1
University of Amsterdam Faculty of Economics and Business
What factors contribute to the price discount of cross-listing
Chinese stocks?
By Wenting Deng
Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of
Bachelor of Science in Economics and Business
2
Statement of Originality
This document is written by Student Wenting Deng who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
1. IntroductionIn recent years, the advantages of lower transaction and information costs, improving corporate governance, overcoming legal restrictions and the ability to enter international markets has led to a dramatic increase in the number of listing companies. At the end of 2013, the number of cross-border listed stocks reached 2963. According to the World Federation of Exchanges, by 2013, the amount of trading non-U.S. stocks in the NYSE increased to 514, counting for 21.98% for the whole market. For Chinese cross-listing companies, the primary choice of listing is in “A+H” (Shanghai and Hong Kong). Chinese investment in Hong Kong remains strong due to the similarities in economy, geography and culture. As the biggest economy with one of the most developed financial systems, the US is the second largest choice for cross listing. At the end of 2012, more than 60 Chinese companies listed both domestically and abroad, 49 of which were listed in Hong Kong, and 11 of which were listed on the NYSE. Recently, the Chinese government announced plans to widen its financial market due to the benefits of solving market segmentation issues, increasing stock liquidity and decreasing information asymmetry seen when establishing cross listing in the Shanghai Stock exchange.
Therefore, an analysis of how Chinese cross-listing firms work has theoretical value and practical relevance.
This paper will focus on what factors contributes to the price premium between domestic capital firms and foreign capital firms. The first section will be the introduction, followed by the theoretical framework. In the theoretical framework section, this paper reviews research about violating the law of one price. The data and properties of the sample will be presented in the third section. The fourth section will discuss the research methods used in this paper, together with the statistical results. A conclusion will be given in the fifth section.
2. Theoretical framework
listing describes a situation in which a stock is listed in two or more stock exchanges. Cross-listing can be called cross boarder Cross-listing, since most cross-Cross-listing firms dually list in domestic and foreign capital markets (Doidge, Karolyi, & Stulz, 2004). The historically studied cross-listing often characterizes the initial stages of capital market integration, in which a country with a relatively underdeveloped capital market allows a small number of domestic securities to be listed on a foreign
4
capital market while limiting its citizens to invest in foreign securities (Alexander, Eun, & Janakiramanan, 1988). Normally, the stock is first listed on the domestic market, then is listed on the foreign capital market. The classical finance paradigm predicts that an asset’s price should be unaffected by its location of trade. If international financial markets are perfectly integrated, then a given set of cash flows has the same value and risk characteristics when it is traded across markets. However, from previous researches (Rosenthal & Young, 1990; Froot & Dabora, 1999; Chan et al., 2003), when the equivalent securities trade in different markets, the law of one price is often violated.
In the early 1970s, scholars studied this phenomenon and explained it using the Capital Asset Pricing Model (CAPM) theory. Under the assumption that investors can freely enter any capital markets without facing exchange rate risk, stocks listed on different capital market have different beta, which leads to different risk premium and further results in price differentiation (Cohn & Pringle, 1973). Doidge et al. (2004) argues that in the CAPM, a firm located in a country that is not fully integrated in the world capital markets typically faces a higher cost of capital because its risk has to be born mostly by investors from its country. If the firm finds a way to make it less costly for foreign investors to hold its shares, these investors share some of the firm’s risk and therefore the cost of capital of the stock falls. From this perspective, cross-listing is a way for firms to make their shares more accessible to foreign investors and will be used only by those firms for which doing so can reduce their cost of capital sufficiently to offset the costs of the listing in foreign market. For example, if two firms have the same expected cash flows but one firm has a U.S. listing and the other does not, one would expect the firm with a U.S. listing to have a higher value than the firm without such a listing because the cost of capital is relative lower (Smith & Sofianos, 1997). Inquiring into this theory, Solnik (1974) argues that the prices of cross-listing stocks are also related to differences in country factors. Following this idea, Bekaert and Harvey (1995) present that since the stock volatility varies among different markets, the stock price should be different as well. Later on, according to Mild segmentation hypothesis raised by Errunza and Losq (1985), it assumes that domestic investors can enter and exit any stock market freely, while foreign investors face a barrier to enter the domestic market. The capital market segmentation includes physical segmentation, like investment barriers, and soft segmentation, such as information asymmetry and liquidity difference. Going into the market segmentation theory, liquidity difference plays an important role. In stock market, liquidity describes how easily shares of stock can be converted to cash without large impacts on the stock price. For liquid stocks, trading volumes can be much bigger, the trading speed can be much faster and the bid-ask spread can be much smaller, sometimes negligible. Illiquid stocks face more risk and higher capital cost, thus leading to a lower price. In this way, domestic investors have lower systemic risk and
