آزاد سازی بازار در یک کشور

China's B-share market, which used to be restricted to foreign investors, was partially opened up in February 2001 to Chinese local investors. We take this as a controlled experiment in cross-border trading on a small scale. We find mild but positive effects on the B-share market, with higher volumes, lower levels of volatility, lower bid–ask spreads and more liquidity after liberalization. Between A- and B-shares, price disparities narrowed; the correlation and the co-integration relationships became stronger; and the flow of information became more balanced. More new individual investors entered into the B-share market without crowding out existing institutional investors. Even though the liberalization measure is partial and one-way, it has helped to improve the quality of the B-share market, and our results lend no support to the popular claim that liberalization does nothing but help the existing foreign shareholders to cash out.

The integration of financial markets through the opening up of domestic markets to foreign investors and cross-border listings is a part of today's mega trend of globalization. Academic studies find substantial benefits in relation to such financial integration. For instance, the cost of capital tends to decline (Henry, 2000a and Bekaert and Harvey, 2000); the information environment tends to improve (Bae et al., 2006); real economic growth increases and consumption growth volatility decreases (Bekaert et al., 2005a, Bekaert et al., 2006 and Bekaert et al., 2007); and cross-listing to high quality foreign markets helps to mitigate agency problems and enhances governance quality (Doidge et al., 2004). However, the reaction to this opening up is not without skepticism, especially after the Asian Financial Crisis. Stiglitz (2003) argues that much of the instability in East Asia in the Crisis period was because these countries pursued capital market liberalization prematurely.1 Specifically, “allowing unfettered flows of speculative capital is extremely risky.”2 Indeed, one main concern for emerging economies when it comes to opening up their markets is the possible contagion effect of disturbances spilling over from developed markets. For instance, Ng (2000) finds significant volatility spillovers from Japan and the US to six Pacific-Basin equity markets. She shows that liberalization events affect the relative importance of the global and regional market factors of spillovers over time. With globalization and regional integration, Baele (2005) finds that even European equity markets face contagion from the US market during periods of high world market volatility. Bekaert et al. (2005b) set up a two-factor model to disentangle return correlation and contagion (residual correlation in the model). They were still able to find economically significant contagion in Asia during the Crisis period. To shed more light on the possible impacts of liberalization, in this paper we examine an interesting partial liberalization event within China, in which the Chinese government opened up its B-share market to local investors in 2001. Chinese companies can issue both A-class and B-class shares with identical features. Initially, A-shares were restricted to local Chinese only, and B-shares were restricted to foreigners only, so that the A-share and B-share markets were completely segmented. On February 19, 2001, the Chinese government announced that the B-share market would open up to local Chinese with foreign currency accounts in Chinese banks. The policy was implemented on February 28. The market reaction was electric at the start, as investors immediately began to redeploy their funds and thousands of Chinese investors poured into the B-share market. However, speculative trading became so intense in the first two months that it was said to scare away overseas institutional investors. Viewed as a small scale, emerging market liberalization phenomenon, we can ask a series of interesting and important questions in relation to this incident: Is market liberalization really beneficial to the liberalized market? Are the benefits transitory or long lasting? Does liberalization benefit all markets, or does it benefit only one at the expense of others? Rigorous study of these questions will help to elucidate the pros and cons of the liberalization policy, and shed light on fundamental issues of financial market integration. This Chinese experience has another unique and interesting feature: it is a “reverse” liberalization. The typical liberalization setting is that less-developed domestic markets open up to foreign investors; in our case, however, the “foreign” B-share market opens up for “domestic” investors.3 Analyzing the impacts of this type of reverse liberalization is of potential importance as it is not clear whether reverse liberalization would lead to different consequences. For instance, in the typical liberalization setting, foreign investors are usually more sophisticated and likely to be institutional investors. Their entrance may improve the price discovery process and the efficiency of the less-developed, domestic market. In contrast, in our setting, it is the less sophisticated retail investors entering into the B-share market dominated by foreign institutional investors. As will be seen, however, our results indicate that “reverse” liberalization has the same effect as those predicted for the normal liberalization cases. As long as trading barriers are reduced or eliminated, there are benefits to be gained. Our investigation is quite timely, as a much larger scale reverse liberalization will soon take place, as the China Securities Regulatory Commission (CSRC) issued “Tentative Measures for the Administration of Overseas Securities Investment by Qualified Domestic Institutional Investors” in June 2007, under which Chinese fund management firms and securities companies are allowed to invest in overseas equity markets.4 