In a series of field experiments, six municipalities investigated whether a different approach in social assistance could get more people into work. The study focused on the effects of an exemption from the obligation to seek work, more intensive counselling or higher earnings disregards for additional income. A different approach can increase the number of people finding jobs. More research is needed.

Why the experiments were conducted: increasing the effectiveness of the Participation Act
Following the introduction of the Participation Act in 2015, various municipalities were keen to test new ways of supporting social assistance recipients in finding paid work or participating in society in other ways. The experimentation provision in the new law allowed for experiments with social assistance. In total, six municipalities − Deventer, Groningen, Nijmegen, Utrecht, Tilburg and Wageningen − conducted their own experiment in cooperation with local knowledge institutions. ZonMw supervised the experiments and CPB conducted an overarching assessment.

Types of intervention: exemption, more intensive counselling and higher earnings disregards
Three different interventions were tested, sometimes combined:
• an exemption from obligations to reintegrate and work: this intervention focused on self-direction and intrinsic motivation. During the experiment, social assistance benefit recipients in this group were obligated to apply for jobs, accept suitable work and follow reintegration activities.
• more intensive, personalised counselling: benefit recipients received more attention from caseworkers, with more contact moments compared to regular guidance and assistance. Further, services were adapted to the recipient's needs and wishes. New reintegration tools were occasionally utilised.
• higher earnings disregards for income on top of benefits: a purely financial incentive. A maximum of 50% (instead of 25%) of earned income was exempt from deduction from benefit payments, up to a maximum of 200 euros per month.

Method used: Randomised Controlled Trial
In each municipality, participants were randomly allocated to either a control group or one of the intervention groups. Participation was voluntary. Between 300 and 900 social assistance recipients took part, depending on the municipality. The duration of the experiment was between 17 and 27 months. Using CBS data, the employment rates of all participants and non-participants were tracked over that period. In addition, effects on reliance on the provisions of the Social Support Act, health and problematic debts were investigated − also referred to as broader welfare effects.

Result obtained: intensive counselling and higher earnings disregards can lead to more individuals finding work; an exemption has no effect
In Utrecht, more intensive counselling and the possibility of increased additional earnings during the experiments led to more benefit recipients finding work (for at least 12 hours a week). No clear effects were observed in the other municipalities. No effect of any kind, including a negative effect, on the number of individuals finding work could be observed from an exemption from obligations in any of the municipalities. In Utrecht, the interventions also had an effect on reliance on the provisions of the Social Support Act. The exemption group made greater use of the guidance available under the Act. Increased participation in daytime activities provided by the Act was observed amongst the counselling group and the group with greater opportunities to keep additional income. Greater reliance on the provisions of the Social Support Act means higher costs, but can also bring social benefits, which can ultimately contribute to greater participation in society.

Impact: services modified and lessons learned for future experiments
Some experimenting municipalities modified the services they provide based on the findings. Also, methodological lessons can be drawn for future policy experiments:
• a larger sample size of participants is needed for clearer findings and conclusions regarding effects of the experiment
• testing fewer interventions at a time increases the sample size and makes implementation easier
• uniformity of experiments across different municipalities may allow for data to be pooled and analysed together, rendering the results more useful and adding to their value
• ensuring a high level of comparability between the control and intervention groups at the outset of the experiment enhances the ability to establish a clear causal relationship between the intervention and the effects
For a more detailed discussion of these and other lessons, see the CPB publication 'Evaluatie experimenten Participatiewet: efecten op brede baten'.

Source: https://www.binnl.nl/home+-+en/knowledge/publications/bin+nl+publications/HandlerDownloadFiles.ashx?idnv=2719979