower than that of permanent employees. This suggests that temporary employees were being used as numerically flexible labour, as they were likely employed for specific, single projects. The project allocation of permanent employees was far more diffuse, with 50% of permanent survey respondents reporting allocation to four or more projects.

Figure 1.

Number of projects by employment status.

This work allocation system was in use at GSO prior to the adoption of the matrix as a method for managing project workloads and budgets. However, in seeking to facilitate organisation-wide allocation, the possibility of scientists being allocated to projects outside their home department or research group increased. The precarious internal and external employment contexts weakened scientists’ capacity to resist unfavourable work allocations. Individuals were expected to be 100% allocated or risk redundancy. Organisational flexibility and agility are often accompanied by fear and insecurity (Gillies, 2011) and this was indeed the case at GSO.

Working under the matrix: impact on professional status security

The deconstruction of professional activity and fragmentation of professional relationships were the primary criticisms of the matrix structure. By drawing much of the decision-making around research development into a central arena, the matrix reduced some scientists’ capacity to be complete producers of scientific work. This posed a significant risk to the professional status security and career prospects of affected scientists as the autonomy to develop research problems is not just a prized professional reward but a key foundation of professional status. Professionals have long been managed under systems of responsible autonomy; such systems balance organisations’ desire to control production with the need to maintain worker motivation (Friedman, 1977). In research science, individuals must demonstrate conceptual capabilities to advance their careers through the achievement of professional output such as publication, particularly at mid- to senior career stages. Numerous research participants at these career stages reported substantial impediments to their professional autonomy under the matrix which hampered their capacity to produce scientific knowledge rather than enact applied scientific analysis.

I have no control over what I do since the introduction of the matrix. GSO work is all dictated from the top. There is no scope for independent thought, new ideas or independent research. There are certainly no resources available for independent work. Science is stuffed at GSO. I am reduced to cranking the handle of a calculator to churn out numbers. Desperate! Real research at GSO died a decade ago. (Mid-career survey respondent)

PROs have long been increasing the centralisation of decision-making around research strategy, but typically around research themes only (Clark, 1996). The transition to the matrix structure at GSO differs markedly as it shifted some conceptual responsibility for research projects away from individual scientists. Under the matrix restructure, some mid- to senior-level scientists lost their autonomy to generate new science projects. The quote below illustrates how resistance to this loss could prevent the securing of full allocation (a risk to ongoing employment):

I think the other problem that we are just identifying too is it’s difficult to start off new work in the matrix structure because there’s no projects that want you because you want to do new work, so it’s actually very difficult … it’s almost impossible to start a new work. (Senior focus group participant)

The precarious employment context limited individual resistance to the organisation’s intrusions on scientists’ professional autonomy. Job tenure insecurity and limited external employment opportunities meant many scientists did not see individual resistance to unwanted changes in their work environment as a feasible option.

The most substantial challenge experienced by scientists under the matrix was allocation to an area outside their area of current research specialisation. Specialisation remains the basis of professional status for research scientists and the foundation of the professional reward system (Aldrich, 2014). For scientists to maintain professional membership and succeed in professional status contestation they must make ongoing contributions to a knowledge specialisation (Latour and Woolgar, 1979; Trowler and Becher, 2001). The scale of difference between an individual’s existing expertise base and that required in their allocated work influenced how negative the impact would be. This varied across the organisation, but most were affected in some way: only 38% of survey respondents reported that the matrix did not require them to work in an area outside their expertise. Allocation to an unfamiliar area of research could stagnate a scientist’s professional status development due to the time required to develop the familiarity sufficient to contribute new knowledge to that field. The following narrative from a mid-career research scientist illustrates the personal and professional significance of such allocation.

