Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/35299 
Year of Publication: 
2009
Series/Report no.: 
IZA Discussion Papers No. 4084
Publisher: 
Institute for the Study of Labor (IZA), Bonn
Abstract: 
Experimental studies have consistently shown that cooperative outcomes can emerge even in finitely repeated games. Such outcomes are justified by existing reputation building models, which suggest that cooperative outcomes can be sustained if some subjects have other-regarding preferences. While the existence of other-regarding preferences is typically used to justify experimental outcomes, we are unaware of empirical studies that explicitly examine the interaction between cooperators (those with other-regarding preferences) and selfish subjects in sustaining cooperation. In this paper, we classify subjects as either selfish or cooperative using simple social preference games and then test for behavioral differences between the two types in a finitely-repeated labor market with unenforceable worker effort. Theory predicts, and our data confirms, that (1) selfish players mimic the actions of cooperators when trading partners can track the individual reputation of past partners and (2) selfish and cooperative types act differently when individual reputations cannot be tracked.
Subjects: 
Contracts
relational contracts
implicit contracts
market interaction
experimental economics
repeated transaction
social preferences
reputation
firm latitude
finitely-repeated games
JEL: 
C91
D31
D86
K12
Persistent Identifier of the first edition: 
Document Type: 
Working Paper

Files in This Item:
File
Size
171.04 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.