5
expected rate, which explains the phenomenon that the foreign capital stock premiums the domestic capital stock would appear when cross-listing in a more developed market. The proposed theory was confirmed by numerous empirical analysis (Abdallah et al., 2011). Except from liquidity difference, information asymmetry is essential to market segmentation theory because foreign investors need to bear higher opportunity cost on information collection when the market is not perfect. Merton (1987) argues that concern about asymmetric information among investors could be an important reason why some investors do not invest at all in certain securities. He proposes a model with incomplete information which elaborates the relationship of shadow cost of incomplete diffusion of information and expected return. Investors’ expected return will increase with the rise of shadow cost. Froot and Dabora (1999) find several explanations of this phenomenon includes the country-specific sentiment shocks and whether the investors are rational or not. Later, along with the gradual development and maturation of behavioral finance, some scholars also attempted to propose their explanation from the perspective of behavioral finance. Scholars put forward ‘noise trader theory’. A noise trader has the following traits: overconfidence -people’s judgements is reliably greater than the objective accuracy of those judgements and they owe the success to their ability, but they blame their failure on bad luck; disposition effect – investors are anxious to sell profitable stock, while they are more willing to hold defective stock; underreaction and overreaction – the former one moves stock price to a certain direction and later one causes the long-term turnover; herding behavior - the tendency for individuals to mimic the actions of a larger group (Grinblatt & Keloharju, 2001). Based on the above-mentioned theory, behavior finance tries to explain the price difference of cross-listing stocks. Unfortunately, they haven’t built the corresponding econometric theory up to now.
When it comes to Chinese cross-listing firm, a similar phenomenon is expected to appear in the cross-listing Chinese companies. Nevertheless, some analysts found that the price of stock in Shanghai Exchange is higher than that in HK (the H share) or in US (the ADR). ADR is equivalent to a specified number of shares of a security trading in its home market. In theory, the price of an ADR or a H share should reflect the value of the underlying security in the Chinese market and the exchange rate between the RMB and the US dollar or HK dollar (Smith & Sofianos, 1997). In practice, the price of ADR and that of H share deviate from parity. More specifically, domestic capital stock premiums the foreign capital stock (Arquette, Brown, & Burdekin, 2008). The abnormal phenomenon can be explained by different reason. Market sentiment appears to have a significant influence on the discount attached to Chinese cross-listing firms (Eun & Sabherwal, 2003). Based on the model with incomplete information, since the shadow cost of foreign investors is higher than that of domestic investors, the expected return
6
of foreign investors will be higher and the stock price will be lower (Merton, 1987). From the liquidity view, domestics Chinese market is more active than foreign markets (i.e US market and Hong Kong market) because of its strong speculative character. Expected exchange rate fluctuation is also a key factor that affects the discounting (Arquette et al., 2008).
From an empirical perspective, the Chinese market is of particular interest. First, different from the typical cross-listing, almost all companies mentioned above went public aboard first and went back to the domestic Chinese market later. The immediate causes of going back to the Chinese market include the purpose of raising more capital, the barrier of refinancing in the foreign market and the government regulation. With the sustained and rapid development of Chinese economy, foreign listed firms seek to enlarge their business scale, which creates a huge demand for capital funding (Kutan & Zhou, 2006). From research, the price/earning ratio can reach 40 in the domestic Chinese market, while in HK or US, the P/E ratio is normally 15. When the earing holds constant, higher P/E ratio means that firms can issue stock at a higher price, raising more funds. Furthermore, it is easier to refinance in the domestic market because the regulation is less strict. A negative reason that triggers the returning of foreign-listed firms is the private benefit of control. The large shareholders sometimes take value-destroying transactions which have a negative impact on the firm value and small shareholders. In order to prevent the impact on the small shareholders from the realization of these benefits of large shareholders, there is usually a policy in transaction area for protection of the small shareholders (La Porta et al., 2000). As mentioned before, most cross-listed firms are large national owed enterprise. Large shareholders make up most of the management board (Jefferson & Su, 2006). Thus, the organization structure in such firm is hard to assure that management will act in the interest of the firm value. Also, a policy aimed at preventing such behavior was not fully established in China at that moment, providing large shareholders with chance and incentives to seize private benefit of control. (Doidge, Karolyi, Lins, Miller, & Stulz, 2009). Second, China is an important emerging market and a study about the abnormal phenomenon in China is worth attention. Meanwhile, most of the Chinese firms listed on the foreign capital market have an important influence on the national economy. They account for big proportion in the Chinese stock market as for the market value and volume (Su & Chong, 2007). Third, China just declared to widen open its financial market to become more integrated with the international market. Consequently, an analysis provides a way to better understand the effects of international competition and it has practical significance to study the price differentiation in domestic market and foreign market.
