Herding

What is herding behaviour?

A behavioural similarity stemming from the interactive observations of individuals.

特征：Individuals who suppress their own beliefs and base their investment decisions solely on the collective actions of the market, even when they disagree with its prediction

Form of collective investor/crowd behavior? 表现形式

Recall previous examples of bubbles and crashes

South Sea share price bubble 1720s; Tulipmania in the 17th century; Stock market in 1920s and collapse in 1929; Internet bubble in late-1990s and subsequent post-2000 crash.

Causes of Herding

1.Behavioural biases (部分有些牵强)

Conformity: feeling comfort going with the crowd (Hirshleifer, 2001)

Congruity: seeking out information that matches your beliefs (Prast, 2001)

Conservatism bias: when you are slow to update your priors because you are under the impression that things follow a certain direction (Barberis, Shleifer & Vishny, 1998)

Home-bias: the proximity of individuals to the investment object can lead to behavioural similarities

Representativeness heuristic: when a series of recent and limited evidence prompts you to see a pattern that may or may not be true (Barberis, Shleifer & Vishny, 1998)

Rumour-heuristic: ambiguity leads people to resort to word-of-mouth in order to rationalize uncertainty

2. Informational reasons

Imitate decisions of others because

1.Possess no private information(未拥有私人信息)

2.Possess private information yet uncertain about its value(虽然有私人信息，但是不确定其价值)

3.Believe information-processing abilities to be inadequate(相信自己没能力处理这些信息)

4.Perceive others as better-informed(觉得别人信息渠道更充分)

3.Career/reputational reasons

“Bad manager follows good manager”. 

If times are good, all managers look good and everyone is happy. If times are not good, any failure can be ascribed to the misfortune of the times. 

Evidence of herding 

Christie and Huang (1995)

Examined herding behaviour in US market and use cross-sectional dispersion of stock returns（股票收益的横截面分散性）to estimate herding

If herding present return-dispersion should decline （分散度会降低）during “extreme” market periods

Found no evidence of herding behavior in the US market

Nofsinger and Sias (1999)

Measure herding by the relationship between change of institutional ownership and excess return（机构所有权变更和超额收益之间的关系）

Find evidence of herding behaviour in the US market

Chang, Cheng and Khorana (2000)

Significant herding only for the emerging markets (Korea, Taiwan); no herding for US, UK, Japan, Hong Kong. 

Caparrelli, D’Arcangelis and Cassuto (2004)

Investigate herding behavior in the Italian stock market

Find evidence herding exists in extreme market conditions

Impact on Financial Markets

Herding may lead to a poor information aggregation

Information cascade(a process of information is successively passed on 信息的连续性) – price forecasts are made sequentially, so each prior forecast becomes part of the next forecasters information set. Herding inhibits information cascade and so limits revelation of private information. Consequence is informational inefficiency. (herding阻碍了信息的连续性)

Herding may lead to over-reaction

Financial markets exhibit a herding behaviour in times of crises(在金融危机时刻，herding 行为就会显现出来), which amplifies the effects of an initial shock to the system. Consequence is over-reaction and informational inefficiency