Informational Contagion in the Laboratory*

Abstract

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. Terms of use: Documents in AbstractWe study the informational channel of financial contagion in the laboratory. In our experiment, two markets with correlated fundamentals open sequentially. In both markets, subjects receive private information. Subjects in the market opening second also observe the history of trades and prices in the first market. We find that although in both markets private information is only imperfectly aggregated, subjects are able to make correct inferences based on the public information coming from the market that opens first. As a result, we observe financial contagion in the laboratory: Indeed, the correlation between asset prices is very close to that predicted by the theory. Finally, as theory predicts, there is no contagion when asset fundamentals are independent: That is, subjects only react to the history of prices and trades in the first market when it is rational to do so because they convey information.

Informational Channels of Financial Contagion

“…() performed a meta‐study of various experiments on social learning, showing that subjects put too much weight on their private signal in situations where it would be rational to herd. In most social learning studies (and similar to Brown, Trautmann, and Vlahu (), Chakravarty, Fonseca, and Kaplan (), and Cipriani et al ()) subjects perfectly observe the decisions of their predecessors, unlike in the present study. Celen and Kariv () relaxed this assumption by adding some noise to the social signal and find a positive relationship between the level of uncertainty in the environment and the emergence of behavioral anomalies, which is consistent with our findings.…”

“…Some experimental papers study financial contagion, but none with the use of global games. Brown, Trautmann, and Vlahu (), Chakravarty, Fonseca, and Kaplan (), and Cipriani, Guarino, Guazzarotti, Tagliati, and Fischer () studied contagion with both fundamental links and social learning . In all of these papers, fundamentals are either perfectly correlated or not correlated at all and the behavior of agents in the first market is perfectly observed and does not constitute a treatment variable.…”

“…In all of these papers, fundamentals are either perfectly correlated or not correlated at all and the behavior of agents in the first market is perfectly observed and does not constitute a treatment variable. Cipriani et al () found that contagion occurs in the second market only when fundamentals are perfectly correlated. Their results are consistent with the case in this paper (the only comparable treatment) where we do not observe the emergence of biases.…”

Bank Instability: Interbank Linkages and the Role of Disclosure

“…In contrast to Chakravarty et al (2014), they only find evidence of contagion in the presence of economic linkages between financial institutions. The results of Brown et al (2017) are supported by Cipriani et al (2018) who consider the informational channel of financial contagion. They find evidence of contagion between two markets, but only as long as asset fundamentals are correlated.…”

Financial Contagion and the Wealth Effect: An Experimental Study

“…Contagion due to portfolio re-balancing occurs because subjects' payoffs depend not only on the return to their investment, but also on the composition of their portfolios. Cipriani et al (2017) investigate experimental asset markets where financial contagion occurs when asset returns in two markets are correlated, but no contagion effect is present in markets with independent fundamentals. These models, however, describe a different nature of "crises", in which contagion is reflected in asset prices.…”