In virtually every country, most people hold traditional religious beliefs. Across Africa, these beliefs are described as being strong and important for peoples’ choices. Despite this, we do not have empirical evidence showing how such beliefs impact peoples’ decisions, nor about their effect on economic outcomes. This paper aims to fill this gap. 


The authors focus their study on beer sellers in rural villages of North Kivu, located in the eastern Democratic Republic of the Congo (DRC). They use a smartphone-based reporting system to collect data on business activities among 81 beer sellers, which they use to document the following facts: First, beer sellers are concerned about theft from their stores, and tend to overestimate its likelihood. In an effort to prevent theft, they tend to stock less beer, and, as a result, they often experience stock-outs that limit their sales. Second, beer sellers believe in and occasionally practice religious rituals aiming at preventing theft.  These facts motivate the central question in the authors’ paper: What is the causal effect of religious rituals on beer sellers’ business performance?

To answer this question, the authors designed an experiment in which a local organization offered subsidized access to anti-theft rituals for all beer sellers who had previously demanded it. They rolled out access to the rituals randomly so that they could measure their impacts. The rituals, which are common across sub-Saharan Africa, are believed to reduce the occurrence of theft. Most participants in the study had previously taken part in a theft-protection ritual, but, given the high cost of the rituals most went unprotected at the time of the experiment. 

The authors evaluate their experiment using data on beer sellers’ economic choices, expectations, and beliefs. They use two methods in their evaluation: an event study, in which they examine outcomes before and after each seller’s ritual, and a waitlist design, in which they compare sellers’ average outcomes on treated vs. non-treated days. They find the following:

  • The rituals increased sellers’ inventory purchases, which resulted in fewer stock-outs, more revenues, and higher profits. These impacts were only observed among sellers who reported believing in the protection rituals, suggesting that rituals increase sellers’ revenues by altering their perceived probability of theft.
  • Further supporting this notion, the authors find that following the ritual, those who believe in its power – approximately 66% of the sample – were more likely to feel protected from theft and less likely to believe theft will occur. 
  • On average, sellers who participate in the rituals earn an additional $78 in profits during the following three months. Twenty-two percent of sellers (nine in total) experience an increase in profits that exceeds the total cost of the ritual. 

This research demonstrates that traditional African religious beliefs have important effects on standard economic outcomes. This result is crucial for more accurate social science research concerning populations that adhere to traditional African religions. More broadly, this research is pathbreaking within economics for expanding the field’s purview to include a more expansive portrait of the factors that shape economic behavior.