Abstract
We introduce stochastic loss into a gift exchange game to study how information on intentions affects reciprocity. In one treatment, the respondent observes the amount received and whether a loss occurred, so both the consequential outcome and the sender’s original intention are known. In the other two treatments, information about whether a loss occurred is hidden, and the respondent is only informed of the amount received (outcome) or the amount initially sent (intention). Using both regression-based approaches and non-parametric tests, we find greater reciprocity in the two treatments that reveal intentions. These differences arise even in a simple one-shot setting without reputational benefits and are economically meaningful; they are similar in magnitude to the difference attributable to a full point reduction in the amount received. Our findings show the impact of the information environment on reciprocity in settings with uncertainty and suggest that transparency is important to reciprocity.
| Original language | English |
|---|---|
| Pages (from-to) | 51-65 |
| Number of pages | 15 |
| Journal | JESA - Journal of the Economic Science Association |
| Volume | 9 |
| Issue number | 1 |
| DOIs | |
| Publication status | Published - Jun 2023 |
Bibliographical note
Published online: 17 January 2025Fingerprint
Dive into the research topics of 'Reciprocity with stochastic loss'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver