No choice? Hiring agency social workers in the Swedish personal social services

ABSTRACT

The practice of hiring agency social workers, i.e. social workers who are employed by for-profit staffing agencies but rented out to the social services, has been questioned and depicted as legally complex, costly and unsustainable. Nevertheless, it continues to be an institutional way of handling vacancies and understaffing in the social services. In this article, professionals’ reasoning about the use of agency workers is investigated through qualitative interviews with 21 social workers/managers in the social services. According to the interviewees’ descriptions, agency social workers are hired in order to achieve numerical stability, i.e. to maintain services notwithstanding staff absence and vacancies. Similar to findings in previous literature, the interviewed professionals describe that workload, turnover and recruitment difficulties are reasons behind the use of agency workers. However, they also highlight other aspects that are associated with the use, such as chronic understaffing and a large proportion of newly qualified social workers. Throughout, the use of agency workers is described as reactive, i.e. as an acute (and often unwanted) solution, and not as deliberate and planned.

ABSTRAKT

Användningen av socionomkonsulter har ofta framställts som juridiskt komplicerad, kostsam och ohållbar. Trots detta har användandet ökat sedan början av 2000-talet, och idag är det ett vanligt sätt att hantera personalbrist inom socialtjänstens individ- och familjeomsorg. I den här artikeln undersöks hur professionella inom individ- och familjeomsorgen resonerar kring anledningarna till användandet av inhyrd personal. I artikeln visas att individ- och familjeomsorgen använder socionomkonsulter som ett sätt att uppnå numerisk stabilitet, dvs. som ett sätt att fortsätta bedriva sin lagstadgade verksamhet trots vakanser och personalbrist. I likhet med resultat från tidigare forskning framkommer att arbetsbelastning, personalomsättning och rekryteringssvårigheter ligger bakom användningen av socionomkonsulter. De professionella i denna studie lyfter dock också fram andra faktorer, som exempelvis den höga andelen nyexaminerade socialarbetare och brist på erfarna kollegor, samt en kronisk underbemanning. Genomgående beskrivs användningen av socionomkonsulter som reaktiv, dvs. som en (oönskad) lösning på akuta situationer och inte som en medveten strategi.

KEYWORDS:

• Agency social workers
• staffing agencies
• personal social services

NYCKELORD:

• socionomkonsulter
• bemanningsföretag
• socialtjänst

Introduction

In Sweden and elsewhere, it has been duly noted that working conditions in the social services are harsh, that turnover is high, and that local authorities have difficulties recruiting staff (Smith & Moore, 2020; Tham, 2016). To remedy problems with understaffing and recruitment difficulties, the social services have come to rely on agency social workers (i.e. social workers that are employed by staffing agencies but rented out on short-term contracts to social services) to maintain service during periods of vacant positions and high workloads (Hoque & Kirkpatrick, 2008; Jones, 2018; Shanks & Mejdell Bjerland, 2021).

At the beginning of the 2000s, there were few agency social workers in Sweden. Those that existed were mainly used in specific circumstances, e.g. situations of conflict of interest, etc. However, over time, agency social workers have increasingly been used to fill vacancies and handle high workloads (Järkestig Berggren et al., 2021). The use has increased in later years, and agency workers are now used widely (Hoque & Kirkpatrick, 2008; Järkestig Berggren et al., 2021). Nevertheless, the extensive use of agency workers in the social services has been questioned. Several sources, such as media and research, have depicted it as legally complex, costly, unsustainable and sometimes unwanted (Cornes et al., 2012; Hyde, 2020; Jones, 2019; Sjöberg, 2021; Tun Hedfors, 2018), but for reasons yet not fully explored, it continues to be an institutional way to handle problems with understaffing.

This article aims to investigate how professionals in the social services (managers and social workers) reason about the use of agency workers. Through analysing the reasons for hiring agency social workers, I intend to develop the understanding of the continuation of a practice that has been discussed as potentially problematic. The previous research regarding this phenomenon is scarce and there is little to no research that qualitatively analyse how the use of agency social workers is discussed and understood by staff within the social services.

