Price Transmission Effect between GDRs and Their ...
Transcript of Price Transmission Effect between GDRs and Their ...
Review of Quantitative Finance and Accounting, 19: 181–214, 2002©C 2002 Kluwer Academic Publishers. Manufactured in The Netherlands.
Price Transmission Effect between GDRs and TheirUnderlying Stocks—Evidence from Taiwan
SHEN-YUAN CHEN∗Department of Finance, Ming Chuan UniversityE-mail: [email protected]
LI-CHUAN CHOUDepartment of Finance, Ming Chuan University
CHAU-CHEN YANGDepartment of Finance, National Taiwan University
Abstract. In this paper we examine the price transmission effect between ADRs or GDRs and their respectiveunderlying stocks. This linkage is investigated for Granger causality using difference form and VECM. Resultsreveal unidirectional causality from Taiwan’s capital market to the foreign market. This asymmetry suggests thedomestic market plays a dominant role in price transmission relative to the foreign market. Besides, the prices ofboth markets will make adjustment to establish a long run cointegrated equilibrium. An additional finding is thatboth the premium and net buy have significant impacts on international price transmission for over twenty percentsamples. Empirical outcomes also provide the evidence that our model is quite robust.
Key words: price transmission, ADRs, GDRs, premium, net buy
JEL Classification: G15, F21, F23, C22
1. Introduction
Among the emerging equity markets, Taiwan’s capital market is an increasingly importantone for global institutional investors due to the government’s incessant revolution andliberalization policy in past decade. For example, in 1991, foreign institutional investorswere allowed to directly invest in Taiwan stock market and from September 1996, Taiwanmarket was included in its indices by the Morgan Stanley Capital International Inc. (MSCI).As Taiwan capital market continuously deregulated, foreign investors are getting more activeto this emerging market.
At the same time, as an important Original Equipment Manufacturer for worldwidefamous enterprises in recent years, Taiwan companies, especially for high-tech industries,are attracting more attention from the global investors. Rapid growth in competitive abilityhas engendered the result that a large number of Taiwan firms have their stocks cross-listed
∗Address correspondence to: 250, Chung Shan N. Rd. Sec. 5, Taipei, Taiwan. Tel.: (886) 2-28824564-2390, Fax:(886) 2-28809769.
182 CHEN, CHOU AND YANG
on international exchanges successfully. The stock price linkage between Taiwan market andforeign market has become an important issue for local and foreign institutional investorsbecause of the price interaction and arbitrage opportunities provided by dually listing.
The most commonly used vehicles of dual listing by Taiwan companies are AmericanDepositary Receipts (ADRs) and Global Depositary Receipts (GDRs). The possible advan-tages for such cross listing are promoting a firm’s reputation in large capital markets, theavailability of capital, lower capital costs and elimination of investment barriers such as do-mestic accounting and tax practices (see Karolyi (1998) survey on why and how companieslist abroad).
DRs can be created in one of the two ways: sponsored DR and unsponsored DR. In asponsored DR, the underlying corporation pays a fee to the depositary institution to coverthe cost of DR program. By regulation, the underlying corporation must provide periodicfinancial reports to the holders of DRs. A sponsored DR is often issued by a public companyto seek to have its stock traded in foreign country and to raise capital from a foreign market.In contrast to sponsored DR, an unsponsored DR is issued by one or more banks or securitybrokerage firms that assemble a large block of the shares of a foreign corporation withoutthe participation of the underlying corporation. Most DRs issued by Taiwanese listingcompanies are sponsored DRs.
The first issuance of DRs sponsored by a Taiwan company was the GDR of China SteelCorporation in May 1992. At the end of 1999, 36 Taiwanese listing companies have issuedDRs and the total amount of issuance had reached to 6.243 billion US dollars. Amongthese DRs, high technology companies are the major sponsors, which record 22 or 61.11%of issuances. The capital raised by these high technology companies was 4.475 billionUS dollars, or 71.68% of all issuances. Besides, the frequency of DRs issued by Taiwancompanies dramatically increased from 1994 to 1999. The status of Taiwan-listed companiesissuing DRs is summarized in Table 1.
In recent years, financial deregulation and international financial integration have resultedin a large amount of research on the dynamics of international transmission between GDRsor ADRs and their underlying securities. For example, Barclay et al. (1990) report thatdual listing of sixteen New York Stock Exchange (NYSE) listed companies on the TokyoStock Exchange (TSE) has no impact on the variances of NYSE close-to-close returns onthe stocks. Kato et al. (1991) and Wahab et al. (1992) try to find arbitrage opportunitiesbetween the prices of ADRs and underlying securities. They generally support the notationthat, after transactions costs, few profitable opportunities exist in these markets, implyingthat both markets are efficient. Jayaraman, Shastri and Tandon (1993) suggest that the listingof ADRs are associated with permanent increases in the return volatilities of the underlyingstocks. Kim, Szakmary and Mathur (2000) use both a vector autoregressive (VAR) modelwith a cointegration constraint and a seemingly unrelated regression (SUR) approach toexamine the relative importance of, and the speed of adjustment of ADR prices to, theseunderlying factors. Their results show that the ADRs appear to initially overreact to the USmarket index but underreact to changes in underlying share prices and exchange rate.
Multiple listing offers a unique opportunity to study the transmission of pricing infor-mation across markets. Neumark, Tinsley and Tonsini (1991) find that the foreign marketreacts to domestic price changes more quickly than the domestic market reacts to foreign
PRICE TRANSMISSION EFFECT BETWEEN GDRS 183
Tabl
e1.
Stat
usof
Taiw
anlis
ted
com
pany
issu
ing
spon
sore
dD
Rs
Com
mon
Num
ber
ofSt
ock
Lis
ting
Issu
ing
DR
s(i
nU
nitP
rice
Tota
lAm
ount
Lea
ding
Cor
pora
tion
Nam
eC
ode
Indu
stry
Loc
atio
nD
ate
thou
sand
)(U
Sdo
llar)
(in
thou
sand
)U
nder
wri
ter
Chi
naSt
eel
2002
TT
Stee
lG
loba
l05
/28/
9218
,000
18.2
032
7,60
0G
oldm
anSa
chs
Asi
aC
emen
t(1)
1102
TT
Cem
ent
Glo
bal
06/2
3/92
2,40
027
.50
66,0
03M
orga
nSt
anle
yPr
esid
entE
nter
pris
es12
16T
TFo
odL
X11
/24/
924,
993
16.5
182
,426
CS
Firs
tBos
ton
Chi
aH
sin
Cem
ent
1103
TT
Cem
ent
LI
05/2
5/93
2,10
016
.90
35,4
90Ja
rdin
gFl
emin
g
Tun
tex
Dis
tinct
1462
TT
Text
ileG
loba
l05
/04/
947,
000
12.1
284
,840
Bar
ing
Bro
ther
sM
icro
elec
tron
ics
2314
TT
Ele
ctro
nic
Glo
bal/L
X05
/24/
943,
900
12.7
048
,260
CS
Firs
tBos
ton
Tech
nolo
gyH
oche
ng18
10T
TG
lass
and
LX
06/2
9/94
2,80
031
.50
85,4
00B
ZW
Cer
amic
Tun
gH
oSt
eel
2006
TT
Stee
lL
X08
/09/
946,
000
17.2
010
3,20
0Ja
rdin
gFl
emin
gY
ageo
2327
TT
Ele
ctro
nic
Glo
bal/L
X09
/28/
944,
000
22.9
011
4,50
0Sc
hrod
er
Aur
ora
2373
TT
Ele
ctro
nic
LX
01/2
7/95
1,87
516
.00
30,0
00N
AG
VC
2322
TT
Ele
ctro
nic
LI
04/0
3/95
5,00
015
.30
76,5
00G
oldm
anSa
chs
ASE
2311
TT
Ele
ctro
nic
Glo
bal/L
X07
/13/
958,
600
15.2
513
1,15
0M
orga
nSt
anle
yA
.D.I
.23
04T
TE
lect
roni
cL
X09
/28/
952,
500
16.9
642
,400
Ban
kers
Tru
stW
alsi
nL
ihw
a16
05T
TE
lec.
Cab
leG
loba
l/LX
10/0
3/95
10,0
0012
.18
121,
800
Dai
wa,
Ban
kers
and
Wir
eT
rust
Silic
onw
are
Prec
isio
n23
25T
TE
lect
roni
cG
loba
l/LI
10/0
4/95
6,00
015
.20
91,2
00B
ZW
Ace
r(1
)23
06T
TE
lect
roni
cG
loba
l/LI
11/0
1/95
17,0
0012
.99
220,
830
Nom
ura
Int’
l
Mac
roni
xIn
t’l
2337
TT
Ele
ctro
nic
NA
SDA
Q05
/14/
9610
,000
17.7
617
6,70
0C
SFi
rstB
osto
nE
verg
reen
Mar
ine
2603
TT
Mar
ine
Glo
bal/L
I07
/30/
9610
,800
18.0
519
4,94
0G
oldm
anSa
chs
Asi
aC
emen
t(2)
1102
TT
Cem
ent
Glo
bal/L
I09
/12/
963,
750
20.0
060
,000
SBC
War
burg
(con
tinu
ed)
184 CHEN, CHOU AND YANGTa
ble
1.(C
onti
nued
)
Com
mon
Num
ber
ofSt
ock
Lis
ting
Issu
ing
DR
s(i
nU
nitP
rice
Tota
lAm
ount
Lea
ding
Cor
pora
tion
Nam
eC
ode
Indu
stry
Loc
atio
nD
ate
thou
sand
)(U
Sdo
llar)
(in
thou
sand
)U
nder
wri
ter
Lite
-on
Tech
nolo
gy23
46T
TE
lect
roni
cL
I09
/25/
964,
900
14.5
571
,295
BZ
WY
ung
Min
gM
arin
e26
09T
TM
arin
eL
I11
/14/
9610
,000
11.6
411
6,39
2U
BS
Acc
ton
Tech
nolo
gy23
45T
TE
lect
roni
cE
urop
e/L
I02
/01/
9712
,000
7.51
90,1
20Ja
rdin
gFl
emin
gTe
coE
lec.
&M
ach.
1504
TT
Eng
inee
rG
loba
l/LI
03/2
7/97
5,54
020
.08
111,
241
SBC
War
burg
Asu
stek
Com
pute
r23
57T
TE
lect
roni
cG
loba
l/LI
05/3
0/97
21,0
0011
.23
235,
830
Nom
ura
Int’
lSt
anda
rdFo
ods
1227
TT
Food
Glo
bal/L
X06
/19/
973,
000
9.69
29,0
70Sc
hrod
erSy
nnex
Tech
nolo
gy(1
)23
47T
TE
lect
roni
cG
loba
l/LX
07/0
3/97
431
9.81
4,22
5B
arin
gB
roth
ers
Synn
exTe
chno
logy
(1)
2347
TT
Ele
ctro
nic
Glo
bal/L
X07
/03/
976,
270
22.2
313
9,38
2B
arin
gB
roth
ers
Ace
r(2
)23
06T
TE
lect
roni
cG
loba
l/LI
07/2
3/97
10,0
0016
.06
160,
600
Gol
dman
Sach
sT
SMC
(1)
2330
TT
Ele
ctro
nic
NY
SE10
/08/
9724
,000
24.7
859
4,72
0G
oldm
anSa
chs
Fubo
nIn
sura
nce
2817
TT
Insu
ranc
eG
loba
l/LI
04/1
7/98
8,00
020
.07
160,
560
CS
Firs
tBos
ton
D-L
ink
2332
TT
Ele
ctro
nic
Glo
bal/L
X09
/18/
985,
000
10.1
350
,650
Salo
mon
Smith
Bar
ney
Win
bond
(1)
2344
TT
Ele
ctro
nic
LX
02/0
5/99
14,6
0011
.45
167,
170
AB
NA
mro
Ace
rPe
riph
eral
s23
52T
TE
lect
roni
cL
I06
/29/
992,
700
23.2
262
,694
Nom
ura
Int’
lT
SMC
(2)
2330
TT
Ele
ctro
nic
NY
SE07
/15/
9912
,094
24.0
029
6,50
0G
oldm
anSa
chs
Synn
exTe
chno
logy
(2)
2347
TT
Ele
ctro
nic
Glo
bal/L
X08
/12/
995,
463
18.9
310
3,41
5Ja
rdin
gFl
emin
gT
SMC
(3)
2330
TT
Ele
ctro
nic
NY
SE09
/09/
995,
486
28.9
615
8,89
7G
oldm
anSa
chs
Mos
elV
itelic
2342
TT
Ele
ctro
nic
LX
09/1
6/99
9,98
08.
7086
,867
Nom
ura
Int’
lH
ouH
aiPr
ecis
ion
2317
TT
Ele
ctro
nic
LI
10/0
7/99
30,0
0013
.89
416,
700
War
burg
Dill
ion
Rea
dR
itek
2349
TT
Ele
ctro
nic
LX
10/1
5/99
27,5
0011
.86
326,
150
AB
NA
mro
Pow
erch
ipSe
mic
ond
5346
TT
Ele
ctro
nic
LX
10/2
1/99
27,0
0010
.70
288,
900
Nom
ura
Int’
lFa
rE
astT
extil
e14
02T
TTe
xtile
LX
10/2
5/99
13,5
0014
.00
189,
000
Gol
dman
Sach
sC
ampa
l23
24T
TE
lect
roni
cL
X11
/09/
998,
000
15.2
712
2,16
0N
omur
aIn
t’l
Win
bond
(2)
2344
TT
Ele
ctro
nic
LX
11/1
2/99
10,0
0016
.70
167,
000
War
burg
Dill
ion
Rea
d
Tota
l6,
242,
778
Dat
aso
urce
:Se
curi
ties
and
Futu
res
Com
mis
sion
Min
istr
yof
Fina
nce,
R.O
.C.;
LX
:Lux
embo
urg;
LI:
Lon
don.
PRICE TRANSMISSION EFFECT BETWEEN GDRS 185
price changes. This asymmetry, confirmed for price indices by Eun and Shim (1989) andHamao, Masulis and Ng (1990), is interpreted by Garbade and Silber (1979) as evidence thatthe foreign market acts as a satellite to the domestic market. Hauser, Tanchuma and Yaari(1998) investigate five companies based in Israel whose stocks are listed on both the TelAviv Stock Exchange and NASDAQ. Their empirical tests of causality in price changes usethe side-by-side Box-Jenkins ARIMA models and the Sims VAR model. Overall, the resultsshow that price causality in dually listed stocks is unidirection from the domestic marketto the foreign market. Jithendranathan, Nirmalanandan and Tandon (2000) evaluate marketsegmentation and its effect on the pricing of cross-listed securities using Indian Global De-positary Receipts (GDRs). They report that capital flow barriers existing in India lead to theGDRs being priced at a premium over the exchange rate adjusted prices of the underlyingIndian securities. And GDR index returns are affected by both domestic and internationalfactors, while the underlying Indian securities are affected only by domestic variables.
Earlier studies on international capital asset theory assume that international dually listedsecurities should sell at the same price in the absence of transaction costs and restrictionto capital flows. Garbade and Silber (1979) reported that prices may differ between marketcenters for short intervals of time in imperfectly integrated market. The adjustment be-tween prices in market A and market B can be characterized in one of two ways: (1) theadjustment may be symmetrical; (2) the adjustment may be one-sided. Hence, this paperuses Granger tests to examine causal relations between the returns on GDRs or ADRs andtheir respective underlying Taiwanese securities. We use error-correction model to analyzethe long run causal relations where the stock returns data is nonstationary. In addition, thispaper further discusses the impact of premium or discount in overseas-listed stocks on theprice transmission effect.1 The net buy by QFIIs2 in Taiwan is also one of important factorsto be measured in price transmission effect because QFIIs play an increasingly importantrole in Taiwan market since the MSCI indices including this emerging market. So we alsoexamine the effect of the net buy variable on causal relations.
Similar to the most existing research, our empirical results show the return causality ismostly unidirection from the domestic market to foreign market for dually listed Taiwanstocks. Besides, the prices of both markets will make adjustment to establish a long runcointegrated equilibrium. Unlike prior studies, this paper finds that both the premium ordiscount and QFII’s net buy have significant impacts on international price transmission forover twenty percent samples. Empirical tests show that our model is robust.
The remainder of this article is organized as follows: the next section presents the datadescription and the related empirical methodology. Our empirical results are described insection three. The final section concludes the paper.
2. Data and methodology
2.1. Data description
Thirty-six listed companies issued GDRs or ADRs by the end of 1999 in Taiwan; ninety-fiveGDRs or ADRs were issued and listed on exchange or over-the-counter (see Table 2). This
186 CHEN, CHOU AND YANGTa
ble
2.L
isto
fTa
iwan
liste
dco
mpa
nies
issu
ing
GD
Rs
orA
DR
s
Blo
ombe
rgC
ode
Cor
pora
tion
Cod
eH
igh
Liq
uid
Low
Liq
uid
No
Quo
tatio
nIs
suin
gD
ate
Shar
esof
QFI
Is’
Hol
ding
(in
thou
sand
)
Rat
ioof
QFI
Is’
Hol
ding
(%)
Rat
ioof
QFI
Is’
&N
on-Q
FIIs
’H
oldi
ng(%
)
2002
TT
CSG
DS
LI
CH
CG
LX
CIS
EY
US
05/2
8/92
547,
186
6.39
6.82
CN
SG
R(E
UR
$)C
ISX
FU
S11
02T
TA
CG
DS
LI
AIA
GR
(EU
R$)
06/2
3/92
69,5
893.
7211
.88
1216
TT
PEN
TL
XU
PEZ
YU
S11
/24/
9219
0,14
86.
5013
.62
1103
TT
CH
CD
SL
IC
HSN
FU
S05
/25/
937,
436
1.06
11.4
3
1462
TT
TT
XS
LI
TU
NG
LX
05/0
4/94
23,8
871.
0712
.01
2314
TT
ME
TG
LX
05/2
4/94
56,6
3017
.15
46.3
818
10T
TH
CD
RL
IH
CD
GL
X06
/29/
948,
589
2.05
2.57
2006
TT
TH
SGG
LX
08/0
9/94
1,08
22.
282.
9423
27T
TY
AG
GL
XY
AG
SP;Y
GE
QY
US
09/2
8/94
79,4
076.
6716
.43
2373
TT
AU
RD
LI
AU
RG
LX
AU
RSP
01/2
7/95
38,8
106.
249.
4323
22T
TG
VC
DL
I;25
52Q
US
04/0
3/95
17,0
302.
2711
.78
2311
TT
ASE
DL
IA
SEG
LX
ASE
SP07
/13/
9525
6,88
812
.97
39.2
823
04T
TA
DO
DL
IA
DIC
LX
09/2
8/95
22,8
393.
3815
.37
1605
TT
WLW
DL
IW
LWD
LX
10/0
3/95
25,8
120.
8213
.53
2325
TT
SIL
DL
ISL
CZ
FU
S;SP
IAF
US;
10/0
4/95
110,
158
9.77
16.3
4SL
CW
YU
S23
06T
TA
CID
LI
AC
EH
FU
SA
CE
RY
US
11/0
1/95
320,
816
10.3
225
.69
AC
IGG
R(E
UR
$)12
80Q
US
2337
TT
MX
ICY
US
MSI
DL
IM
XIT
YU
S05
/14/
9620
1,89
09.
4919
.91
MX
ICG
R(E
UR
$)M
XIA
LI
2603
TT
EG
MD
LI
07/3
0/96
119,
344
6.43
33.1
4E
MA
GR
(EU
R$)
2346
TT
LTT
DL
I09
/25/
9632
,403
6.57
7.04
2609
TT
YM
TD
LI
1848
QU
S11
/14/
9617
5,53
810
.45
11.5
2
2345
TT
AT
OD
LI
AC
TG
LX
;AC
TV
FU
S;02
/01/
9726
,491
11.2
913
.69
AT
HY
YP
US
PRICE TRANSMISSION EFFECT BETWEEN GDRS 187
1504
TT
TE
CD
LI
03/2
7/97
35,8
062.
054.
9923
57T
TA
SKD
LI
05/3
0/97
200,
007
17.4
517
.73
1227
TT
SFT
DL
I06
/19/
9721
,007
6.95
12.3
123
47T
TSY
XD
LI
SYX
TY
LX
;SY
XT
YU
S;07
/03/
9736
,112
10.3
237
.87
2678
QU
S23
30T
TT
SMU
S10
/08/
971,
777,
349
17.7
934
.69
TM
SDL
IT
SFA
GR
(EU
R$)
2817
TT
FBN
DL
IFU
IZF
US;
FUIS
YU
S;04
/17/
9886
,170
5.00
9.57
2030
QU
S23
32T
TD
LK
DL
ID
LN
KG
LX
09/1
8/98
90,7
2729
.63
31.9
923
44T
TW
BD
DL
IW
BD
AL
X;W
BE
KY
US
02/0
5/99
272,
733
7.73
12.2
023
52T
TA
CE
RL
X;A
CE
DL
I06
/29/
9913
7,78
715
.41
15.8
123
42T
TM
SVD
LI
2671
QU
S09
/16/
9957
,125
2.25
14.2
623
17T
TH
HPD
LI
HN
HPF
US;
1092
ZL
X;
10/0
7/99
326,
998
29.7
334
.53
2685
QU
S23
49T
TR
KC
DL
IR
TK
GR
(EU
R$)
RIT
KL
X10
/15/
9912
,776
2.01
3.42
5346
TT
POSD
LI
POSD
LX
10/2
1/99
303,
432
17.8
439
.16
1402
TT
FET
DL
IFA
RE
LX
;269
9QU
S10
/25/
9946
0,97
716
.76
23.7
623
24T
TC
PED
LI
CPE
DL
X11
/09/
9915
2,96
99.
8310
.06
Ave
rage
175,
110
9.10
17.8
7
Dat
aso
urce
s:B
loom
berg
Info
rmat
ion
Syst
em;S
ecur
ities
and
Futu
res
Com
mis
sion
Min
istr
yof
Fina
nce,
R.O
.C.;
EnT
rust
Secu
ritie
sC
ompa
ny.
188 CHEN, CHOU AND YANG
Table 3. List of selected samples
Bloomberg Code
Corporation GDRs or Issuing SampleCode ADRs Code Date
Shares ofQFIIs’Holding (inthousand)
Ratio ofQFIIs’Holding (%)
Ratio of QFIIs’& Non-QFIIs’Holding (%) Days
2317 TT HHPD LI 10/07/99 326,998 29.73 34.53 1515346 TT POSD LI 10/21/99 303,432 17.84 39.16 1412330 TT TSM US 10/08/97 1,777,349 17.79 34.69 6232357 TT ASKD LI 05/30/97 200,007 17.45 17.73 6261402 TT FETD LI 10/25/99 460,977 16.76 23.76 1392311 TT ASED LI 07/13/95 256,888 12.97 39.28 6262345 TT ATOD LI 02/01/97 26,491 11.29 13.69 6262609 TT YMTD LI 11/14/96 175,538 10.45 11.52 6262306 TT ACID LI 11/01/95 320,816 10.32 25.69 6262347 TT SYXD LI 07/03/97 36,112 10.32 37.87 6262324 TT CPED LI 11/09/99 152,969 9.83 10.06 1282325 TT SILD LI 10/04/95 110,158 9.77 16.34 6262337 TT MXICY US 05/14/96 201,890 9.49 19.91 5392344 TT WBDD LI 02/05/99 272,733 7.73 12.20 3071227 TT SFTD LI 06/19/97 21,007 6.95 12.31 6262346 TT LTTD LI 09/25/96 32,403 6.57 7.04 6262603 TT EGMD LI 07/30/96 119,344 6.43 33.14 6262002 TT CSGDS LI 05/28/92 547,186 6.39 6.82 6261102 TT ACGDS LI 06/23/92 69,589 3.72 11.88 6262342 TT MSVD LI 09/16/99 57,125 2.25 14.26 1621605 TT WLWD LI 10/03/95 25,812 0.82 13.53 626
Average: 261,658 10.71 20.73 Sum: 10,328
paper chooses twenty-one GDRs or ADRs from the ninety-five, representing twenty-onelisted companies, for total data of 10,328 sample days (see Table 3). The principal of makingselections is as below:
1. Forty-two GDRs or ADRs which have no quotations and no trading are eliminated.Eighteen GDRs or ADRs which have quotations but traded light are also excluded.
2. We deleted some listed company samples because their shares held by QFII are lessthan or equal to 6.3%. Exceptions are Asia Cement, Mosel Vitelic, and Walsin Lihwa,because these stocks are included in MSCI Taiwan Index.
3. We chose GDRs or ADRs on exchange for some companies issuing more than one andbeing liquid, for example, Taiwan Semiconductor Manufacturing (TSM US). GDRs orADRs in dollar quotation are selected if they have dollar and eurdollar quotations, forexample, ACID LI, EGMD LI, and CSGDS LI.
4. Finally D-Link is also deleted because of late issuance and there only being forty-sevencollected days.
The data for this paper are taken from three sources: the daily close price for GDRs orADRs and NT exchange rate are collected from the Bloomberg information system; the
PRICE TRANSMISSION EFFECT BETWEEN GDRS 189
underlying stock close price and adjusted price for ex-dividend are provided by TaiwanEconomic Journal (TEJ); the volume of net buy by foreign institutions is offered by theEnTrust Securities Company.
The period of selected samples is from October 8, 1997 to May 31, 2000 on the basis ofthe three points below:
1. In September 1997, Taiwan stocks were first included in the MSCI indices as a result ofincreasing foreign institution investment in the Taiwan equity market. It happened thatthe Asian Financial crisis began at the same time, which sharply increased systematicrisk in Taiwan stock market. This event didn’t end until the fourth quarter of 1997.
2. Taiwan Semiconductor Manufacturing (TSM) ADR made an epochal entry for Taiwancompanies issuing GDRs or ADRs because TSM is at the head of Taiwan’s high-techindustry, the first Taiwan company listed in NYSE, and the issuing amount being thelargest among all GDRs or ADRs.
3. These two years and seven months of selected samples cover a bearish market in 1998and bullish market in 1999 and 2000. So this period represents a complete business cyclein the Taiwan stock market.
2.2. Methodology
The methodology employed in this study is based on Granger (1969). Other causalitytesting methods reported in the literature include the test proposed by Sims (1972) and theprocedure suggested by Pierce and Haugh (1977). However, Granger’s tests are employedbecause they are superior to Sims’ (see Geweke, Meese and Dext (1983), and accordingto Hardouvelis (1988)), they perform well for small samples. However, it is necessary totest if the variables are stationary or not before Granger tests. If they are nonstationary, it isappropriate to specify by means of the vector error-correction models (Engle and Granger,1987) to explore Granger causality relationship between GDRs or ADRs and the prices ofunderlying shares.
2.2.1 Unit root test. The assumptions of the classical regression model necessitate thatthe time series be stationary and the errors have a zero mean and finite variance. In thepresence of nonstationary variables, there might be what Granger and Newbold (1974) calla spurious regression.3 Thus, the first step in the analysis is to check if the structure of thereturns series is stationary by the augmented Dickey-Fuller test (ADF).
The augmented Dickey-Fuller test can be applied both in the case of a lower and a higherautoregressive (AR) process. The following equation presents a higher AR process version(with a constant and a time trend) of the Dickey Fuller test:
�yt = a0 + a1t + γ yt−1 +P∑
i=2
βi · �yt−i+1 + et , et ∼ iid(0, σ 2)
190 CHEN, CHOU AND YANG
where yt represents a time series, � implies first difference, and t is the time trend. Accordingto Said and Dickey (1984) the ADF test procedure is valid for a general ARMA process inthe errors. The null hypothesis in the ADF test is unit root (γ = 0). For yt to be stationary,γ should be negative and significantly different from zero.
2.2.2 Cointegration tests. A system of nonstationary individual stock price in levels can,however, share common stochastic trends. Put simply, two nonstationary time series arecointegrated if a linear combination of two variables is stationary, that is, converges to anequilibrium over time. The main idea behind cointegration is a specification of models thatincludes beliefs about the movements of variables relative to each other in the long-run,such as the price of Taiwan’s stock and GDRs or ADRs. Thus a common stochastic trendin a system of stock prices can be interpreted to mean that the stochastic trend in Taiwan’sstock price is related to the GDRs or ADRs trend. There exists more than one method ofconducting cointegration tests. The long-run relationship tests in this paper are conducted bymeans of the method developed by Johansen (1988) and Johansen and Juselius (1990). TheJohansen maximum likelihood approach sets up the nonstationary time series as a vectorautoregressive (VAR). The model is also called vector error-correction model (VECM):
�Xt = c +N∑
i=1
i�Xt−i +∏
Xt−1 + ηt , ηt ∼ niid(0, δ)
where Xt is a vector of nonstationary (in levels) variables, � implies first difference andc is the constant term. The information on the coefficient matrix between the levels of theseries
∏is decomposed as
∏ = αβ ′ where the relevant elements of the α matrix are theadjustment coefficients and the β matrix contains the cointegrating vectors. α and β arep × r matrices of full rank. If r = 0, then
∏ = 0, and there exists no linear combination ofthe elements of Xt that is stationary. At the other extreme, if rank (
∏) = p, Xt is itself a
stationary process. In the intermediate case, when 0 < r < p, there exist r stationary linearcombinations of the elements of Xt . The constant term is included to capture the trendingcharacteristic of the time series involved. The Johansen method provides the trace test andto determine the number of cointegrating vector. It is defined as:
trace statistic = −Tp∑
i=r+1
ln(1 − λi )
for r = 0, 1, 2, . . . , p − 1 where λi is the i th largest eigenvalue. The critical values forthe trace statistic are reported by Osterwald-Lenum (1992), not those tabulated in Johansenand Juselius (1990). The trace statistic generally has greater power when the λi s are evenlydistributed.
2.2.3 Granger causality test. The objective of this section is to investigate causal relationsbetween the returns on GDRs or ADRs and their respective underlying Taiwan securities.The methodology employed in this study is based on Granger (1969). The Granger Causality
PRICE TRANSMISSION EFFECT BETWEEN GDRS 191
tests with difference form involve the estimation of the following equation:
R fi,t = λ
f0 + λ
f1 R f
i,t−1 + λf2 Rd
i,t + εf
i,t (1)
Rdi,t = λd
0 + λd1 Rd
i,t−1 + λd2 R f
i,t−1 + εdi,t (2)
where
Ri,t = log(p f (d)i,t ) − log(p f (d)
i,t )
p f (d)i,t : p f
i,t denotes the close price of GDRs or ADRs in day t;pd
i,t denotes the close price of the underlying security in day t.
Within the same calendar day, Taiwan market closes earlier than US and European markets,so foreign investors can observe the returns of both markets on the same day. On theother hand, the domestic investors observe the returns of the preceding day overseas. Theregression model is set up so that R f
t regresses on R ft−1 and Rd
t , but Rdt regresses on Rd
t−1 andR f
t−1. The Granger causality tests with the VECM involve the estimation of the followingequation while the data exist cointegration relationship:
R fi,t = k f
0 + k f1 R f
i,t−1 + k f2 Rd
i,t + k f3
(log p f
i,t−1 − c0 − c1 log pdi,t−1
) + εf
i,t (3)
Rdi,t = kd
0 + kd1 Rd
i,t−1 + kd2 R f
i,t−1 + kd3
(log pd
i,t−1 − d0 − d1 log p fi,t−1
) + εdi,t (4)
The lambdas and kappas are the parameters to be estimated. In the above estimation ofequation (1) to equation (4), if the estimated coefficients λ
f2 and k f
2 of equations (1) and (3)are statistically significant while the estimated coefficients λd
2 and kd2 of equations (2) and
(4) are not statistically significant, then the results suggest a uni-directional causality, in theGranger sense, from the Taiwan stock returns to change GDR stock returns. In terms coinedby Garbade and Silber (1979), the underlying security market is dominant and the overseassecurity market is a satellite. If, on the other hand, the estimated coefficients λd
2 and kd2 of
equations (2) and (4) are statistically significant while the estimated coefficients λf2 and
k f2 of equations (1) and (3) are not statistically significant, then uni-directional causality
exists from changes in GDRs or ADRs to Taiwan’s stock returns. If the four coefficientsare statistically significant in equations (1) and (4), then the data provide evidence of bi-directional causality. Absence of directional causality is indicated when the set of parametersλd
2 , kd2 , λ
f2 and k f
2 are statistically insignificant. Finally, both kd3 and k f
3 represent the speedof adjustment coefficient for reflecting the long-run disequilibrium in the prices betweenthe underlying stock and the GDRs or ADRs. c0, d0, c1, d1 are cointegrating coefficients.
2.2.4 The impact of premium and net buy on Granger causality
The premium effect. Several studies on the pricing behavior of dual listed internationalsecurities do not find any significant difference between the domestic price and exchangerate adjusted price of the same security listed in an overseas market.4 Park and Tavakkol
192 CHEN, CHOU AND YANG
(1994) examine the exchange rate adjusted returns of Japanese ADRs and the underly-ing stocks and find no significant differences between the two. On the other hand, Millerand Morey (1996) find intra-day pricing differences between the British ADRs and theunderlying securities. This paper examines whether the price transmission effect is signifi-cantly amplified while the magnitude of the premium or discounts in overseas-listed stocksincreases. We set up the estimation of the following equations:
R fi,t = λ
f0 + λ
f1 R f
i,t−1 + (β
f0 + β
f1 ∗ prem f
t
)Rd
i,t + εf
i,t (5)
Rdi,t = λd
0 + λd1 Rd
i,t−1 + (βd
0 + βd1 ∗ premd
t−1
)R f
i,t−1 + εdi,t (6)
R fi,t = k f
0 + k f1 R f
i,t−1 + (φ
f0 + φ
f1 · prem f
t
)Rd
i,t
+ k f3
(log p f
i,t−1 − c0 − c1 log pdi,t−1
) + εf
i,t (7)
Rdi,t = kd
0 + kd1 Rd
i,t−1 + (φd
0 + φd1 · premd
t
)R f
i,t−1
+ kd3
(log pd
i,t−1 − d0 − d1 log p fi,t−1
) + εdi,t (8)
where
prem ft = p f
i,t−1 · EXt − pdi,t
pdi,t
(9)
premdt = p f
i,t−1 · EXt−1 − pdi,t−1
pdi,t−1
(10)
EXt : exchange rate in day t
We further discuss the above equations in two cases: one, both the premium and Rdt
have the same sign. For example, if foreign investors observe that there exists positivepremium and the underlying stock price rises, then through the market mechanism ofarbitrary transactions, the prices of ADRs or GDRs will not change or even go down inorder to shrink the price gap between the underlying stock and ADRs or GDRs. Therefore,both β
f1 and φ
f1 will be expected to be equal to or less than zero. Second, the premium and
Rdt have the opposite sign. In this case, if foreign investors find that there exists positive
premium but the underlying stock price is down, then the prices of ADRs or GDRs willdecline. Thus, both β
f1 and φ
f1 will be expected to be greater than or equal to zero. According
to the above, if both the premium and R ft−1 are the same sign, for example, domestic investors
observe that there exists positive premium and ADRs or GDRs price is also up, then theunderlying stock price will go up and both βd
1 and φd1 will be expected to be greater than or
equal to zero. On the contrary, when both the premium and R ft−1 are the opposite sign, for
example, domestic investors observe that there exists positive premium but ADRs or GDRsprice is down, the underlying stock price will not change or even go up in order to shrinkthe price gap. Therefore, both βd
1 and φd1 will be expected to be equal to or less than zero.
PRICE TRANSMISSION EFFECT BETWEEN GDRS 193
Net buy effect. During these years QFIIs are more aggressive to increase their investmentpositions in Taiwan. In consequence of the substantial growth of QFIIs’ portfolio holdingin Taiwan’s equities, the net buy or net sell of QFIIs’ daily trading has become an importantinvestment signal for all investors participating in the Taiwan market. Stock with positivenet buy by QFIIs is often followed by an increase in share price on next trading day, inparticular, for those stocks underlying ADRs or GDRs. Thus, we also add the variable ofQFIIs’ daily net buy into our empirical models to further examine the impact of net buy onGranger causality relationship between the underlying stock and the ADRs or GDRs. Themodels are as follows:
R fi,t = λ
f0 + λ
f1 R f
i,t−1 + (δ
f0 + δ
f1 · BS f
t
)Rd
i,t + εf
i,t (11)
Rdi,t = λd
0 + λd1 Rd
i,t−1 + (δd
0 + δd1 · BSd
t−1
)R f
i,t−1 + εdi,t (12)
R fi,t = k f
0 + k f1 R f
i,t−1 + (ω
f0 + ω
f1 · BS f
t
)Rd
i,t
+ k f3
(log p f
i,t−1 − c0 − c1 log pdi,t−1
) + εf
i,t (13)
Rdi,t = kd
0 + kd1 Rd
i,t−1 + (ωd
0 + ωd1 · BSd
t−1
)R f
i,t−1
+ kd3
(log pd
i,t−1 − d0 − d1 log p fi,t−1
) + εdi,t (14)
where
BS ft : the volume of net buy for underlying stock by QFIIs in Taiwan observed
by foreigners in day t; i.e., the net shares of total bought minus total sold forunderlying stock by all QFIIs in day t
BSdt−1: the volume of net buy for underlying stock by QFII in Taiwan observed by
the domestic investors in day t
In the above estimation of equations (11) and (13), if QFIIs buy net shares for theunderlying stock in Taiwan’s market and at the same time the return of the underlyingstocks rises, the prices of GDRs or ADRs will rise, fall or hold steady, so the signs of δ
f1
and w f1 can’t be determined. There are two reasons to explain the above phenomenon. One
is that the prices of the underlying stocks are relatively undervalued as a consequence ofQFIIs’ net buy in order to get arbitrage profit. Of course the gap between both prices can beshrunk by arbitrage trading. The other reason is that the prospect of the underlying companydemonstrates such potential that QFIIs purchase these underlying stocks, and as a result theprices of the underlying stocks and ADRs or GDRs simultaneously rise. On the other side, inequations (12) and (14), if the return of GDRs or ADRs is negative and QFIIs buy net sharesof the underlying stock, the return of the underlying stock will be positive, negative or zero,so the signs of δd
1 and wd1 are also uncertain. The reason for the uncertainty is that the
prices of the underlying stocks are undervalued and QFIIs buying net shares will lead to theprice of the underlying stocks rising. However, the underlying stock prices may also dropto reflect the falling price of GDRs or ADRs.
194 CHEN, CHOU AND YANG
Robust test. Finally, our model simultaneously includes the premium and net buy factors toexamine if the model is robust. This is reflected in the estimation of equations (15) to (18):
R fi,t = λ
f0 + λ
f1 R f
i,t−1 + (θ
f0 + θ
f1 · prem f
t + θf
2 · BS ft
)Rd
i,t + εf
i,t (15)
Rdi,t = λd
0 + λd1 Rd
i,t−1 + (θd
0 + θd1 · premd
t + θd2 · BSd
t−1
)R f
i,t−1 + εdi,t (16)
R fi,t = k f
0 + k f1 R f
i,t−1 + (γ
f0 + γ
f1 · prem f
t + γf
2 · BS ft
)Rd
i,t
+ k f3
(log p f
i,t−1 − c0 − c1 log pdi,t−1
) + εf
i,t (17)
Rdi,t = kd
0 + kd1 Rd
i,t−1 + (γ d
0 + γ d1 · premd
t + γ d2 · BSd
t−1
)R f
i,t−1
+ kd3
(log pd
i,t−1 − d0 − d1 log p fi,t−1
) + εdi,t (18)
3. Empirical results
3.1. Unit root and cointegration tests results
As stated earlier, all series are applied in logarithmic form. As required of all cointegrationtests, the series of stock price must first be inspected for the presence of unit roots. Table 4
Table 4. Augmented Dickey-Fuller tests for a unit roota
Corporation Code Taiwan GDRs or ADRs
1102 −2.721650 −2.8850681227 −2.532691 −2.3829611402 −1.494700 −1.2529381605 −1.586543 −1.8456392002 −2.816041 −2.4042932306 −1.864972 −2.0642312311 −1.910316 −2.0500462317 −1.981890 −1.3528532324 −1.791862 −1.9354912325 −3.505806** −3.729858**2330 −1.978247 −2.5876202337 −1.401737 −1.5307772342 −2.154823 −2.2527262344 −3.093396 −3.0082042345 −2.522986 −2.4944032346 −1.172046 −1.1136042347 −3.763689** −3.686703**2357 −2.081670 −2.1916812603 −2.337686 −1.7147612609 −3.341800** −3.717536**5346 −2.465527 −2.501067
aThe entry in each cell is the ADF statistic.**Implies rejection of the null at 5% level.
PRICE TRANSMISSION EFFECT BETWEEN GDRS 195
Table 5. Cointegration test results (Lags in the VAR = 1)
r a = 0 r =< 1
Corporation Code Eigenvalue Trace Statistic Eigenvalue Trace Statistic
1102 0.028652 26.32853*** 0.013037 8.188700***1227 0.018837 12.46983 0.000966 0.6033291402 0.071200 12.34016 0.016081 2.2210291605 0.034779 23.29036*** 0.001924 1.2020122002 0.031511 24.74242*** 0.007604 4.763040**2306 0.199605 14.02069*** 0.002038 1.2730442311 0.033467 24.49023*** 0.005194 3.2493192317 0.090624 16.04527** 0.020229 2.8407112324 0.083378 14.61861 0.028546 3.6490782330 0.044065 28.63125*** 0.001040 0.6459712337 0.215139 13.00895*** 0.000003 0.0018612342 0.114581 20.31530*** 0.005262 0.8442012344 0.065481 20.70226*** 0.000153 0.0465712345 0.011082 8.226734 0.002038 1.2729492346 0.013422 11.19552 0.004419 2.7636382357 0.047425 32.22369*** 0.003050 1.9060042603 0.022752 18.79700** 0.007083 4.435474**5346 0.060651 10.65297 0.013973 1.955989
ar is hypothesized number of cointegrating relationships.*** and **imply rejection of the null at 1% and 5% level, respectively.
presents the results from ADF tests. We employed Akaike’s information criterion to selectthe appropriate lag lengths. For most of the series, we are unable to reject the unit roothypothesis, but there are some exceptions. In other words, most of time series data are I (1),but some series data are I (0). When the series are stable, they don’t need the cointegrationtest.
Table 5 presents the results from the cointegration tests. In this paper, we used trace testto determine the number of cointegrating vectors. Results show a cointegration relationshipexisting between most Taiwan stock prices and their GDR’s or ADR’s prices. In otherwords, we are able to find a stationary long run relationship between both. So, we shouldemploy VECM to test Granger causality relationship for existing cointegration time series.It is appropriate to simultaneously consider long- and short-term effects. For time serieswithout cointegration, they directly take the first difference form to test Granger causalityrelationship. As reported in Table 5, trace tests indicate that at least one cointegrationrelationship exists for twelve of the firms. Each firm’s cointegrating vector is calculated andincorporated in the VAR model estimation to capture the long run equilibrium relationship.
3.2. Granger causality tests results
Table 6 shows the results from Granger Causality with difference form. Only sevensamples have bi-directional causality but fourteen samples among twenty-one total show
196 CHEN, CHOU AND YANG
Table 6. Granger causality tests results—difference form
R fi,t = λ
f0 + λ
f1 R f
i,t−1 + λf2 Rd
i,t + εf
i,t
Rdi,t = λd
0 + λd1 Rd
i,t−1 + λd2 R f
i,t−1 + εdi,t
λf0 λ
f1 λ
f2
Corporation Code D. V.a λd0 λd
2 λd1
1102 R ft −0.0002 −0.0381 0.8993
(0.7954) (0.1667) (0.0000)***Rd
t −0.0005 0.1077 −0.0446(0.6563) (0.0241)** (0.4457)
1227 R ft −0.0002 0.0844 0.7185
(0.8390) (0.0073)*** (0.0000)***Rd
t −0.0006 0.1447 0.0154(0.5680) (0.0010)*** (0.7632)
1402 R ft 0.0001 0.0508 0.7535
(0.9609) (0.4282) (0.0000)***Rd
t 0.0008 0.1383 −0.0683(0.8085) (0.1808) (0.5582)
1605 R ft 0.0001 −0.0282 0.9315
(0.9272) (0.1644) (0.0000)***Rd
t 0.0005 0.0931 −0.1169(0.6575) (0.2052) (0.1407)
2002 R ft −0.0000 0.0164 1.0482
(0.9792) (0.5628) (0.0000)***Rd
t 0.0002 0.1391 0.1770(0.8430) (0.0003)*** (0.0018)***
2306 R ft 0.0000 −0.1074 0.9934
(0.9536) (0.0000)*** (0.0000)***Rd
t 0.0009 0.1744 −0.1645(0.5259) (0.0146) (0.0422)**
2311 R ft 0.0001 0.0156 0.9501
(0.9152) (0.5719) (0.0000)***Rd
t 0.0004 0.1021 −0.1039(0.7498) (0.0216) (0.0742)∗
2317 R ft −0.0003 0.1373 0.8626
(0.8942) (0.0305)** (0.0000)***Rd
t 0.0021 0.1098 0.0003(0.4069) (0.1669) (0.9975)
2324 R ft 0.0002 0.0530 1.0029
(0.9188) (0.2725) (0.0000)***Rd
t −0.0006 0.0938 −0.1403(0.8474) (0.5136) (0.4079)
2325 R ft 0.0000 −0.0440 0.9086
(0.9892) (0.1629) (0.0000)***Rd
t 0.0003 0.0594 −0.0370(0.8375) (0.0896)* (0.4684)
2330 R ft 0.0009 −0.1244 0.7192
(0.5526) (0.0008)*** (0.0000)***Rd
t 0.0011 0.2096 −0.0637(0.3371) (0.0000)*** (0.1398)
PRICE TRANSMISSION EFFECT BETWEEN GDRS 197
Table 6. (Continued)
λf0 λ
f1 λ
f2
Corporation Code D. V.a λd0 λd
2 λd1
2337 R ft 0.0005 −0.1373 0.7759
(0.7222) (0.0000)*** (0.000)***Rd
t 0.0011 0.1300 −0.0548(0.5018) (0.0080)** (0.3338)
2342 R ft 0.0007 −0.0334 0.9720
(0.6420) (0.3512) (0.0000)***Rd
t 0.0056 0.0230 0.0473(0.0815)* (0.8873) (0.7873)
2344 R ft 0.0004 −0.0840 1.0020
(0.7229) (0.0074) (0.0000)***Rd
t 0.0033 0.1238 −0.1679(0.0767) (0.1577) (0.1094)
2345 R ft −0.0006 −0.0753 1.0214
(0.4457) (0.0001)*** (0.0000)***Rd
t 0.0004 0.0116 0.0874(0.7739) (0.8720) (0.2932)
2346 R ft −0.0004 −0.0876 0.9839
(0.6363) (0.0000)*** (0.0000)***Rd
t −0.0000 0.0091 0.1163(0.9796) (0.9094) (0.1848)
2347 R ft 0.0000 −0.1148 0.9800
(0.9753) (0.0000)*** (0.0000)***Rd
t 0.0004 0.0769 0.0748(0.7415) (0.2383) (0.3125)
2357 R ft 0.0003 −0.0628 1.0568
(0.7993) (0.0355)** (0.0000)***Rd
t 0.0010 0.1374 −0.0739(0.3419) (0.0001)*** (0.1711)
2603 R ft −0.0000 0.0439 0.8715
(0.9898) (0.1034) (0.0000)***Rd
t −0.0002 0.0722 0.0435(0.8058) (0.1542) (0.4660)
2609 R ft 0.0006 −0.0094 0.8723
(0.4126) (0.7111) (0.0000)***Rd
t −0.0001 0.0642 −0.0199(0.9626) (0.2578) (0.7549)
5346 R ft 0.0021 −0.0297 0.9594
(0.2426) (0.4854) (0.0000)***Rd
t 0.0033 0.0259 0.0199(0.3150) (0.8617) (0.9020)
aD. V. represents dependent variable.*Significant at the 10% level.**Significant at the 5% level.***Significant at the 1% level.(.) represents p-value.
