Moral Hazard, Networks and Risk Sharing: Evidence from a Lab Experiment in the Field

We have participants will play three risk sharing games with partners. We vary how income is generated (luck only, luck and effort) and, in the games with effort, whether effort is observable to partners.

2015-03-19

2015-05-15

Transfers promised, for each possible set of income (and, in effort observable game, efforts).
Effort, defined as whether or not the participant completes the task and the number of correct answers in the task.

Participants will play three risk sharing games with partners. Individuals will play a luck only game, a game with luck and observable completion of a task and a game with luck and unobservable completion of a task. Subjects will play all three games and game order will be randomized. Participants are randomly assigned to play each game with a partner. Partnerships are non-anonymous - meaning that participants will know their partner's name and may know their partner outside of the laboratory.

Game order is randomized (by a computer in advance of the session). Due to the fact that the number of participants per session varies and thus the number of experiment sessions cannot be known in advance, I use "block" randomization. There are 6 possible game orders of the 3 risk sharing games. I ensure that in every block of 6 sessions, each possible game order is used exactly once. I randomize order separately for each block of 6 sessions.

The design is within-subject as participants play all game. Due to concerns about game order effects, I randomize order of the 3 games resulting in 6 combinations of game orders. The game orders are randomized at the experimental session level.

No

500 individuals

500 individuals

If there are substantial game order effects, we might use only the data from the first game played, effectively changing the design into between subject design. In this case, there would be 3 arms and so effectively approximately 166 subjects by treatment arm.

Abstract: This paper examines whether social ties impact informal insurance in the presence of asymmetric information. Using games with residents of informal settlements in Kenya, I vary the observability of task completion and partners. While individuals are less likely to make transfers as a result of imperfect monitoring, socially close individuals are 30% more likely to make transfers. As a result, socially close individuals are 47% more likely to make transfers than socially distant individuals when task completion cannot be observed. I then explore the mechanisms by which social connections influence risk sharing. I find evidence that social connections increase participants’ motivation to work only when task completion cannot be observed, potentially due to social collateral and/or intrinsic motivation.

Prachi Jain, Imperfect monitoring and informal insurance: The role of social ties, Journal of Economic Behavior & Organization, Volume 180, 2020, Pages 241-256, ISSN 0167-2681, https://doi.org/10.1016/j.jebo.2020.09.028.

https://doi.org/10.1016/j.jebo.2020.09.028