Formation of Natural Trading Blocs in East Asia - 麻省理工学院人工智能研究专业

Formation of Natural Trading Blocs in East Asia
Formation of Natural Trading Blocs in East Asia

An Examination of the Formation of
Natural Trading Blocs in East Asia

Chang-Soo Lee
Korea Institute for International
Economic Policy
P.O. Box 235
Yomgok-dong, Seocho-gu
Seoul 137-747
韩国
cslee@kiep.go.kr

Soon-Chan Park
Division of Economics and
Commerce
Kongju National University
Shinkwan-dong 182
Kongju 314-701
韩国
spark@kongju.ac.kr

抽象的
This paper seeks to identify the appropriate form of a regional
trading agreement in East Asia by estimating the intra-regional
trade bias of various informal regional groups. The major conclu-
sion is that ASEAN(西德:1)3 would be the natural policy choice for the
formation of a regional trading agreement in East Asia. 而且,
ASEAN(西德:1)3 will not diverge from the principles of open regional-
ism and multilateralism, when we consider its inherently open
character and the positive influence of the United States and
日本.

1. 介绍

With a second wave of regionalism sweeping the world,
East Asian countries have come to acknowledge the need
for some form of regional economic cooperation. 日本
and Singapore recently concluded a bilateral free trade
协议 (FTA), and various other forms of regional co-
operation are being proposed in East Asia, such as those
among ASEAN(西德:1)3 (ASEAN plus South Korea, 日本, 和
中国) and ASEAN and China, as well as an FTA for
中国, 日本, 和韩国. 然而, there have been
no serious efforts to answer the following questions:
Which regional integration agreement is the most natural
choice for East Asian countries from an economic perspec-
主动的? Which regional integration agreements should East
Asian countries pursue?

We start from the presumption that the institutionalization
of a natural trading region, rather than the deliberate cre-
ation of preferential arrangements, would be more effec-
tive in accelerating trade among members because it
would exploit existing economic and sociocultural connec-

Asian Economic Papers 4:1

Institute of Technology

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Formation of Natural Trading Blocs in East Asia

系统蒸发散, production networks, and other ties.1 Wonnacott and Lutz (1989) introduced
the concept of “natural trading partners,” emphasizing two deªning criteria: close
trade relations and geographical proximity. This concept of natural trading partners
was popularized as “natural trading blocs” by Krugman (1991), who emphasized
the importance of geographical proximity.

在本文中, we focus on the empirical identiªcation of natural trading regions us-
ing the gravity model of Frankel (1997). We test various combinations of regional
groups as possible natural trading blocs, including the various pairings of China, Ja-
pan, 和韩国; China–Japan–South Korea; ASEAN; South Korea–ASEAN,
China–ASEAN, and Japan–ASEAN; and ASEAN(西德:1)3 as a whole. We ªnd no evi-
dence that the three Northeast Asian countries (中国, 日本, 和韩国)
form a natural trading bloc. 相比之下, the tests suggest that ASEAN(西德:1)3 would be a
natural choice for a regional trading agreement.

We explore further the characteristics of ASEAN(西德:1)3, focusing on two questions. 这
ªrst question is whether ASEAN(西德:1)3 is centered on Japan or the United States. 这
second question is whether there is any possibility that the informal ASEAN(西德:1)3
trading bloc is diverting trade with nonmembers. The evidence suggests that Japan
and the United States serve together as centers for intra-regional trade. The evidence
also shows that the creation of ASEAN(西德:1)3 as a formal trading bloc can be character-
ized as an example of nondiscriminatory market-driven integration.

