Technology and Audit Interventions to Combat Theft in Pakistan's Power Sector

Last registered on March 05, 2026
View Trial History

Pre-Trial

Trial Information

General Information

Title
Technology and Audit Interventions to Combat Theft in Pakistan's Power Sector
RCT ID
AEARCTR-0015084
Initial registration date
February 25, 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
March 05, 2026, 6:14 AM EST

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
Muhammad Usman Naeem
Affiliation
Tufts University
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Tim Dobermann
PI Affiliation
London School of Economics and Political Science
Contact Investigator
PI Name
Robin Burgess
PI Affiliation
London School of Economics and Political Science
Contact Investigator
PI Name
Michael Greenstone
PI Affiliation
University of Chicago
Contact Investigator
PI Name
Faraz Hayat
PI Affiliation
International Growth Centre
Contact Investigator

Additional Trial Information

Status
In development
Start date
2024-02-01
End date
2026-12-31
Keywords
Behavior, Crime, Violence, & Conflict, Environment & Energy, Governance
Additional Keywords
Electricity theft in Pakistan, Aerial bundled cables (ABC), Randomized Controlled Trial (RCT), Energy Pakistan, Revenue Recovery, Advanced Metering Infrastructure (AMI), Kunda, Hook
JEL code(s)
O13 (Economic Development: Agriculture; Natural Resources; Energy; Environment; Other Primary Products), Q48 (Energy: Government Policy), C93 (Design of Experiments: Field Experiments), L94 (Industry Studies: Utilities and Transportation: Electric Utilities), K42 (Illegal Behavior and the Enforcement of Law), O33 (Technological Change: Choices and Consequences; Diffusion Processes), D12 (Consumer Economics: Household Analysis)
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
Pakistan's power sector faces a severe fiscal crisis, with high rates of electricity theft and non-payment contributing to a circular debt at approximately 2.1% of GDP, that threatens the country's economic stability. This situation underscores the urgent need for scalable solutions to reduce losses and improve revenue recovery. Conducted in partnership with the Lahore Electric Supply Company (LESCO) and the Ministry of Energy, this study evaluates the comparative effectiveness of technological versus enforcement-based interventions in high-loss areas of Lahore.

Transformers are randomly allocated into three experimental groups to evaluate distinct anti-theft strategies. The first group (T1) receives the technological intervention of Aerial Bundled Cables (ABCs), a theft-resistant infrastructure designed to physically prevent unauthorized hook connections. The second group (T2) is subject to an enforcement intervention termed "enhanced combing," involving an increased frequency of audits where utility personnel actively identify theft, disconnect illegal users, and issue penalties. Finally, the control group serves as a baseline, maintaining status-quo operations to allow for comparative analysis of intervention effectiveness.

The primary objective is to measure both the short-term and sustained impacts of these interventions on electricity theft (losses), bill payment rates, and consumer regularization. By leveraging high-frequency administrative data from smart metering, the study tracks whether these strategies lead to lasting behavioral change. Key outcomes assessed include the rate of electricity theft (calculated as the difference between electricity supplied and billed), the non-payment rate, and shifts in consumer numbers and behaviors. Additional secondary outcomes, such as accumulated arrears, will also be examined using detailed administrative data provided by LESCO.

The findings from this evaluation will directly inform decisions by the Ministry of Energy and LESCO regarding the most cost-effective strategies for combating electricity theft and enhancing fiscal stability.
External Link(s)

Registration Citation

Citation
Burgess, Robin et al. 2026. "Technology and Audit Interventions to Combat Theft in Pakistan's Power Sector ." AEA RCT Registry. March 05. https://doi.org/10.1257/rct.15084-1.0
Sponsors & Partners

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

Request Information
Experimental Details

Interventions

Intervention(s)
(i) Technological Intervention (ABC): This arm involves the replacement of existing bare low-voltage distribution lines with insulated Aerial Bundled Cables (ABC). This infrastructure is designed to physically prevent the attachment of illegal hook connections (kundas) by shielding the live conductor, thereby increasing the cost and difficulty of electricity theft.

