Do Markets Drive Out Lemmings—or Vice Versa?

Abstract

This paper investigates experimentally a market inspired by two strands of literature: on herd behaviour in non-market situations, and on the aggregation of private information in markets. The first strand suggests that socially undesirable herd behaviour may result when information is private; the second suggests that in a market context the price mechanism may cause the private information to be aggregated correctly and efficiently. This latter therefore suggests that socially undesirable behaviour may be eliminated through the market. We test this experimentally, and find that socially undesirable behaviour may result: the market is misled by agents privately optimizing.Economica (2004) 71, 637-659

Financial markets in the laboratory: an experimental analysis of some stylized facts

“…Moreover, in a recent work, Hey and Morone (2004) showed that whenever complexity increases, noise increases as well. Indeed, their results suggest that the volatility of prices is lower (with the implication that herding is less likely) when the quality and quantity of information in the market are higher.…”

Behavioral finance: insights from experiments I: theory and financial markets

“…In an early study of herd behavior in markets, Hey and Morone (2004) examine whether the market may act as a disciplining device, removing socially undesirable herd behavior through market forces that promote the efficient aggregation of private information. They find evidence of herds developing; either due to bad information, a wrong herd developing, or associated with bubbles/crashes.…”

“…Informational cascades can be viewed as a social phenomenon, hence individual behavior may depend on how the rationality of others is viewed. To this end, Alevy et al (2007) examine a standard informational cascade game, similar to Hey and Morone (2004), to see how market professionals and students respond to uncertainty about the quality of information due to deviations from Bayesian rationality of other participants.…”

Социологические И Экономические Теории Группового Поведения И Их Применимость Для Объяснения Стадного Поведения На Финансовых Рынках

“…В 2004 году английские ученые [Hey, Morone, 2004] определенным количеством «акций» и «денег», за плату могли купить (а могли и не купить) зашумленную информацию о том, какими же эти дивиденды будут. При этом они понимали, что полученный ими сигнал верен лишь с определенной вероятностью.…”