The Transmission of Pricing Information of Dually-Listed between Hong Kong and New York Stock Exchange

doi:10.4236/jssm.2009.24041

Paper Menu >>

Journal Menu >>

J. Service Science & Management, 2009, 2: 348-352

doi:10.4236/jssm.2009.24041 Published Online December 2009 (www.SciRP.org/journal/jssm)

The Transmission of Pricing Information of

Dually-Listed between Hong Kong and New

York Stock Exchange

Shuangfei LI, Shou CHEN

School of Business Adiministration, Hunan University, Changsha, China

Email: [email protected]

Received July 31, 2009; revised September 5, 2009; accepted October 29, 2009.

ABSTRACT

The study investigates the transmission of pricing information between Hong Kong Stock Exchange and New York Stock

Exchange. Using the opening and closing stock prices of these two markets from Jan. 2003 to Apr. 2007 with the

method of Seemingly Unrelated Regression, we draw the conclusions that: 1) intraday returns of Chinese dually-listed

stocks is influenced more obviously by Hang Seng Index than Dow-Jones Average; 2) transmission of pricing informa-

tion is only from New York to Hong Kong; 3) intraday returns of stocks from New York Stock Exchange has a remark-

able influence on that of the next day in Hongkong market, but the stocks price of Hong Kong Stock Exchange has no

relation with which of New York Stock Exchange.

Keywords: Information Tra n smi ssi o n, D ually-List ed, St o ck Price

1. Introduction

In recent years, there have been quite a number of Chi-

nese companies listed in foreign capital markets. Ac-

cording to the China Securities Regulatory’s Statistics,

up to early 2006, there are a total of 120 Chinese compa-

nies listed in foreign capital markets. The market value

of China's A-Shares reaches 9 trillion RMB, while the

market value of overseas listed enterprises reaches 11

trillion RMB. At the mean time, Along with China's fi-

nancial industry is opened to external and to deepen fi-

nance reformation, more and more people begin to pay

attention to the relationships between Chinese market

and overseas securities market. The interdependence of

international equity markets has been examined exten-

sively.

The interdependence of international equity markets

has been examined extensively. Using stock market in-

dices, Koch and Koch examined relationships between

daily closing values of eight national stock indices for the

years 1972, 1980 and 1987.Their result suggested that

international stock market has grown an more interdep-

endent over time, and increased equity market interdep-

endence has been concentrated primarily between mark-

ets in neighboring countries and between markets whose

trading hours overlap [1].

Using multivariate GARCH models, Zhao L. and

Wang Y. point out there is anasymmetry in pred ictability

of the volatility of A share verses B share. Before the

openness to domestic inverstors of B share in Feb. 2001,

the volatility of B share and A share are relatively inde-

pendent. After that, there is a prominent volatility spill-

over effect from A share to B share [2].

However, using the dually-listed stock prices will de-

scribe transmission of pricing information better than

using the indices [3–4]. Using the daily opening and

closing stock prices of seven Japanese corporations that

are dually listed on the New York Stock Exchange

(NYSE) and the Tokyo Stock Exchange (TSE), Lau and

Diltz conclude that market imperfections th at may inhibit

information transfer between TSE and NYSE stock re-

turns are not readily apparent and that international list-

ings do not give rise to arbitrage opportun ities [5].

Bae investigates the transfer of pricing information

between the Hong Kong Stock Exchange (HKSE) and

the London Stock Exchang e (LSE). They find that HKSE

overnight returns respond significantly to changes in LSE

intraday returns, but the transmission pro cess is not com-

pleted at the opening of the SEHK; LSE overnight re-

turns respond significantly to changes in HKSE intraday

returns, but the transmission process is not completed at

the opening of the LSE, either; the impact is stronger

The Transmission of Pricing Information of Dually-Listed between Hong Kong and New York Stock Exchange349

moving from the LSE to the SEHK [6].

Kalok et al. using high frequency data to study how

overnight price movements in local markets affect the

trading activity of foreign stocks on the NYSE. They

found that local price movements affect not only the

opening returns of foreign stocks , but also their returns in

the first 30-min interval. The magnitude of local price

movements is positively related to price volatility of for-

eign stocks, and this relation is stronger at the NYSE

open but weaker afterward. However, local price move-

ments cannot account for intraday variations in trading

volume. They interpret the result as evidence that the

trading activity of foreign stocks on the NYSE is related

more to liquidity trading of US investors and less to local

market information [7].

Chinese enterprises have a shorter history in listing in

foreign capital markets, Research on the relationships

between Chinese Capital Market are in progress. This

study is important due to the growing number of firms

that choose for their common stock to be traded on for-

eign stock exchanges. By examining the transmission of

pricing information, optimal investment decision for in-

vestors from home and abroad can be made.

