Which Chinese Markets to Diversify into?

Transcription

1 Front. Econ. China 2013, 8(2): DOI /s RESEARCH ARTICLE Leo H. Chan Which Chinese Markets to Diversify into? Abstract This paper investigates the correlation and feedback relationships between the Hong Kong Hang Seng Index (HSI), the Hang Seng Chinese Enterprise Index (CEI) and the S&P 500 Index (SP). We divide the indexes into two separate periods, from the inception of the CEI in 1994 to the stock market crash in 2000, and from 2001 to Our results show that the feedback relationship between the CEI and the SP is stronger after As the feedback relationship grows stronger, the diversification benefit reduces for US investors who utilizes the CEI as a tool for diversifying into Chinese markets. Keywords market correlation, diversification, Chinese stock market JEL Classification G14, G15, C32 1 Introduction As Chinese mainland s economy continues to expand at an unprecedented rate, there are increasing interests from outside investors (particularly those from the US) to partake in the growth potential of this emerging economy s stock market. Participating in Chinese mainland s stock markets not only allows investors the chance to capture the growth potential of Chinese mainland s expanding market, it also allows investors to diversify their risk exposures. Empirical studies have shown that emerging markets typically have a smaller correlation relationship with the US markets. Therefore, to diversify into emerging markets is an appropriate choice for the US investors. However, the task of making the right asset allocation decision is not an easy one even in developed markets, the same task is even more difficult for emerging equity markets such as the ones in Chinese mainland. When making asset allocation decisions to diversify into emerging markets, investors must balance the need for diversification and higher returns with the higher level of risk involved. The level of risk is generally negatively correlated with the degree of openness, maturity of the market and regulations of the country where the assets are being traded. In this regard, Hong Received October 15, 2011 Leo H. Chan ( ) Department of Finance and Economics, Woodbury School of Business, Utah Valley University, Orem, UT 84058, USA leohchan@yahoo.com

2 Which Chinese Markets to Diversify into? 221 Kong and Chinese mainland represent two vastly different options for international investors who wish to partake in the equity markets in China. Hong Kong has a well-developed capital market and the regulations of human capital in Hong Kong s financial markets are of a similar standard to the other mature financial markets around the world. The financial markets in Chinese mainland, on the other hand, are relatively young and have gone through various changes in the first two decades of development. Furthermore, foreign investors only have restricted access to the equity markets in Chinese mainland. The decision to give foreign investors limited access to internal equity markets is intended to protect the growth and development of the internal capital markets from speculators. Such protective regulation, however, has also impeded the growth and development of the equity markets in Chinese mainland. Therefore, Chinese mainland s equity markets are not the most ideal for international investors who wish to tap into the growth potential of Chinese mainland s expanding economy. One of the unintended consequences of Chinese mainland s capital markets relative under development and restriction is that many large Chinese companies list their stocks on the Hong Kong stock market (the H-shares Index) (Johansson, 2009). Therefore, investors who are interested in diversifying into the Chinese capital market could utilize the H-shares market rather than directly investing in Chinese mainland s equity market. This is one possible reason why the Hong Kong Stock Exchange s H-shares Index has been so popular since its inception in Still, existing companies in the H-shares Index are mostly large companies in Chinese mainland whose largest shareholder is the Chinese government. This factor could limit the interest of some foreign investors who are concerned about corporate governance and transparency. The alternative to directly investing in Chinese mainland s capital market is to invest in Hong Kong s equity market. Hong Kong s economy depends heavily on the economic performance of Chinese mainland, and so investing in the stock of Hong Kong companies has the benefit of being able to capture Chinese mainland s economic growth without taking on much of the risks associated with the stocks of Chinese companies. However, with increasing participation from foreign investors, the behavior of the equity markets involved will also change. For example, a large fund company based in the US that has investments in Chinese mainland and the US might make the same buying (or selling) decisions in both locations when facing an influx (or withdraw) of funds. Such action would cause similar pressure on equity prices on both markets, thereby increasing the correlation between the two markets. In this paper, we investigate the correlation between Hong Kong s and the mainland s capital markets (using the H-shares Index as our proxy) with the US capital market to determine which of these two markets provides more diversification benefit for those interested in capturing the growth potential of