I was summoned to the [boss’s] office and he just told me: ‘I want you to work on [this research topic]’. Just like blank; he didn’t say, ‘[Alex], do you want to?’ And it was really hard for me to take because we scientists, you know we put a lot of ourselves in our research. And my work on [the previous research topic] was very much stuff that I’d initiated all by myself and essentially I was told to drop all this and to go and work on that [new research topic]. I also had ethical issues about the new research and the impacts are very minimal so essentially we’re talking about you know a few dollars less to very rich [industry partners] … but I was never given a say on the subject. So decisions affecting people’s careers have been made increasingly, I wouldn’t say random but almost. It’s going to be very damaging to my career because all my expertise on [the previous research topic] is not going to help me with my new work on [the new topic]. And conversely all the things that I do in [the new field] will mean that my existing expertise just gathers dust …

You know I’m in science not for the money, not for the glory, the girls, the women, no. I’m in science because I love science and I love research and that’s been motivating me from, you know, from my student days. Now it seems to me at like 42 years old I am finding myself more or less … more and more stripped of any intellectual control in my research. So, more and more of what I say and what I do is stuff that I’ve been told to do. (Mid-career interview participant)

The process through which work was allocated also limited individual resistance as it largely excluded individual researchers from allocation decisions. An analogy employed by a mid-career survey respondent aptly captures scientists’ overall perspectives of work allocation under the matrix; she described the process as ‘horse trading’, in which managers met to place and trade individuals for project work. Under this system, many scientists felt that their labour was being commodified and that the impersonal, mechanistic nature of staff identification and allocation led to alienation. As a result, many felt that their commitment to GSO and its mission were being betrayed, and that their work efforts were being exploited. This discontent was particularly strong among those with long organisational tenure who often reflected on the transition to the more remote, depersonalised system of work allocation of the matrix. The quote below from a senior scientist illustrates how the quantification of staff for centralised allocation facilitated a mechanistic allocation of their labour.

We always had identity number, as soon as we were signed on, we were given identity number, but we are always known by the name rather than the ident. Ident was something that was in the system. Now we are addressed by ident rather than the name…. Hence the ease of moving staff from one block section to another section, because we’ve become a number. (Senior interview participant)

Increasing centralisation of decision-making can exacerbate employees’ sense of powerlessness (Blauner, 1964; Mottaz, 1981; Tummers et al., 2009). Alienation can also exacerbate work burnout and emotional exhaustion (Karger, 1981; Shantz et al., 2014). The following comment illustrates these outcomes:

This [matrix] system results in output managers having little or no long-term interest in the career of the project participants: I am taken off the shelf when I am useful and then put back until the next time feeling drained and unhappy. (Mid-career survey respondent)

The fracturing of research groups and departments under the matrix was contributing to many scientists’ feeling of isolation. Creating an environment that fractures professional networks is particularly harmful for research scientists, who rely on interaction with disciplinary peers for professional development (Aldrich, 2014; Becher and Trowler, 2001). The following quote illustrates how the organisational restructure into a matrix had fractured workgroups and individuals’ sense of organisational membership.

The matrix has done more to fracture the organisation and people’s places in it than anything else. People are ‘tribal’ by nature and need a ‘home’ base for their everyday working lives. [GSO] is the warm fuzzy brand we cling to because our divisional foundations have been degraded/dismembered over the years. (Early-career survey respondent)

As the matrix structure was designed to enhance cross-departmental collaboration, research participants were also asked about their experiences of this under the matrix. Those participating in cross-departmental projects generally reported a lack of time to develop enduring cross-disciplinary relationships and embed cross-disciplinary learning. The quote below illustrates this sentiment. It also reinforces existing research about the challenges of geographically dislocated multidisciplinary teams (Cummings and Kiesler, 2005).

There’s a couple of things. One, it’s hard to work with people that you’re not located with because you just don’t have that ongoing interaction that makes working together easier. And there’s the lack of time that you’re allocated. You’re not given enough time, both in terms of percentage of your time now, and a time horizon, to establish the relationship that you really need to for that to be productive. (Mid-career interview participant)

The quote also extends evidence about the fragmentation of professional peer relationships under the complex organisational structure. The ease with which fragmentation could occur was evident in the mechanistic work allocation system used. The nature of its impact on content and context of scientists’ work is illustrated in the quote below.

