Our selection of research on Policy Evaluation

We have prepared a short summary of two academic articles that might be of interest to you.

The first one is about how choices between academic and vocational tracks affect dropouts and inactivity in Sweden, while the second is about the consequences of welfare cuts on longer-term educational outcomes of children of refugees in Denmark.

Academic or Vocational? How Track Choice Affects Dropout and Inactivity in Sweden

Taghizadeh & Österman (2025), Economics of Education Review

Link to the publication here

This study sheds light on a critical decision faced by students in many education systems: whether to enrol in a general (academic) or vocational upper secondary track. In Sweden, students apply to both types of programmes through a centralised, grade-based system. The authors leverage this system to conduct a regression discontinuity analysis that isolates the effect of track assignment on educational and early adulthood outcomes.

The findings show that among students who applied to both tracks, those who were admitted to general (academic) programmes were less likely to graduate on time and spent more time as NEETs (Not in Education, Employment, or Training) during early adulthood.

This research adds nuance to the debate over academic versus vocational education. While general education is often seen as the “higher” track, for many students, vocational education may be more beneficial for those with mid-range academic profiles who are at the margin between the two paths.

Overall, the study suggests that better matching students to the right educational path, rather than pushing general programmes as a default, could reduce dropout rates and youth inactivity.

(a) Graduation within 3 years - 1

(a) Graduation within 3 years - 2

Unintended Consequences of Welfare Cuts on Children and Adolescents

Dustmann, Christian, Rasmus Landersø, and Lars Højsgaard Andersen. 2024. American Economic Journal: Applied Economics 16 (4): 161–85.

Link to the paper here and to the IZA DP series here.

In 2002, Denmark introduced the "Start Aid" reform, which drastically cut financial assistance to newly arrived refugees by an average of 40%, and up to 50% for families. A recent academic paper has meticulously analyzed the long-term consequences of this reform on refugee children over nearly two decades, revealing a series of severe negative impacts across various aspects of their lives.

Immediately after the reform, refugee families saw their disposable income plummet by 50%, a reduction that averaged 30% over the first five years following residency. For young children, exposure to the reform meant a 24% lower chance of attending preschool, which alarmingly widened the existing achievement gap between refugees and native Danes by 40%. As these children grew older, those who were school-aged when the reform took effect experienced a 20% decrease in the total years of completed post-compulsory education. The study followed these children into adulthood, finding that the reform led to significantly lower employment rates and earnings, consequently increasing the earnings gap between native Danes and immigrants. Furthermore, for teenagers who immigrated with their parents, the reduction in aid surprisingly resulted in a doubling of crime rates, encompassing both property and violent offenses.

Although the researchers identified a short-term positive return of roughly $2,500 per family, primarily due to adults entering the workforce, this initial gain was completely overshadowed by the significant long-term detriments to the children. As these unforeseen, longer-term consequences materialized, the overall financial impact of the reform turned negative, ultimately costing approximately -$12,000 per family.

Figure 3. Refugees’ GPA, crime, education, and earnings by timing of residency before/after reform.

A) GPA, standardized (mean=0, standard deviation=1 in full population), children aged 0-7 at residency

B) Probability of receiving a crime conviction, year 1–10 after residency, children aged 14–18 at residency

C) Years of schooling in 2020, children aged 8–14 at residency

D) Earnings in $1,000, 15–16 years after residency, children aged 8–14 at residency

Note: The figure shows raw plots, by timing of residency in bins, of GPA (standardized) for refugees age 0-7 at residency in Fig A), the fraction of refugees aged 14-18 at residency with a crime conviction during their first 10 years after residency in Fig. B), average years of schooling measured in 2020 for refugees aged 8-14 at residency in Fig. C), and average earnings measured 15-16 years after residency for refugees aged 8-14 at residency in Fig. D). All figures contain linear slopes of the predictions before and after the reform based on Eq. (1).