198 CHEN, CHOU AND YANG
uni-directional causality. This demonstrates that most of the time the Taiwan stock re-turns change GDRs or ADRs returns. In addition, the estimated coefficient λd
2 approaches100 percent, despite the estimated coefficient λ
f2 being near to zero. So our empirical re-
sults mostly support Garbade and Silber’s (1979) idea that the underlying security marketis dominant and the overseas security market is a satellite.
To continue, the cointegration relationship data employ the Granger causality test withVECM which simultaneously considers long run and short run effects between the returnsof Taiwan stock and GDRs or ADRs. Table 7 presents the empirical results with VECM. Inthe short run, it shows similar results that Taiwan stock returns significantly influence GDRsor ADRs returns but not the reverse. In the long run, all the coefficients for the speed of
Table 7. Granger causality tests results—VECM
R fi,t = k f
0 + k f1 R f
i,t−1 + k f2 Rd
i,t + k f3 (log p f
i,t−1 − c0 − c1 log pdi,t−1) + ε
fi,t
Rdi,t = kd
0 + kd1 Rd
i,t−1 + kd2 R f
i,t−1 + kd3 (log pd
i,t−1 − d0 − d1 log p fi,t−1) + εd
i,t
κf
0 κf
1 κf
2 κf
3
Corporation Code D.Va κd0 κd
2 κd1 κd
3
1102 R ft −0.0002 −0.0240 0.9088 −0.0545
(0.8056) (0.3817) (0.0000)*** (0.0000)***Rd
t 0.0927 −0.0324 −0.0302(0.0558)* (0.5812) (0.0857)*
1605 R ft 0.0038 −0.0274 0.9311 −0.0007
(0.6412) (0.1784) (0.0000)*** (0.6455)Rd
t 0.0187 0.0925 −0.1125 −0.0064(0.2620) (0.2078) (0.1565) (0.2749)
2202 R ft −0.0000 0.0277 1.0570 −0.0551
(0.9763) (0.3236) (0.0000)*** (0.0000)***Rd
t 0.0002 0.1333 −0.1699 −0.0146(0.8434) (0.0007)*** (0.0031)*** (0.3940)
2306 R ft −0.0001 −0.0062 1.0453 −0.5855
(0.8883) (0.7336) (0.0000)*** (0.0000)***Rd
t 0.0009 −0.0207 0.0012 −0.3905(0.5105) (0.8064) (0.9889) (0.0000)***
2311 R ft −0.0001 0.0318 0.9659 −0.0425
(0.9250) (0.2527) (0.0000)*** (0.0004)***Rd
t 0.0004 0.0803 −0.0930 −0.0437(0.7441) (0.0724)* (0.1084) (0.0008)***
2317 R ft −0.0005 0.1613 0.8939 −0.0467
(0.8620) (0.0134)** (0.0000)*** (0.1231)Rd
t 0.0021 0.0461 0.0350 −0.1167(0.3888) (0.5669) (0.7434) (0.0047)***
2330 R ft 0.0009 −0.1134 0.7351 −0.0191
(0.5712) (0.0028)*** (0.0000)*** (0.1663)Rd
t 0.0012 0.1740 −0.0648 −0.0607(0.2966) (0.0000)*** (0.1255) (0.0000)***
PRICE TRANSMISSION EFFECT BETWEEN GDRS 199
Table 7. (Continued)
κf
0 κf
1 κf
2 κf
3
Corporation Code D.Va κd0 κd
2 κd1 κd
3
2337 R ft 0.0003 −0.0133 0.8312 −0.5796
(0.8124) (0.6487) (0.0000)*** (0.0000)***Rd
t −0.0048 0.0454 0.0288 −0.1681(0.0790)* (0.4330) (0.6542) (0.0073)***
2342 R ft 0.0004 −0.0023 0.9973 −0.2655
(0.7952) (0.9475) (0.0000)*** (0.0000)***Rd
t 0.0057 −0.1091 0.1679 −0.2772(0.0712)* (0.5285) (0.3598) (0.0418)**
2344 R ft 0.0003 −0.0490 1.0169 −0.1630
(0.8205) (0.1131) (0.0000)*** (0.0000)***Rd
t 0.0006 0.0931 −0.1434 −0.0629(0.8471) (0.3081) (0.1794) (0.2395)
2357 R ft 0.0003 −0.0396 1.0849 −0.0663
(0.8356) (0.1880) (0.0000)*** (0.0001)***Rd
t 0.0010 0.1030 −0.0470 −0.0799(0.3304) (0.0045)*** (0.3828) (0.0001)***
2603 R ft −0.0000 0.0362 0.8796 −0.0200
(0.9899) (0.1782) (0.0000)*** (0.0018)***Rd
t −0.0003 0.0691 0.0580 −0.0329(0.8078) (0.1720) (0.3332) (0.0338)
aD. V. represents dependent variable.*Significant at the 10% level.**Significant at the 5% level.***Significant at the 1% level.(.) represents p-value.
adjustment, kd3 ’s and k f
3 ’s, are negative and most of them are significant. This result indicatesthat once the return relationship of the underlying stock and the GDRs or ADRs deviatesfrom the long run cointegrated equilibrium, both markets will make opposite adjustment toreestablish the equilibrium in next period. In a word, Table 7 shows that price causality indually listed stocks is mostly unidirectional from the domestic market to the foreign marketand the prices of both markets will adjust to a long run cointegrated equilibrium.
3.3. Results of the premium and net buy effect
Table 8 reveals the results from Granger Causality involving the premium in overseas-listed stocks. The estimated coefficient β
f0 or φ
f0 is significant and approaches to 1, but
the estimated coefficient βd0 or φd
0 is mostly not significant and near to zero. So the pricetransmission effect is still unidirectional from the domestic market to the foreign market.The transmission effect of seven samples is influenced by the premium in overseas-listed
200 CHEN, CHOU AND YANGTa
ble
8.C
hang
ein
Gra
nger
caus
ality
resu
lts—
Prem
ium
Rf i,t=
λf 0+
λf 1
Rf i,t−
1+
( βf 0+
βf 1∗p
rem
f t
) Rd i,
t+
εf i,t
(5)
Rd i,
t=
λd 0
+λ
d 1R
d i,t−
1+
( βd 0
+β
d 1∗p
rem
d t−1
) Rf i,t−
1+
εd i,
t(6
)
Rf i,t=
kf 0+
kf 1
Rf i,t−
1+
( φf 0+
φf 1·p
rem
f t
) Rd i,
t+
kf 3
( log
pf i,t−
1−
c 0−
c 1lo
gpd i,
t−1
) +ε
f i,t
(7)
Rd i,
t=
kd 0+
kd 1R
d i,t−
1+
( φd 0
+φ
d 1·p
rem
d t
) Rf i,t−
1+
kd 3
( log
pd i,t−
1−
d 0−
d 1lo
gp
f i,t−
1
) +ε
d i,t
(8)
Cor
pora
tion
λf 0(κ
f 0)
λf 1(κ
f 1)
βf 0(φ
f 0)
βf 1(φ
f 1)
κf 3
Cod
eD
.V.a
λd 0(κ
d 0)
βd 0(φ
d 0)
λd 1(κ
d 1)
βd 1(φ
d 1)
κd 3
1102
Rf t
−0.0
009
−0.0
133
0.95
80−1
.207
1−0
.057
2(0
.297
0)(0
.627
9)(0
.000
0)**
*(0
.000
3)**
*(0
.000
0)**
*R
d t−0
.000
70.
0347
−0.0
107
0.63
73−0
.029
2(0
.514
9)(0
.540
8)(0
.858
1)(0
.052
5)*
(0.0
961)
*
1227
Rf t
−0.0
007
0.08
110.
8006
−0.5
703
(0.5
076)
(0.0
098)
***
(0.0
000)
***
(0.0
187)
**R
d t−0
.000
60.
1650
0.01
57−0
.164
7(0
.588
4)(0
.001
7)**
*(0
.758
4)(0
.474
5)
1402
Rf t
−0.0
000
0.05
220.
7713
−0.1
063
(0.9
902)
(0.4
231)
(0.0
000)
***
(0.8
913)
Rd t
0.00
03−0
.241
6−0
.009
41.
5740
(0.9
375)
(0.1
719)
(0.9
352)
(0.0
095)
***
1605
Rf t
0.00
38−0
.027
40.
9315
−0.0
000
−0.0
007
(0.6
419)
(0.1
798)
(0.0
000)
***
(0.9
608)
(0.6
462)
Rd t
0.01
880.
1359
−0.1
383
−0.7
969
−0.0
064
(0.2
596)
(0.1
022)
(0.0
946)
*(0
.265
3)(0
.275
0)
2002
Rf t
−0.0
000
0.02
941.
0946
−0.0
000
−0.0
551
(0.9
902)
(0.2
967)
(0.0
000)
***
(0.4
570)
(0.0
000)
***
Rd t
0.00
010.
0863
−0.1
410
0.37
27−0
.010
5(0
.945
8)(0
.077
0)*
(0.0
189)
**(0
.109
8)(0
.545
8)
PRICE TRANSMISSION EFFECT BETWEEN GDRS 20123
06R
f t−0
.000
1−0
.006
41.
0349
0.00
00−0
.584
5(0
.865
6)(0
.722
3)(0
.000
0)**
*(0
.506
4)(0
.000
0)**
*R
d t0.