Now that it’s all computerised, and we are all numbers, they can just unplug one person and plug them into another project; they do that a lot. At least we used to be able to stay within the same research program. Each program tends to occupy a particular building or a particular area of the building. So if were moved from one project within a program to another one, it’s not that much of a disastrous move, but when we are plucked from one program to another program it can be, because the direction completely changes … really completely…. The research emphasis changes, not just the project per se. (Senior interview participant)

The low levels of individual resistance to the matrix structure in general and the work allocation system in particular was countered somewhat by passive and/or informal resistance by allocation managers. The following section will investigate managerial perspectives on the implementation of the work allocation system under the matrix.

Managerial perspectives on work allocation

The allocation of work under matrix structures often involves highly politicised placement negotiations between managers (Engwall and Jerbrant, 2003) and this was indeed evident in GSO. Input line managers and output line managers negotiated the purchasing of labour units based on documented capabilities (research skills) and cost (pro-rata salary level). Staff were distributed as units of labour across projects and costed accordingly. Scientist managers involved in project allocation highlighted a range of challenges in allocating work under the matrix. Senior managers and scientists with long organisational tenures discussed the systemic risks for researchers and for project management under the matrix. The quote below from an output leader illustrates concerns regarding the job tenure and job status risks associated with the allocation system:

I had a strong sense from the very beginning that with the matrix we had just created a market economy in the organisation, and with potent, potent, unleashing forces of supply and demand in a market economy. Even though no-one was … it was distasteful to talk about purchaser and provider, I thought it was more useful to think of it just as a straight market. And we set up a market that didn’t really have any market regulations systems, you know they were fairly loose, there weren’t very clear rules on how to operate. We had no social welfare system and we were assuming that basically everybody would get allocated …

The deep issue, the much deeper issue in it is because we had set up this market, for me a deep problem was the commodification of people. The people suddenly became you know a percentage that you had to get mapped into your [department]. Or you thought first and foremost have I got them allocated, have I got the funding there, and even the most compassionate soul on either side starts to lose the sense of that, the integrated people management side of things. And I think we found that very hard to even recognise this happening. The fact that we talk about allocation or the way that even the language we use around that to me sends, you know it is a subliminal message that goes out through the place. (Senior interview participant)

Managerial mechanisms enacting more direct control in research science organisations are typically delegated to scientist managers in attempt to secure legitimacy (Anderson, 2008). Yet these managers are often not well prepared for or supported in their role, are offered few professional rewards for undertaking them, and may resent the bureaucratic controls that they are tasked with implementing (Winter, 2009). In universities, academic managers have been identified as important participants in academic resistance to managerial controls (Anderson, 2008; Benmore, 2002).

At GSO, the challenges of work allocation for scientist managers were both practical and personal. Scientist managers assigned responsibility for work allocation were challenged by the need to meet organisational requirements while wishing to protect the jobs and careers of employees whose labour they were managing. Matrix structures separate the priorities of project outcomes and staff capability. This separation can generate conflict unless a mutual, long-term perspective is developed between input and output managers, as output managers do not have immediate incentive to protect capability development (Katz and Allen, 1985). This conflict was relevant to GSO as a result of short-termism in available research funding, external demand for stakeholder responsiveness, and of the broader context of organisational decline. Being assigned responsibility for scientific careers, input managers prioritised full allocation of their staff because without this, individuals’ employment with the organisation might be threatened. Output managers prioritised project performance and hence securing the best and most appropriate scientific skills to achieve project objectives.

Interview data from input and output managers illustrated the conflict that emerged during work allocation and the potentially personal nature of their role in allocation. Input managers in particular, reported substantial concern and feelings of personal responsibility about the potential consequence of redundancy for staff members if they were unable to allocate their time fully. Numerous managers described the negotiations about staff allocation as based on good intentions and recognition of the structural rather than personal source of the conflict. Others, however, reported that the systemic failures led to hostile and manipulative behaviour. The following quote from an output manager illustrates the conflicting priorities in allocation management and the associated pressures for managers on both sides. It also highlights the potential mismatch between demand for and supply of particular research skills. Traditional, department-based organisational structures better support alignment between research skills and research projects as a result of more integrated research planning.

