Swedish welfare is often characterized as comprehensive and generous. Social assistance constitutes the final safety net in the Swedish model. Unlike most other benefits, eligibility and subsidy levels are here subject to extensive means-testing, with eligibility determined by individual caseworkers. In this article, we explore the extent to which eligibility determination and generosity of benefits have changed in social assistance assessments over three decades, relating these to changes in regulations over the same period.

The article presents data from three independent vignette studies conducted in the mid-1990s, mid-2000s and in 2018 respectively. The data consist of assessments of a total of six vignettes, made by individual caseworkers (n=1 337). The results indicate that overall generosity in social assistance has decreased considerably over the last 25 years and that, despite both national and local initiatives to reduce individual variation in assessments, disparities have instead increased.

Swedish social assistance aims to ensure a ‘reasonable level of living’ for the least well-off in society. The benefit is, however, administered with considerable discretion on both the municipal and individual level. While numerous studies have shown substantial variations with regard to eligibility assessments, knowledge that pertains to differences in benefit levels, that is, the extent to which such variations are present and, if so, which factors co-vary with assessment outcomes, is quite scarce. This article presents analyses of generosity in eight so-called vignettes, collected among just over 1,000 social workers in 19 Swedish municipalities, including Sweden’s three major cities. Variations in generosity are analysed using multiple linear regression analysis (OLS), examining how basic social worker characteristics, professional factors and individual attitudes influence both overall generosity levels and generosity toward individual vignettes. The results show that, although variations in level of generosity are extensive, much of the difference remains unexplained. Among the significant outcomes that are identified, the fact that higher caseloads among social workers result in more generous decisions and that individual attitudes that favour more requirements and control lead to increased restrictiveness in assessments are worthy of note.

In recent years, the use of digital support systems in the administration of Swedish social assistance (SA) has become increasingly widespread. Current surveys show that around 10 percent of Swedish municipalities make use of so-called Robot Process Automation (RPA) in their everyday practice. As key arguments for the use of RPA-support, aspects such as unburdening employees of pure administrative tasks, as well as increased transparency and enhanced legal certainty for applicants, are often highlighted.

The present paper presents results from a larger research project addressing the practical use of RPA-support in Swedish Personal Social Services. The analyses are based on cross-sectional quantitative data from 800 cases collected in four medium-sized Swedish municipalities. Half of the cases reflect assessment outcomes before RPA-support was implemented; the other half show outcomes after implementation. The paper aims, first, to describe the character and contents of the different RPA-tools utilised in municipalities. Second, the paper aims to describe and analyse the outcomes of SA decisions before and after the implementation of RPA-support, respectively.

Our main findings are as follows: (a) while RPA can partially alleviate the administrative burden on social workers, human involvement remains essential for individual assessments; (b) even though the use of RPA-support does not establish any significant relation to the outcomes of SA eligibility assessments, the assessments are far less generous after the implementation of RPA than before; and (c) the changes in generosity are particularly notable with regard to applications outside of the National Benefit Standard, which is a strong predictor for the rejection of applications.

In line with the global trend towards digitalization, digital automation has become a politically endorsed strategy to enhance efficiency and transparency in public service delivery. One such example is the adoption of Robotic Process Automation (RPA) in the context of social assistance (SA) in Sweden. While SA is a municipally organized means-tested subsidy for the most financially vulnerable, RPA functions as highly structured software executing administrative tasks based on predetermined rules. Employing a sociomaterial perspective, this article analyses group interviews conducted in four Swedish municipalities, to explore if and how RPA adoption configures the dynamics of SA casework administration and practice. Findings suggest that RPA adoption does not inherently lead to expected all-encompassing enhancements of faster and fairer eligibility determinations as well as a more client-centric approach in SA casework. Instead, the instrumentalist approach of leveraging digital tools to achieve specific outcomes, combined with the task delineation required by RPA, appears incongruent with the nature of SA casework practice. As caseworkers attest to organizational adjustments and a more instrumental approach towards clients’ situations alongside RPA adoption, this study underscores the inadequacy of the rationalist notion of segmenting SA casework in order to achieve efficiency in casework that, by law, should be based on individual judgements by professional caseworkers.

This issue of the International Journal of Social Welfare contains contributions coming from all corners of the world, covering a wide range of subjects within the field of social welfare. In the first article, Jussi Tervola (Finland), Merita Mesiäislehto (Finland), and Joonas Ollonqvist (Finland) analyse why the number of social assistance recipients varies between Sweden (lower) and Finland (higher), two countries with social assistance systems that share many features. Within a microsimulation approach they decompose cross-country variations in receipt by using eligibility simulations and policy swaps with administrative register data from EU-SILC. Their results show that more generous national norms and earning disregards results in higher receipt rates in Finland, while non-take up rates in Sweden were estimated to be higher. The latter could possibly be explained by more harsh discretionary measures.

This study contributes to an understanding of the role of uncertainty in frontline decision-making about social assistance (SA) eligibility. Assessments by social workers of two vignettes (n=927; n=936) describing fictional SA applicants were analysed. The names of the fictional applicants were randomly varied so that the respondents assessed vignettes with either Swedish- or Arabic-sounding names. The respondents were asked to assess eligibility for SA, and to specify if any alternative assessment would be possible to make. Data was analysed by means of directed content analysis of free-text answers on the one hand, and crosstabulations and multilevel regression models on the other. 