The benefits we are investigating in this study are improvements of market quality. Specifically, we examine changes in liquidity, turnover, volatility, trading cost, price disparity, information linkage, and information flow between Chinese A- and B-shares a year before to a year after the liberalization event. Our results generally show that the B-share market has exhibited higher turnover, lower volatility, lower bid–ask spread, and more liquidity after liberalization. In addition, the turnover, volatility, bid–ask spread, and liquidity of the two markets tend to converge over time. In fact, after liberalization, the price disparities between A- and B-shares have narrowed. Yet, stocks with a relatively larger disparity before liberalization continue to have a relatively larger disparity after liberalization. We interpret this as inconsistent with the view that the original B-share investors had been cashing out after the opening up of the B-share market. Further indirect evidence against the cashing-out view comes from the aggregate data. We observed a surge in the number of individual investor accounts entering the B-share market, and also a mild increase in the number of individual investors entering the A-share market after liberalization. Hence, there is no sign that the increase in the number of investors in the B-share market crowded out its existing institutional investors. Finally, we find that the return correlation and the co-integrating relationships between the two markets have become stronger and tighter. The flow of information between the two markets has also become more balanced. These results indicate that liberalization has benefited the B-share market without adversely affecting the A-share market, and there is no evidence that foreign investors have systematically abandoned the B-share market. All in all, China's experience supports the view that partial liberalization, even the “reverse” liberalization, of a stock market, leads to improvements in the quality of the market in general. However, such improvements are limited, probably due to the partial nature of the liberalization. Viewed as market liberalization within a country, we have a policy change situation close to a controlled experiment. Different countries have different macro conditions, including legal, political, social and economic environments. These are difficult to control when examining the cross-country liberalization effect. In fact, Henry (2000b) considers the ability to control different macro economic stabilizations to be an important contribution to the study of the market liberalization effect across countries. As both A-share and B-share markets exist within the same country, the legal, political, social and economic environments are the same. In addition, our sample only consists of firms with both A- and B-shares. This means that even firm-level differences across the two markets are controlled. This “perfectly matched” sample allows us to focus on how a “pure” lifting of trade barriers impacts the trading in these two markets. The downside of focusing on one country, of course, is that data observations are fewer and the results may be country-specific. Our study also relates to the literature on market segmentation and price differentials. Stulz and Wasserfallen (1995), Bailey and Jagtiani (1994), and Domowitz et al. (1998) show that market segmentation can lead to differential demand for identical stocks in different markets, and thus result in different prices. In addition, trading volume, liquidity, volatility, and even the information set associated with the same stocks in the two segmented markets can be very different. Our study also relates to the cross-market microstructural studies by Hasbrouck (1995), Domowitz et al. (1998), and Arnold et al. (1999). They show how the lifting of trade barriers impacts on the order flow dynamics through time, and subsequently, affects the quality and development of the two markets. There is a series of studies focusing specifically on China's market segmentation phenomenon. Bailey (1994) first documented that B-share prices generally traded at discounts relative to their corresponding A-share prices. Subsequent studies found, in general, that the cross-sectional differences of the B-share discount was related to differences in the trading attributes of stocks, such as: volatility, turnover, and liquidity (Chen et al., 2001); differences in general supply and demand conditions of A- and B-shares (Sun and Tong, 2000); speculative trading of local investors (Mei et al., 2005); and/or differences in information availability (Chan et al., 2008). Mei et al. (2005) argue that the prohibition of short-sales of stocks, non-existence of equity derivatives markets, severe constraint of equity issuance and buy-backs, and relatively short history of the stock market in China (opened in the early 1990s) have all contributed to the excitement and interest in stock trading among Chinese local investors, which has led to the speculative nature of their trading. Mei et al. (2005) use A-share turnover rate to capture the speculative trading activities of domestic investors and find it can explain, on average, 20% of the monthly cross-sectional variation of the A–B-share price premium in the period of 1993–2000. They take this as evidence supportive of their hypothesis that domestic investors' speculative motive generates a speculative component in A-share prices. In postulating some of our hypotheses, we also assume the speculative trading of domestic investors and interpret our results accordingly. The paper proceeds as follows. A brief history of the Chinese equity market is presented in Section 2. Hypotheses regarding the liberalization effect are developed in Section 3. Section 4 explains the data and market quality proxies used in the tests. Section 5 presents the testing techniques and the corresponding results. Section 6 concludes the paper.