TY - JOUR
T1 - Cooperation in Community Interaction Without Information Flows
AU - Ghosh, Parikshit
AU - Ray, Debraj
N1 - Funding Information:
which is necessary and sufficient for as to beequal to a*(Jr) in the social equilibrium. Since, by Proposition 5, VS goes up with an increase in 8, it follows that the above inequality continues to hold at higher values of 8. This completes the proof. 1\ Acknowledgements. We wish to thank Tilman Borgers, Yeon-Koo Che, Douglas Gale, Karla Hoff, Deborah Minehart, Robert Rosenthal, Kunal Sengupta, Chun-Lei Yang and two anonymous referees for helpful comments on earlier drafts of the paper. Financial support under National Science Foundation Grant SBR-9414114 and Grant no. PB90-0172 from the Ministerio de Educacion y Ciencia, Government of Spain, is gratefully acknowledged.
Copyright:
Copyright 2018 Elsevier B.V., All rights reserved.
PY - 1996
Y1 - 1996
N2 - We study cooperative behaviour in communities where the flow of information regarding past conduct is limited or missing. Players are initially randomly matched with no knowledge of each other's past actions; they endogenously decide whether or not to continue the repeated relationship. There is incomplete information regarding player types: a subset of the population is myopic, while the remainder have discount factors that permit cooperation, in principle. We define social equilibrium in such communities. Such equilibria are characterized by an initial testing phase, followed by cooperation if the test is successful. It is precisely the presence of myopic types that permit cooperation, by raising barriers to entry into new relationships. We examine the implications of increased patience, which takes two forms: an increase in the number of non-myopic types, and an increase in the discount factor of non-myopic types. These two notions turn out to have strikingly different implications for the degree of cooperation that can be sustained.
AB - We study cooperative behaviour in communities where the flow of information regarding past conduct is limited or missing. Players are initially randomly matched with no knowledge of each other's past actions; they endogenously decide whether or not to continue the repeated relationship. There is incomplete information regarding player types: a subset of the population is myopic, while the remainder have discount factors that permit cooperation, in principle. We define social equilibrium in such communities. Such equilibria are characterized by an initial testing phase, followed by cooperation if the test is successful. It is precisely the presence of myopic types that permit cooperation, by raising barriers to entry into new relationships. We examine the implications of increased patience, which takes two forms: an increase in the number of non-myopic types, and an increase in the discount factor of non-myopic types. These two notions turn out to have strikingly different implications for the degree of cooperation that can be sustained.
UR - http://www.scopus.com/inward/record.url?scp=0001203076&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0001203076&partnerID=8YFLogxK
U2 - 10.2307/2297892
DO - 10.2307/2297892
M3 - Article
AN - SCOPUS:0001203076
SN - 0034-6527
VL - 63
SP - 491
EP - 519
JO - Review of Economic Studies
JF - Review of Economic Studies
IS - 3
ER -