Job Matching for Displaced Contractual Meter Readers under India's Smart Metering Transition: A Randomised Controlled Trial

Last registered on April 13, 2026

Pre-Trial

General Information

Title

RCT ID

AEARCTR-0018298

Initial registration date

April 08, 2026

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

April 13, 2026, 9:23 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Primary Investigator

Name

Srikant Shaw

Affiliation

Indian Institute of Technology Madras

Contact Primary Investigator

Other Primary Investigator(s)

Additional Trial Information

Status

In development

Start date

2026-08-01

End date

2028-05-31

Keywords

Behavior, Environment & Energy, Labor, Welfare

Additional Keywords

JEL code(s)

J63, J65, Q41, O13, L94

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

India's Revamped Distribution Sector Scheme (RDSS) is deploying AI-assisted automated metering infrastructure across 42 electricity distribution utilities (DISCOMs) to target 250 million smart meter installations. The tasks automated by these systems, including physical meter reading, manual billing, and field-based fault detection, define the routine occupational core of an estimated three to four million contractual workers in the Indian electricity distribution sector. Unlike permanent DISCOM employees protected by the Industrial Disputes Act 1947, contractual workers have no statutory redundancy protection and are the primary channel through which AI-induced displacement will materialise. This study evaluates whether a structured job-matching intervention can mitigate the employment and earnings costs faced by contractual meter readers in DISCOM distribution circles scheduled for imminent smart meter activation. Contractual workers in treatment circles will receive individualised labour market information, active referral to local employers in adjacent electrical and data-entry occupations, and facilitated registration on India's National Career Service Portal. Workers in control circles receive only the current DISCOM worker communication standard, which provides no active transition support. The primary outcome is employment status at 6 months post-activation. Secondary outcomes include monthly earnings, occupational transition, and subjective well-being at 6 and 18 months. The study is embedded within a complementary quasi-experimental difference-in-differences analysis of RDSS rollout effects on DISCOM employment structure at the utility level, which provides the macro-level displacement estimates that motivate and contextualise the RCT.

External Link(s)

Registration Citation

Citation

Shaw, Srikant . 2026. "Job Matching for Displaced Contractual Meter Readers under India's Smart Metering Transition: A Randomised Controlled Trial." AEA RCT Registry. April 13. https://doi.org/10.1257/rct.18298-1.0

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

Contractual meter readers in treatment distribution circles receive a structured job-matching package comprising three components: (1) an individualised labour market information session covering local employer demand, wage benchmarks, and occupational transition options in adjacent electrical and data-oriented roles; (2) active referral to a curated list of local employers in electrical contracting, solar installation, building wiring, and MSME data-entry sectors; and (3) facilitated registration and profile completion on India's National Career Service Portal (NCS Portal, Ministry of Labour and Employment). The intervention is delivered by trained field enumerators approximately 4 to 8 weeks before the worker's circle goes live with smart meter activation. Workers in control circles receive only a general information sheet about government retraining schemes, equivalent to the current DISCOM practice. Each treated worker will receive referrals to at least 3–5 potential employers identified through the baseline employer mapping exercise.

Intervention Start Date

2026-11-01

Intervention End Date

2027-05-31

Primary Outcomes

Primary Outcomes (end points)

Employment status at 6 months post smart meter activation (employed, self-employed, or unemployed), measured by structured phone survey.

Primary Outcomes (explanation)

Employment status is a binary indicator (1 = employed or self-employed, 0 = unemployed or exited labour force) collected at the 6-month endline phone survey. A worker is classified as employed if they report having worked for pay, profit, or family gain for at least one hour in the 7 days preceding the survey, consistent with the PLFS Current Weekly Status definition.

Secondary Outcomes

Secondary Outcomes (end points)

1. Monthly earnings at 6 months post-activation
2. Monthly earnings at 18 months post-activation
3. Occupational transition: whether the worker is employed in an adjacent electrical or data-oriented occupation at 6 months
4. Employment status at 18 months post-activation
5. Subjective wellbeing (Cantril Self-Anchoring Scale, 0 to 10) at 6 months

Secondary Outcomes (explanation)

Monthly earnings are self-reported total income from all sources in the preceding 30 days, expressed in nominal INR. Occupational transition is coded 1 if the worker reports employment in any of the following NCO 2015 categories: electrical and electronics trades workers (NCO 741), data entry clerks (NCO 412), solar photovoltaic installation workers, or similar adjacent roles identified in the baseline employer mapping exercise. Subjective wellbeing is measured using the Cantril ladder question: "On a scale of 0 to 10, where 0 is the worst possible life and 10 is the best possible life, where do you stand today?"
Attrition will be addressed using inverse probability weighting and robustness checks comparing baseline characteristics of attriters and non-attriters.

Experimental Design

This study uses a cluster-randomised controlled trial in which distribution circles within a single RDSS-participating Indian DISCOM serve as the unit of randomisation. The primary analysis will estimate intention-to-treat (ITT) effects based on assignment to treatment, with treatment-on-the-treated (TOT) effects estimated using take-up as an endogenous variable where applicable.

Experimental Design Details

Not available

Randomization Method

Computer-generated random number sequence, stratified by circle size and activation schedule, conducted in the study office prior to any worker contact.

Randomization Unit

Distribution circle (cluster). Each DISCOM distribution circle constitutes one cluster. Individual workers within circles are enrolled but not individually randomised; all enrolled workers within a circle receive the same treatment assignment.

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

16 distribution circles (8 treatment, 8 control). Geographically proximate circles will not be assigned to different treatment arms, where feasible, to minimise contamination across clusters.

Sample size: planned number of observations

320 contractual meter readers (approximately 20 per circle). The sampling frame will be constructed using contractor workforce lists provided by the participating DISCOM or contractor firms.

Sample size (or number of clusters) by treatment arms

8 circles treatment (approximately 160 workers), 8 circles control (approximately 160 workers)

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

MDE of 8 percentage points (absolute) on the primary outcome (employment rate at 6 months). Assumptions: control group employment rate 65%, power 0.80, significance level 0.05, intracluster correlation coefficient (ICC) 0.05, average cluster size 20 workers. Design effect = 1 + (20 - 1) × 0.05 = 1.95. Effective sample size per arm = 160 / 1.95 ≈ 82 workers. This yields 80% power to detect an 8 percentage point difference in employment rates between treatment and control. The 8 pp MDE is consistent with effect sizes reported in comparable job-matching interventions in developing countries (Abebe et al., 2021, Review of Economic Studies; Beam, 2016, Journal of Development Economics).

Supporting Documents and Materials

IRB

Institutional Review Boards (IRBs)

IRB Name

IRB Approval Date

IRB Approval Number

Analysis Plan