Social services in Sweden and the use of agency social workers

In Sweden, the social services have the ultimate responsibility for ensuring that citizens get the help and support that they need. Social services are administered by local authorities in each of Sweden’s 290 municipalities, and although the municipalities have the autonomy to organise the delivery of services as they see fit, they often have departments dealing with (a) personal social services, (b) elderly care, and (c) care for disabled persons. The departments dealing with personal social services often include child welfare, social assistance (i.e. a means-tested cash benefit that in Sweden is administered by trained social workers) and substance abuse treatment. It is these departments, i.e. the personal social services, that are in focus in this article.

In Sweden, around 70% of all municipalities hired agency workers during 2017 and/or 2018, often in their child welfare units. Hence, the use of agency social workers is common and widespread (Shanks & Mejdell Bjerland, 2021). We have little information of the general length of the contracts or the proportion of agency workers in the municipalities. However, official statistics indicate that in child welfare, the proportion of agency workers varied between zero and 67% (measured at the 1st of February 2021; Socialstyrelsen, 2021). Reports from England show that 15% of all social workers in local authorities were employed through for-profit companies in 2019 (Jones, 2019). The use of agency social workers has been linked to the deteriorating working conditions – particularly in the child welfare services (Järkestig Berggren et al., 2021). Additionally, in Sweden, legal changes that made it obligatory to investigate all children suspected to have been subjected to violence were introduced in 2014, and during the following years there was also a great increase of unaccompanied asylum-seeking children arriving in Sweden (Järkestig Berggren et al., 2021; Shanks & Mejdell Bjerland, 2021). Both of these changes have put additional strain on child welfare which, as noted above, has had the most extensive use of agency social workers in the personal social services.

Although the above-mentioned aspects have fuelled the use of agency social workers, the phenomenon must also be understood against the backdrop of legal changes in the 90s that allowed private sector involvement in public sector functions, as well as the legislative exemptions concerning the exercise of public authority that made it possible for the social services to rent agency workers to perform investigations of client’s needs (Järkestig Berggren et al., 2021; Shanks & Mejdell Bjerland, 2021). According to the Social Services act, the Swedish social services are allowed to hand over administrative functions to other actors. However, the possibilities to hand over functions involving exercise of public authority (e.g. assessments of clients’ need and decision-making) to private actors is not permitted (SFS, 2001:453, Ch. 2 §5). Nevertheless, it is possible for the local authorities to circumvent regulations by e.g. refraining from delegating the actual decision-making (JO, 2001). This means that in practice, it is the agency worker who investigates the client’s needs, but the decision should be made (signed) by a manager or someone that is permanently employed by the local authority.

Perspectives on use of agency social workers

The literature regarding the use of agency workers in social services is scarce, but there are a few important studies that should be noted. Several of these are from Britain and were published about a decade ago, but there are also some more recent contributions. Several studies touch upon the pros and cons in relation to the use of agency social workers. For example, studies suggest that agency workers can be beneficial both for the organisations and the employed. For the organisations, hiring agency social workers can be a way to maintain service during periods of vacant positions, high workloads and requirement difficulties. In some circumstances, it may be economically beneficial as it has the potential to limit the costs of recruitment and training (Hoque & Kirkpatrick, 2008; Järkestig Berggren et al., 2021). For professionals, staffing agencies may increase flexibility and expand the labour market (Hoque & Kirkpatrick, 2008). In some contexts, agency work has also been understood as a way to find employment in a tight labour market (Hyde, 2020). For those who possess public sector positions, the use of agency staff may decrease workloads and boost moral during times of understaffing (Hoque & Kirkpatrick, 2008). However, the literature also highlights negative consequences for all concerned parties. For organisations, long-term reliance on agency staff may be costly (Järkestig Berggren et al., 2021). It may also lead to an unequal distribution of tasks. As a result of agency workers’ lack of organisational-specific knowledge, the permanently employed may end up with the most demanding tasks. However, the opposite tendency has also been noted, i.e. agency social workers feeling compelled to take on the most demanding cases out of fear of being laid off if they do not (Cornes et al., 2012). In addition, agency workers risk being viewed as peripheral, may get less access to supervision or in-house training, and may face unreliable pay (Carey 2011; Cornes et al., 2012; Hyde, 2020). As for possible consequences for the quality of work, these issues have been less studied. However, concerns in terms of lack of continuity and difficulties with relations with clients have been raised (Carey 2011), as well as difficulties for agency workers to consult colleagues when faced with ethical challenges (Hyde, 2020). All in all, previous research provides some information regarding the reasons for professionals to move into agency work and how the use of agency workers may affect the user organisations. However, we know little about how professionals in the social services understand and describe the use. This will be developed below.

