Behavioural Economics Seminar: Linking dishonest behavior from the lab and the field

Toke Reinholt Fosgaard, IFRO


In this project we explore if dishonest behavior measured in the laboratory is associated with dishonest behavior in real-life settings. We bought a random sample of the population to the lab and had them perform the so-called die-in-the-cup paradigm, which is giving them a possibility to cheat for additional earnings. We utilize that the data from the lab can be linked to register data provided by Statistic Denmark. The combined dataset allow us to validate if behavior in the lab relates to the real life behavior listed in the registers. We find that those with a criminal record is more likely to be cheating in the lab. Furthermore, we observe a deterrence effect: Given a crime conviction, the larger punishment given, the less likely it is that we observe dishonest behavior in the lab. The findings is providing support for the massive literature on dishonesty mainly collected in the lab.

Open to all.