3 222 Leo H. Chan Chinese mainland s economy. The remainder of the paper is organized as follows: Section 2 provides a brief review of the development of the financial markets in China and literature related to the financial markets and trade relations between Chinese Hong Kong, Chinese mainland and the US. The methodology and data description is set out in section 3. Section 4 presents the empirical results, and a conclusion is provided in section 5. 2 Literature Review 2.1 Brief Review of China s Economic Transformation There are many books and articles that document the details of China s amazing economic transformation. Chow (2001) provides a great review and detailed analysis of the transformation process. China s success is in direct contrast to her counterparts in Eastern Europe since the fall of the Soviet Union (EEFSU). Many studies (e.g., Benzinger, 1997; Brown and Earle, 2000; and Woo, 2001) compare the different paths taken by Eastern European countries and China in the transformation from being centrally planned economies to becoming market economies. The Big Bang path (adopted by EEFSU) to reform has been to model a central-planned economy on a market economy with similar characteristics. Firms and institutions are assumed to function well in a newly transformed economy just as they would do in the market economy that it is modeled upon. The Gradualism approach (adopted by China) is similar to the learning by doing, or trial-and-error approaches. Such approaches allow the newly transformed economies to learn which policies work and which policies do not. The Chinese government s approach to the stock market mirrored its approach to the reform of its economy. The intentions for creating the stock markets are to tap into the domestic savings for sources of capital for investments and help the reform of state-own-enterprises. Two exchanges were created: the Shanghai Stock Exchange and the Shenzhen Stock Exchange, which were created in December 1990 and April 1991, respectively. Like the Special Economic Zones that were created as an experiment for market-based economy, the stock markets were also intended as an experiment. Due to the lack of experience with stock markets and wanting to protect domestic stock markets from speculators, two types of stocks were created: A-shares and B-shares. The A-shares are limited to domestic trading only, while the B-shares are strictly for foreign investors. In effect, there are two markets within two equity markets in China. In its early stages of development, the stock market experienced a lot of turmoil. Relatively inexperienced with how the stock market functions, Chinese investors in the early 1990s assumed that investing in the stock market was a sure way to make money and that their investments were protected by the government.

4 Which Chinese Markets to Diversify into? 223 That led to a speculative bubble in Investors learned a painful lesson when the government, after many of the listed companies failed, imposed stricter regulations for listing companies which caused a large correction in market value. Still, the excessive under-pricing of IPO documented by Tian (2002) persists in A-shares markets. In other words, investors in China still rush in to buy new shares issued in the hope of higher returns on their investments. While Tian (2002) suggests that political rent seeking is one of the main driving forces of excessive under-pricing in the IPO market, it alone could not explain the overall inefficiency of China s stock markets. The main cause for inefficiency in the Chinese stock markets could be excessive restrictions and macroeconomic policy control that limits the openness of the markets. More than 10 years after the creation of the stock markets over 60% of the stocks have limited tradability, and less than 5% of the stocks are denominated in foreign currencies (the B-shares markets). The reason for limited tradability is that most of the stocks are partial shares of State-Owned Enterprises (SOEs) and by limiting their tradability the government will have more control over the stability of the SOEs. This type of partial privatization creates ample opportunities for corruption and under-pricing. The fact that SOEs dominate the A-shares markets is another possible justification for not allowing foreign investors to trade in the A-shares markets. On the other hand, foreign exchange control policies also limit the development of the B-shares markets. The inefficiency of the A-shares markets is taking its toll. Many foreign investment banks are not utilizing their QFII licenses 1 to trade stocks in the A-shares markets due to the lack of quality stocks. Even China s pension fund system is seeking to invest in foreign stock markets, citing a lack of good investment opportunities in the domestic stock markets. The need for reforms in the stock market is evident. An increasing number of companies are able to issue H-shares (traded in Hong Kong) and ADRs (traded in the US) suggests that Chinese companies are increasingly able to meet the listing standards in overseas markets. However, the availability of H-shares and ADRs, while increasing the exposure of Chinese companies to international investors, also limits the development of the domestic financial markets because regulatory reforms can be carried out at an even slower pace. As such, the equity markets in China will remain an inferior alternative to the H-shares market in Hong Kong. 1 Foreign institutional investors are allowed a limited trade of A-shares under the new Qualified Foreign Institutional Investor (QFII) regulations. Under QFII, foreign institutional investors can purchase shares in the A-shares markets with a limited capital range from US$ 50 million to US$ 800 million. The risk associated with poor accounting and reporting standards in China limits the interest in A-shares stocks (other than those which already have good reputations, such as the H-shares or ADRs, however, H-shares and ADRs are found to be trading at a discount to A-shares).

