What Determines Chinese Stock Returns?

This study applied the three-factor model to A-shares in the Chinese equity market, one of the fastest growing markets ever. The sample period is July 1996 through June 2002. Size was found to explain the cross-sectional differences in returns, but contrary to findings for the U.S. market, the book-to-market ratio was not helpful. As in the U.S. experience, beta did not account for return differences among individual stocks. Because of the speculative nature of Chinese capital markets, the large proportion of government-owned shares, and the low quality of the companies' accounting information, the free float (that is, the ratio of shares in a public company that are freely available to the investing public to total company shares) was added to the study to serve as a proxy for company fundamentals. The three-factor model that included proxies for size and free float significantly increased the explanatory power of the market model—from 81 percent to 90 percent. The Chinese equity market is one of the fastest growing markets ever. Although the majority of stocks are those of small-cap companies by U.S. standards, the speed of these companies' growth is astonishing. Like many emerging markets, the Chinese markets suffer from unsatisfactory corporate governance, dubious accounting practices, market manipulation, and insider-trading problems. Not only are institutional investors lacking, but most investors trade speculatively with short holding periods.Our study concerned A-shares, which are available only to domestic investors, for nonfinancial companies. These stocks are traded on either the Shanghai Stock Exchange or the Shenzhen Stock Exchange. Because of restrictions on trading, about half of the A-shares in the study were not tradable.Indeed, the issue of nontradable shares is the paramount issue facing the Chinese stock markets. It has generated inefficiencies and problems of corporate governance.We set out to apply the Fama and French three-factor model to the Chinese market. The unique aspects of the Chinese equity market suggested, however, that alternatives to the three factors might be at work in China. In particular, investors in a speculative, emerging market may understand that book values are very inaccurate and thus it is difficult to evaluate future cash flows or growth opportunities by comparing a company's book value with its market value. Therefore, we sought a proxy that would signal a company's future success, and we argue that free float (that is, the ratio of shares in a public company that are freely available to the investing public to total company shares)serves that purpose.A large proportion of tradable shares should lead to better corporate governance by outsiders, which should lead to company growth and profitability.Using A-shares for theperiod July 1996, when price stabilization was implemented on the Chinese exchanges,through June 2002, we performed cross-sectional tests on the conventional three factors and the proposed free float factor. Then, we constructed our proxies for size (market cap or price times the number of tradable shares) and free float; we ranked all stocks in our sample by size and divided them into two groups (small size and big size). Independently of the size sorting, we also ranked stocks on the residual free float calculated from regressing free float on the log market cap of all stocks in June of each year. We divided the stocks into three groups (high, medium, and low) according to the free float sorting and formed six portfolios with equal weights from the intersection of the two size and three residual free float groups, which were then used to construct proxies—“small minus big” for size and “high residual FF minus low residual FF” for free float.In the cross-sectional study, size was found to explain the differences in returns among individual companies, but contrary to findings for the U.S. market, the book-to-market ratio was not helpful. As in the U.S. experience, beta did not account for return differences among individual stocks. Our primary finding is that free float is useful in understanding cross-sectional return differences for Chinese stocks.The time-series evidence from 25 size+float-sorted portfolios suggests that a three-factor model consisting of the market factor, size, and free float can explain 90 percent of the variation in portfolio returns, which is a 10 percentage point improvement over the result of using a simple market model.