To understand the reasoning about the use of agency social workers in the Swedish social services, this article draws from concepts from literature discussing the organisational use of temporary staff in other types of organisations. Here, several antonymic concepts such as flexibility/stability, periphery/core, reactive/strategic have been used (Atkinson, 1984; Håkansson & Isidorsson, 2015; Svalund et al., 2018). In an oft-cited text investigating the use of agency workers in private sector firms, Atkinson (1984) finds that temporary workers are used in order to achieve numerical flexibility, that is as a strategy to adjust the number of workers to match actual demand and hence being able to quickly cut down on staff if needed. In the same text, he also differentiates between core and periphery staff. The first category is made up of permanent, highly skilled staff that are essential to the organisation as their skills are hard to replace. Instead of numerical flexibility, they provide functional flexibility to the organisation through being able to take on different tasks within the organisation. The latter category, the periphery workers, are instead made up mainly by temporary staff and perform tasks that require less skills and training. These employees are hired or laid off depending on demand, thereby providing numerical flexibility (1984).

Atkinson’s model has met some criticism. It has been accused of being unclear regarding to what extent it is empirical, theoretical or normative. Also, the assumptions of organisations’ rational considerations in relation to the way they decide to contract employees have been questioned (Mollitt, 2006; Nesheim, 2004). However, a lot of current research and newer ideas build on this model, and the concepts are of relevance in this study. Later scholars have noted that, on top of motives related to numerical flexibility, motives related to numerical stability are important. Numerical stability refers to the use of temporary staff to maintain stability notwithstanding staff absence, vacancies etc. In certain organisations, such as education and care services, stability motives have been found more common than flexibility ditto (Håkansson & Isidorsson, 2016; Svalund et al., 2018). Furthermore, in Atkinson’s model, it is presumed that organisations use temporary staff in a strategic manner, meaning that they are used in a deliberate and planned way to for example handle peaks in workload. However, it has also been shown they may be used in a reactive fashion, dealing with unexpected demand fluctuations and sick leaves (Svalund et al., 2018). In (the scarce) previous research focusing on use of agency social workers, it has been indicated that the main motives behind the use mainly are in line with the motives of numerical stability (Shanks & Mejdell Bjerland, 2021). Also, anecdotal evidence appears to suggest that the agency workers are used in a reactive fashion, although this remains to be investigated. The above-mentioned concepts will be used together with the knowledge from previous research to understand the use of agency workers in the Swedish social services.

Method

This article builds on data from 21 interviews with managers (on different levels) and permanently employed social workers from six municipalities in Sweden (Table 1). The social workers and lower levels of managers were either employed in child welfare units (n = 12), in social assistance units (n = 3), or were generalists (working with all target groups, n = 2). Hence, none of the interviewees on a social work level came from specialised substance abuse units, but managers on higher levels (n = 4) were responsible also for such units and the generalists also worked with these clients. Since the use of agency social workers is considerably more common in child welfare, this selection mirrors the use of agency workers quite well. Employees working on different levels and in different areas of social work were selected to ensure that potentially diverging views would be represented. For example, it was considered likely that the understanding of reasons for use of agency workers could differ between managers and social workers or between different areas of social work. The experiences of agency workers were not included in this article, as the focus here is the permanent employees’ view of the reasons behind and continuation of a phenomenon that have been depicted as problematic.

Table 1. The interviewees, field of social work and position in the social services.