5 224 Leo H. Chan 2.2 Literature Review on Financial Market Integration in Asia There are numerous studies that document the relationship between economies in Asia and the US. These literatures can be classified into two main categories. The first group of literature deals with the interdependency (or correlations) between these financial markets (e.g., Chan, 2002; Tay and Zhu, 2000; Cheung and Ho, 1991; Ko and Lee, 1991; Lin, Engle and Ito, 1994; Chowdhury, 1994; Phylaktis and Ravazzolo, 2002; Yu et al., 2007; Johansson, 2009). The second group of literature deals with the linkages between macroeconomic factors such as exchange rates and rates of inflation (e.g., Granger, Huang and Yang, 2000; Cheung and Yuen, 2002; and Devereux, 2003). Ko and Lee (1991) examine the daily stock returns and relationships between the US and five Pacific Basin countries/areas (Japan, Chinese Hong Kong, Singapore, Chinese Taiwan and South Korea). They show that the cross-correlations among Japan, the US, Chinese Hong Kong and Singapore are significant in a statistical sense. However, Chinese Taiwan and South Korea show little relationship with these markets. They also find that the one-day lagged correlations of the Asian countries with the US are larger than contemporary cross-correlations. Phylaktis and Ravazzolo s (2002) findings show that during the 1990 the Asian NICs financial markets were integrated with each other. However, they also report that only Hong Kong and Singapore s markets are integrated with the US market. Since Hong Kong and Singapore s markets are more developed and open, their finding suggests that only the more developed markets in Asia will have a significantly correlated relationship with the US market. More recently Fung et al. (2008) and Yu et al. (2007) report that while the market integration among Asian markets improved from 1994 to 2001, the process has stalled since 2002 and integration remained weak through Noting that the 1987 market crash had changed the correlation relationships for many markets, Chan (2006) reexamines the interdependencies between Japan, Chinese Hong Kong, Singapore and the US and finds stronger interdependencies among these markets after the 1997 Asian Crisis. The increases in the interdependencies between the US and the Asia NICs might reduce the gain in diversification from investing in the Asian NICs markets. Granger, Huang and Yang (2000) provide an analysis of the relationships between exchange rates and stock market price movements in the Asian NICs. Empirical evidences for the US markets suggest only a weak connection between exchange rates and stock market movements. Therefore, the foreign exchange control practice of China should not be a major factor in analyzing equity market behaviors. When considering the relationship between the economy and equity markets in Hong Kong and China, one must understand the following. First, the Hong Kong Dollar is pegged to the US Dollar and the RMB under a target-zone system