The remainder of this study is organized as follows.

Section 2 describes the data and method used in the study,

while Section 3 summarizes the empirical results. Finally,

Section 4 presents the conclus i ons.

2. Data and Method

2.1 Data

Using opening and closing stock prices of Chinese firms

which are dually-listed on the Hong Kong and New York

stock exchanges over the period of 2003 to April 2007.

The firms’ New York listings are in the form of Ameri-

can Depository Receipts (ADRs). We eliminate two

firms due to unavailability of prices for the entire period.

Therefore, the final sample consists of seven firms, the

length of time series is 1043. Hong Kong Stock

Exchange and New York Stock Exchange opening and

closing prices are obtained from www.yahoo.com.cn.

(Table 1)

2.2 Method

The specification and methodology used in this study are

a refined version of those employed by Lau, S T (1994)

and Becker K H. (1990). Intraday and overnight returns

are calculated in the following manner. and

are the daily opening and closing stock prices,

and are the intraday return and overnight re-

turns:

open

oc t

100*log(/ )

occlose open



;

100*log( /)

coopenclose t



;

All return series are adjusted for dividend payment and

stock split. More specifically, we define the overnight

return and the intraday return for the HKSE and NYSE

as follows:

K: The SEHK intraday return on day t,

K: The SEHK return for the overnight period

ending on day t,

NY : The NYSE intraday return on day t,

NY : The NYSE return for the overnight period end-

ing on day t.

The Hong Kong Stock Exchange opens at 10:00 hours

and closes at 16:00 hours, and New York Exchange pre-

cedes Hong Kong by half an hour. Thus, on a given

trading day, the trading times are overlap, and the

Table 1. The description of sample firms

Firms Name Exchange Code Transform ProportionIndustry Listing Date

SINOPEC SHANGHAI ADS SHI (0338) 1:100 Energy 07/26/1993

CHINA TELECOM CP LTD CHA (0728) 1:100 Communication 07/22/2002

BRILLIANCE CHI CBA (1114) 1:100 Autocar manufacture 10/22/1999

GUANGSHEN RAIL CO LT GSH (0525) 1:50 Transportation 05/14/1996

ALUMINUM CP CHIN ADS ACH (2600) 1:100 Metallurgy 12/12/2001

HUANENG POWER INTL HNP (0902) 1:40 Electric power 01/22/1998

CHINA EASTERN AIRLNE CEA (0670) 1:100 Aviation 02/05/1997

Note: The ADR in Exchange code is the code on NYSE and corresponding code on HKSE each. Transform proportion is the number which one

ADR respond to H share.

The Transmission of Pricing Information of Dually-Listed between Hong Kong and New York Stock Exchange

350

open

Hong Kong

K



NY

American

open closeopen

close open

Figure 1. The chronological sequence of events in the New York Stock Exchange (NYSE) and the Hong Kong Stock Exchange

(HKSE)

chronological sequence is as shown in Figure 1.

First of all, we give our attention to the transmission of

pricing information between NYSE and HKSE. We

specify four equations that relate to the transmission of

pricing information between market participants in HK-

SE and NYSE. The first two equations model the trans-

mission of pricing information from the NYSE to the

HKSE.

Specifically, we first examine the effect of changes in

the intraday NYSE returns,, on the following

day’s HKSE overnight return, by estimating the

following regression specification:

NYt

1) Regression Equation (1) illustrates the effect of

changes in the intraday NYSE return on day t-1, on the

following day's HKSE overnight return,

11 11

co oc

HKNY t







 



(1)

2) Equation (2) illustrates the effect of changes in

the intraday NYSE return on day t-1, on the following

day’s HKSE intraday return,

2212

oc oc

HK NY







 



(2)

3) Equation (3) illustrates the effect of changes in

the intraday HKSE return on day t-1,on the same day’s

NYSE overnig ht ret ur n,

133 13

co oc

NYHK t







 





(3)

4) Equation (4) illustrates the effect of changes in

the intraday HKSE return on day t-1,on the same day’s

NYSE intraday return,

144 14

oc oc

NYHK t







 





(4)

where it



（i＝1,2,3,4）is a stochastic disturbance term.

If the one stock exchange responds to changes in re-

turns on the other, then estimation of equation should

produce a significan t positive slope coefficient.

The empirical specifications (1) and (3) each contain

variables that, while not strictly simultaneously deter-

mined, nevertheless share a considerable amount of

common trading time. Upon consideration of our em-

pirical specification and the types of hypotheses we are

testing, we employ a four-step instrumental variables

procedure that differs slightly from that of Koch [1] and

Lau S.T. [5].