7
3. Data and properties of the sampleThe sample used to investigate the research question comprises 9 US-listed companies and 9 Hong Kong-listed companies, both with data available on DataStream, Shanghai Stock Exchange and WRDS. The reason why the sample size is relatively smaller than most researches is that cross-listing is not very common no matter in which countries. This paper has used all the stocks available from the database. Data will be gained on a monthly basis. The time range runs from January 2003 to December 2016. The time period ends in the December of 2016 because the data for the year 2017 is currently not accessible. One thing that needs to be addressed is that for the domestic Chinese stock market, this study will focus on the more liquid A-shares that are sold to domestic investors rather than the B-shares that were originally available only to foreigners. Because of the particularity of Financial companies, this paper will exclude them from the sample (Walled et al., 2015).
4. The model and the result.
The price discount will be defined in the following way; 𝑑𝑖𝑠𝑐𝑜𝑢𝑛𝑡𝑖,𝑡=
𝑝𝑖,𝑡𝑐×𝑒𝑖,𝑡𝑐
𝑝𝑖,𝑡𝐴 − 1 (1)
In the formula, 𝑑𝑖𝑠𝑐𝑜𝑢𝑛𝑡𝑖,𝑡 refers to the discount ratio (company 𝑖 in trading period 𝑡) of the price
in cross-listing market to that in the domestic Chinese market (A-share). 𝑒𝑖,𝑡𝑐 is the exchange rate in that
period. If the 𝑑𝑖𝑠𝑐𝑜𝑢𝑛𝑡𝑖,𝑡< 0, it means that domestic stock price premiums foreign stock price. The
more negative the 𝑑𝑖𝑠𝑐𝑜𝑢𝑛𝑡𝑖,𝑡，the bigger the price differentiation is.
As for the regression model, this paper will first consider the market sentiment. First, from previous research, the stock price in different markets can vary with its correspondence market index. If the index of Shanghai Exchange is stronger than S&P 500 or Hang Seng index (i.e. HK stock exchange index), it is reasonable for investors to demand domestic A share more. Together with the fact that investors in Shanghai Stock Exchange are less rational, the domestics price for cross-listing frim tends to be higher, which causes a price discount of ADR or H-share (Miller, 1999). This paper will use log𝑝𝑒 𝑟𝑎𝑡𝑖𝑜 𝑜𝑓 𝑆𝐻 𝑖𝑛𝑑𝑒𝑥
𝑝
𝑒 𝑟𝑎𝑡𝑖𝑜 𝑜𝑓 𝑆&𝑃 500
to measure the market difference (MI.n) between NYSE and SHSE. SH index is the market index in SHSE. The market difference (MI.h) between HKSE and SHSE will be measured by
8
log 𝑝𝑒 𝑟𝑎𝑡𝑖𝑜 𝑜𝑓 𝑆𝐻 𝑖𝑛𝑑𝑒𝑥𝑝
𝑒 𝑟𝑎𝑡𝑖𝑜 𝑜𝑓 𝐻𝑎𝑛𝑔 𝑆𝑒𝑛𝑔 𝑖𝑛𝑑𝑒𝑥
(Arquette et al., 2008). For example, if the Shanghai price-earnings ratio is higher than that in the United States then this should indicate more positive market sentiment in mainland China and the ADR should sell at a lower price than its Shanghai counterpart, resulting in a more negative ADR price discount. Thus, a negative relationship between market difference (MI) and the discount ratio will be hypothesized. Chart 1 and 2 provide some details on the P/E ratio in different markets. Chart 1 depicts the relative difference in P/E ratio between US market and Shanghai market. The P/E ratio of S&P500 was significantly high from May 2009 to December 2009, reaching 123 in the end of 2009. The best explanation for that is the reported earnings were so low. During the financial crisis, the banks were writing down all of the bad debt associated with the mortgage-backed securities that had lost so much value. This meant that the banks were reporting negative earnings. Since the financial sector is a large part of the S&P500, this alone had an enormous effect on the P/E ratio. Chart 2 reveals that both Shanghai and Hong Kong experienced a decrease in P/E ratio in 2008, followed by fluctuations. While the Shanghai index normally has a higher P/E ratio, the spread between these two ratios steadily declines from 30 in 2008 to almost 0 in 2013 and 2014. In 2015, the P/E ratio trended upward and the spread widens a little bit. The second factor used in this model is the risk-free interest rate (RRR). Risk-free rate level measures the opportunity cost of buying one stock. If the risk-free rate in the domestic market is higher than that in the foreign market, the capital cost of purchasing domestic stock will be higher, decreasing the demand for it and lowering the price difference. log 𝑟𝑖𝑠𝑘 𝑓𝑟𝑒𝑒 𝑟𝑎𝑡𝑒 𝑡𝑆𝐻
