In this paper, we develop a model in the spirit of Diamond and Dybvig (J Polit Econ 91(3): [401][402][403][404][405][406][407][408][409][410][411][412][413][414][415][416][417][418][419] 1983), with bank-specific sequential-service constraints and commitment to a redistribution policy at the terminal date. In the good equilibrium, some insurance is provided against the risk of being affiliated with a bank with too many impatient depositors. We also find, however, that an unpleasant arithmetic takes place in a bad equilibrium, with patient depositors realizing they can be heavily taxed since runs are reducing savings in other banks, which in turn changes the ability of their own bank to induce truth-telling. We prove that even the disclosure device proposed by Green and Lin (J Econ Theory 109(1):1-23, 2003) may not remove bank-run equilibria.
Keywords Sequential service • Deposit insurance • Green-Lin modelThis paper was presented at the SAET 2016 meeting and the conference Liquidity, Market Frictions and the Economy, at the Bank of France, in Paris (June 2016), under the title 'Unifying bank-run theories and (the trap of) financial integration'. We thank seminar participants for helpful comments, specially Régis Breton and Neil Wallace (who suggested the new title). The usual disclaim applies. Cavalcanti and Monteiro acknowledge financial support from CNPq-Brazil.