6 Which Chinese Markets to Diversify into? 225 against the US Dollar. Second, the trade relationship between Chinese Hong Kong and mainland is very strong. If there is any change in correlation between the financial markets of these two, there must be a fundamental change in perception about China s economic condition. Since China relied heavily on trade with the US during the 1980s and 1990s, though decreasing slowly in the 2000s, the correlation between the markets of China and the US over these two different periods is likely to be different as well. 3 Methodology There are many options when it comes to modeling financial market integration. The traditional cointegration analysis or Granger Causality test provides only unit-directional effects. Simple correlation analysis utilizes only level (return) data and ignores the interaction of variance. Furthermore, empirical studies in recent decades all point to time-varying correlation relationships among financial markets. GARCH-type models are the most popular choices for studying time-varying interactions among variance (see Johansson, 2009). In this study, we utilize alternative statistics proposed by Geweke (1982) to test for bi-directional relationship changes. There are four important statistics in Geweke s Measure, they are the following: (1) The linear dependence (or total information flow) between two data series, (2) The linear feedback from the first series to the second series, (3) The linear feedback from the second series to the first series, and (4) The instantaneous linear feedback measure between the two data series. The two statistics (2) and (3) are unidirectional feedbacks. The total of the feedback statistics is the test statistic, which has the null hypothesis that there is no information flow between the two data series. The directional feedback measures test the hypothesis that the first series has no effect on the behavior of the second series. Similarly, the instantaneous feedback measure tests the hypothesis that the second series has no contemporaneous relationship between the two data series through the co-movement of the shocks to the two data series. By comparing the value of the Geweke Measures before and after December 2000, we see if there are any changes in the correlation relationship between the US stock market and the markets in Hong Kong and China since the dot-com bubble. 4 Data Analysis and Results 4.1 Data and Methodology We obtained our data for the H-shares Index from its inception in August 8, 1994

7 226 Leo H. Chan to December Data from the Hang Seng Index and the S&P 500 Index of the same period are also collected. We use the S&P 500 Index instead of the Dow Jones Index because the S&P 500 Index represents a larger number of companies in the US. It is also a value-weighted index, which is the same as the Hang Seng Index and the H-shares Index. We divide the data into three sub-periods: 1994 to 2000, 2001 to 2006 and an extended sub-period from 2001 to It is difficult to make an argument for a specific date to separate the data. We therefore choose the end of the Dot-Com (Tech) Bubble in 2000 as a starting point because investors usually make drastic reevaluations of their investment portfolio and adjust their asset allocation after major downward adjustments in the financial markets. Studying the market correlation between the Hang Seng Index and the H-shares Index from 2007 to 2011 alone is inappropriate because the Hang Seng Index added more stocks from the H-shares Index to its composition in January As only large market value stocks are added from the H-shares Index to the Hang Seng Index, the impact of the H-shares on the Hang Seng Index increases after By extending the data series beyond 2006, we should expect to see a much higher degree of correlation between the H-shares Index and the Hang Seng Index. In order to calculate the Geweke Measures, a standard VAR model setup is required. A calculation of the true relationship between the two data series (first series F t and second series S t ) is as follow: n F = β + β S + β F + ε t 0 1i t i 2i t i Ft i= 1 i= 1 n n t = α0 + α1i t i + α2i t i + εst i= 1 i= 1 n, (1), (2) S S F where the variance/covariance matrix of ( ε Ftε St ) is 2 εft σf σ FS Σ= cov ε =. 2 St σsf σs If, instead, there is no inter-temporal relationship between these two series, then equations (1) and (2) are reduced to 2 uf n t = β0 + β2i t i + Ft i= 1 n St = α0 + α1ist i + ust i= 1 2 ust σ us F F u, (3), (4) with var( uft ) = σ and var( ) =. The equation system (1) (2) can be estimated by a seemingly unrelated regression method, whereas, the equation system (3) (4) can utilize the ordinary least squares method. The Geweke

