Optimal Pricing Policy when Consumers Form Habits: The Case of Piped Water in Vietnam

How to cite

Do, Quy-Toan; Jacoby, Hanan. 2020. "Optimal Pricing Policy when Consumers Form Habits: The Case of Piped Water in Vietnam." AEA RCT Registry. AEARCTR-0006589. https://doi.org/10.1257/rct.6589-1.1

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

Quy-Toan Do* , Hanan Jacoby

* Primary investigator

Registry ID
AEARCTR-0006589
DOI
https://doi.org/10.1257/rct.6589-1.1
First registered
Last updated
Status
Completed
Trial dates
2015-07-01 – 2018-06-01
Countries
Viet Nam ()
JEL codes
H42, L95, D91, O18
Registry record
http://www.socialscienceregistry.org/trials/6589

Abstract

We ran a two-and-a-half year field experiment in a commune in the Red River Delta Region served by a single water utility. About 1,500 households connected to piped water were randomly divided into three groups. We then rolled out a price subsidy scheme and collect monthly information on water consumption use.

We use these data to estimate a dynamic structural model of water demand with habit formation. Under our maintained assumption that consumers do not internalize the effect of their future habits in deciding on their current piped water usage, consumption depends only on the current price and on past usage. Our experiment, by generating random variation in price, not only allows us to identify the (short-run) price elasticity of piped water demand, but also delivers valid instruments for estimating the effect of past consumption on current consumption and, by extension, the effect of current consumption on future consumption. With these preference parameter estimates, we can compute both ex-ante and ex-post willingness-to-pay as well as short-run and long-run price elasticities of demand.

We numerically solve for the optimal pricing scheme given consumer preferences as well as information on fixed investment and operating costs provided to us by the water utility.

Intervention

Prior to the price randomization, starting in July 2015, we `linearized' the water tariff schedule. At the time, the official schedule had three blocks: For consumption below 10 m3, the price was VND5,300 per m3, increased to VND7,200 per m3 for any consumption above 10 m3 and below 20 m3, and again to VND8,000 per m3 for any higher consumption. With the linearization, every household in the three villages was charged a uniform VND5,300 per m3 price and the water utility was reimbursed for the revenue loss relative to the block-price schedule. This linearized price schedule prevailed throughout the experiment, with randomized discounts applied at various points as described below. In June 2017, however, provincial authorities raised the water price to a uniform VND7,950 per m3, an increase which we passed on to all consumers and which was broadly announced by An Thinh. Throughout the experiment, the An Thinh water utility shared their administrative data on household water bills, which includes monthly water usage (in m3) and payment from July 2015 onward.

The actual price experiment started in January 2016, 6 months into the price linearization. The intervention consisted of offering households rebates on their (linearized) water price: 50 percent or 75 percent.

Experimental design

About 1,500 households connected to piped water were randomly divided into three groups. We then rolled out a price subsidy scheme. Group 1 first received a 50 percent discount on piped water for a period of 6 months, with Group 2 and Group 3 serving as controls. Beginning in the second 6-month period, Group 1 got an additional 50 percent off, bringing its discount to 75 percent, while households in Group 2 received a 50 percent discount and Group 3 remained a control group. Eventually, households in Group 3 began receiving 50 and then 75 percent discounts just as the other two groups had their discounts progressively phased out. At any point in time, therefore, at least one group did not receive a subsidy.

Design details

Randomization method
Randomization was done in office by a computer.
Randomization unit
Household
Clusters
1488 households
Observations
1488 households
Arms
Three groups for roll out subsidy: Group 1 = 496 households; Group 2 = 496 households; Group 3 = 496 households
Primary outcomes
household water consumption

Papers from this trial

  1. Quy-Toan Do and Hanan Jacoby, "Optimal Pricing of a New Utility Service: The Case of Piped Water in Vietnam", Policy Research Working Paper 9207: World Bank, April 2020. https://documents.worldbank.org/en/publication/documents-reports/documentdetail/321751586281645715/optimal-pricing-of-a-new-utility-service-the-case-of-piped-water-in-vietnam

Cited by

No citation data available for this registration yet.

Keywords