Lab experiments should become a powerful tool for understanding and fighting corruption, according to research by Olivier Armantier and Amadou Boly, published in the December 2013 issue of the Economic Journal.
Their study finds that regardless of the environment – whether it is in a developed country or a developing country; in the lab or in the field – experiments on corruption can produce similar outcomes, and corrupt behaviour can be linked to the same individual characteristics, such as age or religious commitment.
Corruption is now recognised as one of the most detrimental factors to economic and social development. Yet because it is illegal and hidden, analysis of corruption has been a challenge for applied economists. To circumvent this obstacle, lab experiments on corruption are increasingly being conducted, as they offer a set-up in which a researcher can unambiguously observe corrupt behaviour at the individual level.
At the same time, lab experiments on corruption raise the question of external validity − the extent to which it is possible to generalise from lab results to other relevant settings, particularly to the field. Lab experiments sometimes fail to incorporate some relevant features from the field – for example, appropriate stakes, environment and population.
The new study analyses the effects of two factors on the external validity of lab experiments on corruption:
· First, scrutiny: a typical difference between lab and field experiments is that on entering the lab, subjects know that their behaviour will be monitored and analysed by an experimenter. Because of this scrutiny, lab subjects may be more inclined to make the ''moral'' choice when morality and wealth are competing objectives, as they are with corruption.
· Second, location: lab experiments are typically conducted in developed countries (where most labs are located), while understanding and fighting corruption are generally considered a priority for developing countries.
To investigate the external validity of corruption experiments relative to scrutiny and location, the researchers conducted the same experiment in three different environments: a lab in a developed country, Canada; a lab in a developing country, Burkina Faso; and the field in the same developing country.
In the experiment, a candidate proposes a bribe to an exam grader in order to obtain a better result: the paper comes with a money offer and a message saying: ''Please, find few mistakes in my examination paper''. To determine whether subjects behave differently when they know they are being observed (the scrutiny hypothesis), subjects in the field were informed they had participated in an experiment only after grading was completed.
The researchers conducted four different treatments in each of the three environments by varying factors susceptible to promote or deter corruption: the amount of the bribe; the wage paid to the exam graders; and the level of monitoring and punishment. They then tested whether the direction and the magnitude of the observed treatment effects differed across the three environments.
The study finds that the direction and magnitude of several treatment effects are statistically indistinguishable across the three environments. In addition, regardless of the environment, corrupt behaviour can be linked to the same individual characteristics such as age, religious commitment or exam grading ability.
The authors conclude:
''Our results provide evidence first, that lab experiments on corruption conducted in a developed country and a developing country can produce similar outcomes; and second, that lab experiments on corruption can have empirical relevance in the field.
''If these results are confirmed in future research, then lab experiments should become a powerful tool to understand and fight corruption.''
''Comparing Corruption in the Lab and in the Field in Burkina Faso and in Canada'' by Olivier Armantier and Amadou Boly is published in the December 2013 issue of the Economic Journal. Olivier Armantier is at the Federal Reserve Bank of New York. Amadou Boly is at the United Nations Industrial Development Organization.