8 Which Chinese Markets to Diversify into? 227 Measures can then be calculated from the varian-covarian matrix in equations (1) (2), and the variance estimates from equations (3) (4). 4.2 Empirical Results The simple correlation among these data series is reported in Table 1. In all 3 data samples, the correlation between the Hang Seng Index and the H-shares Index are much stronger than the correlation between the Hang Seng Index and the US market. Likewise, the US market and the Hang Seng Index had a much stronger relationship than of that between the H-shares and the US market. Note that the simple correlation indicates a (marginal) decline in correlation between the Hang Seng Index and the H-shares market, as well as the Hang Seng Index and the S&P 500 Index from 2001 to This is in contradiction with the findings of most empirical studies, which indicate increasing integrations among these markets in recent years. Thus, if an investor is relying on a simple return correlation alone, he/she might come to a wrong asset allocation decision. Table 1 Simple Correlation Coefficients Period HK-CEI HK-SP CEI-SP Tables 2A to 2C report the results of the VAR model layout in equations (1) and (2). The number of lags for the VAR model is determined by AIC. In Table 2A, the results for the Hong Kong and US markets show that these two markets have a positive lead-lag relationship for all periods. The impact of a positive return in the US market has a much higher impact on the Hong Kong market prior to 2001 than it has afterwards; to , respectively. Extending the data to 2011, the lead-lag relationship between the two markets remains below the Table 2A Relationships between the Returns of the Hong Kong and the US Markets Variable HKR 1 SPR 1 HKR 2 SPR 2 HKR 3 SPR ** Constant ( ) 2 (3.1638) ) ( ) (0.490) ( 0.121) HKR{1} ** ** ** ( ) ( ) ( ) ( ) ( 7.311) ( 1.983) ** ** ** ** ** SPR{1} (16.272) (0.2587) ( ) ( ) (23.347) ( ) Note: (1) HKR 1 and SPR 1 are the dependent variables for the structural relationship equation before the Dot-Com Bubble. HKR 2 and DIR 2 are the dependent variables for the structural relationship equation after the Dot-Com Bubble. (2) 1 The number in { } denotes the value of the lag; 2 The values in parenthesis are the associating t-statistics; ** denotes statistical significance at the 5% level.

9 228 Leo H. Chan Table 2B Relationships between the Returns of the Hong Kong and the H-Shares Index Variable HKR 1 CER 1 HKR 2 CER 2 HKR 3 CER Constant (0.8373) HKR{1} ( ) CER{1} (0.4076) ( 0.904) (1.5056) ** (3.7432) (0.1397) ) (1.8302) (1.8101) ** ( ) ** (5.2704) (0.3487) ( ) (0.0329) (1.44) ** ( ) ** (4.1456) Note: HKR 1 and CER 1 are the dependent variables for the structural relationship equation before the Dot-Com Bubble. HKR 2 and CER 2 are the dependent variables for the structural relationship equation after the Dot-Com Bubble. Table 2C Relationships between the Returns of the H-Shares Index and the US Markets Variable SPR 1 CER 1 SPR 2 CER 2 SPR 3 CER 3 Constant ** ** SPR{1} ** ** ** ** SPR{2} ** SPR{3} ** SPR{4} SPR{5} ** SPR{6} ** SPR{7} SPR{8} SPR{9} SPR{10} CER{1} ** ** CER{2} CER{3} CER{4} CER{5} * CER{6} CER{7} CER{8} 0.02** CER{9} CER{10} Note: (1) CER 1 and SPR 1 are the dependent variables for the structural relationship equation before the Dot-Com Bubble. CER 2 and DIR 2 are the dependent variables for the structural relationship equation after the Dot-Com Bubble. (2) We only report the coefficients in the table to conserve space. * denotes the coefficient is significant at the 5% level and ** is significant at the 1% level.