2. Empirical framework and data

2.1 Analytical framework
The empirical framework proposed by Frankel (1997) for detecting and quantifying
intra-regional trade bias is the gravity model, which captures not only the inºuences
of geographical proximity and economic size on trade, but also the trade effects of
formal trading blocs and informal relations. Our basic speciªcation is as follows:

日志 (Tij) (西德:2) (西德:3) (西德:1) (西德:4)
(西德:1) (西德:4)

1 日志 (GDPi
4 (ADJACENTij) (西德:1) (西德:4)

(西德:5) GDPj) (西德:1) (西德:4)

2 日志 (TARIFFj) (西德:1) (西德:4)
(西德:7)

5 (LANGUAGEj) (西德:1) (西德:6)

k BLOCk,

3 日志 (DISTANCEij)

(1)

1 Natural trading regions have been conceptualized as “natural trading partners,” “natural in-

tegration regions,” or “market integration” by Wonnacott and Lutz (1989), Lorenz (1992,
1993), and Drysdale and Garnaut (1993). These analysts argue that growth and efªciency in
natural trading regions are likely to be greatly enhanced if they are supported by regional
trade agreements. Intentionally creating a trading bloc is more costly because extra measures
are required to create close common political and economic preferences, which are already
present in a natural trading bloc.

Asian Economic Papers

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Formation of Natural Trading Blocs in East Asia

where Tij is the bilateral volume of trade (imports) between importing (我) 和前-
移植 (j) 国家, GDPi and GDPj are the importing (我) and exporting (j) 县-
tries’ GDPs, TARIFFj indicates the average import tariff in country j and DISTANCEij
is the distance between countries i and j. The various formal and informal trading
k BLOCk , where k (西德:2) 欧洲联盟, NAFTA, MERCOSUR (Mercado
blocs are represented by (西德:7)
Común del Sur), AFTA (ASEAN Free Trade Agreement), and ASEAN(西德:1)3. 下列的
the tradition of the gravity model, bilateral distance (which represents transporta-
tion costs) is measured by the log of physical distance between the capital cities of
the two countries. 此外, we introduce the dummy variable ADJACENTij into
the gravity model to indicate when the two countries share a common land border
and the variable LANGUAGEj to capture the effects of a common language on bilat-
eral trade.2

Methodologically, we extend the usual gravity equation by explicitly introducing
import tariffs. The theoretical gravity models deªne trade costs as all trade barriers
impeding bilateral trade and assume that geographical distance represents all trade
成本. Baier and Bergstrand (2001) extend the usual gravity equation by introducing
explicit import tariffs and ªnd that tariffs have a great negative impact on trade.
(See also Deardorff 1998 and Anderson and van Wincoop 2003.)

Once the norm has been established by the gravity model, a dummy variable can be
added to represent the case in which both countries in a given pair belong to the
same regional trading bloc (either formal or informal). A statistically signiªcant pos-
itive coefªcient for the bloc dummy can be interpreted to mean that a trading bloc is
forming naturally if there are no formal regional integration agreements.

There has been a long debate on the major reasons for regional trade concentration.
Krugman (1991) and Summers (1991) point out that regional concentration in trade
is attributable to geographical proximity, whereas Bhagwati (1992) and Panagariya
(1995) emphasize “artiªcial” preferential trade policies as the dominant explanation
for high trade concentration ratios. Frankel, 斯坦因, and Wei (1997) refer to the real-
world situation between the two extremes of “natural” and “artiªcial” trade blocs as
a “super-natural” bloc. Frankel (1997) uses the term “informal trading bloc” as a
contrasting concept to the existing idea of a “formal trading bloc.” In this paper, 我们
focus on the existence of a regional trade bias; 因此, the bloc dummy in our gravity
equation represents all the regional trade biases: formal and artiªcial factors, 地理-
graphic natural factors, and other unidentiªed natural factors.

2 We allow for countries in which more than one language is spoken and consider English,

Chinese, 德语, 法语, Japanese, 西班牙语, and Portuguese. A common language reduces
transaction costs and enhances exporters and importers’ mutual understanding of their re-
spective cultures and legal systems.

Asian Economic Papers

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Formation of Natural Trading Blocs in East Asia

2.2 数据
Our data set covers 51 countries3 for the 1990–2001 period. The source for bilateral
trade ºows is the International Monetary Fund’s Direction of Trade Statistics. We use
current GDPs (expressed in U.S. 美元) obtained from the World Bank’s World De-
velopment Indicators.