0001
−0.0
315
1.02
412.
9732
−0.3
836
(0.9
620)
(0.7
080)
(0.7
873)
(0.0
118)
**(0
.000
0)**
*
2311
Rf t
0.00
04−0
.092
70.
0783
0.01
01−0
.043
7(0
.752
0)(0
.110
6)(0
.197
3)(0
.960
1)(0
.002
9)**
*R
d t0.
0004
0.07
83−0
.092
70.
0101
−0.0
437
(0.7
520)
(0.1
973)
*(0
.110
6)(0
.960
1)(0
.002
8)**
*
2317
Rf t
−0.0
005
0.15
830.
8665
0.00
01−0
.046
0(0
.846
7)(0
.016
2)**
(0.0
000)
***
(0.6
891)
(0.1
305)
Rd t
0.00
230.
0889
0.03
65−0
.118
2−0
.114
8(0
.369
6)(0
.609
1)(0
.734
0)(0
.780
8)(0
.006
2)**
*
2324
Rf t
0.00
020.
0665
0.95
320.
0004
(0.9
002)
(0.1
722)
(0.0
000)
***
(0.0
948)
*R
d t−0
.000
50.
1294
−0.1
587
−0.7
208
(0.8
618)
(0.4
380)
(0.3
661)
(0.6
715)
2325
Rf t
0.00
00−0
.037
50.
9880
−0.0
004
(0.9
947)
(0.2
338)
(0.0
000)
***
(0.0
080)
***
Rd t
0.00
040.
0814
−0.0
418
−0.0
969
(0.7
754)
(0.1
392)
(0.4
199)
(0.6
033)
2330
Rf t
0.00
07−0
.121
00.
6540
0.00
02−0
.019
2(0
.666
3)(0
.001
6)**
*(0
.000
0)**
*(0
.183
2)(0
.163
3)R
d t0.
0012
0.18
50−0
.065
3−0
.028
8−0
.060
7(0
.282
2)(0
.000
2)**
*(0
.123
0)(0
.782
1)(0
.000
0)**
*
2337
Rf t
0.00
01−0
.015
90.
7948
0.00
02−0
.574
8(0
.938
9)(0
.587
0)(0
.000
0)**
*(0
.069
7)*
(0.0
000)
***
Rd t
−0.0
055
0.04
280.
0331
1.35
87−0
.172
3(0
.045
0)**
(0.4
587)
(0.6
071)
(0.0
902)
*(0
.005
9)**
*
(con
tinu
ed)
202 CHEN, CHOU AND YANG
Tabl
e8.
(Con
tinu
ed)
Cor
pora
tion
λf 0(κ
f 0)
λf 1(κ
f 1)
βf 0(φ
f 0)
βf 1(φ
f 1)
κf 3
Cod
eD
.V.a
λd 0(κ
d 0)
βd 0(φ
d 0)
λd 1(κ
d 1)
βd 1(φ
d 1)
κd 3
2342
Rf t
0.00
040.
0011
1.01
85−0
.000
2−0
.264
5(0
.784
5)(0
.973
9)(0
.000
0)**
*(0
.240
7)(0
.000
0)**
*R
d t0.
0054
−0.0
751
0.16
611.
5598
−0.2
759
(0.0
936)
*(0
.673
2)(0
.365
7)(0
.408
7)(0
.043
1)**
2344
Rf t
0.00
03−0
.049
11.
0055
0.00
00−0
.163
1(0
.812
1)(0
.112
6)(0
.000
0)**
*(0
.732
5)(0
.000
0)**
*R
d t0.
0006
0.12
34−0
.159
3−0
.956
2−0
.069
4(0
.844
7)(0
.207
3)(0
.141
8)(0
.385
2)(0
.199
1)
2345
Rf t
−0.0
006
−0.0
753
1.01
780.
0165
(0.4
784)
(0.0
001)
***
(0.0
000)
***
(0.8
725)
Rd t
0.00
050.
0364
0.08
35−0
.099
2(0
.758
7)(0
.663
0)(0
.316
6)(0
.558
3)
2346
Rf t
−0.0
004
−0.0
875
0.99
15−0
.000
2(0
.586
2)(0
.000
0)**
*(0
.000
0)**
*(0
.194
0)R
d t−0
.000
00.
0073
0.11
720.
0202
(0.9
794)
(0.9
331)
(0.1
893)
(0.9
564)
2347
Rf t
0.00
00−0
.114
10.
9831
−0.0
001
(0.9
570)
(0.0
000)
***
(0.0
000)
***
(0.5
909)
Rd t
0.00
030.
0632
0.08
260.
3805
(0.8
264)
(0.3
601)
(0.2
725)
(0.5
483)
PRICE TRANSMISSION EFFECT BETWEEN GDRS 203
2357
Rf t
0.00
02−0
.041
31.
0726
0.00
02−0
.066
3(0
.868
5)(0
.171
5)(0
.000
0)**
*(0
.450
7)(0
.000
1)**
*R
d t0.
0011
0.12
19−0
.055
5−0
.101
9−0
.080
8(0
.295
6)(0
.006
5)**
*(0
.314
5)(0
.472
3)(0
.000
1)**
*
2603
Rf t
−0.0
000
0.04
010.
9174
−0.0
002
−0.0
210
(0.9
859)
(0.1
361)
(0.0
000)
***
(0.0
522)
*(0
.001
0)**
*R
d t−0
.000
30.
0267
0.07
980.
2721
−0.0
330
(0.7
938)
(0.6
683)
(0.2
038)
(0.2
457)
(0.0
329)
**
2609
Rf t
0.00
04−0
.013
30.
8217
0.00
03(0
.588
4)(0
.598
0)(0
.000
0)**
*(0
.002
6)**
*R
d t−0
.000
1−0
.006
7−0
.001
10.
3824
(0.9
444)
(0.9
237)
(0.9
863)
(0.0
823)
*
5346
Rf t
0.00
20−0
.026
90.
9496
0.00
01(0
.260
8)(0
.535
8)(0
.000
0)**
*(0
.702
4)R
d t0.
0036
0.05
070.
0435
1.15
38(0
.280
6)(0
.736
1)(0
.789
7)(0
.290
2)
a D.V
.rep
rese
nts
depe
nden
tvar
iabl
e.*S
igni
fican
tatt
he10
%le
vel.
**Si
gnifi
cant
atth
e5%
leve
l.**
*Sig
nific
anta
tthe
1%le
vel.
(.)
repr
esen
tsp-
valu
e.
204 CHEN, CHOU AND YANG
Tabl
e9.
Cha
nge
inG
rang
erca
usal
ityre
sults
—ne
tbuy
Rf i,t=
λf 0+
λf 1
Rf i,t−
1+
( δf 0+
δf 1·B
Sf t
) Rd i,
t+
εf i,t
(11)
Rd i,
t=
λd 0
+λ
d 1R
d i,t−
1+
( δd 0
+δ
d 1·B
Sd t−1
) Rf i,t−
1+
εd i,
t(1
2)
Rf i,t=
kf 0+
kf 1
Rf i,t−
1+
( ωf 0+
ωf 1·B
Sf t
) Rd i,
t+
kf 3
( log
pf i,t−
1−
c 0−
c 1lo
gpd i,
t−1
) +ε
f i,t
(13)
Rd i,
t=
kd 0+
kd 1R
d i,t−
1+
( ωd 0
+ω
d 1·B
Sd t−1
) Rf i,t−
1+
kd 3
( log
pd i,t−
1−
d 0−
d 1lo
gp
f i,t−
1
) +ε
d i,t
(14)
Cor
pora
tion
λf 0(κ
f 0)
λf 1(κ
f 1)
δf 0(ω
f 0)
δf 1(ω
f 1)
κf 3
Cod
eD
.V.a
λd 0(κ
d 0)
δd 0(ω
d 0)
λd 1(κ
d 1)
δd 1(ω
d 1)
κd 3
1102
Rf t
−0.0
003
−0.0
232
0.91
140.
0000
−0.0
544
(0.7
715)
(0.3
999)
(0.0
000)
***
(0.6
190)
(0.0
000)
***
Rd t
−0.0
007
0.09
93−0
.027
60.
0000
−0.0
321
(0.5
349)
(0.0
403)
**(0
.637
6)(0
.028
1)**
(0.0
671)
*
1227
Rf t
−0.0
002
0.08
440.
7185
−0.0
000
(0.8
445)
(0.0
075)
***
(0.0
000)
***
(0.9
931)
Rd t
−0.0
005
0.13
950.
0188
−0.0
000
(0.6
787)
(0.0
017)
***
(0.7
130)
(0.3
648)
1402
Rf t
0.00
100.
0603
0.74
71−0
.000
0(0
.725
4)(0
.354
1)(0
.000
0)**
*(0
.376
8)R
d t0.
0019
0.11
36−0
.058
1−0
.000
0(0
.579
2)(0
.283
0)(0
.619
2)(0
.288
4)
1605
Rf t
0.00
40−0
.028
50.
9311
−0.0
001
−0.0
007
(0.6
288)
(0.1
618)
(0.0
000)
***
(0.1
673)
(0.6
381)
Rd t
0.01
870.
0934
−0.1
131
0.00
00−0
.006
4(0
.262
2)(0
.205
1)(0
.155
1)(0
.867
1)(0
.275
0)
PRICE TRANSMISSION EFFECT BETWEEN GDRS 205
2002
Rf t
−0.0
000
0.02
871.
0557
0.00
02−0
.054
5(0
.937
5)(0
.305
8)(0
.000
0)**
*(0
.285
9)(0
.000
0)**
*R
d t0.
0001
0.12
60−0
.160
0−0
.000
1−0
.012
6(0
.875
1)(0
.001
4)**
*(0
.005
6)**
*(0
.150
6)(0
.464
3)
2306
Rf t
−0.0
002
−0.0
062
1.04
700.
0001
−0.5
870
(0.8
095)
(0.7
317)
(0.0
000)
***
(0.1
820)
(0.0
000)
***
Rd t
1.00
10−0
.025
52.
0044
−0.0
001
−0.3
961
(0.4
729)
(0.7
628)
(0.9
609)
(0.4
134)
(0.0
000)
***
2311
Rf t
0.00
030.
0315
0.97
17−0
.000
1−0
.042
9(0
.813
8)(0
.257
5)(0
.000
0)**
*(0
.329
5)(0
.000
4)**
*R
d t0.
0001
0.95
020.
0307
0.00
01−0
.042
5(0
.952
4)(0
.000
0)**
*(0
.270
9)(0
.524
9)(0
.000
4)**
*
2317
Rf t
0.00
050.
1742
0.94
76−0
.000
5−0
.050
6(0
.855
8)(0
.006
8)**
*(0
.000
0)**
*(0
.010
4)**
(0.0
894)
*R
d t0.
0016
0.02
970.
0466
0.00
02−0
.117
6(0
.510
8)(0
.711
2)(0
.661
3)(0
.074
6)*
(0.0
041)
***
2324
Rf t
0.00
000.
0649
1.01
290.
0004
(0.9
906)
(0.1
700)
(0.0
000)
***
(0.0
059)
***
Rd t
−0.0
006
0.15
39−0
.192
4−0
.000
2(0
.844
9)(0
.316
0)(0
.273
9)(0
.265
9)
2325
Rf t
0.00
04−0
.046
20.
9171
−0.0
003
(0.8
030)
(0.1
420)
(0.0
000)
***
(0.0
377)
**R
d t0.
0002
0.05
90−0
.036
50.
0000
(0.8
526)
(0.0
923)
*(0
.474
6)(0
.805
7)
(con
tinu
ed)
206 CHEN, CHOU AND YANG
Tabl
e9.
