To Charge or Not to Charge: Pricing Health Products in Northern Uganda

How to cite

Karlan, Dean; Raffler, Pia; McConnell, Margaret; Fischer, Greg. 2016. "To Charge or Not to Charge: Pricing Health Products in Northern Uganda." AEA RCT Registry. AEARCTR-0001414. https://doi.org/10.1257/rct.1414-1.0

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

Dean Karlan* , Pia Raffler , Margaret McConnell , Greg Fischer

* Primary investigator

Registry ID
AEARCTR-0001414
DOI
https://doi.org/10.1257/rct.1414-1.0
First registered
Last updated
Status
Completed
Trial dates
2011-10-01 – 2011-12-31
Countries
Uganda (Africa)
JEL codes
D11, D12, D83, I11, I18, O12
Registry record
http://www.socialscienceregistry.org/trials/1414

Abstract

In a field experiment in Uganda, a free distribution of three health products lowers subsequent demand relative to a sale distribution. This contrasts with work on insecticide-treated bed nets, highlighting the importance of product characteristics in determining pricing policy. We put forward a model to illustrate the potential tension between two of these important factors, learning and anchoring, and then test this model with three products selected specifically for their variation in the scope for learning. We find the rank order of percentage change of shifts in demand matches theoretical predictions, although the differences are not statistically significant, and only two of three pairwise comparisons match when the reductions are specified in percent terms. These results highlight the importance of understanding pricing policy with respect to product, market, and household characteristics.

Experimental design

This project sought to evaluate the tension between two important factors, learning and anchoring, in determining the demand for health products in less-developed countries, and help settle the debate in the development aid community over the pricing of these products. A field experiment in northern Uganda was designed where health products were distributed door-to-door either for free or for sale at market prices. Three products were offered that differed in their scope for learning: Panadol, a pain reliever widely known to consumers and for which no scope for learning was expected; Elyzole, a deworming drug that was moderately well-known and for which negative learning due to side effects was expected; and Zinkid, an improved treatment for childhood diarrhea that was largely unknown and for which positive learning was expected. Approximately ten weeks after the initial distribution, representatives from an unrelated, for-profit firm offered the households either the same or a new health product for sale at market prices. The key outcome measure is households’ purchase decisions in this second distribution.

Design details

Randomization method
The first wave of marketing (Wave 1), conducted in October-November 2011, employed a two-level clustered randomization design, with randomization both at the village and individual level. First, villages were randomly assigned to one of four treatment groups in a two-by-two design. The first treatment dimension was the price of the product, either free (“Free”) or sold (“Sale”). The second dimension was the type of distributing organization: either a not-for-profit, non-governmental organization (“NGO”) or a for-profit business (“For-Profit”). Thirty villages were assigned to each of the four treatment cells. Then, at the household level, one of three products was randomly assigned to be offered to each household: Panadol (paracetamol, a painkiller), Elyzole (albendazole,a deworming medication), and a combination pack of Restors and Zinkid (oral rehydration salts, “ORS” and zinc supplements, the World Health Organization’s recommended treatment for childhood diarrhea). For the Sale treatment group, the same price was used for the entire study.
Randomization unit
villages, households (randomization was conducted at both the village and individual level)
Clusters
120 villages
Observations
5,707 households
Arms
33% of households offered a given product in Wave 1 25% of households offered a given product in Wave 2
Primary outcomes
Demand in Wave 1 by Product: (Panadol, Elyzole, Zinkid) - Take-Up - Quantity Demand in Wave 2 by Product: (Panadol, Elyzole, Zinkid, Aquasafe) - Product same as in Wave 1?

Papers from this trial

  1. Fischer, Greg, Dean Karlan, Margaret McConnell, and Pia Raffler. “To Charge or Not to Charge: Evidence from a Health Products Experiment in Uganda. Working Paper June 2016. https://www.povertyactionlab.org/sites/default/files/publications/477_544_To%20charge%20or%20not%20Charge_Greg%26Dean%26Margaret%26Pia_June2016.pdf

Cited by

No citation data available for this registration yet.

Keywords