We faced some difªculties in obtaining bilateral tariff rates. Prewo (1978) 提供
annual bilateral tariff rates for 18 OECD countries for 1958 通过 1974, 和
Deardorff and Stern (1990) also provide measures for pre– and post–Tokyo Round
average tariff rates for 18 OECD countries. We could not obtain bilateral tariff rates
for other countries, so we use the means of the national tariff rates given in the
World Trade Organization’s Integrated Data Base.

3. Are trading blocs forming in East Asia?

3.1 Do South Korea, 日本, and China constitute a regional trading bloc?
To test whether an informal regional trading bloc is forming in Northeast Asia, 我们
estimate the gravity equation (a panel estimation using a random-effects model) 到
include a Northeast Asian bloc dummy in addition to contemporary regional trad-
ing blocs, such as the EU, NAFTA, MERCOSUR, and AFTA.

The estimation results are summarized in table 1.4 We focus on the bloc effect. 这
coefªcient for the EU shows a statistically signiªcant level of 0.233 in 1990–2001 for
the case of China–Japan–South Korea. The coefªcients for NAFTA and MERCOSUR
are not statistically signiªcant for any years tested. The regional dummy of AFTA
(ASEAN) exhibits a strong inward trade bias during 1990–2001. The coefªcient on
the Northeast Asian dummy (China–Japan–South Korea) shows statistical sig-
niªcance for this period at the 10 percent level. 因此, we ªnd weak evidence to sug-

3 Countries in the data set are Algeria, Angola, 阿根廷, 澳大利亚, 奥地利, Bahamas, 呸-
rain, Bangladesh, 比利时, 巴西, 加拿大, 智利, 中国, 哥伦比亚, 丹麦, 埃及, Fin-
土地, 法国, 德国, 加纳, 希腊, 危地马拉, Honduras, 香港, 印度, 印度尼西亚,
爱尔兰, 意大利, 日本, 肯尼亚, 韩国, Malaysia, 墨西哥, 摩洛哥, 荷兰, 秘鲁,
菲律宾, 波兰, Portugal, 新加坡, 南非, 西班牙, 斯里兰卡, 瑞典, Switzer-
土地, Thailand, Turkey, 英国, 美国, Uruguay, and Venezuela.

4 The sum of two countries’ log-GDPs is highly signiªcant statistically in the regressions and
generally less than 1, indicating that trade increases with a country’s GDP but does so less
than proportionally. The coefªcient for distance implies that each percentage point increase
in distance reduces trade by about 0.89 百分. The coefªcient on the dummy variable for
language also shows high statistical signiªcance, but the coefªcient for a common border is
not statistically signiªcant. The introduction of import tariffs as the new independent vari-
able leads to a great increase in the explanatory power of the regressions.

Asian Economic Papers

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Formation of Natural Trading Blocs in East Asia

桌子 1. Gravity estimation: China–Japan–South Korea

Basic model
((西德:8)32.766***
((西德:8)55.987)

0.881***

(100.098)

0.894***

((西德:8)23.198)

(西德:8)0.142***
((西德:8)9.814)

0.776***

(10.382)

0.296
(1.586)

0.221*
(1.673)

0.607
(1.090)

0.209
(0.366)

0.754***
(2.579)

0.945

26,620

Intercept

Log (GDPi)

Log (DISTANCEij)

Log (TARIFFj)

LANGUAGEj

ADJACENTij

欧洲联盟

NAFTA

MERCOSUR

AFTA (ASEAN)

China–Japan–South Korea

Adjusted R2

不. of observations

笔记: Figures in parentheses are t-statistics.

***Statistically significant at the 1 percent level.

**Statistically significant at the 5 percent level.

*Statistically significant at the 10 percent level.

China–Japan–South Korea
(西德:8)32.788***
((西德:8)56.081)

0.881***

(99.988)
(西德:8)0.889***

((西德:8)23.009)

(西德:8)0.143***
((西德:8)9.855)

0.779***

(10.440)

0.306*
(1.645)

0.233*
(1.645)

0.612
(1.101)

0.210
(0.369)

0.758***
(2.599)

1.019*
(1.866)

0.945

26,620

gest that a trading bloc is forming among the three Northeast Asian countries (坳-
umn 2 of table 1).