10 Which Chinese Markets to Diversify into? 229 level. On the other hand, there is no significant lead-lag relationship from the Hong Kong market to the US market prior to However, when extending the data to 2011, we saw a statistically significant effect from the lag return in the Hong Kong market to the US market. This result is consistent with other studies that show a gradual decline of influence of the US market on the Hong Kong market, while the world s financial markets are increasingly integrated. Table 2B reports the results for the Hong Kong Hang Seng Index and the H-shares Index. The returns of the Hang Seng Index have very little influence over the returns of the Chinese Enterprise Index prior to After 2001, the relationship changed drastically. The returns of the Hang Seng Index have a strong negative impact on the returns of the CEI. When the data is extended to 2011, the impact of the returns of the Hang Seng Index on the returns of the H-shares Index return is even more significant. This could be an indication that investors are participating in both indexes. An investor who invests in both markets would shift his/her fund to (from) the Hang Seng Index if there is a positive (negative) return the day before. Table 2C shows the results for the H-shares Index and the S&P 500 Index. The results of the first period could be unreliable due to a low level of participation in the H-shares Index by US investors. After 2001, the result mirrors the relationship between the Hong Kong market and the US market. The value of the estimated coefficient for the HKR{1} SPR{1} is , while the value between CER{1} SPR{1} was Extending the data to include 2011, both numbers increase. In other words, based on the VAR result, there is a larger benefit for diversifying into the H-shares Index for US investors prior to However, after 2001, the diversification benefit of the H-shares is similar to that of the Hang Seng Index. So far, both the static correlation coefficients and VAR point to a reduction in diversification benefits for the US investors who are interested in investing in the H-shares Index. The bi-directional results reported in Table 3, panels A to C suggest a different story. First, there is a reduction in instantaneous feedback between the Hang Seng Index and the H-shares Index (from to ). A much more drastic reduction is reported for the H-shares Index and the S&P 500 Index (from to ). On the other hand, the instantaneous feedback between the Hong Kong and the US market improved after The total feedback also shows a decline between the Hang Seng Index and the H-shares Index (from to 0.375). Similarly, a decline in total feedback is observed between the H-shares Index and the S&P 500 Index (from to ). While there is also a small decline in the total feedback between the Hang Seng Index and the S&P 500 Index, the reduction is at a much smaller scale compared to the decline for the other two pairs.

11 230 Leo H. Chan Table 3 Value of the Geweke Measure Panel A: Geweke Measures for the Hong Kong Market and the Chinese Enterprises Index Period 1 Period 2 Period 3 F ** ** ** ˆHK CE FˆHK CE ** ** F ˆCE HK F ˆCE, HK ** ** ** Panel B: Geweke Measures for the Hong Kong Market and the US Markets ˆHK US Period 1 Period 2 Period 3 F ** ** ** FˆHK US F ** ** ** ˆUS HK F ˆUS, HK ** ** ** Panel C: Geweke Measures for Chinese Enterprises Index and the US Markets Period 1 Period 2 Period 3 F ** ** ** ˆCE US FˆCE US F ** ** ** ˆUS CE F ˆUS, CE ** ** ** Note: (1) ** indicates significant at the 1% level; (2) F ˆi US, (i = HK or CE) = the instantaneous feedback between the two markets; (3) F ˆi US = feedback from Hong Kong (or H-Shares Index) market to the US market; (4) FˆUS i = feedback from the US market to Hong Kong (or H-Shares Index) market; (5) F ˆUS, i = total feedback measure between the two markets. The result for the Geweke Measures we obtained by extending the data to include 2011 verifies our initial speculation that the Hang Seng Index and the H-shares index would have a much higher degree of correlation after 2006 due to the larger share of H-shares stocks in the Hang Seng Index. The total feedback measures between these two markets stands at , with the majority of it coming from instantaneous feedback (0.9881). The feedback measures between the H-shares Index and the US market shows a much higher degree of information flow between these two markets. While the increases in the total feedback measure between the Hong Kong and the US markets is about 22%, the percentage change between the H-shares Index and the US market is almost 200%. To summarize, both the static statistics on correlation and VAR model point to declines in the diversification benefit between the H-shares Index and the US