Municipality	Interviewee n=	Position and field of social work
Municipality I, approx. 100,000 inhabitants, urban municipality, low/decreased use of agency workers	IP1	Manager, Personal social services
	IP2	Manager, Child welfare
	IP3	Manager, Child welfare
	IP4	Social worker, Child welfare
	IP5	Social worker, Child welfare
Municipality II, approx. 140,000 inhabitants, urban municipality, intermediate use of agency workers	IP6	Manager, Child welfare
Municipality III, approx. 50,000 inhabitants, urban municipality, high reliance on agency workers	IP7	Manager, Personal social services
	IP8	Manager, Social assistance
	IP9	Group leader, Child welfare
	IP10	Group leader, Child welfare
	IP11	Social worker, Child welfare
	IP12	Social worker, Child welfare
Municipality IIII, approx. 3000 inhabitants, rural municipality, low/decreased use of agency workers	IP13	Manager, Personal social services
	IP14	Social worker, generalist
	IP15	Social worker, generalist
Municipality V, approx. 16,000 inhabitants, urban municipality, intermediate use of agency workers	IP16	Manager, Personal social services
	IP17	Manager, Child welfare
Municipality VI, approx. 150,000 inhabitants, urban municipality, intermediate use of agency workers	IP18	Manager, Child welfare
	IP19	Social worker, Child welfare
Municipality VII, approx. 18,000 inhabitants, municipality close to larger city, intermediate use of agency workers	IP20	Social Worker, social assistance
	IP21	Social Worker, social assistance

Based on findings from a previous study which mapped the use of agency social workers in Sweden during 2017–2018 (Shanks & Mejdell Bjerland, 2021), the municipalities were selected in a way that was intended to allow maximum variation in terms of use of agency social workers, e.g. high use, intermediate use and low use, judged by their own descriptions (cf. Flyvbjerg, 2006, s. 230). The extent of use in the respective municipalities varied over time and from only one or two, to 18 in a team of 50 social workers. Due to difficulties finding enough interviewees from some municipalities (especially those with an intermediate use of agency workers), additional municipalities from the same category had to be selected. In the end, one of the included municipalities was selected on the basis of having contracted a large proportion of agency social workers (six interviewees), two were selected from municipalities where the use had been limited or had decreased (eight interviewees), and finally four municipalities were selected from those having used an intermediate proportion of agency social workers (seven interviewees). This procedure intended to allow analysis of potentially varying attitudes to, and reasons behind, the use of agency social workers, as well as insight into policies and attempts to limit the use of staffing agencies. In addition to the variation in terms of agency social workers, the municipalities also show different characteristics in terms of size and location.

The semi-structured interviews were conducted using a videoconferencing program, took around one hour and were held by the author. The interviewees were invited to participate on the basis of informed consent and that quotes should be anonymised. When constructing the interview guide, I was informed by previous literature regarding agency /social/ workers and thereby, my analytic interest was guided by this literature. Nevertheless, the questions in the interview guide that concerned the reasons behind the use of agency social workers were broad and open; e.g. ‘how would you describe the use of agency social workers in your organisation’, ‘why do you think that your organisation use agency social workers’. Through the interviews, lengthy and rich descriptions were gained, highlighting similarities as well as differences in experiences in the interviewee’s experiences. The interview guide was adjusted depending on the position of the interviewee. The study strictly followed ethical guidelines and national laws (SFS, 2003). Informed consent was obtained and the participants were informed of their right to withdraw at any time during the research process. They were assured of their anonymity and informed that the findings would be presented in a way that would make it impossible to identify any individual. A formal ethical approval from the Swedish Ethical Review Agency is not needed for research that does not include sensitive personal data (Etikprövningsmyndigheten, n.d.) and was therefore not sought for this project.

After the interviews, the recordings were transcribed verbatim and analysed using the software program NVivo. A thematic content analysis was conducted, as this type of analysis is appropriate for examining the perspectives of individuals and shedding light on potential similarities and differences between their experiences (Nowell et al., 2017). For the purpose of this article, the focus was directed towards descriptions of reasons for using agency social workers given by the different levels of workers. As a first step, the interview transcripts were coded and broadly categorised in line with the focus areas of this article – e.g. reasons for use as understood by the permanent employees. Secondly, sub-themes (such as workload, working conditions, recruitment difficulties etc.) were identified. Concepts from Atkinsons model and its further developments (see above) were used as ‘sensitising concepts’, meaning that they guided the attention in the analysis and were used as frames for theoretical understanding (Bowen, 2006). Although the analysis was informed by these concepts, I also tried to be alert to statements indicating diverting views. This yielded understanding of e.g. the importance of the supply-side. Throughout the article, the results are related to previous literature. In addition, I have tried to provide substantial information regarding the research setting as well as rich data, so that readers can decide to what extent the findings are transferable to their settings (see Nowell et al., 2017).

