Aim and development:

The project aimed at studying the organizational shaping of the caseworker in two of Sweden’s largest welfare state bureaucracies: the Public Employment Services (PES) and the Social Insurance Agency (SIA). The project started off with a particular interest in the system of governance by objectives and performance (with its inherent focus on transparency and auditing), and how this culture of performativity and control affects the professional role and identity but also the self and subjectivity of the front-line staff. However, during the period of study, both agencies implemented post-bureaucratic and ‘post-NPM’ organizational reforms, such as trust- and value-based governance. Both agencies have then again shifted back to stricter management by objectives and ‘trust in numbers’. These shifts in governance led to the insight that we had to focus on a variety of forms of organizational governance – not just audit culture – including paying attention to the interplay between hierarchical and horizontal governing mechanisms and processes at the workplace, including teamwork or peer control.


Implementation:

We have done more extensive data collection than envisaged in the application, studying additional local offices than originally planned for. The first reason was scientific as we realized that the organizational culture may differ in different local offices and we thus need to study several local offices in order to be able to assess the role of central governance versus local office cultures and practices. Secondly, for reasons of anonymity in relation to our interviewees (to prevent ’backdoor identification’), we wanted to study local offices in two regions in Sweden. Consequently, we have done ethnographic observations at seven local offices (five SIA and 2 PES offices) located in two regions in Sweden (the same region for both the SIA and the PES); qualitative interviews with the caseworkers in these offices, their local managers, and ‘specialists’ assisting them in case assessment; qualitative interviews at the central PES and SIA offices and with government officials; and studied policy and organizational document. Through cooperation with another research project, we have also been able to make a representative survey to caseworkers in both agencies (N= 1393). Thus, with over 100 qualitative interviews in total, rich and extensive ethnographic multi-site fieldwork, and a representative survey, the project comprises a rather unique data set and will continue to provide the basis for publications in the future.


Research findings and conclusions:

One finding is that management ideologies and governance systems in the public sector may change quickly, which alerted us to the role of normative control and regulation in the organizational governance and control, to make the workforce adaptable to changes. This insight moved us further in the direction of critical management studies of normative governance and subjectivity, as well as in the direction of sociology of emotions approaches. To capture the complex organizational shaping and governance of staff, we analyzed normative governance, management by discourse, management by emotion, management by peers and management by symbols, numbers and colours, among others.

A second finding is that we see major differences in the normative regimes of the PES and the SIA and in their staff’s relation to political targets and audit culture. Numerical targets and audit culture in the SIA serve as a sign that staff are ‘doing the right things’ and evoke organizational pride and pleasure in work. Follow-up is not perceived as control as much as help to improve oneself along with organizational performance targets. The SIA is also an organization that takes pride in adapting to politically set objectives in a smooth way. In the PES, we found more of skepticism, sense of irony and self-irony, and pride in doing a good job according to internalized professional standards regardless of current management ideology. Caseworkers in the two agencies thus display different ways of ‘being professional’, despite the fact that frontline staff in both agencies lack a common education background. The SIA staff display a distinct form of ‘organizational professionalism’ (Evetts 2009), where delivering organizational targets efficiently is what being professional is about, making their staff highly adaptable to changing management directives. However, this orientation also led staff to down-prioritize one of their tasks stipulated by law, namely coordination of rehabilitation for clients in need, in favour of eligibility control and reducing the number of people on sick-benefit. The PES staff in contrast displayed a more pragmatic and experience-based professionalism, oriented to helping the clients in ‘the best way’ which meant a pragmatic orientation to formal rules. Frontline staff here was less adaptable to management norms coming ‘from above’. Agency culture in the two agencies thus differed considerably, and staff at the two agencies tended to look down and criticize the staff of the other agency, making collaboration between the two agencies challenging.

A third finding relates to the SIA, where the transparency ideal is an integral part of the organizational governance in the agency, and implemented through tools, technologies, teamwork, and socio-spatial governance. Our analysis revealed how the transparency ideal penetrates the organizational life in a much more pervasive way than is usually acknowledged in the ‘audit society’-literature, colouring not only the organizational oversight of case management, but also the relations between colleagues and between management and staff. As an internalized ideal, transparency affects caseworker subjectivity, lending the welcoming of audit as a way to self-improvement along with the improvement of organizational performance. However, transparency was much less salient in relation to clients and the outside world. This organizational regime of visibility, as established in social interaction as well as in organizational routines, was designed to facilitate the organizational performances and the smooth implementation of welfare-to-work policies in the health insurance. However, the relation to the clients as well as other agencies/stakeholders was explicitly down-prioritized by management and staff alike, likely resulting in clients in need of coordination to risk ‘falling between the chairs’. It should also be noted that the turnover of caseworkers in the SIA increased rather dramatically during the period of study (from 7 % in 2014 to 19 % in 2019). Our ethnographic research revealed the lack of space for questioning directives ‘from above’ or for having critical discussions on their work practice. With limited ‘voice’ opportunities in this agency, exit remained the main strategy for frontline staff critical of the current priorities of work. In contrast, in the local PES offices, a bottom-up culture was cherished by frontline staff and local managers alike, allowing much more space for criticism and critical reflections on their work practice.

A general conclusion from our research is that in addition to formal rules and output measurement, the informal shaping of frontline staff can be considerable and more attention to the inner life of bureaucracies is needed in future research. Ethnographic research is here essential, if not even necessary.


External collaboration and societal impact:

Apart from 10 presentations at academic conferences and a couple of seminar presentations at different departments in Sweden, we have diffused research findings widely through various channels: Online-presentation at the SIA head office (made available to all SIA staff throughout the country), presentation at the Social Insurance Inspectorate, online-presentation at the SPID-network (for practitioners and social insurance scholars alike), the project leader was interviewed in Swedish public service television (Uppdrag Granskning, four programs 2021), the project members signed a debate article in the biggest Swedish daily (Dagens Nyheter) together with other scholars, among others. We have also made a number of more easily accessible publications in Swedish to ‘reach out’. Research findings will be diffused to the local offices studied in the coming year.

All publications are published Open Access.