Completed projects

Can Tax Classes Build Compliance Culture? Evidence from Randomized Survey Experiments in Cameroon. Submitted.

With Marc Ateba, Miguel Fonseca, Jannesquin Royer.

This paper has been presented at: The Advances with Field Experiments Conference (Chicago), the CIBES Conference (Florence), the IFS/Tax Dev Internal Seminar series (London), and the IIPF Conference (Nairobi) and the Exeter Internal Seminar Series (Exeter).

Revise & Resubmit to International Tax and Public Finance Journal.

We explore how instructing future taxpayers on basic tax information helps build a tax-paying culture under low state capacity. We embedded a randomized survey experiment in a large tax awareness campaign towards young adults in Cameroon. We randomly assigned 1,962 public and private secondary school students from 42 classes to tax information classes. We provide causal evidence of significant effects

on basic tax knowledge and compliance attitudes with differential treatment effects across gender, risk attitudes and family background. Also providing social norms of compliance improves knowledge retention 13 months post-intervention. Our results indicate strong effects of the informational treatments for cohorts of students at both tails of the knowledge and tax morale distributions. These findings highlight a high potential for scalability and offer a new perspective on the political economy of tax educational reforms aimed at improved tax compliance norms among young and potentially entrepreneurial populations.

Working paper available here

Do Salient Negative Externalities Deter Bribing? A Lab-in-the-Field Experimental Evidence from Cameroon.

With Miguel Fonseca.

This paper has been presented at: The Future of Behavioural Ethics Conference (Berlin), Economic Science Association Conference (Nairobi), Exeter PhD Conference. 

Submitted to Experimental Economics Journal.

Small-scale corruption such as bribing has deleterious consequences for millions of people’s daily lives, especially in developing countries, as it undermines institutions and people’s trust in them. In this paper we analyze a lab-in-the-field experiment that provides insight into the effectiveness of informing participants about social costs of bribing in an artefactual three-person queue in which places are either assigned randomly or on a needs basis. In the absence of a message, bribing rates are unaffected by queue assignment. However, providing information about the externality caused by bribing greatly diminishes the incidence of bribing. We also find evidence of an interaction effect: messages are more effective when queue places are allocated on a needs basis. These findings support the argument that, in some contexts, preference for procedural fairness is less prevalent, yet salient awareness campaigns on the social harm caused by antisocial behavior

can be effective.

Keywords: Salience, Negative Externality, Corruption, Experiments 

(JEL: D62, D73, C91, D83, O17)