Findings

As a background for the findings presented below, it should be noted that almost all interviewees describe that their municipalities have attempted to reduce (with varying degrees of success), the use of agency social workers. The majority describe that they have experienced demands from higher organisational levels to do so. Many interviewees associate these demands with high costs related to contracting agency social workers, but some also mentioned that these demands were associated with concerns about quality. The existence of strategies and the demands from higher organisational levels indicate that the use of agency workers is, if not unwanted, then at least a practice that the organisation wishes to decrease. It is against this background that the following should be interpreted. If the practice is unwanted – why is it continued?

Workload, inexperienced staff and turnover

Previous research has indicated that high workloads are connected to the use of agency workers (Hoque & Kirkpatrick, 2008; Jones, 2018; Shanks & Mejdell Bjerland, 2021). This is validated, developed and challenged in the statements below.

In the first quote, the interviewee, a social worker in child welfare, describe a situation in which agency workers are contracted in order to deal with cases that remain unallocated due to high workloads:

IP12: We have too many unallocated cases. The managers must have panicked, I think / … /. They said that we should not use agency social workers, and the use has been stopped from time to time. Then, a pile [of cases] builds up, and they cannot cope with the pile. We, social workers, cannot take on more work because we are all fully occupied. Then they have hired agency social workers to deal with the pile.

This is much in tune with previous research. The services offered by the social services are often statutory and must be delivered irrespective of the work situation or resources. Also, the tasks must be completed within a certain timeframe. Here, the social worker describes a situation in which the managers, responsible for the services in their units, have been unable to hand out cases to the permanent staff, causing investigations to pile up. Eventually, and although they had previously decided not to use agency workers, such workers are contracted to deal with the situation.

Several interviewees from child welfare units highlight how a large proportion of newly qualified staff and a lack of more experienced colleagues has contributed to the use of agency workers. These circumstances have been less emphasised in previous research on use of agency social workers. In the quote below, a senior manager explains how s/he understands the development of the situation with increased use of agency social workers in child welfare units:

IP1: There was also a change in the law / … / and it became a legal requirement to investigate all children who were suspected of having been exposed to violence. So, the pile of cases concerning children and young people increased dramatically. / … / Several years before, the BBIC [similar to the Integrated Children's System] had been introduced, so the work in child welfare just got heavier and heavier, and busier. And all the good workers left. They went to adult social work units, or to treatment units. Then we needed to fill up with newly graduated social workers, and then we slowly but surely made the whole thing [child welfare] collapse / … /. It's like three [bumps on the road]: The introduction of BBIC was the first bump. Then, the change of law and new requirements regarding child welfare investigations came, and that was the second bump. And then the unaccompanied minors came, that was third.

This manager identifies three bumps on the road towards the situation in which many child welfare services find themselves in today. This manager’s explanation shows many similarities with that given by Järkestig Berggren et al. (2021). Both sources highlight the legal changes that made it mandatory to investigate all children suspected of having been subjected to violence, and the increase of unaccompanied asylum-seeking children arriving in Sweden. These changes caused additional pressure on the local authorities. However, this manager also adds two additional developments, namely the introduction of a new assessment instrument (BBIC) and the increased dependency on newly graduated social workers. The first (the assessment instrument) was perceived by some as leading to increased proceduralisation and higher demands on documentation (cf. Horwath, 2011), and according to the manager quoted above, this contributed to a higher workload and to experienced staff leaving their jobs. In turn, this led to the second development; the increased recruitment of newly qualified social workers. For a newly qualified social worker, the work in child welfare may be difficult (cf. Tham & Lynch, 2021). Several interviewees describe that newly qualified social workers do not always stay long in the job, creating something resembling a vicious circle.

