Date Submitted: 15 Sep 2017
This In practice piece gives a practitioner’s perspective on the article “Institutional Investors and Equity Prices: Information, Behavioral Bias and Arbitrage” by Bing Han and Dongmin Kong, working paper. Available at SSRN: https://ssrn.com/abstract=2926401
What’s the Investment Issue?
Since the 1980s, Chinese government has started to restructure state-owned enterprises. Upon the initial public offering, the parent firm receives legal person (LP) shares in return for the assets it inserted. LP shares make up about one-third of the total capital stock of the listing firm. Another one-third are state shares held by local governments and central government. These two third of shares are called non-tradable shares since they are restricted from selling to individual investors and trading on stock exchanges. The remaining one-third is publicly issued and traded by individuals and institutions. Owing to the split–share structure, various levels of government entities controlled the listed companies and had power over the appointment of key managers, making the managers of the listed companies responsible not for the public investors but for the bureaucratic power, putting public investors in an inferior position and leading to greater price inefficiency and volatility.
Owing to the drawback of the split-share structure, Chinese government was committed to a reform. The plan was to align the interest of various types of shareholders including non-tradable shareholders and tradable shareholders. The main reform structure was carried out in 2005. It invites non-tradable and tradable shareholders to negotiate the conditions under which non-tradable shares (NTS) can be converted into tradable shares (TS). The reform process would cause more shares (the past NTS) to float in market and make individuals’ shares be diluted. To compensate individual investors’ loss, corporation founders provide certain amount of compensation to public investors in return for the right to trade their NTS in the secondary market. After the negotiation two parties agree to a compensation ratio and then NTS would be allowed to trade publicly. Taking advantage of this unique natural experiment, the authors try to answer these four questions:
Do institutions have private information and trade on it?
Does institutions’ trading move stock price?
Do institutions act as arbitrageurs and exploit stock mispricing?
Do institutions make abnormal profit from trading?
How Does the Author Tackle This Issue?
Sample period from January 1, 2005 to December 31, 2008 was chosen to encompass the split–share reform (from mid-2005 to 2007). Final sample includes 1,215 firms listed in Chinese stock markets (SSE) and Shenzhen Stock Exchange (SZSE). Several key dates are presents as following: T0
: The company first announces the reform. T1
: Shares of the reform companies resume trading. T2
: Shareholder registration date. T3
: Trading resumes after the reform plan is approved. Then the authors study investors in four types: individual investors, active domestic institutions, passive domestic institutions and qualified foreign institutional investors (QFIIs). They analyze the trading behavior of these investor types in different windows.
What Are the Findings?
The empirical evidence shows that passive institutions possess and then trade based on private information about company-specific compensation ratio. The timing of the reform and company-specific compensation ratio are vital during the process of the reform and are private since these information are not open to public. Compensation ratio determines the wealth transfer from NTS to TS holders.
Their evidence shows that passive institutions buy an abnormal amounts of tradable shares that end up with high compensation ratio prior to the announcement date and sell abnormal amounts of tradable shares that end up with low compensation ratio. However, passive institutions do not make abnormal profits because they are subject to disposition effect: they tend to sell their winners too soon after trading resumes (T1
) – the company announces the split–share structure reform, leaving large future gains ungarnered. The average cumulative abnormal return (CAR) of passive institutions from the ten trading days prior to T0
is 2.22%, while the CAR on T1
is −2.53%. Thus, the informed passive institutions do not make abnormal profit through their information advantage. Individual investors also exhibit disposition effect; they sell an abnormally large amount of stocks with high unrealized gains after the reform is announced (T1
). The disposition effect exhibited by passive institutions and individual investors leads to underpricing in stocks with high compensation ratios.
However, QFIIs and active institutions who have no information advantage turn out to be significantly buyers of the most underpriced stocks. The stocks intensely bought by active institutions and QFIIs outperform the stocks intensely sold by them by approximately 3.5% and 0.05% over the period [T0 − 10
]. The CAR of the reform companies from T1
is positive and significant (5.33%), although CAR on T1
is significantly negative (−2.53%). This result means that if the price on T1
is affected by the disposition effect of individual investors and passive institutions, some types of investors can take advantage of the mispricing by buying on T1
. Those results show that qualified foreign institutions and active institutions make abnormal returns by arbitraging this mispricing and helping improve market efficiency.
What Are the Implications for Investors and Investment Professionals?
This paper provides new insights into the role of institutional investors in asset pricing through
a unique split–share structure natural experiment and the rich dataset with all daily trades of various types of investors. Some institutional investors make abnormal returns, despite of information disadvantage, by arbitraging mispricing created by investors who have information advantage. Thus, investors who have private information may not able to make abnormal return and even suffer a loss if they do not trade in the right way, such as subject to disposition effect. Thus, both institutional and individual investors may be subject to cognitive errors, such as disposition effect, which leads to asset price predictability.
*Danling Jiang is the Associate Professor of Finance at SUNY at Stony Brook and the Chang Jiang Scholar Visiting Professor at Southwest Jiaotong University. Jingyu Cui is a Master of Science in Finance student at SUNY at Stony Brook.