South Carolina Reemployment Services and Eligibility Assessments (RESEA) Randomized Controlled of Automated Job Referrals

How to cite

Jenkins, Ethan; Pepin, Gabrielle; O'Leary, Christopher J.; Sojourner, Aaron. 2026. "South Carolina Reemployment Services and Eligibility Assessments (RESEA) Randomized Controlled of Automated Job Referrals." AEA RCT Registry. AEARCTR-0017038. https://doi.org/10.1257/rct.17038-1.0

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

Ethan Jenkins* , Gabrielle Pepin , Christopher J. O'Leary , Aaron Sojourner

* Primary investigator

Registry ID
AEARCTR-0017038
DOI
https://doi.org/10.1257/rct.17038-1.0
First registered
Last updated
Status
In development
Trial dates
2026-01-26 – 2028-09-01
Countries
Private
JEL codes
J65, J64
Registry record
http://www.socialscienceregistry.org/trials/17038

Abstract

Job loss and long periods of joblessness, in particular, generate persistent earnings losses and negatively affect workers’ physical and mental health (Jacobson, LaLonde, and Sullivan 1993; Couch and Placzek 2010; Krueger and Mueller 2012; Damaske 2021; Fallick et al. 2025). As such, interventions intended to speed reemployment among displaced workers have the potential to substantially improve their long-term labor market outcomes and overall well-being. In this study, we will estimate causal impacts of providing frequent, timely, and customized automated job vacancy referrals to unemployment insurance (UI) beneficiaries on job search, UI receipt and duration, and labor market outcomes. We will randomly assign South Carolina Reemployment Services and Eligibility Assessments (RESEA) participants to receive staff-assisted registration for an automated job referral service called Virtual Recruiter® (VR). While all of South Carolina’s UI beneficiaries may register themselves for VR within the SCWorks Online System portal, participation in this service currently is quite low among RESEA participants. Thus, there is much scope for staff assistance in registering for VR to increase participation in the service, increase engagement with job vacancy postings, and potentially to speed reemployment and UI exit. We plan to enroll the RESEA population to the experimental treatment during a twelve-month period. We expect that the intervention will involve random assignment of about 17,000 UI beneficiaries referred to RESEA, providing power to detect effect sizes of magnitudes that we may reasonably expect.

Intervention

We will randomly assign RESEA participants to receive staff-assisted registration for an automated job referral service called Virtual Recruiter® (VR) provided by the vendor Geographic Solutions, Inc. While all of South Carolina’s UI beneficiaries may register themselves for VR within the SCWorks Online System (SCWOS) portal, participation in this service currently is quite low among RESEA participants: only 5.4 percent of the RESEA population with benefit year begin dates from January 2022 to June 2024 participated in VR (Pepin et al. 2025). Thus, there is much scope for staff assistance in registering for VR to increase participation in the service, increase engagement with job vacancy postings, and speed reemployment and UI exit.

Experimental design

We will randomly assign a full 12 months of RESEA participants to receive staff-assisted registration for an automated job referral service called Virtual Recruiter® (VR) provided by the vendor Geographic Solutions, Inc. This randomization occurs when a claimant is profiled for the first time as part of the existing RESEA profiling process.

While all of South Carolina’s UI beneficiaries may register themselves for VR within the SCWorks Online System (SCWOS) portal, participation in this service currently is quite low among RESEA participants. During the initial RESEA appointment, RESEA analysts will register treatment group participants with VR (if they are not participating already). Treatment group participants will reserve the right to opt out of registration, and control group participants still will be able to register for VR on their own. Additionally, unless the RESEA participant opts out, the RESEA analyst will set treatment participants’ VRs to run daily, explaining that this will allow them to view and apply for jobs as soon as they are posted. The RESEA analyst will recommend and assist treatment participants in opting into email or text message notifications that will alert them to their job referrals in case they do not login to the SCWOS portal every day and, unless the participant objects, will not change the default expiration date of 90 days from the date the VR is saved.

At each subsequent RESEA appointment, RESEA analysts will follow up with treatment participants who registered for VR. Specifically, they will ask each treatment participant whether they have been receiving job referrals from VR and, if so, whether those referrals have been relevant to the jobs in which they are most interested. If the treatment participant tells the RESEA analyst that their VR has not been sending relevant job referrals, the analyst will use their expertise and collaborate with the participant to edit the VR settings to better fit the jobseeker’s preferences. The RESEA analyst also will edit the VR to expire two months after the appointment so that it does not expire before the participant’s next RESEA appointment.

Design details

Randomization method
When a claimant is profiled for the first time as part of the existing RESEA profiling process, a random number will be generated that will be used to assign them into either a treatment or control group with equal probability. The random number with at least seven significant digits will be drawn from the uniform distribution with minimum value 0 and maximum value 1. The claimant will be assigned to the control group if the random number is less than 0.5 and to the treatment group if the random number is greater than or equal to 0.5.
Randomization unit
Individual
Clusters
A full 12 months of South Carolina RESEA participants. We expect that the intervention will involve random assignment of about 17,000 UI beneficiaries referred to RESEA, as 17,200 UI beneficiaries with benefit year begin dates from May 2023 to April 2024 were referred to the program.
Observations
A full 12 months of South Carolina RESEA participants. We expect that the intervention will involve random assignment of about 17,000 UI beneficiaries referred to RESEA, as 17,200 UI beneficiaries with benefit year begin dates from May 2023 to April 2024 were referred to the program.
Arms
We expect there to be approximately 8,500 UI beneficiaries assigned to the control group and 8,500 UI beneficiaries assigned to the treatment group.
Primary outcomes
1) Full-time equivalent weeks of UI compensation 2) Total UI compensation received ($) 3) UI exhaustion rates 4) Employed, 2 quarters after benefit year begin (BYB) date 5) Earnings ($), 2 quarters after BYB date 6) Employed, 6 quarters after BYB date 7) Earnings ($), 6 quarters after BYB date 8) Cumulative earnings and UI compensation combined, 6 quarters after BYB date 9) Cumulative earnings, 6 quarters after BYB date 10) Probability that the first reemployment job is in the same 5-digit Standard Occupational Classification (SOC) code as the pre-claim job (conditional on employment by 6 quarters after BYB date) 11) Probability that the first reemployment job is in the same 4-digit SOC code as the pre-claim job (conditional on employment by 6 quarters after BYB date) 12) Probability that the first reemployment job is in the same 3-digit SOC code as the pre-claim job (conditional on employment by 6 quarters after BYB date) 13) Probability that the first reemployment job is in the same 2-digit SOC code as the pre-claim job (conditional on employment by 6 quarters after BYB date)
Secondary outcomes
See analysis plan.

Cited by

No citation data available for this registration yet.

Keywords