Chapter 15 CHINESE A AND B SHARES YAN HE, Indiana University Southeast, USA Abstract A and B shares exist in the Chinese stock markets. A shareholders are domestic investors and B share- holders are foreign investors. During the early-and mid-1990s, B shares were traded at a discount rela- tive to A shares, and B-share returns were higher than A-share returns. It is found that B-share mar- ket has persistent higher bid-ask spreads than the A-share market and traders in the B-share market bear higher informed trading and other transaction costs. In addition, the higher volatility of B-share returns can be attributed to the higher market mak- ing costs in the B-share market. Keywords: Chinese A shares; Chinese B shares; stock; return; volatility; asymmetric information; bid-ask spread; transaction costs; stock ownership; Shanghai stock exchange; Shenzhen stock ex- change The development of equity markets in China started in early 1990s. Open for business in Decem- ber 1990, the Shanghai Stock Exchange (SHSE) and the Shenzhen Stock Exchange (SZSE) are the two major securities exchanges in China. By 1998, the SHSE had raised a total of RMB140.814 bil- lion for listed companies and the SZSE had raised a total of RMB 128 billion for listed companies. The two exchanges played an important role in promoting the restructuring of state-owned enter- prises. Stock shares in China are divided into two broad categories: untradable and tradable. By the end of 1998, the total untradable equity of the listed companies was 166.484 billion shares (i.e. 65.89 percent of the total equity of the listed companies), allocated as follows: (1) shares owned by government, 86.551 billion; (2) shares owned by legal persons, 71.617 billion; (3) shares owned by employees and others, 8.317 billion. Outstand- ing tradable shares totaled 86.193 billion shares (i.e. 34.11 percent of the total equity of the listed companies), allocated as follows: (1) Class A shares, 60.803 billion; (2) Class B shares, 13.395 billion; and (3) Class H shares, 11.995 billion. Class A shares are owned by domestic investors and traded in the domestic markets. Class B shares are owned by foreign investors but traded in the domestic markets. Class H shares are listed on the Hong Kong Stock Exchange. China has tightly restricted foreign stock own- ership throughout the 1990s. The ownership re- striction creates two distinct groups of investors: the domestic and foreign investors. Class A shares are domestic shares and class B shares are foreign shares. In 1991, the Shanghai Stock Exchange (SHSE) and Shenzhen Stock Exchange (SZSE) began to offer B shares, providing foreign investors with a legal channel to invest in China’s equity markets. B shares are also known as Renminbi Special Shares. B shares are issued in the form of registered shares and they carry a face value de- nominated in Renminbi. B shares are subscribed and traded in foreign currencies, but they are listed and traded in securities exchanges inside China. The B share market has attracted a consid- erable amount of foreign investors. The Market provides an additional channel for foreign capital to invest in China. Since March 2001, China has opened its B-share market – previously reserved for overseas investors – to Chinese individuals with foreign currency de- posits. However, the impact of the opening up of the B-share market to Chinese is limited, because that market is small compared to the number of Chinese people and institutions’ foreign currency holdings. Despite the rising foreign currency de- posits in China, Chinese people who have foreign currency holdings still account for a very small proportion of investors. Tables 15.1 to 15.3 are obtained from the China Securities Regulatory Committee. During the early- and mid-1990s, B shares were traded at a discount relative to A shares, and B-share returns were higher than A-share returns. Su (1999) explains the return premiums on the foreign-owned B shares in the Chinese stock mar- kets by testing a one-period capital asset-price model (CAPM). He concludes that foreign inves- tors are more risk-averse than domestic investors. Sun and Tong (2000) explain the price discount of the B shares by differential demand elasticity. They document that when more H shares and red chips are listed in