Overcoming labor market shortages for the energy transition: Experimental evidence

How to cite

Shukla, Devanshee. 2025. "Overcoming labor market shortages for the energy transition: Experimental evidence." AEA RCT Registry. AEARCTR-0016424. https://doi.org/10.1257/rct.16424-1.1

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

Devanshee Shukla*

* Primary investigator

Registry ID
AEARCTR-0016424
DOI
https://doi.org/10.1257/rct.16424-1.1
First registered
Last updated
Status
In development
Trial dates
2025-07-21 – 2026-05-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/16424

Abstract

Firms worldwide are facing challenges in recruiting and retaining entry-level, low-skilled labor, a situation particularly critical in nascent industries characterized by rapid growth and labor shortages. Such recruitment frictions impose severe constraints on firms, manifesting in outright labor shortages, frequent churn due to mismatched expectations among workers, and persistently low-quality candidate pools. These frictions undermine firm productivity, escalate hiring costs, and jeopardize competitive advantage. This project aims to understand whether varying what job‑training information is shared and how it is delivered can boost potential employees' uptake of entry‑level roles in solar energy sector, by running a village‑level field experiment with a large training provider across Indian states.

Intervention

We will implement a cluster-randomized controlled trial involving youth aged 18–35 across 384 villages. Our experimental design adopts a 2×2 factorial approach, systematically varying the type of information provided (course details alone versus comprehensive salary and career trajectory data) and its delivery modality (digital dissemination versus face-to-face interaction).

Experimental design

Villages are assignment randomly to one of 4 conditions (control, treatment 1, treatment 2 or treatment 3).

Design details

Randomization method
Randomization done in office.
Randomization unit
We first stratify on state and prior exposure to the training organization (proxied by the number of students enrolled in any program of the organization; 0=new village, no student enrolled in the organization's program and 1=old village where upto 5 students may be enrolled in the organization's program), and then randomly assign villages in each strata to one of the treatment arms.
Clusters
The randomization will be clustered at the village level. The planned number of clusters are 384 villages.
Observations
After discussion with the organization, we expect to find between 10-15 eligible youth in each village, making our expected sample size between 3840-5760 (exact number will depend on how many youth we are actually able to find in every village).
Arms
We have 384 villages in our sample. Of these, we have 96 villages in the control arm and 96 villages in each of the 3 treatment arms.
Primary outcomes
Enrollment into the solar training course
Secondary outcomes
Subject to data availability, we aim to also track how many of those enrolled complete the course, and their performance in the course. Depending on data availability, we will also measure the cognitive and technical performance of the enrolled candidates in the course assessments that are part of the online course. We will also explore which candidates are most likely to enroll and complete the course* (subject to data availability) based on the prior education levels of youth, their work experience and their ex-ante exposure to the solar sector, their expected starting wage in the sector after training and their initial reservation wages. We will also examine how the beliefs and expectations of youth in terms of their expected salary in the sector update after the intervention. Subject to budget and data availability, we will also try to track the youth and examine the course completion and subsequent placement outcomes of the youth.

Cited by

No citation data available for this registration yet.

Keywords