One manager challenges the idea that high workloads per se (at least if workload is measured in number of cases) is the reason for contracting agency social workers. Rather, s/he understands the high proportion of newly qualified social workers as the main reason for the use of agency social workers:

IP16: As we have a very low workload, I can confidently say, I think I have enough experience to be able to say this: we have a very small number of cases per social worker. But, the perceived workload [is high], I cannot question that. And I think that is the case because they are so inexperienced. They are inexperienced and they work with very important things. No one likes to lie in bed and twist and turn and feel ‘have I made a mistake?’ / … / There is a fear of making mistakes and there are not enough experienced and stable [staff] to support, help and explain [to the less experienced] what actually is a dangerous situation and what is not.

As indicated above, this senior manager understands the use of agency workers not as a result of a high workload in terms of number of cases, but the result of an unfortunate combination of the nature of the work (complex and including many difficult decisions) and high proportion of inexperienced staff. This manager continues by explaining that in lack of experienced employees, they have hired agency social workers to gain the much-wanted experience.

In general, the situation with high workload (real or perceived) and few experienced social workers in the child welfare units has, according to several interviewees, led to sick leaves and staff turnover. Agency workers are therefore hired to ensure that the statutory services are delivered on time and in accordance with the law. Although other branches of the personal social services (substance abuse and social assistance) also use agency social workers (Shanks & Mejdell Bjerland, 2021), the situation in these units appears to be slightly different. In the quote below, a manager from a social assistance unit gives his/her view on this:

IP8: I think that there is a difference in the heaviness of cases [between child welfare and social assistance]. Many newly graduated social workers come straight from school and want to do a good job in child welfare, and then it becomes too heavy. It is not the same thing in social assistance, they stick around a bit longer. The regulations are clearer in this work. There is more to hold on to.

This manager, who used to work in child welfare but moved to social assistance, describes that the casework in social assistance has a clearer structure, which makes it easier for newly qualified staff to cope with the work. This means that they stay longer, creating a more stable workgroup. This is of course not the case in all social assistance units, but given the lower proportion of agency social workers in these units, it may have some explanatory power.

Recruitment difficulties

On top of the turnover, several municipalities appear to have difficulties recruiting new staff. In some cases, it is not only difficult to recruit experienced social workers, but social workers at all. In some cases, this can be the result of the location, as in the case of this small and remote municipality:

IP13: In Municipality IIII, it has been very difficult to recruit social workers. It has a lot to do with it being a very small municipality. / … / It is quite isolated, and the nearest town is far away. / … / So, it is difficult to recruit. / … / It has been difficult to get people to apply for the jobs, and it has also been difficult to get them to stay.

This particular municipality is quite extreme in terms of location, but it shares the problem with many, less remote, municipalities. Other municipalities see other reasons for the recruitment difficulties, such as re-organisations, conflicts and bad reputation, few qualified applicants etc.

In some cases, the interviewees describe that it is not particularly difficult finding applicants with the right basic qualification, but staffing the units with (at least some) experienced staff and making the staff stay is more difficult:

IP16: It has not been difficult to find someone, but it has been very difficult to find experienced [staff], and then it is difficult to get people to stay. We share the difficulties of finding experienced staff with most municipalities.

A colleague from the same municipality offers an explanation to why this is the case:

IP17: The challenge is the lack of experience and the turnover within the profession which makes employees feel that they can handle a maximum of two years in one workplace before they have to move. It is bad for the workplace and for professional development.

Again, the combination of a high proportion of inexperienced staff and the difficulties of the work is highlighted. According to the interviewee above, this leads to turnover and a need to use agency social workers.

Need created by the supply side?

The statements above give the impression that the local authorities have no choice but to use agency social workers. To be able to get the job done, i.e. to deliver the statutory services to the citizens, they have developed an unwanted dependency on agency workers. In one interview, a respondent – a senior manager – voluntarily addresses the somewhat paradoxical situation with the continuing use of the ‘unwanted’ agency workers. This topic is brought up at the end of the interview, as an answer to the question ‘is there anything else that you think that I should know?’. The senior manager starts with a rhetorical question, and then moves on to his/her understanding of why the use continues:

IP16: Why is the phenomenon of agency social workers needed? Let's assume that it is more negative than positive to have agency social workers, why is it needed then? I know that several municipalities in the county talked about this a few years ago. We played with the idea that – if it was not people we worked with, which means that there is a risk that some are not cared for properly – otherwise we played with the idea that we would just ban [agency workers] and shake hands with all municipalities that management would forbid the hiring of agency workers. It would be absolutely awful for a while but then those who are agency workers either have to do something else, like building cars at Toyota, or those who want to work with this [social services] would have to take on ordinary employment. Then the staffing agencies would not survive. We really wanted to do it, but it is not possible to jump the gap when risking not following the law. That is why we are powerless.