Hong Kong, the B-share discount becomes larger. In addition, Chui and Kwok (1998) show that the returns on B shares lead the returns on A shares, which induces an asymmetric positive cross-autocorrelation between the returns on B and A shares. They argue that A- and B-share investors have different access to information, and information often reaches the B-share market be- fore it reaches the A-share market. The Chinese stock markets have grown very rapidly during the late 1990s and early 2000s. A number of studies investigate the return and risk in the newly developed markets. For example, Lee et al. (2001) examine time-series features of stock returns and volatility, as well as the relation between return and volatility in four of China’s stock exchanges. On the one hand, test results provide strong evidence of time-varying volatility and show volatility is highly persistent and predict- able. On the other hand, the results do not show any relation between expected returns and expected risk. Table 15.1. Trading summary of A and B shares during 11=2000–11=2001 A or B share listed Total market capitalization Stock turnover Stock trading volume No. of transactions No. of companies (100 000 000 Yuan) (100 000 000 Yuan) (100 000 000) (10 000) 2000=11 1063 46061.78 5012.27 365.02 5013 2000=12 1088 48090.94 3737.6 271.35 3719 2001=01 1100 48497.99 3013.63 220.08 3082 2001=02 1110 46228.75 1950.05 151.92 2197 2001=03 1122 50908.44 5095.17 488.33 4335 2001=04 1123 51006.9 5395.87 422.43 4720 2001=05 1129 53205.49 4452.16 328.33 3739 2001=06 1137 53630.58 4917.12 355.5 4449 2001=07 1140 46440.83 3100.68 228.25 2983 2001=08 1151 48054.63 2490.85 221.31 2507 2001=09 1154 45831.36 1766.64 154.67 1858 2001=10 1152 43742.14 1951.5 181.03 1914 2001=11 1153 45431.59 2092.26 200.31 2374 436 ENCYCLOPEDIA OF FINANCE The development in the Chinese markets may affect the risk and return of A- and B-share classes. He and Wu (2003) provide two interesting find- ings: (1) the daily returns of domestic shares (A shares) and foreign shares (B shares) were almost identical in the late 1990s, while the B-share re- turns were much higher than the A-share returns during the mid-1990s; (2) the volatility of B-share daily returns was higher than that of A shares, while previous studies have often documented higher return volatility for A shares. (For example, Su and Fleisher (1999) report that A shares have higher volatility than B shares based on the data of mid-1990s.) Since A and B shares are entitled to the same cash flows of a firm and have similar returns, the higher return volatility of B shares is puzzling. The market microstructure theory suggests that both volatility and bid-ask spreads are positively related to asymmetric information (see Kyle, 1985; Easley et al., 1996). According to this theory, higher vola- tility is caused by higher degree of information asymmetry and participation rate of informed traders in the market, which, in turn, lead to higher trading costs. Thus, the higher volatility of B shares may be due to a more severe asymmetric information problem in the B-share market. If so, we should observe higher trading costs for B shares. Furthermore, Easley et al. (1996) show that spreads and volatility are negatively related to liquidity. Since the order processing cost is the cost of providing liquidity and immediacy, lower liquidity results in higher order processing cost and higher volatility. A recent study by Green et al. (2000) on the London Stock Exchange shows that changes in transaction costs have a significant effect on share price volatility. More- over, Chordia et al. (2002) document that return volatility is significantly related to quoted spreads. These findings confirm the theoretical prediction that volatility and trading costs are positively correlated. Therefore, the higher volatility in the B-share market may reflect higher idiosyncratic risk (rather Table 15.2. A and B shares offering (1987–1998) 87 88 89 90 91 92 93 94 95 96 97 98 Total Shares issued (100MM) 10 25 7 4 5 21 96 91 32 86 268 102 746 A share 10 25 7 4 5 10 43 11 5 38 106 79 343 H share 40 70 15 32 137 13 307 B share 11 13 10 11 16 25 10 96 Capital raised (RMB 100MM) 10 25 7 4 94 375 327 150 425 1,294 837 3,553 A share 10 25 7 4 50 195 50 23 224 655 440 1,687 H share 61 89 