Is risk rewarded on the Chinese A-share market? MSCI recently announced it will include 222 Chinese A-shares in its Emerging Markets Index in 2018. While this is just a small fraction of the stocks listed on China’s two domestic markets, in Shanghai and Shenzhen, this decision is an important milestone for the local stock market.
However, many still worry about the risks often associated with A-shares, especially the risk of trading suspensions. One thing is clear, Chinese local stocks tend to be rather volatile. We therefore decided to test if the low volatility effect can be found in the Chinese market. Or, to use a well-known parable, if the tortoise beats the hare in China as well.
The evidence for the existence of the low volatility effect across numerous markets is overwhelming: low-risk stocks give high risk-adjusted returns. A large number of academic studies analyze and confirm the existence of the low volatility effect. This effect has been extensively documented for US, European, Japanese and 19 emerging stock markets 1&2. For early international evidence see also the paper by Blitz and van Vliet3, (2007).
The current Chinese A-share market is much younger than its European and US counterparts. The Shanghai and Shenzhen Stock Exchanges were only opened, or in the case of Shanghai re-opened, in 1990. The Shanghai Sharebrokers’ Association was originally founded in 1891 and flourished in the following decades, but was closed in 1949. Despite that, the total market capitalization of stocks listed on Chinese exchanges is increasing at a phenomenal pace. The total A-shares market capitalization now makes it the second largest stock market in the world after the US.
To the best of our knowledge, the low volatility effect for the Chinese stock market has not yet been tested. Recently, Hsu, Viswanathan, Wang and Wool4 documented that value investing works well for A-shares, but the momentum factor produces mixed results. A-shares are good testing ground for factors, especially since these markets are not integrated and ownership patterns are clearly different. For example, the majority of Chinese stocks are owned by individuals, while the US market is far more dominated by institutional investors.
In our new research, we focused on the 1,000 largest A-shares. All stocks were sorted according to their 36-month historical return volatility and placed in ten decile portfolios. The least volatile stocks were put in Portfolio 1, and the most volatile stocks ended up in Portfolio 10. These portfolios were rebalanced on a quarterly basis and returns were equally weighted. The research sample starts in January 2001 and ends in December 2016.
Figure 1 shows that A-shares tend to be more volatile than stocks in other equity markets. The spread in volatilities ranged from 28% to 37%. But it was also persistent and predictable since we observed a monotonic increase in volatilities across the ten volatility-sorted portfolios. However, unlike risk, there was not a linear increase in returns. Quite the opposite, in fact. The portfolio with the lowest risk, generated the highest returns. Returns ranged from +13% for the low risk portfolio to slightly below 0% for the high risk portfolio. A strategy that systematically bought the portfolio containing the most risky stocks, would have made a small loss over the 2001-2016 period.
Overall, the low volatility anomaly seems to be somewhat stronger than in other markets
Overall, the low volatility anomaly seems to be somewhat stronger than in other markets, such as the US. Many explanations have been put forward for the low volatility effect5. One of them is the ‘lottery ticket’ explanation, which assumes that many investors, especially individuals, buy risky stocks for their option-like payoff. This risk-loving behavior drives up the prices of risky stocks and brings down expected returns pushing the risk-return relationship into negative territory. The high trading volumes typical of A-shares suggest a link with another behavioral phenomenon, the ‘winner’s curse’.
Turnover in Chinese stocks is four to five times as high as for US stocks, despite the fact that trading costs are higher too. For high-volume volatile stocks there is less consensus on the price, which means that the buyers of these stocks might overpay them. As the A-share market is driven more by retail investors, we should not dismiss these two behavioral explanations.
At the same time, the negative relationship between risk and return poses a particular challenge, when it comes to other frequently mentioned explanations for the low volatility effect. Two of these are related to investment constraints, such as leverage restrictions and benchmarks as well as limits to arbitrage. Even though it is quite likely that these ‘investment constraint’ explanations also play a significant role here, they cannot explain the outright negative relationship between risk and return. How to disentangle these different explanations remains an interesting question for further research.
Since 2006, Robeco has been successfully running various Conservative Equity strategies, that exploit the low volatility effect. The model we use is the same for both developed and emerging markets. Besides volatility, it incorporates several other factors. We also find this model works very well for the Chinese A-share market. In fact, the results are remarkably similar to those seen in other markets. Our enhanced approach also adds value in this market, despite the fact that the stand-alone low risk factor is so strong for this sample and momentum gives mixed results.
Investors interested in A-shares should be aware of these new results. As in other markets, prudent investing, with a focus on risk reduction seems to be a long-term winning strategy. Once again, the tortoise beats the hare.
1‘The Volatility Effect in Emerging Markets’, Blitz, D., & Pang, J., Van Vliet, P., (2011), Emerging Markets Review, 16, 31-45.
2‘Low Risk Stocks Outperform within All Observable Markets of the World’, Baker, N., & Haugen, R.A., (2012)
3‘The Volatility Effect: Lower Risk Without Lower Return’, Blitz, D., & Van Vliet, P., (2007), Journal of Portfolio Management, Fall issue, 102-113.
4‘Anomalies in Chinese A-Shares’, Hsu, J C. and Viswanathan, V., Wang, C. and Wool, P. (2017), SSRN working paper 2955144.
5‘Explanations for the Volatility Effect: An Overview Based on the CAPM Assumptions’, Blitz, D., Falkenstein, E. & Van Vliet, P., (2014), Journal of Portfolio Management, Spring 2014, 61 -76.
当資料は情報提供を目的として、Robeco Institutional Asset Management B.V.が作成した英文資料、もしくはその英文資料をロベコ・ジャパン株式会社が翻訳したものです。資料中の個別の金融商品の売買の勧誘や推奨等を目的とするものではありません。記載された情報は十分信頼できるものであると考えておりますが、その正確性、完全性を保証するものではありません。意見や見通しはあくまで作成日における弊社の判断に基づくものであり、今後予告なしに変更されることがあります。運用状況、市場動向、意見等は、過去の一時点あるいは過去の一定期間についてのものであり、過去の実績は将来の運用成果を保証または示唆するものではありません。また、記載された投資方針・戦略等は全ての投資家の皆様に適合するとは限りません。当資料は法律、税務、会計面での助言の提供を意図するものではありません。
商号等： ロベコ・ジャパン株式会社 金融商品取引業者 関東財務局長（金商）第２７８０号
加入協会： 一般社団法人 日本投資顧問業協会