Can Good Products Drive Out Bad?

Malaria continues to be one of the major killer-diseases in Africa, despite existence of effective prevention methods and effective cures. While the reason for this public health failure is likely to be multifaceted, recent evidence from retail markets for Artemisinin-based Combination Therapy (ACT) drugs suggest that poor quality of the medicine is one important factor.

Antimalarial medicine is an experience good, meaning that consumers observe neither the quality of the medicine sold nor the utility it will yield before purchase. Learning about quality thus has to be based on health experience after purchase and treatment. However, because malaria mimics several other diseases – both in initial symptoms and in signs of severe illness – and because symptomatic diagnosis rather than parasite-based diagnostic testing is the norm in most of Africa, misdiagnosis is common. Misdiagnosis, in turn, is likely to hamper learning about both the effectiveness of ACTs and their quality.

Fighting the selling of fake ACTs

Can the entry of a retailer selling high quality medicine affect the market equilibrium in such an environment and if so how? We investigate this question using data from a randomized controlled trial, embedded within the scale-up of a new health delivery program in Uganda. We identify two mechanisms that can improve the market equilibrium: Exit of incumbent drug shops selling poor quality and a shift to higher quality drugs by outlets remaining. We find that the reduced form treatment effects are large: Approximately a year after the new market actor entered the share of incumbent firms selling fake ACTs dropped by more than 50 percent in the intervention compared to the control group.

These results have clear policy implications. The negative productivity consequences of weak incentives for building reputation are quite clear. Thus, understanding the feasibility and cost-effectiveness of alternative interventions to improve reputation-building mechanisms in these markets are important from a policymaking perspective, and to this end more research is needed.

open close
Program area
ACT, Asymmetric Information, Counterfeit Medicine, Field Experiment, Malaria
Swedish Research Council, Program for Development Research, SIDA, J-PAL, William F. Milton Fund at Harvard Medical School, Harvard Center for Population and Development Studies
Study type
Randomized interventions (controlled trial)
open close
Martina bjoerkman-nyqvist
Stockholm University
Jakob Svensson
IIES, Stockholm University
David Yanagizawa-Drott
Zurich CED

see all affiliates

Involved partners


Similar research

Effect of a community health delivery program on under-five mortality in Uganda