The Economics of the Public Option: Evidence from Local Pharmaceutical Markets

How to cite

Cuesta, Jose Ignacio; Atal, Juan Pablo; Otero, Cristobal; Gonzalez, Felipe. 2021. "The Economics of the Public Option: Evidence from Local Pharmaceutical Markets." AEA RCT Registry. AEARCTR-0008403. https://doi.org/10.1257/rct.8403-1.1

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

Jose Ignacio Cuesta* , Juan Pablo Atal , Cristobal Otero , Felipe Gonzalez

* Primary investigator

Registry ID
AEARCTR-0008403
DOI
https://doi.org/10.1257/rct.8403-1.1
First registered
Last updated
Status
Completed
Trial dates
2016-10-12 – 2017-01-19
Countries
Chile ()
JEL codes
D72, H4, I16, L3
Registry record
http://www.socialscienceregistry.org/trials/8403

Abstract

This field experiment informs the broader question about the effects of competition by state-owned firms on market outcomes. The study design is motivated by two facts. On the one hand, substantial price differences have been documented across private and public pharmacies. On the other hand, few consumers have registered and purchased products from the latter, despite being eligible to it. The goal of the field experiment was to study whether (lack of) information about public pharmacies affected awareness about it, pharmacy choice patterns, and perception about local politicians (who are accountable for their implementation).

Intervention

We designed an intervention to address information frictions. First, it could be the case that households are simply unaware of the existence of public pharmacies as an option in their choice sets. Lack of awareness of public pharmacies, may explain why families are not taking
advantage of them. Second, it could be the case that, while aware of its existence, households are already familiar with private pharmacies location and attributes, while not so for public pharmacies. If searching those characteristics is costly, then it might deter households from taking advantage of public pharmacies. The treatment takes the form of informational flyers that are specific to each local market. The flyer provides information about the existence of a public pharmacy in the county and two general features of it, namely the fact that it offers lower prices than private pharmacies and that it may take longer to deliver the products. It also includes information about location, contact information, hours of attention and eligibility requirements.

Experimental design

We designed a field experiment to study whether the availability of public pharmacies affected consumer shopping behavior. To induce variation in awareness about the public pharmacy in their local market, we implemented an informational intervention. The decision to provide information was based on a survey we conducted before the experiment, which revealed that consumers were only partially informed along two dimensions. First, some households were unaware of the existence of a public pharmacy in their county. Second, even when households knew about the pharmacy, they were not perfectly informed about the lower prices and other attributes. The existence of imperfect information provides us with a unique opportunity to randomly expose consumers to public pharmacies using our experiment, and thus to measure individual responses to them.

The treatment consisted of an informational flyer. It provided information about the existence of a public pharmacy in the county, and stated that it offered lower prices and longer waiting times than private pharmacies. Additionally, it included its location, contact information, opening hours, and eligibility requirements. We delivered the flyer to consumers coming out of private pharmacies in the 20 counties with public pharmacies in Santiago. The information was tailored to each county.

In terms of recruitment, enumerators approached consumers leaving a private pharmacy in each county and assessed their eligibility. Eligible participants were those who (i) lived and were registered to vote in the county, (ii) had purchased a prescription drug, and (iii) were not registered in the public pharmacy. Those who accepted to be part of the study, answered the baseline survey. Once the survey was completed, participants were randomly assigned to treatment and control groups. The enumerator only learned the assignment of the individual after completing the survey. We implemented this survey between October 12 and 20, 2016, right before mayoral elections. Two months after the baseline survey, we conducted a follow-up survey to measure the same variables as in baseline. We implemented this survey by phone.

Design details

Randomization method
Randomization was implemented at the point of survey, within the electronic form.
Randomization unit
Individual level randomization.
Clusters
826 individuals enrolled in the study.
Observations
826 individuals enrolled in the study.
Arms
507 individuals were assigned to the treatment group, 319 were assigned to the control group.
Primary outcomes
On the economic side, we focus on consumer knowledge, perceived prices, perceived waiting times, enrollment, purchase behavior. On the political side, we focus on self reported voting behavior for the incumbent mayor or her party.

Cited by

No citation data available for this registration yet.

Keywords