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

桌子 2 tests whether there is any tendency toward a special trade bias in the three
possible combinations of regional groupings in the Northeast Asian region: 南
Korea–Japan, South Korea–China, and China–Japan. None of the coefªcients on the
regional trading bloc dummies shows statistical signiªcance in the panel estimation,
corroborating the assertion that there is no regional trade bias among China, 日本,
和韩国. 换句话说, the trade relations between them can be simply
explained with the gravity variables.

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

3.2 Does ASEAN 3 constitute a regional trading bloc?
This section investigates the existence of a regional trade bias within the boundary
of ASEAN(西德:1)3. 第一的, we add a new bloc dummy, ASEAN(西德:1)3, to the baseline model of
桌子 1 and test whether a regional trading bloc is forming from the perspective of
the ASEAN(西德:1)3 框架 (桌子 3). 第二, we examine whether there is any ten-

Asian Economic Papers

Formation of Natural Trading Blocs in East Asia

桌子 2. Gravity estimation: South Korea–Japan, South Korea–China, and China–Japan

Intercept

Log (GDPi)

Log (DISTANCEij)

Log (TARIFFj)

LANGUAGEj

ADJACENTij

欧洲联盟

NAFTA

MERCOSUR

AFTA (ASEAN)

South Korea–Japan

South Korea–China

China–Japan

Adjusted R2

不. of observations

South Korea–Japan
(西德:8)32.767***
((西德:8)56.001)

South Korea–China
(西德:8)32.778***
((西德:8)55.994)

China–Japan
(西德:8)32.766***
((西德:8)56.003)

0.881***

(100.117)
(西德:8)0.895***

((西德:8)23.213)

(西德:8)0.142***
((西德:8)9.812)

0.776***

(10.387)

0.298
(1.597)

0.223*
(1.694)

0.608
(1.094)

0.210
(0.368)

0.757***
(2.592)

0.093
(0.665)

0.945

26,620

0.881***

(100.060)
(西德:8)0.892***

((西德:8)23.068)

(西德:8)0.142***
((西德:8)9.825)

0.777***

(10.394)

0.299
(1.605)

0.224*
(1.701)

0.608
(1.092)

0.209
(0.367)

0.755***
(2.584)

0.605
(0.640)

0.945

26,620

0.881***

(99.984)
(西德:8)0.893***

((西德:8)23.148)

(西德:8)0.142***
((西德:8)9.824)

0.777***

(10.393)

0.298
(1.598)

0.223*
(1.691)

0.608
(1.093)

0.209
(0.367)

0.755***
(2.583)

0.526
(0.558)

0.945

26,620

笔记: Figures in parentheses are t-statistics.

***Statistically significant at the 1 percent level.

**Statistically significant at the 5 percent level.

*Statistically significant at the 10 percent level.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

dency to promote regional trade between ASEAN as a whole and each of the three
Northeast Asian countries, 那是, 韩国(西德:1)ASEAN, 日本(西德:1)ASEAN, 和
中国(西德:1)ASEAN (桌子 4).

桌子 3 shows that ASEAN(西德:1)3 has a signiªcant and apparent intra-regional trade
bias in two models, with statistically signiªcant coefªcients in the models of 1.304
和 1.096, respectively.5

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

5 The estimated coefªcient is 1.096, implying that two countries of this bloc trade about 200

percent more with each other (after holding constant the variables for GDP, distance, and tar-
iffs, as well as the other gravity variables) than two otherwise similar countries [经验值(1.096) (西德:2)
2.99]. This suggests that although ASEAN(西德:1)3 countries have no formal trade agreement,
there are unidentiªed factors that promote bilateral trade in this Northeast Asian region.