This manager’s statement explicates his/her view on the perceived dependency on agency social workers. However, his/her statement also indicates a somewhat different description of the situation, or at least it raises the question if the ‘need’ for agency workers rather is created by the staffing agencies (the supply side). The existence of staffing agencies means alternative work opportunities (and better pay) for social workers who have chosen to opt out of public employment. Previous research has indicated that both push and pull factors may be at play in this context. Social workers may be tempted by the flexibility and, as the situation is in Sweden, the better pay offered by staffing agencies. However, they may also flee from permanent positions due to deteriorating working conditions (Kirkpatrick & Hoque, 2006; Kunda et al., 2002)

Although many quotes imply a situation in which agency social workers are unwanted but needed to maintain services in times of difficulties, not all interviewees appear to agree to the descriptions of local authorities as helpless victims in search for stability. Below, the interviewee offers a reply to the question ‘why is there a need for agency social workers?’.

IP4: Because I think that many municipalities save money by [reducing the number of] regular staff. Then there is a lot to do and the staff has not got time, and the staff gets sick. They [managers] see that they are unable to fulfil their primary assignments and they see that cases pile up. Then they get scared that something may happen to the children. Therefore, emergency solutions are used, although the problems really are organisational. If – in municipalities where there are a lot of agency social workers – they would have expanded each team with one permanently employed, the problem would have been solved. You wonder – the money spent on four employees in our unit versus all the agency social workers we have had – how will the balance sheet look in the end?

In the quote above, the interviewee – who is a social worker in child welfare – implies that there may be (what s/he sees as misguided) economic incentives that cause local authorities to refrain from recruiting enough staff to be able to cope with the tasks at hand. In the end, this calls for hiring agency social workers as an emergency solution when work piles up.

Discussion

Choosing the lesser of two evils in order to achieve numerical stability?

This article sets out to investigate how professionals in the social services reason about the use of agency workers. All in all, the picture that emerges is that most interviewees are under the impression that their municipalities would rather do without agency social workers if they could. More or less all interviewees describe that higher management levels had conveyed a wish to reduce the use of agency social workers. It is also clear from the interviews that some municipalities had launched rather comprehensive measures to reduce the use of agency social workers. Nevertheless, the use apparently continues – at least on a general level. The obvious question that arises is therefore: If the use of agency social workers really is unwanted, why is it still such a common way of handling problems with understaffing?

Previous literature on social work, and literature dealing with perspectives on the use of temporary staff in general, offer some general explanations to the use of agency /social/ workers. On a contextual level, the increased marketisation, the legislative changes and legislative exemptions that have affected social work organisations can help in understanding the use (Järkestig Berggren et al., 2021). Also, on an organisational level, high strain, recruitment difficulties and vacancies have been highlighted as reasons (Shanks & Mejdell Bjerland, 2021).

This study validates such findings, but also highlights some other influential aspects, such as the effects of a high proportion of newly qualified staff and the difficulties in recruiting and keeping experienced staff. Without the presence of experienced staff in a workgroup, newly qualified social workers are left to their own devices and have no one to turn to for help and guidance. This, according to some interviewees, causes the social workers to experience that they are overwhelmed by workload, perhaps as a result of being inexperienced and having to deal with complex cases. In turn, this may cause turnover. Here, it should be noted that different levels of employees may have somewhat diverging understandings. Statements from some social workers suggest instead that the use of agency social workers is a result of municipalities refraining from recruiting enough staff (in order to save money). Hence, through the eyes of someone on the frontline, the issue at hand is a real (not perceived) too high workload due to a shortage of staff.