(Con
tinu
ed)
Cor
pora
tion
λf 0(κ
f 0)
λf 1(κ
f 1)
δf 0(ω
f 0)
δf 1(ω
f 1)
κf 3
Cod
eD
.V.a
λd 0(κ
d 0)
δd 0(ω
d 0)
λd 1(κ
d 1)
δd 1(ω
d 1)
κd 3
2330
Rf t
0.00
10−0
.115
80.
7581
−0.0
005
−0.0
198
(0.5
326)
(0.0
023)
***
(0.0
000)
***
(0.0
426)
**(0
.149
9)R
d t0.
0013
0.20
31−0
.081
2−0
.000
4−0
.059
5(0
.234
8)(0
.000
0)**
*(0
.052
1)*
(0.0
000)
***
(0.0
000)
***
2337
Rf t
0.00
02−0
.012
70.
8331
−0.0
001
−0.5
776
(0.8
766)
(0.6
630)
(0.0
000)
***
(0.4
058)
(0.0
000)
***
Rd t
−0.0
048
0.04
570.
0286
0.00
00−0
.167
9(0
.081
8)*
(0.4
318)
(0.6
582)
(0.9
477)
(0.0
075)
***
2342
Rf t
0.00
04−0
.001
20.
9929
0.00
01−0
.265
3(0
.790
8)(0
.971
2)(0
.000
0)**
*(0
.493
0)(0
.000
0)**
*R
d t0.
0056
−0.1
101
0.16
270.
0001
−0.2
724
(0.0
758)
*(0
.525
7)(0
.376
8)(0
.592
7)(0
.046
4)**
2344
Rf t
0.00
03−0
.047
31.
0162
0.00
01−0
.164
1(0
.830
8)(0
.127
5)(0
.000
0)**
*(0
.436
9)(0
.000
0)**
*R
d t0.
0008
0.07
04−0
.123
70.
0002
−0.0
598
(0.7
932)
(0.4
528)
(0.2
538)
(0.2
943)
(0.2
638)
2345
Rf t
−0.0
008
−0.0
756
1.01
910.
0000
(0.3
488)
(0.0
001)
***
(0.0
000)
***
(0.2
919)
Rd t
−0.0
001
−0.0
117
0.10
130.
0001
(0.9
473)
(0.8
719)
(0.2
227)
(0.0
196)
**
PRICE TRANSMISSION EFFECT BETWEEN GDRS 207
2346
Rf t
−0.0
003
−0.0
892
0.98
280.
0001
(0.6
558)
(0.0
000)
***
(0.0
000)
***
(0.4
070)
Rd t
−0.0
000
0.01
130.
1124
0.00
01(0
.995
0)(0
.888
5)(0
.201
3)(0
.569
3)
2347
Rf t
0.00
00−0
.114
70.
9797
0.00
00(0
.997
6)(0
.000
0)**
*(0
.000
0)**
*(0
.863
7)R
d t0.
0004
0.07
620.
0751
0.00
00(0
.761
1)(0
.243
9)(0
.311
3)(0
.831
7)
2357
Rf t
0.00
06−0
.038
91.
0834
−0.0
002
−0.0
643
(0.6
252)
(0.1
950)
(0.0
000)
***
(0.0
858)
*(0
.000
2)**
*R
d t0.
0009
0.10
38−0
.044
10.
0001
−0.0
770
(0.4
311)
(0.0
042)
***
(0.4
128)
(0.1
710)
(0.0
003)
***
2603
Rf t
−0.0
000
0.03
540.
8802
0.00
01−0
.019
9(0
.957
6)(0
.188
2)(0
.000
0)**
*(0
.577
9)(0
.001
9)**
*R
d t−0
.000
20.
0673
0.05
94−0
.000
0−0
.032
5(0
.820
4)(0
.186
6)(0
.323
3)(0
.762
0)(0
.036
3)**
2609
Rf t
0.00
07−0
.013
00.
8748
−0.0
002
(0.3
496)
(0.6
081)
(0.0
000)
***
(0.1
131)
Rd t
−0.0
001
0.06
70−0
.022
60.
0001
(0.9
247)
(0.2
381)
(0.7
226)
(0.2
796)
5346
Rf t
0.00
22−0
.029
80.
9597
0.00
00(0
.235
0)(0
.486
0)(0
.000
0)**
*(0
.814
0)R
d t0.
0028
0.02
050.
0237
−0.0
002
(0.4
082)
(0.8
904)
(0.8
837)
(0.3
851)
a D.V
.rep
rese
nts
depe
nden
tvar
iabl
e.*S
igni
fican
tatt
he10
%le
vel.
**Si
gnifi
cant
atth
e5%
leve
l.**
*Sig
nific
anta
tthe
1%le
vel.
(.)
repr
esen
tsp-
valu
e.
208 CHEN, CHOU AND YANG
Tabl
e10
.C
hang
ein
Gra
nger
caus
ality
resu
lts—
prem
ium
and
netb
uy
Rf i,t=
λf 0+
λf 1
Rf i,t−
1+
( θf 0+
θf 1·p
rem
f t+
θf 2·B
Sf t
) Rd i,
t+
εf i,t
(15)
Rd i,
t=
λd 0
+λ
d 1R
d i,t−
1+
( θd 0
+θ
d 1·p
rem
d t+
θd 2
·BSd t−
1
) Rf i,t−
1+
εd i,
t(1
6)
Rf i,t=
kf 0+
kf 1
Rf i,t−
1+
( γf 0+
γf 1·p
rem
f t+
γf 2·B
Sf t
) Rd i,
t+
kf 3
( log
pf i,t−
1−
c 0−
c 1lo
gpd i,
t−1
) +ε
f i,t
(17)
Rd i,
t=
kd 0+
kd 1R
d i,t−
1+
( γd 0
+γ
d 1·p
rem
d t+
γd 2
·BSd t−
1
) Rf i,t−
1+
kd 3
( log
pd i,t−
1−
d 0−
d 1lo
gp
f i,t−
1
) +ε
d i,t
(18)
Cor
.λ
f 0(κ
f 0)
λf 1(κ
f 1)
θf 0(γ
f 0)
θf 1(γ
f 1)
θf 2(γ
f 2)
κf 3
Cod
eD
.V.a
λd 0(κ
d 0)
θd 0(γ
d 0)
λd 1(κ
d 1)
θd 1(γ
d 1)
θd 2(γ
d 2)
κd 3
1102
Rf t
−0.0
010
−0.0
127
0.95
97−1
.202
60.
0000
−0.0
572
(0.2
850)
(0.6
440)
(0.0
000)
***
(0.0
003)
***
(0.7
092)
(0.0
000)
***
Rd t
−0.0
009
0.04
34−0
.006
90.
6120
0.00
00−0
.031
1(0
.418
6)(0
.444
4)(0
.907
9)(0
.062
0)*
(0.0
331)
**(0
.075
9)*
1227
Rf t
−0.0
007
0.08
120.
8007
−0.5
711
0.00
00(0
.506
7)(0
.009
8)**
*(0
.000
0)**
*(0
.018
7)**
(0.9
338)
Rd t
−0.0
005
0.15
380.
0185
−0.1
088
−0.0
000
(0.6
745)
(0.0
050)
***
(0.7
185)
(0.6
547)
(0.4
754)
1402
Rf t
0.00
070.
0646
0.79
02−0
.260
4−0
.000
0(0
.828
2)(0
.332
2)(0
.000
0)**
*(0
.743
6)(0
.352
7)R
d t0.
0015
−0.2
796
0.00
361.
6147
−0.0
000
(0.6
548)
(0.1
191)
(0.9
751)
(0.0
078)
***
(0.2
178)
1605
Rf t
0.00
40−0
.028
40.
9316
−0.0
000
−0.0
001
−0.0
007
(0.6
294)
(0.1
631)
(0.0
000)
***
(0.9
572)
(0.1
677)
(0.6
388)
Rd t
0.01
880.
1373
−0.1
393
−0.8
032
0.00
00−0
.006
4(0
.259
8)(0
.100
0)(0
.093
0)*
(0.2
623)
(0.8
310)
(0.2
751)
PRICE TRANSMISSION EFFECT BETWEEN GDRS 209
2002
Rf t
−0.0
000
0.03
031.
0918
−0.0
001
0.00
02−0
.054
6(0
.951
5)(0
.281
4)(0
.000
0)**
*(0
.475
9)(0
.295
8)(0
.000
0)**
*R
d t0.
0000
0.08
95−0
.138
80.
3041
−0.0
000
−0.0
097
(0.9
511)
(0.0
673)
*(0
.020
9)**
(0.2
092)
(0.2
970)
(0.5
761)
2306
Rf t
−0.0
002
−0.0
065
1.03
760.
0000
0.00
01−0
.586
0(0
.791
1)(0
.721
4)(0
.000
0)**
*(0
.546
1)(0
.192
0)(0
.000
0)**
*R
d t0.
0001
−0.0
385
0.02
933.
1092
−0.0
001
−0.3
908
(0.9
193)
(0.6
481)
(0.7
429)
(0.0
088)
***
(0.2
745)
(0.0
000)
**
2311
Rf t
0.00
020.
0306
0.95
840.
0001
−0.0
001
−0.0
429
(0.8
434)
(0.2
725)
(0.0
000)
***
(0.6
022)
(0.3
659)
(0.0
004)
***
Rd t
0.00
040.
0818
−0.0
944
−0.0
053
0.00
00−0
.043
7(0
.768
1)(0
.188
8)(0
.106
8)(0
.980
0)(0
.794
8)(0
.002
9)**
2317
Rf t
0.00
050.
1759
0.96
18−0
.000
0−0
.000
5−0
.051
0(0
.844
2)(0
.006
9)**
*(0
.000
0)**
*(0
.848
1)(0
.011
5)**
(0.0
887)
*R
d t0.
0018
0.09
430.
0491
−0.1
790
0.00
02−0
.114
8(0
.468
7)(0
.584
5)(0
.645
8)(0
.671
9)(0
.071
1)*
(0.0
058)
***
2324
Rf t
0.00
010.
0778
0.96
450.
0004
0.00
04(0
.971
6)(0
.102
4)(0
.000
0)**
*(0
.095
5)*
(0.0
061)
***
Rd t
−0.0
005
0.24
81−0
.247
2−1
.560
2−0
.000
3(0
.874
5)(0
.188
8)(0
.187
1)(0
.388
1)(0
.180
4)
2325
Rf t
0.00
04−0
.039
61.
0009
−0.0
004
−0.0
003
(0.7
704)
(0.2
079)
(0.0
000)
***
(0.0
054)
***
(0.0
249)
**R
d t0.
0004
0.08
45−0
.041
9−0
.112
90.
0000
(0.7
871)
(0.1
294)
(0.4
192)
(0.5
554)
(0.7
097)
(con
tinu
ed)
210 CHEN, CHOU AND YANG
Tabl
e10
.(C
onti
nued
)
Cor
.λ
f 0(κ
f 0)
λf 1(κ
f 1)
θf 0(γ
f 0)
θf 1(γ
f 1)
θf 2(γ
f 2)
κf 3
Cod
eD
.V.a
λd 0(κ
d 0)
θd 0(γ
d 0)
λd 1(κ
d 1)
θd 1(γ
d 1)
θd 2(γ
d 2)
κd 3
2330
Rf t
0.00
08−0
.120
90.
6981
0.00
02−0
.000
4−0
.019
8(0
.602
8)(0
.001
6)**
*(0
.000
0)**
*(0
.353
4)(0
.073
8)*
(0.1
496)
Rd t
0.00
150.