First, we estimate market model regressions for

and

K using ordinary least squares:

oc oc

NYDJA 5tt







 (5)

66 6

oc oc

HKHSI t









(6)

Where is intraday return on the Dow Jones In-

dustrials Index, and

DJA

SI is intraday return on the Hong

Kong Hang Seng Index.

Secondly, the resulting parameter estimates are used to

create instruments (i.e. predictions), and

for variables and Equa-

tions (1) and (3), respectively.

KHˆoc

tDJAYN 55 ˆ



 (5’)

tHSIKH 66ˆ



 (6’)

Third, the instrumental variables are substituted into

Equations (1) and (3), yielding:

tYNHK 7177 ˆ



  (1’)

tKHNY 81881 ˆ



 (3’)

Finally, the parameters of the system of Equations (1’),

(2), (3’) and (4) are estimated jointly using the Seem-

ingly Unrelated Regression (SUR) method. By using the

estimates of the covariance of residuals across equations,

the SUR method improves the efficiency of the parame-

ter estimates.

3. Empirical Results

We believe that the instrumental variables produced in

the manner described above are appropriate instruments,

as they satisfy the two criteria for valid instruments. Ta-

ble 2 reports the parameter estimates for the market

models specified in Equations (5) and (6), all regression

slopes are statistically significant at the one per cent level,

indicating that all sample firms from NYSE and HKSE

return respond to changes of Share Market Indexes.

For each firm, the results give us an attention that



(0.56～0.87) is more than 5



(0.97～1.52), which

indicates the fluctuation range of Chinese stocks returns

of NYSE are narrower than Dow Jones Industrials Index,

and which from HKSE are wider than Hong Kong Hang

The Transmission of Pricing Information of Dually-Listed between Hong Kong and New York Stock Exchange351

Table 2. Results of the market model for the NYSE and the HKSE

New York Stock Market Hong Kong Stock Market

Stock 5





Stock 6





SHI 0.026

(0.73) 0.618***

(13.76) 0.154 0338.HK 0.004

(0.06) 1.475***

(15.72) 0.192

CHA -0.003

(-0.10) 0.781***

(20.83) 0.294 0728.HK 0.069

(1.41) 1.094***

(16.41) 0.206

CBA -0.090*

(-1.78) 0.759***

(11.99) 0.121 1114.HK -0.084

(-1.36) 0.998***

(11.88) 0.119

GSH 0.017

(0.40) 0.603***

(11.38) 0.111 0425.HK

-0.137**

(-1.98) 1.222***

(13.00) 0.140

ACH 0.011

(0.25) 0.870***

(15.51) 0.188 2600.HK -0.080

(-1.58) 0.968***

(14.08) 0.160

HNP 0.040

(1.20) 0.590***

(14.18) 0.162 0902.HK -0.139*

(-1.76) 1.214***

(11.26) 0.109

CEA -0.055

(-1.24) 0.560***

(9.97) 0.087 0670.HK -0.024

(-0.34) 1.519***

(16.13) 0.200

Note: t-values appear in parentheses.

***Significant at the 1 per cent level. **Significant at th e 5 pe r cent level. *Significant at the 10 per cent level.

Table 3. SUR estimates of Equations (1’), (2), (3’) and (4)

Transfer direction From New York to Hong Kong From Hong Kong to New York

Stock 7

















SHI

(0338.HK) 0.039

(0.85) 0.972***

(10.77) 0.040

(0.52) -0.106*

(-1.74) 0.066

(0.95) 0.000

(0.05) 0.066

(0.96) 0.002

(0.55)

CHA

(0728.HK) -0.029

(-0.86) 0.928***

(17.99) 0.090*

(1.65) -0.015

(-0.30) 0.055

(1.08) 0.002

(0.35) 0.055

(1.08) 0.002

(0.82)

CBA

(1114.HK) 0.240***

(5.83) 0.671***

(9.95) -0.072

(-1.10) -0.098***

(-2.61) 0.063

(0.85) 0.002

(0.51) 0.063

(0.85) 0.001

(0.51)

GSH

(0425.HK) 0.151***

(3.33) 1.002***

(10.69) -0.114

(-1.53) 0.016

(0.30) 0.088

(1.59) 0.000

(-0.06) 0.088

(1.60) 0.000

(0.07)

ACH

(2600.HK) 0.049

(0.63) 0.617***

(5.45) -0.063

(-1.14) -0.015

(-0.43) 0.006

(0.04) 0.000

(-0.01) 0.006

(0.04) 0.000

(0.02)

HNP

(0902.HK) 0.051

(1.09) 1.138***

(11.51) -0.104

(-1.24) -0.146**

(-2.05) -0.050

(-0.62) 0.000

(-0.11) -0.050

(-0.61) 0.000

(0.15)

CEA

(0670.HK) 0.241***

(4.77) 1.667***

(14.70) 0.006

(0.08) 0.090*

(1.76) 0.106

(1.92) 0.000

(0.03) 0.106

(1.92) 0.000

(0.11)

Note: t-values appear in parentheses.