31 84 360 38 763 B share 44 38 38 33 47 81 26 307 Rights offering of A and B shares 82 50 63 70 198 335 797 Table 15.3. Number of listed companies (1990–1998) Companies 1990 1991 1992 1993 1994 1995 1996 1997 1998 Issuing A share 10 14 35 140 227 242 431 627 727 Issuing B share 0 0 0 6 4 12 16 25 26 Issuing A and B shares 0 0 18 34 54 58 69 76 80 Issuing A and H shares 0 0 0 3 6 11 14 17 18 Total 10 14 53 183 291 323 530 745 851 CHINESE A AND B SHARES 437 than higher systematic risk) of B-share stocks. The trading risk associated with asymmetric informa- tion can be diversified away and therefore it is not systematic risk (see Chordia et al. 2001). Asset- pricing models (e.g. CAPM and APT) suggest that expected returns should be determined by systematic risk. Since higher volatility does not necessarily imply higher systematic risk, it may not be accompanied with higher returns. Su (1999) finds that market risk (measured by market betas) can explain returns of A and B shares, but nonmarket risk variables, such as the variance of returns and firm size, do not systematically affect returns. Thus, the difference in return vola- tility between the A- and B-share markets may be caused by the difference in idiosyncratic risk. Trad- ing cost, which reflects asymmetric information and liquidity of trading, may explain the B-share market anomaly. For example, if B-share investors incur higher trading costs than A-share investors, the return volatility of B shares would be higher than that of A shares, other things being equal. In line with the above arguments, He and Wu (2003) examine whether the difference in trading costs (or market making costs) between the Chinese A and B shares can explain the difference in return vola- tility between the two classes of shares. They esti- mate the end-of-day bid-ask spread and its informed trading and noninformed trading cost components for each stock using daily data in the late 1990s. Their results show that the B-share market has persistent higher bid-ask spreads than the A-share market, and traders in the B-share market bear higher informed trading and other transaction costs. Furthermore, they find that the higher volatility of B-share returns can be attrib- uted to the higher market making costs in the B-share market. REFERENCES Chordia, T., Roll, R., and Subrahmanyam, A. (2002). ‘‘Order imbalance, liquidity, and market returns.’’ Journal of Financial Economics, 65: 111–131. Chui, A. and Kwok, C. (1998). ‘‘Cross-autocorrelation between A shares and B shares in the Chinese Stock Market.’’ Journal of Financial Research, 21: 333–354. Easley, D., Kiefer, N., O’Hara, M., and Paperman, J. (1996). ‘‘Liquidity, information, and infrequently traded stocks.’’ Journal of Finance, 51: 1405–1436. Green, C.J., Maggioni, P., and Murinde, V. (2000). ‘‘Regulatory lessons for emerging stock markets from a century of evidence on transactions costs and share price volatility in the London Stock Exchange.’’ Journal of Banking and Finance, 24: 577–601. He, Y., Wu, C., and Chen, Y M. (2003). ‘‘An explan- ation of the volatility disparity between the domestic and foreign shares in the Chinese Stock Markets.’’ International Review of Economics and Finance, 12: 171–186. Kyle, A. (1985). ‘‘Continuous auctions and insider trad- ing.’’ Econometrica, 53: 1315–1335. Lee, C.F., Chen, G., and Rui, O.M. (2001). ‘‘Stock returns and volat ility on China’s stock markets.’’ The Journal of Financial Research, 24: 523–544. Su, D. (1999). ‘‘Ownership restrictions and stock prices: evidence from Chinese markets.’’ Financial Review, 34: 37–56. Su, D. and Fleisher, B.M. (1999). ‘‘Why does return volatility differ in Chinese stock markets?’’ Pacific- Basin Finance Journal, 7: 557–586. Sun, Q. and Tong, W. (2000). ‘‘The effect of market segmentation on stock prices: the China syndrome.’’ Journal of Banking and Finance, 24: 1875–1902. 