𝑟𝑖𝑠𝑘 𝑓𝑟𝑒𝑒 𝑟𝑎𝑡𝑒 𝑡𝑓𝑜𝑟𝑒𝑖𝑔𝑛 will be used to
measure the relative risk-free rate in the model. The higher this factor, the more negative the discount ratio (Eun & Janakiramanan, 1986). Third, in the foreign listing market, investors can choose various forms of capital products, while in the domestic Chinese market, A share is the only investment option for most individual investors. Meanwhile, individual investors account for more than 50 % of the market (Su & Chong, 2007). These two reasons make Shanghai Stock Exchange have higher liquidity level. Based on previous studies, if Shanghai Stock Exchange has higher liquidity level, the price differentiation between domestic Chinese stock and foreign stock will be bigger. Relative turnover ratio (log𝑇𝑅𝑖,𝑡𝐴
𝑇𝑅𝑖,𝑡𝐶) will be used to access the liquidity level of each market. In the model, RTR refers to the relative
turnover ratio. 𝑇𝑅𝑖,𝑡𝐴 is the turnover ratio of stock 𝑖 in trading period 𝑡. Similarly, 𝑇𝑅𝑖,𝑡𝐶 refers to the
turnover ratio of the same stock in its foreign listed market (Levine & Schmukler, 2006). Arquette et al. (2008) states that the tendency for foreign share prices to remain substantially lower than A shares of the
9
same company reflected not just liquidity issues but is also consistent with the fact that mainland Chinese investors being ‘risk-lovers’. But so far, this factor is hard to quantify and thus will not be used in the regression model. The above mentioned four variables can measure the overall market sentiment.
Chart 1 Chart 2 0 20 40 60 80 100 120 140 1/ 1/ 08 5/ 1/ 08 9/ 1/ 08 1/ 1/ 09 5/ 1/ 09 9/ 1/ 09 1/ 1/ 10 5/ 1/ 10 9/ 1/ 10 1/ 1/ 11 5/ 1/ 11 9/ 1/ 11 1/ 1/ 12 5/ 1/ 12 9/ 1/ 12 1/ 1/ 13 5/ 1/ 13 9/ 1/ 13 1/ 1/ 14 5/ 1/ 14 9/ 1/ 14 1/ 1/ 15 5/ 1/ 15 9/ 1/ 15 1/ 1/ 16 5/ 1/ 16 9/ 1/ 16
P/E ratio for S&P 500 and Shanghai A share index
PE.S&P PE.SE 0 10 20 30 40 50 60 1/ 1/ 08 5/ 1/ 08 9/ 1/ 08 1/ 1/ 09 5/ 1/ 09 9/ 1/ 09 1/ 1/ 10 5/ 1/ 10 9/ 1/ 10 1/ 1/ 11 5/ 1/ 11 9/ 1/ 11 1/ 1/ 12 5/ 1/ 12 9/ 1/ 12 1/ 1/ 13 5/ 1/ 13 9/ 1/ 13 1/ 1/ 14 5/ 1/ 14 9/ 1/ 14 1/ 1/ 15 5/ 1/ 15 9/ 1/ 15 1/ 1/ 16 5/ 1/ 16 9/ 1/ 16
P/E ratio for Hang Seng index and Shanghai A share index
10
There is not only substantial variation across markets in the discount, but also difference across companies. The data in table1 shows that the actual discount ranging from -64% to 115%. Such wide difference suggests that it is likely there are some company-specific differences in sentiment
as well as overall market sentiment effects. The company sentiment (CS) is measured by log( 𝑃/𝐸 𝑟𝑎𝑡𝑖𝑜 𝑡𝑐𝑜𝑚𝑝𝑎𝑛𝑦 𝑖
𝑃/𝐸 𝑟𝑎𝑡𝑖𝑜 𝑡𝑆ℎ𝑎𝑛𝑔ℎ𝑎𝑖 𝑖𝑛𝑑𝑒𝑥), which is the natural logarithm of P/E ratio of the company divided by the
natural logarithm of P/E ratio of the Shanghai market index. If the company’s P/E ratio is higher than that of the market index, it indicates a high local Chinese sentiment (Su & Chong, 2007). Besides the P/E ratio, this paper also uses market capitalization in billions (MC.b) as a variable. The larger firm is expected to have better information exposure and lower barriers to trade, resulting in the discount ratio narrowing to zero. Even without the participation of the market sentiment and company sentiment, the discount can still be different because of the presence of exchange rate (Grammig et al., 2005). Hence, this paper includes expected exchange rate change into the model.