Mostly [we started] off with shrapnel … you know, bits of people that weren’t doing something else. And then it was a very protracted negotiation process with the other divisions because of course what they did was give me all the people that weren’t allocated and often they didn’t have the skills to match or they weren’t very good deliverers. My job is to get out an economics report and an ant expert is not going to be able to do it. I mean you know Blind Freddy could see there’s a problem with the way they’ve structured it but I don’t know if Blind Freddy up the top is seeing it.

[The input manager’s] job is to put bums on seats otherwise he’s going to have to make people redundant. So he’s doing his job too. But he’s not responsible for science outcomes, he’s responsible for bums on seats. So he will do anything he can to threaten, bully and make me accept a person that needs a bum on a seat when I know that that person is not going to deliver. (Senior interview participant)

The backdrop of ongoing redundancies represented an important driver of input manager behaviour that could counteract relevant or appropriate allocations. Housed within the same division, input managers were generally personally familiar with the staff for whose allocation they were responsible. Due to a desire to limit the harshness of adverse consequences of the matrix and as a means to deal with the complexity of the system, many scientist managers were engaging in informal organisation. This was most evident in the documentation of allocations in group budgets, for it was there that under-allocations would be most evident, and that administration was most complicated. As the quote below illustrates, some managers did not fully disclose allocations in their internal budget and used nominal work allocation to achieve full allocation of all staff.

I guess the biggest challenge is that we’re picking up costs from other divisions and yet we’re having to meet our own costs for people who aren’t allocated, and that makes it difficult. Last year there was like a gentleman’s agreement where, ok, we’re picking up five full-time equivalents from [that division] and they’re probably picking up five of ours, so we’ll just do it and not worry about charging each other. But then there were the rogue divisions who said ‘no way, I’m charging ya’ so you’d get charges arriving in your project budgets for these people we’re being charged for. We had a person allocated to our group who we didn’t even know. (Mid-career interview participant)

Avoidance and technical compliance are two common forms of resistance identified in universities as mechanisms to deal with unwanted or overly burdensome administrative requirements (Anderson, 2008). The challenges associated with matrix management and the employment risks associated with the organisational context provided motivation for these forms of resistance at GSO. For allocated employees, informal organisation by managers may have added to confusion about the content and location of their work, but all were grateful for managers’ attempts to mitigate the harsh effects of matrix-based work allocation.

Conclusion

This article examined how the organisation of work under a matrix organisational structure at GSO impacted the professional status security of research scientists. Almost universally, the structure inhibited scientists’ opportunities to do work they were passionate about and reduced morale and scientists’ sense of being valued by the organisation. The impact on scientists’ professional status security varied both across and within departments. For scientists whose work was negatively affected by adverse work allocations, working under the matrix reduced their professional autonomy and fragmented their working relationships. For some, the change generated significant disruption in their professional development within their research specialisation. For others, it led to work routinisation and/or an inability to operate as a complete producer of scientific work. All of these factors pose threats to scientists’ capacity to compete for professional status, and in some instances, threatens their professional membership if individuals are unable to maintain professional relationships within and contributions to their research specialisation.

The structural changes within the organisation were experienced in an environment of high job tenure insecurity and limited external job alternatives. This context, and the heterogeneity of scientists’ experiences of the matrix, inhibited affected scientists’ capacity to individually resist unfavourable work allocations or working conditions. A degree of autonomy over research direction-setting is not just a professional luxury for research scientists but a necessity to ensure the maintenance of professional membership as complete producers of research science work (Gibbons et al., 1994). The findings demonstrated that middle managers can mitigate the harshness of organisational interventions and serve as resisters of organisational change through informal organisation.

The scale of intervention into the organisation of scientists’ work at GSO may serve as a warning of things to come. Many PROs are grappling with external pressures for relevance and responsiveness in increasingly austere public funding environments. Many PROs experience similar pressures to GSO to generate stakeholder-relevant cross-disciplinary research and limits on public funding availability. This study illustrates that cross-divisional work allocation can reduce scientists’ job autonomy to the point where it risks undermining their professional membership and status. Many mid-career and senior scientists reported being stripped of the autonomy that had allowed them to be complete producers of scientific research. In these cases, there is evidence of de-professionalisation.

