Non-payment and strategic enforcement: Experimental evidence from Indian electricity

How to cite

Burlig, Fiona; Sudarshan, Anant. 2021. "Non-payment and strategic enforcement: Experimental evidence from Indian electricity." AEA RCT Registry. AEARCTR-0008742. https://doi.org/10.1257/rct.8742-1.0

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

Fiona Burlig* , Anant Sudarshan

* Primary investigator

Registry ID
AEARCTR-0008742
DOI
https://doi.org/10.1257/rct.8742-1.0
First registered
Last updated
Status
On going
Trial dates
2021-11-01 – 2023-12-27
Countries
Private
JEL codes
O13, Q48, H26
Registry record
http://www.socialscienceregistry.org/trials/8742

Abstract

State revenue collection is challenging in developing-country settings, which in turn reduces the ability of the state to provide public goods. Electricity is a key sector facing substantial revenue challenges. Frequent bill non-payment leads utilities to provide low-quality power supply, including imposing regular blackouts (Burgess et al 2020). In this project, we are working with two state utilities in Madhya Pradesh, India to measure the effectiveness of and design an optimal targeting strategy for enforcement. We randomize 30,000 households into a control group and a variety of enforcement treatments, informed by the public finance literature on tax payments (e.g. De Neve et al 2021), ranging from SMS reminders to disconnections. We plan to compute direct effects of these treatments on bill payments and electricity consumption. We will also estimate treatment effect heterogeneity. To consider optimal targeting from the household's perspective, we will conduct an endline survey to measure household finances, consumption, psychological well-being, and other measures, in order to consider optimal targeting from the household's perspective.

Intervention

30,000 households with a record of unpaid electricity bills will be randomly assigned to a control group or a series of utility enforcement treatments starting in December 2021.

Experimental design

We perform a stratified randomization where we randomly assign 30,000 households to the treatments described above. We are collecting administrative data from the utility on electricity consumption, bill payments, and connection status. We conducted a baseline survey in November-December 2021, and plan to conduct an endline survey in summer 2022 to examine impacts on additional outcomes such as household finances, consumption, and mental health.

Design details

Randomization method
Stratified randomization done in Stata
Randomization unit
Household
Clusters
30,000 households
Observations
30,000 households
Arms
7,500 households pure control. 2,500 households business as usual control. 5,000 households SMS reminder. 5,000 households SMS social comparison. 5,000 households mailed disconnection notice. 2,500 households in-person disconnection notice with no follow-up. 2,500 households in-person disconnection notice with subsequent disconnection.
Primary outcomes
Electricity consumption, bill payment, arrears, re-connection to the grid (if applicable). Endline survey outcomes to be added prior to fielding the survey.

Cited by

No citation data available for this registration yet.

Keywords