Dishonesty behavior: a natural field experiment with Tanzanian farmers

Start date

The act of dishonesty affects societies both socially and economically. Dishonesty behavior by employees may imply fraud, lower productivity and financial losses to the firm, dishonesty norms are likely to play a key role in the occurrence of corruption, and dishonest parents are more likely to raise dishonest children suggesting ‘exogenous’ intergenerational transfer of the behavior and its long run implications.

A vast majority of previous studies relies on lab experiments and have been carried out in developed countries. Here, we are adding to the few studies with field evidence and, as far as we know, for the first time from a developing country.

While lab experiments offer a clean research environment, a limitation is that granting anonymity for individual behavior implies data collection only at group level.Further, in a lab setting it cannot be ruled out that subjects know that they are being studied, which may affect behavior. Natural field experiments may address these two weaknesses.

In this study the subjects are Tanzanian farmers that were interviewed on farming activities and socio-economic conditions and participated in risk and time preference experiments during September 2013. The responding farmers in the study received their payments via mobile phone. The extensive questionnaire also included a question how they would act if they received mobile transfer money by mistake, where 44% claimed that they would return the whole amount, 21% some, 27% would return none, and 8% were uncertain. Presumably the subjects are not aware of the link back to the interviews more than a year ago, and will react to the experiment without regards to those. Our first research question is: What is the level of dishonesty in an anonymous natural field experiment and are norms activated that reduces the level of dishonesty from any of the two treatments? Second, we address the external validity of the farmers’ stated behavior a year ago and whether socio-economic variables explain their dishonesty behavior.

Theme 1: Individual behaviour, cooperation and trust.

Project status
Project | 20 March 2015