The Value of Early Labor Market Experience: Evidence from a Large-Scale Summer Job Randomization

How to cite

Lundborg, Nils-Petter. 2025. "The Value of Early Labor Market Experience: Evidence from a Large-Scale Summer Job Randomization." AEA RCT Registry. AEARCTR-0015301. https://doi.org/10.1257/rct.15301-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Nils-Petter Lundborg*

* Primary investigator

Registry ID
AEARCTR-0015301
DOI
https://doi.org/10.1257/rct.15301-1.0
First registered
Last updated
Status
In development
Trial dates
2012-01-01 – 2026-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/15301

Abstract

This project analyzes data from a natural experiment in which the municipality of Malmö used a lottery to randomly allocate approximately 50,000 summer jobs to high school students between 2012 and 2020. The program provided students with four weeks of paid employment during the summer break, with 6.5-hour workdays. This project will link the lottery data to register-based information on education, labor market participation, and crime, to study the long-term effects of summer jobs on these outcomes. Beyond the direct impact on participants, the project will also investigate potential spillover effects on their siblings and parents.

Intervention

The summer jobs program in Malmö provided high school students aged 16–19 with four weeks of paid employment during the summer break, with 6.5-hour workdays. Job placements covered a wide range of sectors, including administration, childcare and youth services, community organizations and recreation, culture, gardening and maintenance, and health and social care. To ensure fairness in allocation, a lottery was used to randomly assign jobs to applicants. The intervention and the lottery were performed by the municipality of Malmö, without the involvement of researchers.

Experimental design

All youth aged 16–19 attending a high school in the municipality of Malmö were eligible to participate in the summer employment lottery, organized by the municipality. The program has been consistently oversubscribed in every year for which we have data, meaning that the number of applicants has exceeded the available job placements. The municipality therefore used a lottery to allocate the summer jobs. In this project, we use the lottery data maintained by the municipality of Malmö to estimate the causal effect of summer employment on long-term outcomes.

As part of the application process, applicants were asked to specify their preferences regarding various aspects of the summer job, including the time period, job type (e.g., administration, childcare and youth services, community organizations and recreation, culture, gardening and maintenance, and health and social care), and geographical location. The lottery was conducted taking into account each applicant's stated preferences, ensuring that assignments aligned as closely as possible with their preferences.

Design details

Randomization method
The municipality of Malmö conducted the lottery, with the randomization process carried out by a computer in an office.
Randomization unit
Individual randomization.
Clusters
No cluster design.
Observations
Between 2012 and 2020, a total of 54,442 lottery particants were recorded. Since some pupils participated multiple times, the number of unique participants was 27,757.
Arms
Between 2012 and 2020, a total of 33,872 participants were offered a summer job through the lottery, of whom 32,915 accepted the offer.
Primary outcomes
Outcomes for lottery participants and siblings: Educational outcomes: high-school grades and national test scores, high-school completion, university enrollment, field of study, years of schooling. Labor outcomes: employment, unemployment, occupation, sector, earnings, social welfare receipt, sick-leave. Crime outcomes: criminal convictions, including the type of crime (Alcohol- and drug-related crimes, as well as violent crimes). Outcomes for parents: Labor outcomes: employment, unemployment, occupation, sector, earnings, social welfare receipt, sick-leave Crime outcomes: criminal convictions, including the type of crime (Alcohol- and drug-related crimes, as well as violent crimes)

Cited by

No citation data available for this registration yet.

Keywords