2390
−0.0
834
−0.0
916
−0.0
004
−0.0
595
(0.1
838)
(0.0
000)
***
(0.0
465)
**(0
.375
9)(0
.000
0)**
*(0
.000
0)**
*
2337
Rf t
0.00
00−0
.015
30.
7971
0.00
02−0
.000
1−0
.573
0(0
.996
3)(0
.600
6)(0
.000
0)**
*(0
.074
5)*
(0.4
442)
(0.0
000)
***
Rd t
−0.0
055
0.04
280.
0331
1.35
86−0
.000
0−0
.172
3(0
.046
3)**
(0.4
607)
(0.6
081)
(0.0
908)
*(0
.999
7)(0
.006
0)**
*
2342
Rf t
0.00
040.
0019
1.01
38−0
.000
20.
0001
−0.2
643
(0.7
814)
(0.9
568)
(0.0
000)
***
(0.2
635)
(0.5
541)
(0.0
000)
***
Rd t
0.00
53−0
.079
20.
1627
1.40
150.
0001
−0.2
728
(0.0
954)
*(0
.658
0)(0
.377
5)(0
.471
2)(0
.723
4)(0
.046
5)**
2344
Rf t
0.00
03−0
.047
41.
0042
0.00
000.
0001
−0.1
642
(0.8
220)
(0.1
271)
(0.0
000)
***
(0.7
198)
(0.4
330)
(0.0
000)
***
Rd t
0.00
080.
0957
−0.1
383
−0.6
757
0.00
02−0
.064
9(0
.801
0)(0
.354
4)(0
.214
2)(0
.557
5)(0
.406
6)(0
.231
8)
2345
Rf t
−0.0
008
−0.0
755
1.01
170.
0332
0.00
00(0
.391
1)(0
.000
1)**
*(0
.000
0)**
*(0
.749
0)(0
.276
2)R
d t−0
.000
10.
0023
0.09
89−0
.054
30.
0001
(0.9
606)
(0.9
783)
(0.2
355)
(0.7
492)
(0.0
226)
**
PRICE TRANSMISSION EFFECT BETWEEN GDRS 211
2346
Rf t
−0.0
004
−0.0
891
0.99
04−0
.000
20.
0001
(0.6
051)
(0.0
000)
***
(0.0
000)
***
(0.2
012)
(0.4
252)
Rd t
−0.0
000
0.00
840.
1138
0.03
150.
0001
(0.9
949)
(0.9
228)
(0.2
036)
(0.9
322)
(0.5
671)
2347
Rf t
0.00
00−0
.114
10.
9829
−0.0
001
0.00
00(0
.978
1)(0
.000
0)**
*(0
.000
0)**
*(0
.593
5)(0
.872
6)R
d t0.
0003
0.06
270.
0827
0.37
610.
0000
(0.8
428)
(0.3
647)
(0.2
720)
(0.5
535)
(0.8
488)
2357
Rf t
0.00
06−0
.040
61.
0715
0.00
02−0
.000
2−0
.064
3(0
.655
4)(0
.178
4)(0
.000
0)**
*(0
.462
9)(0
.087
6)*
(0.0
002)
***
Rd t
0.00
100.
1357
−0.0
576
−0.1
711
0.00
01−0
.077
7(0
.382
8)(0
.002
9)**
*(0
.295
7)(0
.246
9)(0
.100
6)(0
.000
2)**
*
2603
Rf t
−0.0
001
0.03
930.
9181
−0.0
002
0.00
01−0
.020
9(0
.952
1)(0
.144
5)(0
.000
0)**
*(0
.051
4)*
(0.5
598)
(0.0
011)
***
Rd t
−0.0
003
0.02
610.
0805
0.26
82−0
.000
0−0
.032
8(0
.803
0)(0
.676
3)(0
.200
9)(0
.254
5)(0
.832
5)(0
.034
9)**
2609
Rf t
0.00
05−0
.016
70.
8248
0.00
03−0
.000
2(0
.512
3)(0
.509
5)(0
.000
0)**
*(0
.003
0)**
*(0
.132
5)R
d t−0
.000
1−0
.010
4−0
.002
60.
4211
0.00
02(0
.895
9)(0
.881
5)(0
.968
0)(0
.057
8)*
(0.1
854)
5346
Rf t
0.00
21−0
.027
00.
9501
0.00
010.
0000
(0.2
537)
(0.5
353)
(0.0
000)
***
(0.7
105)
(0.8
270)
Rd t
0.00
310.
0445
0.04
581.
0927
−0.0
002
(0.3
639)
(0.7
683)
(0.7
796)
(0.3
183)
(0.4
257)
a D.V
.rep
rese
nts
depe
nden
tvar
iabl
e.*S
igni
fican
tatt
he10
%le
vel.
**Si
gnifi
cant
atth
e5%
leve
l.**
*Sig
nific
anta
tthe
1%le
vel.
(.)
repr
esen
tsp-
valu
e.
212 CHEN, CHOU AND YANG
stocks in equations (5) and (7). On the other side, four samples report that the transmissioneffects have been changed in equations (6) and (8) when the premium is considered. Thatis, the premium has significant impact on international price transmission for one fifth toone third of samples. However the impact may amplify or reduce the transmission effectsince the signs of β
f1 (φ
f1 ) are ambiguous.
The results from Granger Causality involving QFII’s net buy are shown in Table 9. Taiwanstock market is still dominant and the foreign market is a satellite. The result is similar to thestated outcomes above. In addition, QFII’s net buy significantly influences the price trans-mission effect of twenty percent samples and the direction of net buy effect is also obscured.
Finally, Table 10 illustrates the results from Granger Causality, including both the pre-mium and QFII’s net buy factors. The outcomes are quite consistent with the Table 8 andTable 9. The premium and QFII’s net buy still have significant impact on international pricetransmission for the same samples. These results prove the robustness of our models.
4. Conclusion
This paper studies the price transmission effect between ADRs or GDRs and their re-spective underlying stock. The selected samples cover twenty-one sponsored DRs issuedby Taiwan listing companies from October 8, 1997 to May 31, 2000. We use Granger teststo examine causal relations between the returns of both capital markets. Error-correctionmodel is employed to analyze the long run causal relations when the stock returns data isnonstationary.
Results reveal unidirectional causality from Taiwan’s capital market to foreign markets.This asymmetry suggests the domestic market plays a dominant role relative to the for-eign market. At the same time, the prices of both markets will make opposite adjustmentto establish the long run cointegrated equilibrium. An additional finding is that both thepremium and QFII’s net buy have significant impacts on international price movement forover twenty percent samples. Empirical outcomes provide the evidence that the modelspresented in this paper are quite robust.
Notes
1. One anonymous reviewer indicates that the premium or discount is an important variable for closed-endfunds. However, there are a few differences in premium concept between ADRs and closed-end funds. ADRsand underlying securities are two separately traded assets and their prices may exist lead-lag relation in bothdirections, thus causes the possible price premium or discount across different markets. The net asset value(NAV) of a closed-end fund is not marketable. NAV may affect the market price of a closed-end but not thereverse.
2. QFIIs (Qualified Foreign Institutional Investors) refer to foreign banks, insurance companies, securities firms,fund management institutions and other investment institutions who meet the qualifications set by the Secu-rities and Futures Commission, Taiwan, R.O.C.
3. A spurious regression has a high R2, t-statistics that appear to be significant, but the results are without anyeconomic meaning. The regression output “looks good” because the least-squares estimates are not consistentand the customary tests of statistical inference do not hold.
4. For a review of arbitrage and market efficiency of ADR markets, see Karolyi (1998).
PRICE TRANSMISSION EFFECT BETWEEN GDRS 213
References
Barclay, M. J., R. H. Litzenberger and J. B. Warner, “Private Information, Trading Volume and Stock ReturnVariances.” Review of Financial Studies 3, 233–253, (1990).
Engle, R. E. and C. W. J. Granger, “Cointegration and Error-Correction: Representation, Estimation, and Testing.”Econometrica 55, 251–276.
Eun, C. and S. Shim, “International Transmission of Stock Price Movements.” Journal of Financial and Quanti-tative Analysis 24, 241–256, (1989).
Geweke, J., R. A. Meese and W. Dext, “Comparing Alternative Tests of Casuality in Temporal Systems: AnalyticResults and Experimental Evidence.” Journal of Econometrics 21, 161–194, (1983).
Garbade, K. D. and W. L. Silber, “Dominant and Satellite Markets: A Study of Dually-Traded Securities.” Reviewof Economics and Statistics 61, 455–460, (1979).
Granger, C. W. J., “Investigating Causal Relations by Econometric Models and Crosss-Spectral Models.” Econo-metrica 37, 424–438, (1969).
Granger, C. W. J. and P. Newbold, “Spurious Regressions in Econometrics.” Journal of Econometrics 2, 111–120,(1974).
Geweke, J., R. A. Meese and W. Dext, “Comparing Alternative Tests of Casuality in Temporal Systems: AnalyticResults and Experimental Evidence.” Journal of Econometrics 21, 161–194, (1983).
Hamao, Y., R. Masulis and V. Ng, “Correlation in Price Changes and Volatility Across International Stock Markets.”Review of Financial Studies 3, 281–307, (1990).
Hardouvelis, G. A., “Economic News, Exchange Rates and Interest Rates.” Journal of International Money andFinance 71, 23–35, (1988).
Hauser, S., Y. Tanchuma and U. Yaari, “International Transfer of Pricing Information between Dually ListedStocks.” The Journal of Financial Research 21, 139–157, (1998).
Jayaraman, N., K. Shastri and K. Tandon, “The Impact of International Cross Listings on Risk and Return,” Journalof Banking and Finance l7, 91–103, (1993).
Johanse, S., “Statistical Analysis of Cointegration Vectors.” Journal of Economic Dynamics and Control 12,231–254.
Johanse, S. and K. Juselius, “Maximum Likelihood Estimation and Inference on Cointegration with Applicationto the Demand For Money.” Oxford Bulletin of Economics and Statistics 52, 169–209.
Kato, K., S. Linn and J. Schallheim, “Are there Arbitrage Opportunities in the Market for American DepositoryReceipts?” Journal of International Financial Markets Institutions and Money 1, 73–89, (1991).
Kim, M., A. C. Szakmary and I. Mathur, “Price Transmission Dynamics between ADRs and their UnderlyingForeign Securities.” Journal of Banking and Finance 24, 1359–1382, (2000).
Miller, D. P. and M. R. Morey, “The Intraday Pricing Behavior of International Dually Listed Securities.” Journalof International Financial Markets, Institutions and Money 6, 79–89, (1996).
Ng, S. and P. Perron, “Unit Root Tests in ARMA Models with Data Dependent Methods for the Selection of theTruncation Lag.” Journal of the American Statistical Association 90(429), 268–281, (1995).
Osterwald-Lenum, M., “A Note with Quantiles of the Asymptotic Distribution of the Maximum Likelihoodcointegration Rank Test Statistics.” Oxford Bulletin of Economics and Statistics 54, 461–471, (1992).
Park, J. and A. Tavakkol, “Are ADRs a Dollar Translation of their Underlying Securities? : The Case of JapaneseADRs.” Journal of International Financial Markets, Institutions and Money 4, 77–87, (1994).
Pierce, D. A. and L. Haugh, “Causality in Temporal Systems: Characterization and Survey.” Journal of Econo-metrics 5, 265–293, (1977).
Sim, C., “Money, Income and Causality.” American Economic Review 62, 540–552, (1972).Said, S. and D. Dickey, “Testing for Unit roots in Autoregressive Moving Average Models with Unknown order.”
Biometrika 71, 599–607, (1984).Wahab, M., “A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedas-
ticity.” Journal of International Financial Markets Institutions and Money 2, 97–130, (1980).