The role of institutional investors in post-earnings announcement drift: evidence from China

Abstract

We examine how institutional investors influence post-earnings announcement drift (PEAD) in China. Our findings suggest that institutional holdings are positively correlated with PEAD in China, especially when institutional investors herd strongly on earnings news. This positive relationship is more salient for institutional investors with shorter investment horizons and in firms with higher information opacity. We also find that stock prices reverse in the fourth quarter after the earnings announcement. In contrast to the well-established view that institutional investors exploit PEAD and accelerate the speed of information incorporation, our findings suggest that they may instead exacerbate PEAD and slow down price discovery in emerging markets with different institutional backgrounds.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Acknowledgment

We thank the editors and reviewers for their many insightful suggestions.

Notes

1 Agrawal and Mandelker (1990, 1992), Wermers (1999, 2000), Chen et al. (2000), Chung et al. (2002), Hartzell and Starks (2003) and Aggarwal et al. (2011) provide examples.

2 For example, Ball and Brown (1968) and Bernard and Thomas (1989, 1990) find PEAD in the US; Booth et al. (1996) identify PEAD in the Finnish market; Mande and Kwak (1996) show significant PEAD in Japan; Liu et al. (2003) document PEAD in the London Stock Exchange; and Truong (2010) reports PEAD in the New Zealand market.

3 Anecdotal evidence also supports the existence of institutional investors’ herding activities. For example, Leshi Internet and Information Technology Corporation’s stock prices dramatically increased in 2013 (with an annual return of 318%) after institutional investors herded into the stock and then plummeted after 2016. The news media often consider Chinese institutional investors to be short-term speculators who lack their own judgement and are bogged down by short-term events; under significant performance pressure, fund managers choose to herd in hot market sectors, resulting in overvaluations. Reference can be made to http://fund.jrj.com.cn/2018/03/01162724175217.shtml (in Chinese). Google Translate can be used to view the contents of the page.

4 We do not differentiate who the crowd is. We consider it probable that institutional investors trade to follow other institutional investors, or even retail investors, under the significant noise trader risk in China (De Long et al. 1990).

5 Although we do not identify the exact institutional differences in China that result in significant herding, we suggest several possibilities in this section.

6 http:/www.reuters.com/article/us-china-stocks/china-stock-market-freezing-up-as-sell-off-gathers-pace-idUSKCN0PI04Q20150708.

7 A large-scale national survey (China Household Finance Survey 2012: http://www.chfsdata.org/) documents that two thirds of new accounts are opened by individuals who have not graduated from high school.

8 The small CAR is due to aggregating the positive and negative CAR for our sample firms.

9 Our IO percentage is similar to that of Firth et al. (2016). Some studies (e.g., Cheng et al. 2016, Chen et al. 2017b) report higher institutional ownership because their definition of IO includes shares held by non-financial firms in addition to ours. Our results hold when we define IO following their approach.

10 Our result indicating that institutional investors increase PEAD is mainly concentrated in the group of firms with positive earnings surprise. This is not surprising, as herding is much more difficult when earnings surprise is negative, especially with short-sale constraints. In undocumented results, we show that in the firms in the bottom SUE quintile, institutional ownership actually reduces PEAD when short-sale constraints are in place.

11 Our results hold when we use analyst forecast accuracy instead.

12 We thank an anonymous reviewer for concisely pointing out this argument.

Additional information

Funding

We thank the National Natural Science Foundation of China (Project No. 71772181, 71790603, 71872192).