In a recent whitepaper, we assessed the viability of deploying quantitative strategies in the China A-share market. We conclude that the A-share market has many of the ideal characteristics in which quantitative strategies should flourish. At the same time, we note that there are unique investor behaviours and structural nuances which quant investors need to be aware of. We present some of our findings in this article.
As the China A-share market becomes more accessible to offshore investors, and as global index providers increase the weight of China A stocks in regional and global indices, the option for global or regional investors to treat China as a rounding error in their benchmarks is fast disappearing. The size (and growth) of the Chinese market and the fact that it has relatively low correlation with other global markets add further compulsion for global investors.
In our whitepaper “The potential for quant in the China A-share market”, we assessed the viability and potential challenges of deploying quantitative strategies in the China A-share market by comparing it with other global markets where quantitative strategies have been prevalent for decades.
We note that for a start, data – which is key to identify factors and run backtest studies, is bountiful. Global, together with local data providers, offer sufficient market and financial statement data as well as analyst estimates and recommendations across the largest 800 China A-shares. At the same time, while the Chinese stock market is relatively young compared to the developed markets, the number and size of listings in China have grown rapidly over the past decade. There is ample breadth (in terms of number of stocks) for researching and forming quantitative strategies.
The China A market is also liquid. Over 2,4001 stocks have at least USD25m of average daily value traded. Total value traded as a proportion of free float market capitalisation is higher than other global markets and regions, although this is partly due to the greater retail participation in the China market. Meanwhile, transaction costs2 are lower than in Hong Kong or Japan, although higher than in the US.
When we break down the China A market capitalisation by GICS sectors, we get a broad and diverse universe of stocks.
This provides a more optimal backdrop for a stock selection-based quantitative strategy. Notably, not only is the China A-share market diverse, the sector diversity also appears fairly stable over time which would prove very useful when assessing backtested results. See Fig. 1.
Fig. 1. Breakdown of total market cap by GICS sector for the largest 800 China A-shares3
Meanwhile, the high dispersion among stock returns within the China A-share market provides more opportunities for a quantitative strategy to be rewarded for its stock selection ability. The overall market volatility in China is higher than for other regional and global markets, so investing in this market does require a tolerance for risk.
Stock returns are driven by many factors, including exogenous factors which may not be offset via risk management techniques. The Citi Risk Attribute Model (RAM) estimates how much the risk of stocks in different markets is driven by macro factor sensitivity, global market sensitivity, sector (and country for regions) sensitivity as well as other style factors.
The idiosyncratic element represents the residual risk that is unexplained by the model. Markets with a high idiosyncratic risk contribution are indicative of more stock-specific influences and less impact from macro influences, which is an ideal characteristic for stock selection-based quant strategies. The China A market, according to the model, is such a market4. See Fig. 2.
Fig. 2. Average residual (unexplained) proportion of estimated risk from the Citi Risk Attribution Model (RAM)5
Besides considering the characteristics of the China A-share market and the likely implications for quantitative strategies, we also assessed the efficacy of some common quant factors in the China A-share market. Most of these factors (e.g. price to book, earnings revisions and low volatility) are as efficacious in the China A-share market as they are in other regions, and even more so for selected factors. A standout was the strength of the shorter-term reversal factor – this implies that stocks that rally in the previous month tend to reverse in the subsequent month. The Momentum factor was however an exception. It was not only weaker than in other regions but was also negatively rewarded. All these effects warrant further analysis and are likely to lead to more refined factors being formed to capture this market’s unique underlying inefficiencies.
We believe that the China A-share market has many of the ideal characteristics in which quantitative strategies should flourish. That said, there are clear signs of different investor behaviours as well as structural nuances that must be considered and respected by quantitative investors.
The higher proportion of stocks held by retail investors, for example, create observable and persistent market efficiencies that can be exploited by quantitative strategies.
At the same time, the limited availability of hedging instruments, the limited number of approved stocks for short selling and higher borrowing costs for shorting represent challenges for implementing a long-short strategy in China A-shares. Investors are also prevented from selling shares bought on the same day, which results in unique market anomalies.
Our assessment is not a statement of any permanent structure to the Chinese A-share market. Given the dynamism of the market, investors will need to re-evaluate findings to take into account changes in market conditions and structural factors over time. This discipline will be core for any systematic strategy in the Chinese market.
Please click this link to read the complete whitepaper “The potential for quant in the China A-share market”.