The results show, first, that the social workers related the possibility of alternative assessments to varying definitions of what the uncertainty problem was, why it had arisen and how it could be solved. These aspects are discussed in terms of four to some extent overlapping rationales. Second, the results show that ethnicity bias may be involved both in how suitable for eligibility respondents assess SA applicants to be, and how definite or uncertain they perceive their own assessments to be. For both vignettes, the choice to deny SA to applicants with Arabic-sounding names was most clearly linked to perceived uncertainty. 

Swedish social assistance (SA) is administered with substantial discretion among both municipalities and social workers. Although such discretionary power has its merits, it also comes with potential disadvantages. Previous studies show considerable variation in assessments of SA eligibility, not only between municipalities but also between social workers within the same organisation. Further, the differences seem to have increased over time and apply irrespective of the method used for measurement. The present paper analyses eligibility assessments of eight so-called vignettes, collected among just over 1,000 social workers in 19 Swedish municipalities, including Sweden’s three major cities. The assessments are analysed by means of multilevel logistic and linear regression analysis, focusing on the influence that basic social worker characteristics, professional factors and individual attitudes exert over (a) propensity to grant SA in separate vignettes and (b) general tendency to grant applications. The results show that social workers’ caseload and attitudes towards clients and SA in general affect the outcomes of assessments the most. Along with the fact that significant parts of the variations remain unexplained, one main conclusion is that Swedish SA administration seem to be somewhat disengaged from formal rules and regulations.

Det svenska socialbidraget (SA) hanteras med betydande handlingsfrihet bland både kommuner och socialsekreterare. Även om sådan diskretionär makt har sina fördelar, kommer den också med potentiella nackdelar. Tidigare studier visar på stor variation i bedömningar av SA, inte bara mellan kommuner utan även mellan socialsekreterare inom samma organisation. Vidare verkar skillnaderna ha ökat över tid och gäller oavsett vilken metod som används för mätning. Denna artikel analyserar bedömningar av åtta så kallade vinjetter, insamlade bland drygt 1 000 socialarbetare i 19 svenska kommuner, inklusive Sveriges tre storstäder. Bedömningarna analyseras med hjälp av logistisk och linjär regressionsanalys, med fokus på det inflytande som socialarbetares grundläggande egenskaper, professionella faktorer och individuella attityder har över (a) benägenhet att bevilja SA i enskilda vinjetter och (b) generell tendens att bevilja ansökningar. Resultaten visar att socialarbetares ärendemängd och attityder gentemot klienter och SA i allmänhet påverkar utfallet av bedömningar mest. Tillsammans med att betydande delar av variationerna förblir oförklarade är en huvudslutsats att svensk handläggningen av SA verkar vara något frikopplad från formella regler och förordningar.

This issue of International Journal of Social Welfare offers, as usual, a variety of contributions of relevance for welfare policy and social work. In the first article Aino Suomi (Australia), Timothy Schofield (Australia) and Peter Butterworth (Australia) reports from a study of stereotypes held by Australians towards five different social groups (unemployed, people with a mental disability, people with a physical disability, single parents and elderly). Departing from a stereotype content model (SCM), a sample of adult Australians were asked to reflect over their images and presuppositions of the five groups. The results show that receipt of income support payments made no difference in how elderly or people with physical disabilities were perceived, while single parents and unemployed were stereotyped in a more negative manner when the received benefits.

In this article we explore the extent to which Swedish social workers encounter IPV, as well as their readiness to handle these cases. The study draws from data gathered in an online survey answered by 787 caseworkers within the personal social services, exploring the rate of asking clients about intimate partner violence, access to IPV training, level of self-rated competence, and amount and frequency of IPV-cases in the caseload. Our findings do on the one hand confirm that IPV is a widespread problem that a large proportion of social workers come across on a regular basis, and on the other show that substantial parts of Swedish social workers consider themselves to be rather ill-equipped at handling cases of IPV. Results from multiple regression analyses show that training, high self-perceived competence and administrative procedures each tend to increase the likeliness of social workers regularly asking clients about IPV. Social service organisations striving to increase their ability to detect and support victims of IPV are thus likely to benefit by raising the level of staff competence as well as implementing administrative procedures throughout the organisation.

I artikeln undersöker vi i vilken utsträckning svenska socialarbetare kommer i kontakt med våld i nära relationer och deras beredskap att hantera sådana ärenden. Studien bygger på data från en internetbaserad enkät, besvarad av 787 handläggare inom individ- och familjeomsorgen, kring hur ofta de frågar om våldsutsatthet, tillgång till utbildning om våld i nära relationer, självupplevd kompetens i att hantera våldsärenden, samt förekomst av ärenden som rör våld i nära relationer. Studien bekräftar att våld i nära relationer är ett utbrett problem som en stor andel socialarbetare regelbundet kommer i kontakt med, men den visar också att de svarande upplever att de är förhållandevis dåligt rustade för att hantera sådana ärenden. Vidare framkommer att utbildning kring våld i nära relationer, hög självupplevd kompetens inom området och för området specifika administrativa rutiner var för sig tenderar att öka sannolikheten att socialarbetare regelbundet frågar klienter om våldsutsatthet. Socialtjänstorganisationer som strävar mot att öka sin förmåga att upptäcka och ge stöd till våldsutsatta skulle därmed sannolikt gynnas av att höja personalens kompetensnivå samt att implementera administrativa rutiner i organisationen.