12 Which Chinese Markets to Diversify into? 231 market. The Geweke Measures show that information flow among all markets has improved significantly over the last two decades. More importantly, the H-shares index is increasingly behaving like the Hang Seng Index. While investors from the US who wish to diversify into the Chinese equity market should still consider the H-shares Index rather than the Hang Seng Index, due to the lower correlation and information flow between these two markets, the diversification benefit of the H-shares Index over the Hang Seng Index is declining. In effect, in recent years the H-shares Index and the Hang Seng Index are almost substitutes for each other. In order to gain more diversification benefits, investors should look into other sources for accessing the Chinese equity markets, perhaps by participating in the B-shares markets. 5 Conclusion In this paper, we investigate the correlation and feedback relationships between the Hang Seng Index, the Chinese Enterprise (H-shares) Index and the S&P 500 Index to determine which Chinese capital market would result in the best diversification benefits for investors. The static statistics on correlation and the VAR model point to a decreasing benefit for diversifying into the Chinese Enterprise Index (the H-shares Index) over time. However, the bi-directional feedback relationship, the Geweke Measures, still suggests that using the H-shares Index is more useful than the Hang Seng Index in deriving diversification benefits. But the advantageous position held by the H-shares Index over the Hang Seng index is declining. The decision made by the Hang Seng Index to include more stocks from the H-shares Index has made these two indices closer substitutes for each other. If the Hang Seng Index and the H-shares Index are the only choices available for diversifying into the Chinese equity markets, using the H-shares Index is still a better choice. However, investors should consider other alternatives in accessing the Chinese equity markets if diversification benefits are an important concern. Until the liberalization of the equity markets in China is complete, participating in the B-shares market might be more beneficial than utilizing the H-shares market. References Benzinger V (1997). Can China s gradualist reform strategy be applied to Eastern Europe? Working paper, Stanford University Asia Pacific Research Center Chan L H (2006). How does the Asian Crisis affect the interdependencies between major financial markets in Asia and the US. Journal of Emerging Markets, 10(2): Cheng L T W, Fung J K W, Chan K C (2000). Pricing dynamics of index options and index futures in Hong Kong before and during the Asian financial crisis. Journal of Futures Markets, 20:

13 232 Leo H. Chan Cheung Y-L, Ho Y-K (1991). The intertemporal stability of the relationships between the Asian emerging equity markets and the developed equity markets. Journal of Business Finance and Accounting, 18: Cheung Y-W, Yuen J (2002). Effects of US inflation on Hong Kong and Singapore. CESifo working paper, No. 700, UC-Santa Cruz Chow P (2002). China s Economic Transformation. Malden, MA: Blackwell Publishers Inc. Chowdhury A R (1994). Stock market interdependencies: Evidence from the Asian NIEs. Journal of Macroeconomics, 16: Fung L, Tam C-S, Yu I-W (2008). Assessing financial market integration in Asia equity and bond markets. Available at SSRN: Geweke J (1982). Measurement of linear dependence and feedback between multiple time series. Journal of American Statistical Association, 76: Granger C W J, Huang B N, Yang C W (2000). A bivariate causality between stock prices and exchange rates: Evidence from recent Asian flu. The Quarterly Review of Economics and Finance, 40: Johansson C A (2009). China s financial market integration with the World. CERC Working Paper 10 Ko K-S, Lee S B (1991). A comparative analysis of the daily behavior of stock returns: Japan, the U.S. and the Asian NICs. Journal of Business Finance and Accounting, 18: Lin W-L, Engle R F, Ito T (1994). Do bulls and bears move across borders? International transmission of stock returns and volatility. The Reviewof Financial Studies, 7: Phylaktis K, Ravazzolo F (2002). Measuring financial and economic integration with equity prices in emerging markets. Working paper, City University Business School, UK Tay N S P, Zhen Zhu (2000). Correlations in returns and volatilities in Pacific-Rim stock markets. Open Economics Review, 11: Tian L (2002). Excessive IPO underpricing, the government issuer and the flotation time game. Working paper, Peking University Management School Woo W T (2001). Recent Claims of China s Exceptionalism: Reflections inspired by WTO accession. Working paper, UC-Davis Yu I-W, Fung L, Tam C-S (2007). Assessing financial market integration in Asia equity markets. Working Paper 04/2007, Hong Kong Monetary Authority