Abstract
This thesis theoretically, but mostly experimentally, investigates corruptibility and ways to fight corruption. In the first chapter, I investigate if an individual is more likely to behave corruptly if she believes that the majority of his/her peers are behaving in a similar manner.
I also study what motivates an individual to follow the behavior of others more: knowing that the majority behaves corruptly or the willingness to fulfill the expectations of one's peers. I use a one-shot reverse public goods experiment to provide answers to these questions.
The results suggest the existence of a peer effect of corrupt behavior. Expectations of the number of corrupt peers are found to be the main source of subjects' anti-social behavior.
In the second chapter, I question the rationale behind previous research results on the relationship between officials' wages and the level of corruption. In the light of counterintuitive empirical findings by Schulze et al. (2016) and Chen and Liu (2018), I investigate the wages-corruption relationship by means of a laboratory experiment, using a robust version of a bribery game with four different relative wages for public officials.
I also introduce a new method of assigning different wages to participants, which helps to reduce self-selection based on ability, but creates a sense of entitlement to the endowment. The main findings are that (i) the effect of wages on officials' propensity to accept bribes is negative, and (ii) the effect of wages on officials' propensity to act on bribes (i.e., reciprocate a bribe by providing a corrupt service) is U-shaped.
In the third chapter, I propose an explanation of the positive, and U-shaped, wage-corruption relationship found in previous research: When an official's salary rises beyond a certain threshold, their self-perceived socio-economic status (SSES) increases and they become less averse to corrupt behavior. I investigate this theory using a laboratory experiment in which public officials can accept a bribe and then decide whether to act on it.
The act benefits the briber but imposes externalities on social welfare. I manipulate officials' SSES experimentally to test if there is some such effect and when it kicks in.
The results reported in this paper suggest that the effect of SSES on the propensity of officials with the highest wage to accept and reciprocate bribes is positive and significant if a bribe is high enough.