Recognising the inappropriateness of the matrix structure for the organisation, GSO commenced a devolution of the matrix back into a traditional functional structure in 2015 with the exception of some cross-disciplinary units. This will alleviate some of the adverse working conditions associated with the matrix organisational structure. However, the lack of autonomy resulting from stakeholder prioritisation and the structural instability resulting from the politicisation of government science funding risks presenting a disincentive for early-career scientists to pursue government employment. The ongoing degradation of public science employment in Australia generally, and in the government sector particularly, contradicts the federal government’s promotion of science education and careers. More attention should be paid to the working conditions and professional opportunities in this sector as Australia is already a net loser of domestically qualified research scientists (Hugo, 2005) and recent industry research has found that large numbers of publicly funded researchers, up to 80% in some fields, are considering leaving their career in science as a result of limited funding and career opportunities (ASMR, 2016; PSA, 2018).

The cross-departmental nature of primary data collection in this study meant that self-reporting was relied upon to understand the changes associated with the introduction of the matrix structure. Longitudinal research is needed regarding the evolving organisational structures of PROs and their impact on scientists’ professional status and status security. More research is also needed to advance knowledge about the working conditions and professional status security of scientists employed by government science organisations, which remains underdeveloped compared with universities. Indeed, contemporary research on these themes is underdeveloped generally in Australian sociology (Huppatz and Ross-Smith, 2017) despite ongoing degradations to the working conditions of Australian employees.

Footnotes

Acknowledgements

The authors wish to thank the representatives of the GSO Staff Association for the support with primary data collection and the Journal of Sociology reviewers for their feedback.

Contributions

All authors were in involved in research design, data collection and manuscript preparation. Miriam Glennie led data analysis and drafting of the manuscript.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Author biographies

Miriam Glennie is a post-doctoral fellow in the School of Business at the University of New South Wales, Canberra campus. Her research areas are inequality and organisational sociology.

Michael O’Donnell is a professor in the School of Business at the University of New South Wales, Canberra campus. His research areas are labour relations and public sector workers.

John Benson is a professor in the School of Management at Monash University, Kuala Lumpur, Malaysia campus. His research areas are labour economics and industrial relations.

Michelle Brown is a professor in the School of Management at University of Melbourne, Melbourne. Her research areas are employment systems and human resource management.

References

Abbott

(1981) ‘Status and Status Strain in the Professions’, American Journal of Sociology 86: 819–35.

Aldrich

(2014) ‘Interdisciplinarity: Its Role in a Discipline-based Academy. Oxford: Oxford University Press.

Anderson

(2008) ‘Mapping Academic Resistance in the Managerial University’, Organization 15(2): 251–70.

Australian Society for Medical Research (ASMR) (2016) ASMR Health and Medical Research Workforce Survey Report. URL (consulted 2 April 2018): https://asmr.org.au/wp-content/uploads/library/Workforce16.pdf

Bäckstrand

(2003) ‘Civic Science for Sustainability: Reframing the Role of Experts, Policy-makers and Citizens in Environmental Governance’, Global Environmental Politics 3(4): 24–41.

Barry

Chandler

Clark

(2001) ‘Between the Ivory Tower and the Academic Assembly Line’, Journal of Management Studies 38: 88–101.

Becher

Trowler

(2001) Academic Tribes and Territories: Intellectual Enquiry and the Culture of Disciplines. Buckingham: McGraw-Hill Education.

Benedictus

Miedema

Ferguson

(2016) ‘Fewer Numbers, Better Science’, Nature 538: 7626.

Benmore

(2002) ‘Perceptions of the Contemporary Academic Employment Relationship’, International Studies in Sociology of Education 12(1): 43–58.

10.

Blauner

(1964) Alienation and Freedom: The Factory Worker and his Industry. Oxford: Chicago University Press.

11.

Boardman

Bozeman

(2007) ‘Role Strain in University Research Centers’, Journal of Higher Education 78(4): 430–63.

12.

Cannizzo

Osbaldiston

(2016) ‘Academic Work/Life Balance: A Brief Quantitative Analysis of the Australian Experience’, Journal of Sociology 52: 890–906.

13.

Clark

(1996) ‘Substantive Growth and Innovative Organization: New Categories for Higher Education Research’, Higher Education 32(4): 417–30.

