Enhanced Salience of Nonlinear Pricing and Energy Conservation

How to cite

Ko, Minkyong; Ta, Chi; Garg, Teevrat. 2023. "Enhanced Salience of Nonlinear Pricing and Energy Conservation." AEA RCT Registry. AEARCTR-0012178. https://doi.org/10.1257/rct.12178-2.0

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

Minkyong Ko* , Chi Ta , Teevrat Garg

* Primary investigator

Registry ID
AEARCTR-0012178
DOI
https://doi.org/10.1257/rct.12178-2.0
First registered
Last updated
Status
Completed
Trial dates
2023-10-01 – 2024-06-30
Countries
Viet Nam (Hanoi)
JEL codes
Q41, D03, D12
Registry record
http://www.socialscienceregistry.org/trials/12178

Abstract

Digital tools hold promise for scaling energy conservation by giving households real-time information about their electricity use and costs. Yet whether such app-based interventions can meaningfully reduce consumption depends on users’ engagement. We conduct a natural field experiment on a random sample of 45,000 electricity customers in Hanoi, Vietnam, that tested two mobile-app interventions built on the utility’s smart-meter platform. One treatment (“price salience”) displayed each household’s current marginal price tier and consumption to date; the other (“billing salience”) showed consumption and bill to date. Across the full sample, neither intervention reduced electricity use on average, and we can rule out effects as small as one percent. To understand this precise null, we examine engagement with the app and find no effects on the extensive margin, and only limited responses on the intensive margin. Among households that already engage with the app, the price-salience treatment modestly increased engagement and led to small consumption declines late in the billing cycle, when marginal prices rise mechanically under the nonlinear tariff. These results underscore both the promise and limits of digital behavioral tools for demand management -- while low-cost app integrations can inform attentive users, engagement does not necessarily scale with delivery, limiting the ability of such interventions to automatically generate population-level energy savings.

Intervention

he experiment evaluates two information interventions delivered through the electricity utility’s mobile application using smart-meter data. One treatment provides households with real-time information on their current marginal price tier and cumulative electricity consumption, while the second provides real-time information on cumulative consumption and bill-to-date; the control group receives no additional pricing or billing information beyond the app’s standard features.

Experimental design

We are conducting a large-scale experiment covering 45,000 users of a recently launched mobile app from a state-owned electric utility company in Vietnam. The treatment groups receive either a real-time app display of their estimated daily marginal prices or their total estimated bills to date, for a minimum of six months.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
individual household
Clusters
N/A
Observations
45,000 households
Arms
15,000 households in control, 15,000 households in each treatment groups
Primary outcomes
Reduction in electricity consumption at the household level, resulting from the provision of information on the nonlinear pricing system

Cited by

No citation data available for this registration yet.

Keywords