Thus, the regression model is built in the following way:
𝑝𝑟𝑖𝑐𝑒 𝑑𝑖𝑠𝑐𝑜𝑢𝑛𝑡𝑖,𝑡= 𝛼0+ 𝛽1∗ ER + 𝛽2∗ MI + 𝛽3∗ RRR + 𝛽4∗ RTR + 𝛽5∗CS+ 𝛽6∗ MC. b + 𝜀
During the primary investigation, this research finds that most cross-listing firms are from manufacture and transportation industry. Among firms analyzed, 66% of the firms are from manufacturing industry, 22% percent of the firms are in the transportation industry, the rest of them are from communication industry (Chart 3). More importantly, all the manufacture firms are in either petroleum industry or mineral industry. This is because that China is a manufacturing powerhouse. It has the second largest GDP right now. Moreover, because of some historical reasons, many national-owned large-scale enterprises have long experienced low efficiency. To build a modern enterprise system, national-owned enterprises need to adjust their stock structures (Jefferson & Su, 2006). Without a developed domestic capital market, many Chinese national owned enterprises chose to list in foreign market. Later on, with the development of Chinese capital market, the government provided support to attract those firms to return to China. Those firms play important roles in the Chinese stock market and have a big market share in their industry (Chen, 2004).
From the formula (1), this paper calculates the monthly price discount for these firms and then get the average price discount for each firm. In table 1, all the cross-listing firms, except China United Network Communications, have a discounted price in the foreign market. Comparing with the price discount for HK listed stock, that of US-listed stock is slightly bigger, but that difference is not significant.
11
In other words, the stock price in HK almost equals to that in the US. Thus, the law of one price is not violated between ADR and H-share.
Chart 3
Table 1
Company name Industry PD(US) PD(HK)
Aluminum Corporation of China Limited manufacture -0.4684 -0.4674 China Huaneng Group Corporation manufacture -0.2436 -0.2432 China Petroleum & Chemical CorAluminum
Corporation of China Limited
manufacture -0.1928 -0.1919
Sinopec Shanghai Petrochemical Company Limited
manufacture -0.6397 -0.6388
Yanzhou Coal Mining Company manufacture -0.4446 -0.4440
PetroChina Company Limited manufacture -0.2744 -0.2739
China Eastern Airlines Corporation Limited transportation -0.4807 -0.4805 China Southern Airlines Company Limited transportation -0.4356 -0.4347 China United Network Communications
Group Co., Ltd. communication 1.0098 1.0105 manufacturing industry 67% transportation industry 22% other industry 11%
Industry composition
12
Table 2 displays the average price discount in the order of time. Overall, the discount ratio was negative except in 2014. That is to say that the domestics stock price usually premiums its foreign stock price. The price discount ratio continued to approach zero after 2009, reaching a positive ratio in 2014. After that, the ratio dropped from almost 30% to -30% and fluctuated around this level. To investigate why the price discount moves that way, big events happened in that period need to be mentioned. In the beginning of 2008, China experienced a downturn in its stock market and stock price decreased dramatically, which narrowed the price difference between domestic stock and foreign stock. In late 2008, subprime mortgage crisis in America went into a full-blown global financial crisis. Stock markets collapsed and stock prices plummeted. Different from other major capital markets, Chinese capital market ended with moderate losses because it was not a fully opened market and the government could intervene the market relatively easily (Liu, Uchida, & Yang, 2012). Following the financial crisis in 2008, the world economy started to recover in the following years. However, the Shanghai Stock Exchange still experienced a bear market until late 2014. From late 2014 to June 2015, the Shanghai stock exchange rose by almost 60%. After the intervention of China Securities Regulatory Commission on 13th June 2015 to slow down the irrational speculation, the Shanghai stock market began to turn into the downside. By the end of August, the Shanghai stock index dropped by 45%. That can partly explain why 2015 witnessed a drastic increase in price discount.