***Significant at the 1 per cent level. **Significant at th e 5 pe r cent level. *Significant at the 10 per cent level.

Seng Index. Ther efore, contrast to Dow Jones Industr ials

Index, the stocks of mainland firms listed in Hong Kong

are sensitivity to Hang Seng Index, which would be an

overreact.

The first two columns in Table 3 reports the results of

SUR estimates of the effect of changes in the sample

firm NYSE intraday return on the following day’s sam-

ple firm HKSE overnight return. All seven regression

slopes are statistically positive significant at the one

percen t level, indi cating that sa mple firm HKSE retu rns

respond to changes in corresponding firms returns on

the NYSE. If the regression slopes of firms, 7



, higher

than one, indicating that stock returns in Hong Kong

effected by New York higher than that in New York and

vice versa.

The following two columns in Table 3 reports the re-

sults of SUR estimates of the effect of changes in the

sample firm NYSE intraday return on the following

day’s sample firm HKSE intraday return. This regression

allows us to determine whether the transmission of pric-

ing information has been completed at the opening trade.

The insignificance of three of the seven regression slopes

suggests that most of the transmission of pricing infor-

mation is completed at the first trade of the day. The

negative significance of three of that suggests that based

on the opening price closing price have a reversion. Only

one firm have a negative significance slopes at the ten

per cent level indicating that pricing information trans-

mission is continue at the all trade of the day.

Associate Equation (1’) w ith Equation (2), as shown in

The Transmission of Pricing Information of Dually-Listed between Hong Kong and New York Stock Exchange

352

Table 3 is the transmission of pricing information from

New York to Hong Kong. 7



are higher than 2



and

so are the corresponding t-statistics, which suggests that

the transmission from New York to Hong Kong is com-

pleted at the first trade of the day.

The regression slopes in Equations (3’) and (4) are

insignificant, which suggests that the transmission of

pricing information from Hong Kong to New York is

unconspicuous. This result disagrees with Lau [5] and

Bae [6].

4. Conclusions

This study examines th e transmission of pricing info rma-

tion between market participants in the HKSE and the

NYSE. Using the opening and closing stock prices of

seven Chinese firms that are inter-listed on the Hong

Kong and the New York stock exchanges over the period

of 2003 to April 2007, four regression specifications re-

lating to the transmission of pricing information are ex-

amined. The findings of this study suggest that: 1) Con-

trast to Dow Jones Industrials Index, the stocks of

mainland firms listed in Hong Kong are sensitivity to

Hang Seng Index. 2) Sample firm HK SE opening returns

respond to changes in sample firm NYSE intraday stock

performance; the transmission process is apparently

completed at the open ing of the HKSE; 3) The transmis-

sion of pricing information from Hong Kong to New

York are unconspicuous.

REFERENCES

[1] P. D. Koch and W. Koch, “Evolution in dynamic linkages

across Daily National Stock Indexes,” Journal of Interna-

tional Money and Finance, Vol. 10, No. 2, pp. 231–251,

1991.

[2] L. Y. Zhao and Y. M. Wang, “Information flows and

volatility spillover across A share and B share market in

China,” Journal of Finance, Vol. 10, pp. 37–52 (in Chi-

nese), 2003.

[3] D. Long, J. Payne, and C. Feng, “Information transmis-

sion in the Shanghai equity market,” Journal of Financial

Research, Vol. 22, pp. 29–45, 1999.

[4] C. S. Eun and S. Sabherwal, “Cross-border listings and

price discovery: Evidence from US-listed Canadian

stocks,” Journal of Finance, Vol. 58, No. 2, pp. 549–575,

2003.

[5] S. T. Lau and J. D. Diltz, “Stock returns and the transfer

of information between the New York and Tokyo Stock

Exchanges,” Journal of International Money and Finance,

Vol. 13, No. 2, pp. 211–222, 1994.

[6] K. H. Bae, B. Cha, and Y. L. Cheung, “The transmission

of pricing information of dually-listed stocks,” Journal of

Business Finance and Accounting, Vol. 26, pp. 709–723,

1999.

[7] K. Chan, M. Chockalingam, and W. L. Kent, “Overnight

information and intraday trading behavior: evidence from

NYSE cross-listed stocks and their local market informa-

tion,” Journal of Multinational Financial Management,

Vol. 10, pp. 495–509, 2000.