Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Social Learning with Endogenous Information and the Countervailing Effects of Homophily

Yunus C. Aybas, Matthew O. Jackson

TL;DR

This paper develops a dynamic, endogenously informative model of social learning with endogenous information generation under homophily. Using a two-group, graphon-based network and overlapping generations, it shows that homophily can improve information quality but reduce endogenous information in sparse networks, while dense networks can flip the tradeoff so that homophily enhances learning. Key mechanisms include direct and indirect inferences about the risky action’s value $v$, observed payoffs, and the distribution of costs $c$ across groups, with steady states $(g(v),b(v))$ describing long-run behavior. The results yield practical policy insights: initiate cross-group ties to accelerate learning when initial disparities are large, then allow same-type mentorship as learning solidifies, and emphasize cost-differentiated information to promote efficient diffusion.

Abstract

People learn about opportunities and actions by observing the experiences of their friends. We model how homophily -- the tendency to associate with similar others -- affects both the endogenous quality and diversity of the information accessible to decision makers. Homophily provides higher-quality information, since observing the payoffs of another person is more informative the more similar that person is to the decision maker. However, homophily can lead people to take actions that generate less information. We show how network connectivity influences the tradeoff between the endogenous quantity and quality of information. Although homophily hampers learning in sparse networks, it enhances learning in sufficiently dense networks.

Social Learning with Endogenous Information and the Countervailing Effects of Homophily

TL;DR

This paper develops a dynamic, endogenously informative model of social learning with endogenous information generation under homophily. Using a two-group, graphon-based network and overlapping generations, it shows that homophily can improve information quality but reduce endogenous information in sparse networks, while dense networks can flip the tradeoff so that homophily enhances learning. Key mechanisms include direct and indirect inferences about the risky action’s value , observed payoffs, and the distribution of costs across groups, with steady states describing long-run behavior. The results yield practical policy insights: initiate cross-group ties to accelerate learning when initial disparities are large, then allow same-type mentorship as learning solidifies, and emphasize cost-differentiated information to promote efficient diffusion.

Abstract

People learn about opportunities and actions by observing the experiences of their friends. We model how homophily -- the tendency to associate with similar others -- affects both the endogenous quality and diversity of the information accessible to decision makers. Homophily provides higher-quality information, since observing the payoffs of another person is more informative the more similar that person is to the decision maker. However, homophily can lead people to take actions that generate less information. We show how network connectivity influences the tradeoff between the endogenous quantity and quality of information. Although homophily hampers learning in sparse networks, it enhances learning in sufficiently dense networks.
Paper Structure (12 sections, 31 equations, 2 figures)

This paper contains 12 sections, 31 equations, 2 figures.

Table of Contents

  1. Introduction
  2. The Model
  3. Countervailing Effects of Homophily
  4. Full Homophily
  5. Partial Homophily
  6. More Levels of Costs and Values
  7. Concluding Remarks
  8. Proofs
  9. Multiple Costs and Values, Assortativity in Costs, and Incidental Homophily
  10. Perfect cost homophily and complete learning
  11. Incidental homophily from cost assortativity
  12. Detailed Dynamics

Figures (2)

  • Figure 1: Steady state level of green group agents taking risky action when $\pi_g=0.6, \pi_b=0.3$. We show the plot for real values of $d_g$ but the actual values of $d_g$ are discrete.
  • Figure 2: Blue/green homophily as a function of cost when there perfect cost homophily with no attention to green/blue and $F_g$ likelihood ratio dominates $F_b$. The figure is for equal-sized groups.