Table 2
Year Average discount ratio (US) Average discount ratio (HK)
2008 -0.3846 -0.3850 2009 -0.4490 -0.4503 2010 -0.3413 -0.3424 2011 -0.2207 -0.2177 2012 -0.1416 -0.1426 2013 -0.0271 -0.0255 2014 0.0295 0.0329 2015 -0.3417 -0.3413 2016 -0.2934 -0.2921
13
Chart 4
Except these nine firms, this paper also looks into other financial cross-listing firms. However, none of the traditional financial institutions, for example, banks, has cross-listed on the US market. All the financial firms listed in US capital market are a new type of financial institutions, which focus on online third-party payment platform and none of which has returned to Chinese capital yet. Meanwhile, previous studies show that financial firms usually have different price pattern. Therefore, no research will be performed for the Chinese cross-listed companies in the finance industry in this paper.
Table 3 presents the result of employing the model to explain the price differential in Chinese stock and US stock market. The negative coefficients suggest that those variables have the impact of making the discount more negative. In all columns, the exchange rate variable remains statistically and economically significant as adding additional explanatory variables. Its negative coefficient match with the anticipation. Columns (2) -(4) allow for analyzing market sentiment effect in addition to the exchange rate. Surprisingly, the market index does not play a considerable role in price differential of US-listed stock, which deviates from the initial hypothesis. Column (6) shows the estimates after adding all variables. Except that RRRU variable become insignificant and the coefficient of all variables becomes smaller, the findings are similar to before.
A detailed analysis of the full specification (table 3 column (6)) is presented as follow:
-0.5000 -0.4000 -0.3000 -0.2000 -0.1000 0.0000 0.1000 2008 2009 2010 2011 2012 2013 2014 2015 2016
average discout ratio
Average discount ratio (US) Average discount ratio (HK)
14
1. Exchange rate (ERUS): Significant on 1% confidence level, the coefficient of ERUS is -0.358%. It indicates that every 1 point change in the exchange will change the ADR discount by -0.358%, which is consistent with the expectation that change in the expected exchange rate can explain a large portion of the deviation in share price across international market (Arquette et al., 2008).
2. The relative P/E ratio of the market index (MI.n): Different from previous studies, in this model, the market index variable is not statistically significant.
3. The relative risk-free rate (RRRU): After adding company sentiment variables, the RRRU becomes insignificant.
4. The relative turnover ratio (RTRn): the coefficient is -0.0177 and is significant at 1% confidence level. It suggested that every 1% of increasing the relative turnover ratio will lead 0.12 percent of an increase in the price discount (make the price discount more negative). It verifies the liquidity theory: because investors do not have various investment option, which makes A- share the only option for most investors, Shanghai stock market is more active than US stock market.
5. Company sentiment (CS): the coefficient of CS is both economically and statistically significant at 1% confidence level, which is consistent with the hypothesis.
6. Market capitalization in billions (MC.b): whereas the market cap variable is statistically significant, it has a very small coefficient, implying that market cap only explains a very small proportion of the price differential. Still, a positive coefficient indicates that larger companies have better information exposure, confirming the hypothesis.
Table 3
Regression of US-listed stock discount Dependent variable: ADR price discount (PDn)
(1) (2) (3) (4) (5) (6) PDn PDn PDn PDn PDn PDn ERUS -0.414*** -0.419*** -0.512*** -0.452*** -0.421*** -0.358*** (-7.73) (-7.77) (-7.50) (-6.74) (-6.34) (-14.42) MI.n 0.0283 0.00162 -0.0149 -0.0447 -0.0160 (0.85) (0.05) (-0.43) (-1.30) (-1.24) RRRU -0.0335* -0.0612*** -0.0609*** -0.00821
15
(-2.20) (-4.03) (-4.06) (-1.44) RTRn 0.0750*** 0.0795*** -0.0177*** (7.62) (8.16) (-3.39) CS -0.0532*** -0.0388*** (-5.13) (-9.95) MC.b 0.0000627** * (4.54) _cons 2.456*** 2.497*** 3.092*** 2.640*** 2.474*** 1.948*** (7.03) (7.08) (6.97) (6.05) (5.74) (12.05) N 972 972 972 971 969 861 t statistics in parentheses * p < 0.05, ** p < 0.01, *** p < 0.001In table 4, a similar model is used to quantify the price difference for stocks that are cross-listed on Shanghai and Hong Kong market. Similar to the ADR model, the exchange rate variable remains statistically and economically significant in all columns, but when adding other variables into the model, the coefficient drops sharply from -3.337 to -1.153. Columns (2) -(4) allow for analyzing market sentiment effect in addition to the exchange rate. In the H-share discount model, as suggested in previous research, differences in market index explain the price differentials across the two markets. Column (6) shows the estimates after adding all variables. Except relative turnover ratio, all other coefficients remain statistically significant as predicted.
