Academic journal article IUP Journal of Applied Finance

Herding during Market Upturns and Downturns: International Evidence

Academic journal article IUP Journal of Applied Finance

Herding during Market Upturns and Downturns: International Evidence

Article excerpt

(ProQuest: ... denotes formulae omitted.)


Herding behavior is commonly defined as the behavioral tendency of an investor to follow the actions of others in the market. There are several reasons that induce herding. On the one hand, investors may follow each other when it is thought that other agents in the market are better informed (Avery and Zemsky, 1998; Calvo and Mendoza, 2000; and Chari and Kehoe, 2002). On the other hand, herding may occur when investors prefer to suppress their own belief in favor of the market consensus (Devenow and Welch, 1996). Furthermore, managers may exhibit herding behavior as a result of the incentives offered by the compensation scheme, conditions of employment or maybe to save their reputation (Scharfstein and Stein, 1990; Rajan, 1993; Trueman, 1994; and Maug and Naik, 1996).

The suppression of the personal information in favor of the market consensus may lead to a situation in which pricing process moves the market towards inefficiency. As Wermers (1999) underlines, the question of herding is crucial to understanding the way information is incorporated into prices.

There is considerable empirical evidence in social psychology that underlines the suppression of individual opinion to group opinion.1 Despite its common use in the literature, herding can be analyzed as a rational or irrational behavior. The rational view arises when investors ignore their private information and follow others so as to maintain their reputational capital in the market (Scharfstein and Stein, 1990; Bikhchandani et al., 1992; Welch, 1992; and Rajan, 1993). However, the irrational view refers to investors' psychology where they ignore their own beliefs and opinions, and blindly follow other investors (Devenow and Welch, 1996).

Lakonishok et al. (1992), Christie and Huang (1995) [hereafter CH (1995)], Chang, Cheng and Khorana (2000) [hereafter CCK (2000)], Hwang and Salmon (2004), and Hachicha et al. (2008) propose pragmatic measures to detect the herding behavior.

Academic research suggests that herding would be tightly linked to market stress (CH, 1995 and CCK, 2000 among others). However, Hwang and Salmon (2004) find the opposite. In fact, they conclude that herding behavior is more widespread during periods of market calm. Although herding behavior is by now well documented in the literature, there have been a few empirical studies on the implication of this behavior both in downturns and upturns. Particularly, some questions one would like to answer are: Is herding more prevalent in upturns or in downturns? Is there a symmetrical reaction? How does herding affect prices and volatility?

In this paper, we extend the work of CH (1995) and CCK(2000) along three dimensions. First, we study herding both in upturns and downturns. Second, we propose a new and more powerful approach to detect herding based on equity return behavior using EGARCH. Third, in order to verify the existence of this behavior, we use a methodology as in Hwang and Salmon (2004).

Literature Review

Banerjee (1992) developed a simple model for herding behavior. He indicates that herding is a behavioral bias that must be quantified to well understand the financial markets. The author demonstrated in his model that it would be better in some cases for market participants to ignore their personal information and blindly follow the behavior of others.

Following this formulation, several researchers focused on the question to quantify this phenomenon to understand its effect on assets, prices, volatility, etc. Lakonishok et al. (1992) were the first to propose an empirical methodology to detect the herding behavior. They developed a statistical measure that defines herding as the behavioral tendency when a group of fund managers buy or sell stocks at the same time, as compared to a situation where each one reacts independently. This measure is based on the transactions of some groups who are generally the portfolio managers. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.