A recent study, published in the December issue of the renowned American Economic Review, has now investigated this issue by means of an economic experiment and, interestingly, did not find any evidence for herding In collaboration with McKinsey&Company, Professor Jörg Oechssler (University of Heidelberg) and his co-authors designed an internet-based artificial financial market
Hardly any discussion on the exuberance of financial markets misses the opportunity to blame herd behavior by investors. Just as lemmings, financial market participants drive asset prices to extremes, or so the usual story goes. But is this indeed the case?
Perhaps surprisingly, proving that herd behavior is present in financial markets is not that easy. After all, it might simply be the case that some new favorable information regarding some stock has arrived, causing investors to buy independently from each other. On real financial markets one never really knows which information investors base their decisions on. Thus, dismissing this alternative explanation for alleged herd behavior proves to be difficult.
A recent study, published in the December issue of the renowned American Economic Review, has now investigated this issue under controlled conditions by means of an economic experiment and, interestingly, did not find any evidence for herding. In collaboration with the consultancy McKinsey&Company, Professor Jörg Oechssler (University of Heidelberg) and his co-authors Dr. Mathias Drehmann (Bank of England) and Dr. Andreas Roider (University of Bonn) designed an internet-based artificial financial market, where more than 6,000 participants could invest in various assets. Prizes amounting to more than 11,000 € provided the players with incentives to choose wisely.
Contrary to what one might have expected, market participants did not blindly follow the precedent set by earlier traders much to the contrary. When deciding how to invest, players usually followed their own information. At the same time, many participants explicitly chose to break existing trends; thereby stabilizing the market. Such behavior may very well be reasonable if one thinks that current market levels are exaggerated and are the result of irrational behavior by earlier traders. Interestingly, various subgroups of the participants differed in their willingness to trust in the rationality of others. While participants with a background in physics seemed to rely on other players behaving rationally, those with a background in psychology seemed less sure of this. And indeed, psychologists' intuition for the possibly irrational behavior of other investors earned them higher profits.
"Herding and Contrarian Behavior in Financial Markets: An Internet Experiment", American Economic Review, 95 (2005) 1403-1426
Professor Jörg Oechssler
Alfred-Weber-Institut, Department of Economics
phone: ++49-(0)6221 543548
Dr. Michael Schwarz
University of Heidelberg
Public Information Officer
phone: ++49-(0)6221 542310, fax 542317