A detailed analysis of the full specification (table 4 column (6)) will be presented as follow: 1. Exchange rate (ERHK): Significant on 1% confidence level, the coefficient of ERUS is –1.153. It indicates that every 1 point change in the exchange will change the ADR discount by -1.153. Comparing with the ADR model, the China-HK exchange rate matters more to HK market.
2. The relative P/E ratio of the market index (MI.h): Unlike the ADR model, in which the relative P/E ratio of the two markets is not correlated with the price discount, the coefficient in HK model is
16
significant at 1% level. It suggests that when Shanghai stock market goes up relative to Hong Kong market, the price discount will go more negative.
3. The relative risk-free interest rate (RRRH): Consistent with the previous prediction, the coefficient remains significant at 1% level and reveasl a positive relationship between relative risk-free rate and the price discount. If the risk-free rate is higher in the Chinese market, domestics investors face a higher opportunity cost to purchase stocks in Shanghai market, lowering the demand and thus stock price.
4. The relative turnover ratio (RTRh): the coefficient is not significant for the RTRh.
5. Company sentiment (CS): similar to ADR models, the coefficient of CS is both economically and statistically significant at 1% confidence level, which is consistent with the prediction.
6. Market capitalization in billions (MC.b): the coefficient shows a similar relationship as in ADR model.
Table 4
Regression of HK-listed stock discount
Dependent variable: H-share price discount (PDh)
(1) (2) (3) (4) (5) (6) PDh PDh PDh PDh PDh PDh ERHK -3.347*** -1.986*** -1.966** -0.621 -0.173 -1.153*** (-7.90) (-3.81) (-3.16) (-1.01) (-0.28) (-5.09) MIh -0.359*** -0.359*** -0.561*** -0.582*** -0.264*** (-4.40) (-4.38) (-6.91) (-7.26) (-8.99) RRRh 0.000787 -0.00917 -0.00186 0.0131** (0.06) (-0.74) (-0.15) (2.93) PTRh 0.133*** 0.138*** -0.00491 (9.41) (9.94) (-0.83) CS -0.0544*** -0.0397*** (-5.44) (-10.92)
17
MC.b 0.0000879* ** (8.40) _cons 2.566*** 1.542*** 1.525** 0.458 0.136 0.656*** (7.22) (3.65) (2.99) (0.91) (0.27) (3.56) N 972 972 972 971 969 861 t statistics in parentheses * p < 0.05, ** p < 0.01, *** p < 0.001 5. ConclusionFrom the study, even excluding the exchange rate fluctuation, there still exists arbitrage opportunity between Shanghai A share and foreign shares. Theoretically, investors can benefit from conversion ADRs or H-shares into A shares. However, even Shanghai A share and Hong Kong H share represent the same ownership stake in a company, the shares are not convertible, not to mention the ADRs. One alternative strategy is shorting the Chinese security and being long the ADRs or H-shares for most of the firms in the sample. This strategy was unachievable before 2008 because China did not allow securities lending before 2008. This also explains why the price discount is generally smaller after 2008. The behavior of arbitrager is crossing out the arbitrage opportunity.
To conclude, based on the study above, the difference in market sentiment, the fluctuation of the expected exchange rate, the company sentiment and the liquidity difference and information asymmetry caused by segmentation jointly contributes to the price differential across markets. However, there are still many other factors that can influence the price discount. In the future research, stock volatility, dividend and so forth can be included to build a more comprehensive model.
18
ReferenceAbdallah, A. A. N., Abdallah, W., & Saad, M. (2011). The Effect of Cross‐Listing on Trading Volume: Reducing Segmentation Versus Signaling Investor Protection. Journal of Financial
Research, 34(4), 589-616
Alexander, G. J., Eun, C. S., & Janakiramanan, S. (1988). International listings and stock returns: Some empirical evidence. Journal of financial and quantitative analysis, 23(2), 135-151.
Arquette, G. C., Brown, W. O., & Burdekin, R. C. (2008). US ADR and Hong Kong H-share discounts of Shanghai-listed firms. Journal of Banking & Finance, 32(9), 1916-1927.
Bekaert, G., & Harvey, C. R. (1995). Time‐varying world market integration. The Journal of Finance, 50(2), 403-444.
Busaba, W. Y., Guo, L., Sun, Z., & Yu, T. (2015). The dark side of cross-listing: A new perspective from China. Journal of Banking & Finance, 57, 1-16.