Comparing the situation in the social services with those in ‘the flexible firm’ (Atkinson, 1984), there are not many statements that can be interpreted in line with the ideas of numerical flexibility, at least if it is understood as a strategic way to adjust the number of workers to match demand. Interviewees mention economic incentives that cause local authorities to refrain from recruiting permanent staff, but this is depicted more as a hopeful expectation that the staffing levels nevertheless will be sufficient. When agency workers are hired, it appears to be reactive (cf. Svalund et al., 2018) – an acute response to an inability to meet the requirements of the law (e.g. meeting the timeframe) with the staffing levels they have.

Judging from the interviews, it appears more plausible that the local authorities use agency workers in order to achieve numerical stability (Håkansson & Isidorsson, 2016; Svalund et al., 2018). Unlike a firm, local authorities cannot refrain from delivering statutory services in cases of staffing difficulties. Hence, shortages of staff (or high workloads) that risk hindering service delivery are avoided through the use of agency workers. Some interviewees also indicate that agency social workers in some instances are used in order to provide some sort of competence stability, i.e. to provide experience to an inexperienced workgroup. This may be seen as a somewhat more strategic use, but the way it is explained by the interviewees, it appears more reactive. It is not as if the local authority willingly employs inexperienced staff and hires agency workers as experts, or as if agency social workers always provide the needed experience. Particularly during the years of high demand, the experience levels among the agency social workers was described as varied (Järkestig Berggren et al., 2021).

It is also interesting to return to the ideas of core and periphery in this context. As Atkinson (1984) describes it, the core is made up of permanent, highly skilled staff that are essential to the organisation. Temporary staff are used in the periphery, where they perform tasks that require less skills and training (ibid.). Other studies have developed Atkinson's model, showing that organisations may indeed use external staff in the core, when they wish to gain certain expertise that is lacking, or in order to facilitate innovation (Nesheim, 2004). Neither of these reasons fit the case of social services. Unlike Atkinsons’ idea, the agency workers are not placed in the periphery, but at the core of the social services. However, unlike in the developed model, they are not generally used as experts providing new knowledge or innovation, but have the same formal competences as other employees and are expected to perform the same tasks. In accordance with regulations, agency social workers are (sometimes) prohibited from making formal decisions – this is instead done by a permanent employee (generally a manager). However, this is generally just a formality, as the investigation and assessment is already done by the agency worker. All in all, agency social workers appear mainly to be seen as substitutes to the preferred alternative of permanent staff, not expected to bring any new competences or value to the organisations. The social services are not perceived to benefit from the use, not in an economic sense and not in terms of quality. Instead, the beneficiaries appear to be the for-profit companies renting out agency social workers.

Conclusion and implications – stable incontinuity?

This article has shown that the social services’ staff perceive the use of agency social workers more or less solely as a way to reactively and unwillingly counteract staffing difficulties. This use is clearly at odds with ideas on how agency workers may be used strategically in order to increase flexibility, innovation or knowledge in an organisation (Atkinson, 1984). From the results, it is difficult to determine if the use primarily is a consequence of (a) circumstances on the labour market, meaning that recruitment difficulties, a high proportion of newly qualified staff and turnover causes a reliance on agency social workers; (b) that social services are in the hands of the supply side or; (c) chronic understaffing (in order to save money) in the social services making them unable to fulfil their obligations without using agency social workers. It is likely that all of these aspects are at play, perhaps to a varying degree depending on the context of the municipality. All in all, the findings in this study validate (the scarce) previous knowledge about the contextual reasons for the use of agency workers, and furthers our knowledge about how social workers perceive the continuous use.

The results should not be understood as if the professional’s experiences of agency social workers, once they are contracted, necessarily are bad. However, by definition, temporary agency workers are contracted for short periods of time which means that no matter how good they are, they cannot offer continuity. In the long run, this is bound to have consequences both for the workforce and for the clients.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Swedish research council for Health, Working Life and Welfare (Forskningsrådet för hälsa, arbetsliv och välfärd, FORTE) under grant number 2018-01654.

Notes on contributors

Emelie Shanks

Emelie Shanks is an assistant professor at the Department of Social Work, Stockholm University. Her research interests include staffing and management of the personal social services, and she is currently leading a project concerning the use of temporary agency workers in these organisations. She is also involved in research projects concerning threats and violence in social work, and the organisation of residential care for children and youth.