Threshold public good games under uncertainty with different summation technologies

How to cite

Carlsson, Fredrik; Lange, Andreas; Ek, Claes. 2019. "Threshold public good games under uncertainty with different summation technologies." AEA RCT Registry. AEARCTR-0005175. https://doi.org/10.1257/rct.5175-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Fredrik Carlsson* , Andreas Lange , Claes Ek

* Primary investigator

Registry ID
AEARCTR-0005175
DOI
https://doi.org/10.1257/rct.5175-1.0
First registered
Last updated
Status
On going
Trial dates
2019-12-02 – 2019-12-20
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/5175

Abstract

In a laboratory experiment we investigate the role of threshold uncertainty in a threshold public good game. We build on the work by Barrett and Dannenberg (PNAS, 2012). We first attempt to replicate their finding that the introduction of threshold uncertainty reduces contributions and increases the probability of not reaching the threshold in a public good game with a standard summation technology. We then introduce another version of the game with the difference that the technology is weakest link instead of a standard summation. We investigate how uncertainty affects contributions and the probability of not reaching the threshold in a weakest link setting. We then also compare summation with weakest link technology.

Intervention

The main question is what role the summation technology plays in threshold public good games. In particular, we investigate if there are differences between a standard summation technology and a weakest link technology on contribution to the public good and on the probability of not reaching the threshold.

Experimental design

We investigate if there are differences between a standard summation technology and a weakest link technology on contribution to the public good and on the probability of not reaching the threshold.

Design details

Randomization method
By computer
Randomization unit
Individual
Clusters
No clusters
Observations
We sample 120 observations per treatment, 12 groups with 10 subjects in each treatment.
Arms
The sample is equally allocated.
Primary outcomes
Our main dependent variable is individual contributions in each game.
Secondary outcomes
In addition, we will have an interest in the secondary behavioral outcome of whether the group manages to reach the threshold or not. For this variable we will look at both the actual success, and the predicted success rate based on random sampling from the distribution of individual contributions within each treatment. For the treatments with an uncertain threshold the probability of success will depend on if the contributions are below, within, or above the uncertainty range. Thus, for the treatments with a certain threshold, the probability of success will be either 0 or 100%. Finally, we will measure individual pledges and suggestion for group contributions.

Cited by

No citation data available for this registration yet.

Keywords