Asian Economic Papers

Formation of Natural Trading Blocs in East Asia

桌子 3. China–Japan–South Korea, AFTA, and ASEAN 3

模型 1
(西德:8)32.832***
((西德:8)56.512)

0.878***

(99.967)
(西德:8)0.872***

((西德:8)22.727)

(西德:8)0.145***

((西德:8)10.959)

0.783***

(10.586)

0.303*
(1.640)

0.281**
(2.146)

0.662
(1.201)

0.261
(0.464)
(西德:8)0.522
((西德:8)1.407)
(西德:8)0.023
((西德:8)0.041)

1.304***
(5.514)

0.945

26,620

Intercept

Log (GDPi)

Log (DISTANCEij)

Log (TARIFFj)

LANGUAGEj

ADJACENTij

欧洲联盟

NAFTA

MERCOSUR

AFTA (ASEAN)

China–Japan–South Korea

ASEAN(西德:1)3

Adjusted R2

不. of observations

笔记: Figures in parentheses are t-statistics.

***Statistically significant at the 1 percent level.

**Statistically significant at the 5 percent level.

*Statistically significant at the 10 percent level.

模型 2 (模型 1 without ASEAN)

(西德:8)32.847***
((西德:8)56.518)

0.879***

((西德:8)100.060)

(西德:8)0.873***

((西德:8)22.723)

(西德:8)0.145***

((西德:8)10.574)

0.783***

(10.574)

0.308*
(1.668)

0.279**
(2.125)

0.655
(1.189)

0.254
(0.451)

0.147
(0.263)

1.096***
(5.936)

0.945

26,620

When we test for the effect of a China–Japan–South Korea bloc together with the
ASEAN(西德:1)3 影响, the former effect disappears completely. We think that large differ-
ences between economic systems and development levels, as well as lack of experi-
ence in economic integration, are the biggest challenges for an FTA among these
three countries.

相似地, the coefªcient for the AFTA bloc does not show any statistical signiªcance
when estimated with ASEAN(西德:1)3. These results are in line with the conclusion of
Frankel and Wei (1996) that all of the increase in intra-Asian trade can be explained
by the rapid growth of these economies, not by any special ASEAN effect. 这是
convincing when we consider that the three countries (韩国, 日本, 和
中国) accounted for about 70 percent of the total trade of ASEAN(西德:1)3 countries and
关于 92 percent of the combined GDP of ASEAN(西德:1)3 countries in 2000. AFTA might

Asian Economic Papers

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Formation of Natural Trading Blocs in East Asia

桌子 4. South Korea ASEAN, Japan ASEAN, and China ASEAN

Intercept

Log (GDPi)

Log (DISTANCEij)

Log (TARIFFj)

LANGUAGEj

ADJACENTij

欧洲联盟

NAFTA

MERCOSUR

韩国(西德:1)ASEAN

日本(西德:1)ASEAN

中国(西德:1)ASEAN

ASEAN(西德:1)3

Adjusted R2

不. of observations

South Korea ASEAN
(西德:8)33.334***
((西德:8)57.083)

(西德:8)33.314***
((西德:8)57.211)

Japan ASEAN
(西德:8)33.172***
((西德:8)56.928)

(西德:8)33.164***
((西德:8)57.077)

China ASEAN
(西德:8)33.170***
((西德:8)56.714)

(西德:8)33.126***
((西德:8)56.770)

0.884***

(101.016)
(西德:8)0.848***

((西德:8)21.996)

(西德:8)0.140***
((西德:8)9.686)

0.779***

(10.538)

0.246
(1.337)

0.319***
(2.436)

0.704
(1.279)

0.362
(0.642)

2.030***
(8.113)

0.882***

(100.758)
(西德:8)0.841***

((西德:8)21.879)

(西德:8)0.142***
((西德:8)9.852)

0.783***

(10.643)

0.267
(1.452)

0.344***
(2.631)

0.719
(1.313)

0.366
(0.653)

1.692***
(6.366)

0.882***

(100.767)
(西德:8)0.855***

((西德:8)22.190)

(西德:8)0.140***
((西德:8)9.693)

0.777***

(10.510)

0.248
(1.343)

0.309**
(2.354)

0.701
(1.273)

0.346
(0.614)

0.880***

(100.594)
(西德:8)0.848***

((西德:8)22.064)

(西德:8)0.143***
((西德:8)9.866)

0.782***

(10.618)

0.270
(1.467)

0.333**
(2.547)

0.715
(1.304)

0.350
(0.623)

0.882***

(100.658)
(西德:8)0.858***

((西德:8)22.155)