14.

Cole

(1973) Social Stratification in Science. Chicago: University of Chicago Press.

15.

Connell

(2015) ‘Australian Universities under Neoliberal Management: The Deepening Crisis’, International Higher Education 81: 23–5.

16.

Cummings

J.N.

Kiesler

(2005) ‘Collaborative Research across Disciplinary and Organizational Boundaries’, Social Studies of Science 35(5): 703–22.

17.

Currall

Towler

(2003) Research Methods in Management and Organizational Research: Toward Integration of Qualitative and Quantitative Techniques. Thousand Oaks, CA: Sage.

18.

Dearlove

(1997) ‘The Academic Labour Process: From Collegiality and Professionalism to Managerialism and Proletarianisation?’, Higher Education Review 30(1): 56.

19.

Debackere

(2000) ‘Managing Academic R&D as a Business at KU Leuven: Context, Structure and Process’, R&D Management 30(4): 323–8.

20.

Denis

(1986) ‘Matrix Structures, Quality of Working Life, and Engineering Productivity’, IEEE Transactions on Engineering Management 3:148–56.

21.

Engwall

Jerbrant

(2003) ‘The Resource Allocation Syndrome: The Prime Challenge of Multi-project Management?’, International Journal of Project Management 21(6): 403–9.

22.

Etzkowitz

Webster

Gebhardt

Terra

(2000) ‘The Future of the University and the University of the Future: Evolution of Ivory Tower to Entrepreneurial Paradigm’, Research Policy 29(2): 313–30.

23.

Finch

Lewis

(2003) ‘Focus Groups’, pp. 170–98 in Richie

et al . (eds) Qualitative Research Practice: A Guide for Social Science Students and Researchers. London: Sage.

24.

Ford

R.C.

Randolph

W.A.

(1992) ‘Cross-functional Structures: A Review and Integration of Matrix Organization and Project Management’, Journal of Management 18(2): 267–94.

25.

Friedman

(1977) ‘Responsible Autonomy versus Direct Control over the Labour Process’, Capital and Class 1(1): 43–57.

26.

Gallie

Felstead

Green

Inanc

(2017) ‘The Hidden Face of Job Insecurity’, Work, Employment and Society 31: 36–53.

27.

Garrett-Jones

Turpin

Diment

(2010) ‘Managing Competition between Individual and Organizational Goals in Cross-sector Research and Development Centres’, Journal of Technology Transfer 35(5): 527–46.

28.

Gibbons

Limoges

Nowotny

Schwartzman

Scott

Trow

(1994) The New Production of Knowledge: The Dynamics of Science and Research in Contemporary Societies. London: Sage

29.

Gillies

(2011) ‘Agile Bodies: A New Imperative in Neoliberal Governance’, Journal of Education Policy 26(2): 207–23.

30.

GSO (2000–16) GSO Annual Reports. Canberra, Australia: GSO.

31.

Halsey

(1992) Decline of Donnish Dominion: The British Academic Professions in the Twentieth Century. Oxford: Oxford University Press.

32.

Harley

Lee

F.S.

(1997) ‘Research Selectivity, Managerialism, and the Academic Labor Process: The Future of Nonmainstream Economics in UK Universities’, Human Relations 50: 1427–60.

33.

Houston

Meyer

L.H.

Paewai

(2006) ‘Academic Staff Workloads and Job Satisfaction: Expectations and Values in Academe’, Journal of Higher Education Policy and Management 28(1): 17–30.

34.

Hugo

(2005) ‘Some Emerging Demographic Issues on Australia’s Teaching Academic Workforce’, Higher Education Policy 18: 207–29.

35.

Hull

(2006) ‘Workload Allocation Models and “Collegiality” in Academic Departments’, Journal of Organizational Change Management 19(1): 38–53.

36.

Huppatz

Ross-Smith

(2017) ‘A Discipline at the Crossroads? Using a Gender-inspired Paradigm to Reposition the Sociology of Work and Employment’, Journal of Sociology 53: 756–70.

37.

Karger

(1981) ‘Burnout as Alienation’, Social Service Review 55(2): 270–83.