(ii) Enforcement Intervention (Combing): This arm implements enhanced combing, a rigorous enforcement regime characterized by high-frequency audits. Dedicated utility teams inspect the transformer perimeter to identify theft, dismantle illegal connections , and issue detection bills to offenders. This intervention tests the efficacy of administrative enforcement without infrastructure upgrades.

(iii) Control: A passive control group where transformers receive no new infrastructure or special audit teams, continuing with LESCO's standard operating procedures.
Intervention Start Date
2026-03-02
Intervention End Date
2026-08-14

Primary Outcomes

Primary Outcomes (end points)
Transformer Loss Rate (Theft), Bill Payment Rate (Recovery), Number of Active Connections (Regularization)
Primary Outcomes (explanation)
(i) Theft Rate - Transformer Loss Rate (Theft) – defined as the difference between the total energy units supplied to a transformer (measured via AMI) and the total units billed to consumers, divided by units supplied. If the technological (ABC) and enforcement (Combing) interventions are effective, we expect a significant decline in this rate. This serves as the direct measure of the intervention's success in reducing illegal connections.

(ii) Bill Payment Rate (Recovery) – defined as the ratio of total cash collected to the total amount billed for a given month. A critical concern is that if households are prevented from stealing electricity, they may simply refuse to pay their bills, effectively substituting theft with non-payment. Tracking this outcome tests whether the interventions actually translate into increased government revenue or simply shift the type of financial loss.

(iii) Number of Active Connections (Regularization) – defined as the count of active, metered consumers associated with a transformer. We hypothesize that as the cost of theft increases (due to ABCs or enforcement), households previously relying on illegal connections will be forced to apply for legal meters. An increase in this count would indicate successful "regularization" of the user base.

Secondary Outcomes

Secondary Outcomes (end points)
Consumer Arrears, Consumption Units
Secondary Outcomes (explanation)
(i) Consumer Arrears – defined as the total outstanding balance accumulated by consumers on a transformer. We track this to assess whether the interventions inadvertently cause households to accumulate debt rather than clear their dues.

(ii) Consumption Units – defined as the average billed kWh per active connection. We hypothesize that reduced theft and improved reliability may lead to an increase in legal consumption

Experimental Design

Experimental Design
This study employs a cluster-randomized controlled trial (RCT) design to evaluate strategies for reducing electricity theft. The unit of randomization is the distribution transformer, representing a distinct neighborhood network. The study sample consists of 120 high-loss transformers selected from the Lahore Electric Supply Company (LESCO) service area .

To ensure balance, transformers were stratified by administrative circle (Eastern and Southern) and randomly assigned into three experimental arms (N=40 per arm):
(i) Technological Intervention: Replacement of bare wires with theft-resistant Aerial Bundled Cables (ABC).
(ii) Enforcement Intervention: Implementation of "enhanced combing," a high-frequency audit and disconnection drive.
(iii) Control: Continuation of status-quo utility operations.

High-frequency outcome data is collected via Advanced Metering Infrastructure (AMI) installed at the transformer level, complemented by monthly administrative billing and collection records. Evaluation relies exclusively on confidential administrative data already generated by LESCO. We assemble (i) a monthly transformer‑level panel tracking units supplied, units billed, and technical losses from the date each transformer was metered, and (ii) matched household‑level records of billing, payments, arrears, and consumption for every customer connected to study transformers. These data will continue to flow for at least 18  months after installation, allowing us to trace short‑ and medium‑run effects.
Experimental Design Details
Not available
Randomization Method
Computer Generated Randomization
Randomization Unit
Electricity Distribution Transformer
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
120 Electricity Distribution Transformer
Sample size: planned number of observations
120 (Transformers) * 100 (Households per Transformer - estimated) = 12,000
Sample size (or number of clusters) by treatment arms
40 (ABC Treated Transformers), 40 (Audit Only Transformers) and 40 (Control Transformers)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number
Analysis Plan