Working Paper: NBER ID: w13126
Authors: Markus Mobius; Adam Szeidl
Abstract: This paper builds a theory of informal contract enforcement in social networks. In our model, relationships between individuals generate social collateral that can be used to control moral hazard when agents interact in a borrowing relationship. We define trust between two agents as the maximum amount that one can borrow from the other, and derive a simple reduced form expression for trust as a function of the social network. We show that trust is higher in more connected and more homogenous societies, and relate our trust measure to commonly used network statistics. Our model predicts that dense networks generate greater welfare when arrangements typically require high trust, and loose networks create more welfare otherwise. Using data on social networks and behavior in dictator games, we document evidence consistent with the quantitative predictions of the model.
Keywords: Trust; Social Networks; Social Capital; Informal Contracts
JEL Codes: D02; D23
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| higher connectivity in a social network (D85) | increased trust among individuals (Z13) |
| greater ethnic or racial homogeneity (J15) | higher levels of trust (Z13) |
| dense networks (C45) | greater welfare when arrangements require high trust (D69) |
| loose networks (D85) | advantageous for lower-value asset exchanges (G19) |
| trust is a function of network statistics (D85) | trust is determined by the weakest link in the network paths (D85) |
| maximum network flow (D85) | trust (G21) |
| negative correlation between network flow measures of trust and selfish behavior in allocation decisions (Z13) | trust (G21) |