The AEA RCT Registry will be down for maintenance on Tuesday, April 20th, from 9pm EDT to 11pm EDT to perform necessary upgrades. The site will be in maintenance mode during that time. We apologize for any inconvenience this may cause.
Is Supported Employment Effective for Young Adults with Disability Pension? Evidence from a Swedish Randomized Evaluation
Last registered on January 25, 2019


Trial Information
General Information
Is Supported Employment Effective for Young Adults with Disability Pension? Evidence from a Swedish Randomized Evaluation
Initial registration date
January 15, 2019
Last updated
January 25, 2019 3:36 AM EST
Primary Investigator
Stockholm University
Other Primary Investigator(s)
Additional Trial Information
Start date
End date
Secondary IDs
This paper reports results from a large-scale randomized experiment evaluating whether a Supported Employment rehabilitation intervention strategy can improve labor market opportunities for young adults on disability pension better than regular vocational rehabilitation. The Supported Employment intervention utilizes a caseworker as back-up for the individual during training to reduce employers’ risks when hiring an individual with unclear productivity. In total, 1,063 individuals were randomly assigned between interventions. The main results show that 18 months after the start of the project, participants with Supported Employment have work rates that are approximately 10 percentage points higher than participants who received regular rehabilitation.
External Link(s)
Registration Citation
Skogman Thoursie, Peter. 2019. "Is Supported Employment Effective for Young Adults with Disability Pension? Evidence from a Swedish Randomized Evaluation ." AEA RCT Registry. January 25. https://doi.org/10.1257/rct.3794-1.0.
Former Citation
Skogman Thoursie, Peter. 2019. "Is Supported Employment Effective for Young Adults with Disability Pension? Evidence from a Swedish Randomized Evaluation ." AEA RCT Registry. January 25. http://www.socialscienceregistry.org/trials/3794/history/40594.
Experimental Details
Supported Employment
Supported Employment is a vocational rehabilitation strategy distinguished by its early workplace focus. The basic idea is that for an individual, the route to acquire and remain at one job should be shorter than the route to establish a general productivity level sufficient to compete on the regular labor market. Therefore, efforts are focused on getting the individual to a specific workplace at an early stage in the rehabilitation process at the expense of general human capital acquisition. To emphasize the direct acquisition of work, followed by support at the workplace, the strategy is denoted “place then train”, in contrast to the human capital-oriented approach, denoted “train then place”. The general method was developed in the US in the 1980s and introduced in Sweden at the beginning of the 1990s.
Each Supported Employment caseworker is trained in the method and well informed about the local labor market. To enable intensive individual support, each caseworker has a maximum of 30 individuals listed at the same time. Throughout the intervention, the caseworker keeps regular contacts with both the participant and the employer in order to secure a long-term hire. The focus is on regular employment (typically wage-subsidized for our target group), and if the employer does not want to make a regular hire after a workplace introduction, a new workplace is sought out. The follow-up period lasts approximately one year.
The Supported Employment intervention can be thought of as incentivizing employers to hire individuals with disabilities by removing the initial costs of training and reducing uncertainty. For potential employees with disabilities, employers’ expectations about training time and uncertainty about productivity are on average higher than they are for those without disabilities. Thus, for an individual whose wage subsidy is just right in the long term, the hiring cost is on average too large for him or her to compete on the regular labor market.
We argue that Supported Employment should be more effective than an increase in initial wage subsidies because firm productivity exhibits diminishing returns to scale. Large unexpected shocks to productivity can be harmful to employers, which implies that a large variation in the error that employers make when predicting individual productivity during the hiring process is costly for them. A large variation in the prediction error might prevail among our study participants because many of them have psychiatric diagnoses entailing unexpected negative health shocks. When the uncertainty is very high, it might not be possible to subsidize employers enough to compensate for this risk without running into a principal agent problem because if the wage subsidy exceeds the wage, employers have incentives for ‘parking’, i.e., taking on individuals with disabilities without actually giving them workplace training.
Within the Supported Employment intervention, the caseworker handles problematic situations and serves as a back-up in situations when the individual faces an unexpected negative health shock. If the employer knows that the caseworker will be there and will take care of such situations, uncertainty is reduced. Supported Employment is considered a more personal intensive rehabilitation intervention compared to Regular rehabilitation.

Regular vocational rehabilitation
Two Swedish authorities, the Social Insurance Agency and the Public Employment Service, have a joint responsibility to provide vocational rehabilitation to individuals with disabilities. The rehabilitation strategy conducted by the two authorities follows a protocol and is known as Enhanced Cooperation. The rehabilitation intervention that constitutes the main control group consists of the interventions available within Enhanced Cooperation, in which all study participants are enrolled. Caseworkers within Regular Rehabilitation are calculated to have more than 40 individuals listed simultaneously (however, in practice, this figure is higher) and are responsible for both administering their records and coordinating activities.
Following assessment and enrolment, regular rehabilitation continues with follow-up meetings between the participant and the caseworkers from the two government authorities; at these meetings, results are discussed and activities can be planned, revised, or removed. Activities are classified into two types: work preparatory interventions and work-oriented interventions. Work preparatory interventions aim to prepare and empower the individual before he or she participates in work-oriented interventions. Work preparatory interventions are of low intensity and may include a workplace visit or job counseling. Work-oriented interventions may be offered directly or after a period of work preparatory interventions. Work-oriented activities include job search and workplace training.