(西德:8)0.141***
((西德:8)9.760)

0.756***

(10.192)

0.263
(1.420)

0.298**
(2.264)

0.687
(1.243)

0.339
(0.599)

0.881***

(100.384)
(西德:8)0.853***

((西德:8)22.126)

(西德:8)0.143***
((西德:8)9.919)

0.767***

(10.365)

0.283
(1.535)

0.316**
(2.413)

0.696
(1.266)

0.332
(0.590)

1.850***
(7.400)

1.475***
(5.492)

0.685***
(3.625)

0.945

26,620

0.945

26,620

0.714***
(3.733)

0.945

26,620

1.672***
(6.653)

0.945

26,620

1.201***
(4.215)

0.699***
(3.450)

0.945

26,620

笔记: Figures in parentheses are t-statistics.

***Statistically significant at the 1 percent level.

**Statistically significant at the 5 percent level.

*Statistically significant at the 10 percent level.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

be meaningless without the trade with its major Northeast Asian trading partners.
The conclusion seems to be that China, 日本, 韩国, and ASEAN are in fact
functioning as a trading bloc.

桌子 4 summarizes the results of the tests for possible combinations between
ASEAN and each of the three Northeast Asian countries. The trade relations of
ASEAN with Japan, 中国, 和韩国, 分别, 非常相似. 这
coefªcients on bloc dummies for South Korea(西德:1)ASEAN, 日本(西德:1)ASEAN, 和
中国(西德:1)ASEAN show signiªcant and apparent intra-regional trade biases in every
estimation, and the effects of the two dummies are signiªcant even when the
ASEAN(西德:1)3 dummy with a strong statistical signiªcance is included in the gravity
方程 (ªnal row, columns 2, 4, 和 6). This implies that the South Korean, Japa-

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Asian Economic Papers

Formation of Natural Trading Blocs in East Asia

nese, and Chinese networks are in place, as commonly believed, and they are
competing for a close relationship with the ASEAN countries.

4. The characteristics and future of ASEAN 3

4.1 Are there special trade relations with Japan and the United States?
The East Asian economies are generally outward-oriented. The United States and Ja-
pan are major trading partners of particularly vital importance for the growth of
东亚. One question is whether a trading bloc is forming in East Asia that is cen-
tered on Japan and/or the United States. On the issue of regionalization centered on
日本, Frankel (1992), using a separate dummy variable speciªcally for bilateral
Asian trade with Japan, found little evidence in the regional trade patterns that Ja-
pan is transforming itself into a center of trade. 相比之下, Choudhry, Abu-Bakar,
and Wylie (2000), also using a simple gravity model of trade ºows within East Asian
国家 (including Hong Kong and Taiwan), concluded that East Asian intra-
regional trade centered on Japan goes beyond what can be explained by the gravity
变量.

桌子 5 tests for the existence of special trade relations between ASEAN(西德:1)3 and Ja-
pan and between ASEAN(西德:1)3 and the United States by adding two separate dummy
variables for the bilateral trade of countries in ASEAN(西德:1)3 with Japan and the United
状态, 分别. The coefªcient on the Japan dummy is statistically signiªcant at
这 10 percent level, whereas that on the U.S. dummy is signiªcant at the 1 百分
等级. 因此, these results support the claim that ASEAN(西德:1)3 is centered on the
United States and Japan together.

The increase in the intra-regional trade of ASEAN(西德:1)3 might arise from the fact that
East Asian economies are highly dependent on the United States for their export
市场. It is well known that East Asian countries have developed a low-cost pro-
duction network in the region to achieve competitiveness in the world market and
to export many products to other regions. Within this network, Japan provides
countries in the region with capital and intermediate goods, thereby increasing re-
gional trade. The potential for regional growth in trade in East Asia critically de-
pends on growth in ªnal demand for East Asian products in other regions, particu-
larly in North America. The signiªcant and positive economic ties between the
United States and ASEAN(西德:1)3, 一方面, and between Japan and ASEAN(西德:1)3,
在另一, can be understood in this regard. 同时, the weaker sig-
niªcance of the Japan dummy, in comparison with the U.S. dummy, might reºect
the fact that although Japan has constructed a production network in the ASEAN(西德:1)3