438 ENCYCLOPEDIA OF FINANCE Chapter 16 DECIMAL TRADING IN THE U.S. STOCK MARKETS YAN HE, Indiana University Southeast, USA Abstract All NYSE-listed stocks were switched from a frac- tional to a decimal trading system on January 29, 2001 and all NASDAQ stocks followed suit on April 9, 2001. The conversion to decimal trading in the U.S. markets has significantly reduced bid–ask spreads. This decline is primarily due to the drop in market makers’ costs for supplying liquidity. In add- ition, rounding becomes less salient after the deci- malization. The decrease in bid–ask spreads can be ascribed to the decrease in price rounding, when controlling for the changes in trading variables. Keywords: decimal trading; decimalization; NYSE; NASDAQ; clustering; rounding; bid–ask spread; volatility; fractional trading; price improvement The minimum increment of trading prices varies substantially with market and location. For in- stance, pricing of stock, bond, and options markets in the U.S. and Canada had traditionally been denominated in eighths, while in European and Asian markets decimal prices are more common. During the later half of 1990s, the U.S. and Can- adian markets underwent substantial changes. Canadian stocks switched from fractions to deci- mals in April 1996. In the U.S. markets, the min- imum tick size was reduced from one-eighth of a dollar to one-sixteenth of a dollar in June 1997. At the beginning of year 2000, the U.S. equity markets were the only major financial markets in the world that traded in fractional increments. This frac- tional trading practice puts U.S. markets at a com- petitive disadvantage with foreign markets trading the same securities. In addition, individual inves- tors may have a difficulty in determining the dif- ferences between increasingly smaller fractions. To make the U.S. securities markets more com- petitive globally and their prices easier to decipher, the Securities Industry Association and the Secur- ities and Exchange Commission decided to convert the U.S. equity and exchange-traded options mar- kets from fractional to decimal trading. The NYSE selected seven pilot securities for a decimal pricing test on August 28, 2000, another 57 securities were added to the pilot program on September 25, 2000, and another 94 were added on December 4, 2000. The NASDAQ market began its decimal test with 14 securities on March 12, 2001, and another 197 securities were added on March 26, 2001. All NYSE-listed stocks were switched to a decimal trading system on January 29, 2001 and all NAS- DAQ stocks followed suit on April 9, 2001. Recently, a number of studies have generated interesting findings about the effects of decimaliza- tion on return volatility and bid–ask spreads. They report that decimalization affects bid–ask spreads, volatility, quote size, and price improvement frequency (or the probability of trades within the quoted bid–ask spreads). First of all, it was shown that the recent conversion to decimal trading in the U.S. markets has significantly reduced bid–ask spreads (see NYSE, 2001; NASDAQ, 2001; Chak- ravarty et al., 2001a,b; Chung et al., 2001; Gibson et al., 2002). These findings coincide with two earl- ier studies (Ahn et al., 1998; Bacidore, 1997) on the Toronto Stock Exchange (TSE). Bacidore et al., (2001b) examine a wide range of market quality issues on the NYSE post-decimalization, and find that an increase in the aggressiveness of limit order pricing results in narrower bid–ask spreads. Chung et al. (2004) examine the relationship between NASDAQ trading costs and the extent of order preferencing. They document lower order prefer- encing and a positive relationship between the bid– ask spread and the proportion of internalized vol- ume on NASDAQ after decimalization. Second, Bessembinder (2003) and NASDAQ (2001) show that intraday return volatility has declined, and there is no evidence of systematic reversals in quotation changes. Thus, it appears that the NYSE and NASDAQ markets are able to supply sufficient liquidity in the wake of decimalization. Third, Bessembinder (2003) presents that quote size decreases after decimalization. Jones and Lip- son (2001) and Goldstein and Kavajecz (2000) re- port decreases in limit-order book depth after an earlier NYSE tick size reduction, and Bacidore et al. (2001a) report decreases in limit-order book depth after the decimalization on the NYSE. Fi- nally, Bacidore et al. (2001b) and Bessembinder (2003) find evidence that the percentage of orders experiencing price improvement (i.e. executed within the quotes) increases on the NYSE after decimalization, though the dollar amount of price improvement falls. According to Coughenour and Harris (2003), decimal trading effectively relaxes