An experimental evaluation of active labor market programs

How to cite

Zanella, Giulio; Salomone, Riccardo. 2020. "An experimental evaluation of active labor market programs." AEA RCT Registry. AEARCTR-0004978. https://doi.org/10.1257/rct.4978-1.1

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

Giulio Zanella* , Riccardo Salomone

* Primary investigator

Registry ID
AEARCTR-0004978
DOI
https://doi.org/10.1257/rct.4978-1.1
First registered
Last updated
Status
Completed
Trial dates
2017-02-01 – 2017-12-31
Countries
Italy (Provincia di Trento)
JEL codes
J64, J65, J68
Registry record
http://www.socialscienceregistry.org/trials/4978

Abstract

The field experiment consists of a randomized control trial (RCT) in which two active labor market policies were randomly offered to unemployed workers in the fifth month of unemployment insurance receipt: a traditional job search assistance program (JSA, offered to 660 job seekers) or a reemployment voucher (RV, offered to 754 job seekers) consisting of a bonus for a specialized third party service provider (a lump sum payment of up to 5000 euros) if the provider successfully matches a job seeker with a new employer within six months. A third group of 1400 randomly-selected job seekers from the same population was assigned to a control group, which could not received JSA or RV for a period of six months.

Intervention

The RCT involves 2,814 job seekers randomly selected in the Spring of 2017 from the population of unemployed workers in the Province of Trento (Italy) who are receiving unemployment benefits and not benefiting from any active labor market policy at the time of selection. These job seekers were randomized into three distinct groups. First, a group of 660 individuals assigned to receive a job search assistance program consisting of between 10 and 40 hours of personalized coaching in writing an effective resume, searching job openings data bases, managing job interviews, as well as self-assessment and motivation. Second, a group of 754 workers assigned to receive a reemployment voucher which they can spend at a public or private provider to receive personalized reemployment services. Finally, 1400 workers are assigned to a control group which for six months cannot be enrolled in job search assistance activities or be offered a reemployment voucher. The 1,414 unemployed workers assigned to one of the two treatments received an identical letter from the public employment agency informing them of their eligibility for a program aimed at helping them finding a new job. This letter invited the recipients to show up at the local public employment agency office to receive more information and to activate the program. Activation was not mandatory (the letter could be ignored with no consequences for unemployment insurance receipt or future reemployment services).

Experimental design

Given the nature of the randomization, we will compare outcomes by treatment status to identify the intention-to-treat effect. Moreover, by instrumenting take-up of the randomly offered policy by assignment status we will be able to identify the local average treatment effect by 2-stage least squares

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Individual
Clusters
2814 job seekers
Observations
2814 job seekers
Arms
between 50 and 900 job seekers in each local employment center
Primary outcomes
Employment status after four months and after six months from assignment to treatment

Cited by

No citation data available for this registration yet.

Keywords