Bank runs as coordination failures: An experimental study

Garratt, Rod; Keister, Todd

doi:10.1016/j.jebo.2009.03.009

Cited by 101 publications

(92 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As a consequence, no depositor knows if the other two depositors are connected or not. 6 A key element of the model is that when the depositors decide, they may not be sure of the payo¤ they will receive. For instance, if a patient depositor at the …rst position waits, then her payo¤ depends on what the other patient depositor does (i.e., c 1 0 2 fc 00 ; c 01 g).…”

Section: Network Information and Bank Runsmentioning

confidence: 99%

“…Her decision may follow either two withdrawals (Y 3 = f1; 1g) or a waiting and a withdrawal (Y 3 = f1; 0g). In 6 Given the nature of bank runs it seems reasonable to assume that depositors do not know exactly what other depositors know. However, the theoretical result also holds when network structure is common knowledge.…”

Section: De…nition 1 a Bank Run Occurs If At Least One Patient Deposimentioning

confidence: 99%

See 1 more Smart Citation

Do Social Networks Prevent Bank Runs?

2012

View full text Add to dashboard Cite

We develop, both theoretically and experimentally, a stereotypical environment that allows for coordination breakdown, leading to a bank run. Three depositors are located at the nodes of a network and have to decide whether to keep their funds deposited or to withdraw. One of the depositors has immediate liquidity needs, whereas the other two depositors do not. Depositors act sequentially and observe others' actions only if connected by the network. Theoretically, a link connecting the …rst two depositors to decide is su¢ cient to avoid a bank run. However, our experimental evidence shows that subjects'choice is not a¤ected by the existence of the link per se. Instead, being observed and the particular action that is observed determine subjects'choice. Our results highlight the importance of initial decisions in the emergence of a bank run. In particular, Bayesian analysis reveals that subjects clearly depart from predicted behavior when observing a withdrawal.JEL Classi…cation: C70, C90; D85; G21

show abstract

Section: Network Information and Bank Runsmentioning

confidence: 99%

Section: De…nition 1 a Bank Run Occurs If At Least One Patient Deposimentioning

confidence: 99%

Do Social Networks Prevent Bank Runs?

2012

View full text Add to dashboard Cite

show abstract

“…7 Madies (2006) shows that sunspot bank runs can occur and that a suspension of convertibility or a full (as opposed to partial) deposit insurance may be required to prevent bank runs. Garratt and Keister (2009) show that when liquidity demand is not subject to stochastic shocks, panic-driven bank runs are unlikely to occur. With stochastic liquidity shocks, however, self-fulfilling bank runs are frequent.…”

Section: Introductionmentioning

confidence: 97%

Understanding Bank-Run Contagion

2017

View full text Add to dashboard Cite

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. The refereeing process of this paper has been coordinated by a team composed of Gerhard Rünstler, Kalin Nikolov and Bernd Schwaab (all ECB). Terms of use: Documents inThe paper is released in order to make the research of MaRs generally available, in preliminary form, to encourage comments and suggestions prior to fi nal publication. The views expressed in the paper are the ones of the author(s) and do not necessarily refl ect those of the ECB or of the ESCB. Acknowledgements Abstract:We study experimental coordination games to examine through which transmission channels, and under which information conditions, a panic-based depositor-run at one bank may trigger a panic-based depositor-run at another bank. We find that withdrawals at one bank trigger withdrawals at another bank by increasing players' beliefs that other depositors in their own bank will withdraw, making them more likely to withdraw as well. Importantly though, observed withdrawals affect depositors' beliefs, and are thus contagious, only when depositors know that there are economic linkages between their bank and the observed bank.JEL classification: D81; G21; G28

show abstract

“…Notably, in all of these cases the banks that suffered the run were fundamentally healthy, bad news about another bank sparked the run. Experimental evidence also suggests that observability plays an important role in the emergence of bank runs (see, for example, Garratt and Keister, 2009;Schotter and Yorulmazer, 2009;Kiss et al, 2012). Moreover, Kiss et al (2014) study a small-scale environment resembling the Diamond-Dybvig setup in which bank runs are caused by coordination problems.…”

Section: Introductionmentioning

confidence: 99%

“…More information about previous decisions seems to reduce the likelihood of bank runs. In Garratt and Keister (2009) and Kiss et al (2014) there were no fundamental problems with the bank and it was common knowledge, so there fundamental problems or negative information about the bank extracted from the behavior of other depositors cannot be behind the runs.…”

Section: Introductionmentioning

confidence: 99%