the public order precedence rule and gives special- ists more price points within the bid–ask spread on which to quote aggressively. This allows spe- cialist trading firms of all size to trade more often inside the current quote and so the probabil- ity that a trade occurs inside the quotes becomes higher. Almost all the above studies document the changes in ‘‘total’’ return volatility and spreads of transactions. He and Wu (2004) examine the composition of return volatility, serial correlation, and trading costs before and after the decimaliza- tion on the NYSE. Specifically, they decompose the variance of price changes into components associated with public news, rounding errors, and market-making frictions (asymmetric information and liquidity costs). First, the test results show that both variance components due to market-making frictions (or bid–ask spreads) and rounding errors decline considerably after decimalization, while the variance component due to public news shocks remains unchanged. Second, the serial correlation of price changes is significantly reduced after deci- malization, indicating a weakened bid–ask bounce effect as a result of decimal trading. Finally, bid– ask spreads decline substantially after decimaliza- tion and this decline is primarily due to the drop in market makers’ costs for supplying liquidity. In addition to volatility and transaction costs, the recent decimalization also provides an oppor- tunity to revisit the issue of price rounding. Since traders often choose to use a larger price increment than the minimum tick, prices tend to cluster on certain fractions or decimals even when the tick is small. (See Ball et al. (1985) for gold trading; Brown et al. (1991) for silver; Goodhart and Cur- cio (1992) for foreign exchange; and Aitken, et al. (1995) for Australian stocks.) Harris (1999) pre- dicts that the conversion to decimal trading would lead to lower execution costs. Bessembinder (2002) shows that bid–ask spreads have declined after the decimalization. He and Wu (2003) investigate the pattern of price rounding before and after decimal trading and its effect on bid–ask spreads for NYSE stocks by using the second pilot sample which includes 57 NYSE securities. Prior to September 25, 2000, these stocks were traded on sixteenths. Since then, they have been traded on pennies. First, since decimal trading leads to a finer price grid or a set of less discrete prices, it is expected to observe a decline in frequencies of rounding on integers, halves, and quarters. Second, although frequencies of rounding on integers, halves, and quarters may 440 ENCYCLOPEDIA OF FINANCE decline after decimalization, it is expected that cross-sectionally the relationship between round- ing and trading variables and the relationship be- tween execution costs and rounding will stay the same. That is, the sensitivity of trading variables to rounding and the sensitivity of execution costs to rounding should remain unchanged because the fundamentals of the market do not change as a result of decimalization. Finally, consistent with the arguments of Harris (1997, 1999), it is expected to find a significant relationship between the de- crease in execution costs and the decrease in rounding after decimalization, when controlling for the changes in stock features. If fractional pri- cing indeed allows market makers to keep bid–ask spreads artificially high to earn a positive rent, a conversion to decimal trading should reduce price rounding, decrease market makers’ rents, and cause a fall in bid–ask spreads. The empirical results of He and Wu (2003) show that although rounding is pervasive in trans- action prices, bids, and asks in both the pre- and post-decimalization periods, it has become less sa- lient after the decimalization. The cross-sectional relationship between rounding and trading vari- ables is similar before and after the decimalization, and so is the relationship between execution costs and rounding when trading variables are held con- stant for each stock. More importantly, the quoted and effective bid–ask spreads decrease after the decimalization, and this decrease can be ascribed to the decrease in price rounding when controlling for the changes in trading variables. REFERENCES Ahn, H J., Cao, Q.C., an d Choe, H. (1998). ‘‘Decimal- ization and competition among stock markets: Evidence from the Toronto stock exchange cross- listed securities.’’ Journal of Financial Markets, 1: 51–87. Aitken, M., Brown, P., Buckland, C., Izan, H.Y., and Walter, T. (1995). ‘‘Price clustering on the Australian stock exchange.’’ Working Paper, University of Western Australia. Bacidore, J. (1997). ‘‘The impact of decimalization on market quality: an empirical investigation of the To- ronto Stock Exchange.’’ Journal of Financial Inter- mediation, 6: 92–120. Bacidore, J., Battalio, R., and Jennings, R. (2001a). ‘‘Order submi ssion strategies, liquidity supply, and trading in pennies on the New York Stock Ex- change,’’ Working Paper, Indiana University. Bacidore, J., Battalio, R., Jennings, R., and Farkas, S. (2001b). ‘‘Changes in order characteristics, displayed liquidity, and execution quality on the NYSE around the switch to decimal pricing,’’ Working Paper, The New York Stock Exchange. Ball, C.A., Torous, W.N., and Tshoegl, A.E. (1985). ‘‘The degree of price resolution: the case of the gold market.’’ Journal of Futures Markets, 5: 29–43. Bessembinder, H. (2003). ‘‘Trade execution costs and market quality afte r decimalization.’’ Journal of Fi- nancial and Quantitative Analysis, 13: 19–42. Brown, S., Laux, P., and Schachter, B. (1991). ‘‘On the existence of an optimal tick size.’’ Review of Futures Markets, 10: 50–72. Chakravarty, S., Harris, S., and Wood, R. (2001a). ‘‘Decimal trading and market impact,’’ Working Paper, University of Memphis. Chakravarty, S., Harris, S., and Wood, R. (2001b). ‘‘Decimal trading and market impact: The Nasdaq experience.’’ Working Paper, University of Memphis. Chung, K., Van-Ness, B., and Van-Ness, R. (2001). ‘‘Are Nasdaq stocks more costly to trade than NYSE stocks? Evidence after decimalization.’’ Working Paper, Kansas State University. Chung, K., Chuwonganant, C., and McCormick, T. (2004). ‘‘Order preferencing and market quality on Nasdaq before and after decimalization.’’ Journal of Financial Economics, 71: 581–612. Coughenour, J. and Harris, L. (2003). ‘‘Specialist profits and the minimum price increment.’’ Working Paper. Gibson, S., Singh, R., and Yerramilli, V. (2002). ‘‘The effect of decimalization on the components of the bid- ask spreads.’’ Working Paper, Cornell University. Goldstein, M. and Kavajecz, K. (2000). ‘‘Eighths, six- teenths and market depth: Changes in tick size and liquidity provision on the NYSE.’’ Journal of Finan- cial Economics, 56: 125–149. Goodhart, C. and Curcio, R. (1992). ‘‘Asset price dis- covery and price clustering in the foreign ex- change market.’’ Working Paper, London School of Business. Harris, L. (1997). ‘‘Decimalization: a review of the ar- guments and evidence.’’ Working Paper, University of Southern California. DECIMAL TRADING IN THE U.S. STOCK MARKETS 441 Harris, L. (1999). ‘‘Trading in pennies: A survey of the issues.’’ Working Paper, University of Sout hern California. He, Y. and Wu, C. (2003). ‘‘The effects of decimaliza- tion on return volatility components, serial correl- ation, and trading costs.’’ Working Paper. He, Y. and C. Wu. (2004). ‘‘Price rounding and bid-ask spreads before and after the decimalization.’’ Inter- national Review of Economics and Finance, 13: 19–42. Jones, C. and Lipson, M. (2001). ‘‘Sixteenths: direct evidence on institutional trading costs.’’ Journal of Financial Economics, 59: 253–278. NASDAQ. (2001). ‘‘The impact of decimalization on the NASDAQ stock market,’’ prepared by NAS- DAQ Research Department. NYSE. (2001). ‘‘Comparing bid-ask spreads on the NYSE and NASDAQ immediately following NAS- DAQ decimalization,’’ prepared by NYSE Research Department. 442 ENCYCLOPEDIA OF FINANCE . 47 81 26 307 Rights offering of A and B shares 82 50 63 70 198 335 797 Table 15.3. Number of listed companies ( 199 0– 199 8) Companies 199 0 199 1 199 2 199 3 199 4 199 5 199 6 199 7 199 8 Issuing A share. A and B shares offering ( 198 7– 199 8) 87 88 89 90 91 92 93 94 95 96 97 98 Total Shares issued (100MM) 10 25 7 4 5 21 96 91 32 86 268 102 746 A share 10 25 7 4 5 10 43 11 5 38 106 79 343 H share. 1088 48 090 .94 3737.6 271.35 37 19 2001=01 1100 48 497 .99 3013.63 220.08 3082 2001=02 1110 46228.75 195 0.05 151 .92 2 197 2001=03 1122 5 090 8.44 5 095 .17 488.33 4335 2001=04 1123 51006 .9 5 395 .87 422.43