Chang, E. C., Cheng, J. W., & Khorana, A. (2000). An examination of herd behavior in equity markets: An international perspective. Journal of Banking & Finance, 24(10), 1651-1679.
Chen, J. J. (2004). Determinants of capital structure of Chinese-listed companies. Journal of Business research, 57(12), 1341-1351.
Cohn, R. A., & Pringle, J. J. (1973). Imperfections in international financial markets: Implications for risk premia and the cost of capital to firms. The Journal of Finance, 28(1), 59-66.
Doidge, C., Karolyi, G. A., & Stulz, R. M. (2004). Why are foreign firms listed in the US worth more?. Journal of financial economics, 71(2), 205-238.
Doidge, C., Karolyi, G. A., Lins, K. V., Miller, D. P., & Stulz, R. M. (2009). Private benefits of control, ownership, and the cross‐listing decision. The Journal of Finance, 64(1), 425-466. Domowitz, I., Glen, J., & Madhavan, A. (1998). International Cross‐Listing and Order Flow Migration:
Evidence from an Emerging Market. The Journal of Finance, 53(6), 2001-2027.
Errunza, V., & Losq, E. (1985). International asset pricing under mild segmentation: Theory and test. The Journal of Finance, 40(1), 105-124.
Eun, C. S., & Janakiramanan, S. (1986). A model of international asset pricing with a constraint on the foreign equity ownership. The Journal of Finance, 41(4), 897-914.
Eun, C. S., & Sabherwal, S. (2003). Cross‐border listings and price discovery: Evidence from US‐ Listed Canadian stocks. The Journal of Finance, 58(2), 549-575.
19
Froot, K. A., & Dabora, E. M. (1999). How are stock prices affected by the location of trade?. Journal of financial economics, 53(2), 189-216.
Grammig, J., Melvin, M., & Schlag, C. (2005). Internationally cross-listed stock prices during overlapping trading hours: price discovery and exchange rate effects. Journal of Empirical Finance, 12(1), 139-164.
Grinblatt, M., & Keloharju, M. (2001). How distance, language, and culture influence stockholdings and trades. The Journal of Finance, 56(3), 1053-1073.
Harris, L. (1990). Liquidity, trading rules and electronic trading systems (No. 91-8).
Jefferson, G. H., & Su, J. (2006). Privatization and restructuring in China: Evidence from shareholding ownership, 1995–2001. Journal of Comparative Economics, 34(1), 146-166.
Kutan, A. M., & Zhou, H. (2006). Determinants of returns and volatility of Chinese ADRs at NYSE. Journal of Multinational Financial Management, 16(1), 1-15.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (2000). Investor protection and corporate governance. Journal of financial economics, 58(1), 3-27.
Levine, R., & Schmukler, S. L. (2006). Internationalization and stock market liquidity. Review of Finance, 10(1), 153-187.
Li, H. (2007). International linkages of the Chinese stock exchanges: A multivariate GARCH analysis. Applied Financial Economics, 17(4), 285-297.
Liu, C., Uchida, K., & Yang, Y. (2012). Corporate governance and firm value during the global financial crisis: Evidence from China. International Review of Financial Analysis, 21, 70-80. Miller, D. P. (1999). The market reaction to international cross-listings:: evidence from Depositary
Receipts1This paper is based on Chapter 2 of my doctoral thesis at the University of California, Irvine. I am indebted to Philippe Jorion for guidance throughout this project. I would like to thank Geert Bekaert, John Bizjak, Vihang Errunza, Nai-Fu Chen, Charles Cuny, Roberto Gutierrez, Robert Haugen, Neal Stoughton, Eli Talmor, and anonymous referee, the editor, Jerold Warner, and seminar participants at the Berkeley Program .... Journal of Financial economics, 51(1), 103-123.
Merton, R. C. (1987). A simple model of capital market equilibrium with incomplete information. The journal of finance, 42(3), 483-510.
Poon, W. P., Firth, M., & Fung, H. G. (1998). Asset pricing in segmented capital markets: Preliminary evidence from China-domiciled companies. Pacific-Basin Finance Journal, 6(3), 307-319.
20
Smith, K., & Sofianos, G. (1997). The impact of an NYSE listing on the global trading of non-US stocks. New York Stock Exchange.
Solnik, B. H. (1974). An equilibrium model of the international capital market. Journal of economic theory, 8(4), 500-524.
Su, Q., & Chong, T. T. L. (2007). Determining the contributions to price discovery for Chinese cross-listed stocks. Pacific-Basin Finance Journal, 15(2), 140-153.
