Renaud Bourles (GREQUAM, CNRS - École Centrale de Marseille)
En collaboration avec Yann Bramoullé et Eduardo Perez-Richet
Abstract: We analyze in this paper the risk sharing implications of altruistically driven transfers. Agents are embedded in a fixed network and care about the well-being of their network neighbors. When incomes are stochastic and transfers conditional on incomes, we find that altruism networks have a strong impact on risk. We show that informal insurance tends to be better when the average path length is lower and that small shocks are efficiently insured if the graph of transfers is connected. We moreover find that large shocks tend to be well-insured, in contrast to what happens in models with social collateral. We finally uncover rich structural effects: a new link may decrease or increase the risk faced by others.