Case Management
A second control group used in our study is Case Management, which is a high-intensity rehabilitation strategy focusing on individualized resource coordination, i.e. allowing the strengths, weaknesses and preferences of the individual to impact both the shape and content of the coordination efforts. This strategy was developed in the US during the 1970s and expanded during the 1980s into different models, where the role of the case manager, the intensity of the contact, and the degree of coordination may differ. Case management is used extensively for individuals with severe mental illness, with the goal of increasing well-being and reducing hospital time. The growth and spread of the model have been linked to the dismantling of institution-based care for people with long-term mental illness, a dismantling that has taken place throughout most of the western world. The role of Intense Case Management in reducing psychiatric morbidity is well evaluated with mixed but predominantly positive results, although its effectiveness seems most clear for individuals with severe problems and a high level of hospitalization. Much less is known about the effectiveness of the Case Management method as a vocational rehabilitation intervention aimed at participants’ increased labor market participation.
The version of the Case Management intervention model used in this study is the so-called resource model, selected because of its use within several Swedish municipalities. Following the resource model, the case manager plans the intervention based on the individual’s decisions and resources. The case manager has primary responsibility for coordinating and ensuring that the individual receives adequate care and support. The participation of the individual in planning the activities varies depending on his or her ability and preferences. So that the case managers can work independently of the Social Insurance Agency and the Public Employment Service in our study, they are employed by the local municipality. Thus, the Case Management intervention involves co-operation between both state authorities and the municipality. Case management in our study includes an orientation towards labor market participation. While it is up to the individual to select the scope and intensity of the rehabilitation, the case manager in the study setting has the obligation to help the individual navigate towards the long-term goal of labor market participation. To allow individualized high-intensity support, the method manuals emphasize that a case manager should have no more than 30 individuals listed simultaneously.
The general idea of the Case Management method is that continuity of care among treatment agencies is ensured by sufficient support and that the individual can develop coping skills that should allow increased involvement in community life and lead to greater autonomy. The Case Management intervention is thus a staffing intensive vocational rehabilitation strategy with a mechanism that does not specifically target the work place.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
- Non-subsidized employment on the regular labor market
- Subsidized employment on the regular labor market
- Sheltered employment
- Regular education (i.e., education that is not part of rehabilitation)
In the main analysis, these four categories will constitute one main outcome category called employment.
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
The recruitment of participants to the study took place from November 2014 to December 2015. Because the interventions were set to a duration of one year, the last person randomized into the experimental project finished rehabilitation in December 2016. The project was conducted in 25 Swedish municipalities (out of 290 municipalities in total) that were chosen to represent Sweden by including municipalities belonging to urban as well as rural areas. In total, 1,062 individuals participated in the experiment.
The recruitment of participants followed the regular recruitment process of individuals on disability pension. First, a caseworker from the Social Insurance Agency assesses whether the individual could be helped by vocational rehabilitation and asks whether the individual is interested in entering the labor market. The next step is follow-up meetings with further assessments of rehabilitation potential, conducted by caseworkers from both the Social Insurance Agency and the Public Employment Service. The aim of these meetings is to identify the individual’s work ability from a medical and labor market perspective. Outside of the study setting, an individual who is assessed as having the potential to be helped by vocational rehabilitation is to be offered the Regular vocational rehabilitation that constitutes the main control intervention in the study.

Individuals who were assessed to be eligible for rehabilitation received general information about the study and were then invited to participate. Those who declined the offer to participate were given the opportunity to receive Regular rehabilitation (79 individuals declined participation at this stage). Those who agreed to participate had to sign a consent form and were informed about the intervention to which they were randomized. Caseworkers had no prior information on which intervention the participant would receive – the information on the intervention to which the participant was randomized was kept in a closed envelope and opened at the meeting only after the individual had signed the consent. Therefore, how the caseworker presented the study to the participant did not depend on the outcome of the randomization. Information about the two interventions that were not offered to the participant was only provided if he or she asked for this information (34 percent either asked for this information or indicated that they had prior knowledge of the other interventions ). Four individuals declined participation after they had been randomized to the study; they were excluded from the study.
An interesting question is how our study population compares to the general population of young adults on disability pension. The majority of young adults on disability pension do not receive rehabilitation because they are assessed as unlikely to be helped by vocational rehabilitation. This exclusion could be due to a lack of labor market experience and/or very severe disabilities. In Table A in the Appendix, we compare our study population with the general population of young adults with disability pension and with the subpopulation of young adults with disability pension who receive vocational rehabilitation. The three groups resemble each other in many respects in their general demographics. However, we have access to three indicators reflecting different types of support that can be provided to young adults on disability pension. First, an individual can receive a legal decision containing the right to receive different kinds of support and services. Second, the individual can take part in institutionalized daily activities. Daily activities are available to give the person something interesting and meaningful to do during the daytime. Third, an individual can receive special housing support, which includes help with daily activities such as opening and reviewing mail, cleaning, weekly planning and/or communication with government authorities. These three indicators may reflect that a person has difficulties entering the labor market. Comparing the three indicators across the three groups, Table A in the Appendix suggests that our study group is somewhere in between the other two populations in terms of distance to the labor market. Our study population seems to have somewhat stronger attachment to the labor market than does the general population of young adults on disability pension but somewhat weaker attachment than does the subpopulation receiving regular rehabilitation. This conclusion is also reinforced by the fact that the assessment of eligibility for rehabilitation within our project was somewhat more lenient than what is typically the case.
Experimental Design Details
Randomization Method
Randomization done in office by a computer
Randomization Unit
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
Sample size: planned number of observations
Sample size (or number of clusters) by treatment arms
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB Name
IRB Approval Date
IRB Approval Number
Post Trial Information
Study Withdrawal
Is the intervention completed?
Is data collection complete?
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports, Papers & Other Materials
Relevant Paper(s)