38.

Katz

Allen

T.J.

(1985) ‘Project Performance and the Locus of Influence in the R&D Matrix’, Academy of Management Journal 28(1): 67–87.

39.

Kimber

(2003) ‘The Tenured “Core” and the Tenuous “Periphery”: The Casualisation of Academic Work in Australian Universities’, Journal of Higher Education Policy and Management 25: 41–50.

40.

Kornhauser

(1962) (1962) Scientists in Industry: Conflict and Accommodation. Berkeley, CA: University of California Press.

41.

Kuprenas

(2003) ‘Implementation and Performance of a Matrix Organization Structure’, International Journal of Project Management 21(1): 51–62.

42.

Latour

Woolgar

(1979) Laboratory Life: The Social Construction of Scientific Facts. Beverly Hills, CA: Sage.

43.

Macdonald

(1995) The Sociology of the Professions. London: Sage.

44.

Marginson

Considine

(2000) The Enterprise University: Power, Governance and Reinvention in Australia. Cambridge: Cambridge University Press.

45.

McCormick

(2007) ‘Democratizing Science Movements: A New Framework for Mobilization and Contestation’, Social Studies of Science 37(4): 609–23.

46.

Meek

(2002) ‘On the Road to Mediocrity? Governance and Management of Australian Higher Education in the Market Place’, pp. 235–60 in Amaral

Jones

G.A.

Karseth

(eds) Governing Higher Education: National Perspectives on Institutional Governance. Dordrecht: Springer.

47.

Merton

(1973) The Sociology of Science: Theoretical and Empirical Investigations. Chicago: University of Chicago Press.

48.

Morgan

D.L.

(1998) ‘Practical Strategies for Combining Qualitative and Quantitative Methods: Applications to Health Research’, Qualitative Health Research 8(3): 362–76.

49.

Morse

J.M.

(2003) ‘Principles of Mixed Methods and Multimethod Research Design’, pp.189–208. in Tashakkori

Tedlie

(eds) Handbook of Mixed Methods in Social and Behavioural Research. Thousand Oaks, CA: Sage.

50.

Mottaz

C.J.

(1981) ‘Some Determinants of Work Alienation’, Sociological Quarterly 22(4): 515–29.

51.

Patton

M.Q.

(2005) Qualitative Research. Washington, DC: Sage.

52.

Professional Scientists Australia (PSA) ( 2018) ‘The Brain Drain that Is Hurting Australia’. Available at: http://www.professionalsaustralia.org.au/professional-women/wp-content/uploads/sites/48/2018/03/Media-release-The-science-brain-drain-that-is-hurting-Australia-8.3.2018–2.pdf (accessed 2 April 2018).

53.

Ritchie

Lewis

Nicholls

C.M.

Ormston

(2013) Qualitative Research Practice: A Guide for Social Science Students and Researchers. London: Sage.

54.

Rothman

(1984) ‘Deprofessionalization: The Case of Law in America’, Work and Occupations 11: 183–206.

55.

Santiago

Carvalho

(2008) ‘Academics in a New Work Environment: The Impact of New Public Management on Work Conditions’, Higher Education Quarterly 62: 204–23.

56.

Shantz

Alfes

Truss

(2014) ‘Alienation from Work: Marxist Ideologies and Twenty-first-century Practice’, International Journal of Human Resource Management 25(18): 2529–50.

57.

Slaughter

Leslie

(1997) Academic Capitalism: Politics, Policies, and the Entrepreneurial University. Baltimore, MD: Johns Hopkins University Press.

58.

Tummers

Bekkers

Steijn

(2009) ‘Policy Alienation of Public Professionals: Application in a New Public Management Context’, Public Management Review 11(5): 685–706.

59.

Winter

(2009) ‘Academic Manager or Managed Academic? Academic Identity Schisms in Higher Education’, Journal of Higher Education Policy and Management 31(2): 121–31.

60.

Witte

H.D.

(1999) ‘Job Insecurity and Psychological Well-being: Review of the Literature and Exploration of Some Unresolved Issues’, European Journal of Work and Organizational Psychology 8: 155–77.