Asian Economic Papers

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Formation of Natural Trading Blocs in East Asia

桌子 5. The relationship of ASEAN 3 to Japan and the United States

模型 1 (我们. dummy)
(西德:8)32.636***
((西德:8)56.129)

模型 2 (Japan dummy)
(西德:8)32.841***
((西德:8)56.550)

0.876***

(99.759)
(西德:8)0.882***

((西德:8)23.120)

(西德:8)0.147***

((西德:8)10.178)

0.769***

(10.446)

0.306*
(1.666)

0.273**
(2.099)

0.666
(1.216)

0.257
(0.458)

1.114***
(6.290)

2.024***
(4.335)

0.945

26,620

0.878***

(99.971)
(西德:8)0.871***

((西德:8)22.699)

(西德:8)0.145***

((西德:8)10.062)

0.785***

(10.616)

0.313*
(1.699)

0.284**
(2.171)

0.658
(1.196)

0.254
(0.450)

1.068***
(5.941)

0.837*
(1.656)

0.945

26,620

Intercept

Log (GDPi)

Log (DISTANCEij)

Log (TARIFFj)

LANGUAGEj

ADJACENTij

欧洲联盟

NAFTA

MERCOSUR

ASEAN(西德:1)3

我们. dummy

Japan dummy

Adjusted R2

不. of observations

笔记: Figures in parentheses are t-statistics.

***Statistically significant at the 1 percent level.

**Statistically significant at the 5 percent level.

*Statistically significant at the 10 percent level.

框架, there has been less than full demand within the region for a large part
of those ªnal goods produced by that network.

4.2 Openness of ASEAN 3
We investigate whether there is economically important trade discrimination in
ASEAN(西德:1)3. To capture the extra-bloc effects of ASEAN(西德:1)3, we introduce the open-
ness dummy variable for each bloc, deªned by Frankel and Wei, which takes the
value of 1 for the case of imports from all countries by member countries of bloc k.6
The coefªcient of the openness dummy indicates to what extent members of a group
trade with other countries in general, regardless of whether they are in the same
团体. If this variable is negative, it indicates that members of the group in question

6 例如, C (中国)-J (日本)-K (韩国) openness (西德:2) 1 when i of Tij

否则, C-J-K openness (西德:2) 0.

(西德:2) C, J, and K.

Asian Economic Papers

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Formation of Natural Trading Blocs in East Asia

桌子 6. Bloc openness of ASEAN 3

Intercept

Log (GDPi)

Log (DISTANCEij)

Log (TARIFFj)

LANGUAGEj

ADJACENTij

ASEAN(西德:1)3

ASEAN(西德:1)3 Open

欧洲联盟

EU Open

NAFTA

NAFTA Open

MERCOSUR

MERCOSUR Open

Adjusted R2

不. of observations

ASEAN 3
openness
(西德:8)32.513***
((西德:8)55.759)

0.876***

(99.880)
(西德:8)0.903***

((西德:8)23.425)

(西德:8)0.146***

((西德:8)10.131)

0.794***

(10.779)

0.275
(1.500)

0.741***
(3.898)

0.417***
(5.290)

欧洲联盟
openness
(西德:8)32.930***
((西德:8)56.839)

0.873***

(98.556)
(西德:8)0.837***

((西德:8)21.558)

(西德:8)0.151***

((西德:8)10.425)

0.818***

(11.054)

0.361*
(1.959)

0.082
(0.602)

0.344***
(5.151)

NAFTA
openness
(西德:8)32.760***
((西德:8)55.714)

0.877***

(98.334)
(西德:8)0.876***

((西德:8)22.879)

(西德:8)0.146***

((西德:8)10.102)

0.774***

(10.426)

0.312*
(1.695)

0.533
(0.952)

0.132
(1.124)

0.945

26,620

0.945

26,620

0.945

26,620

MERCOSUR
openness
(西德:8)33.099***
((西德:8)57.054)

All dummies
openness
(西德:8)32.464***
((西德:8)55.034)

0.880***

(100.577)
(西德:8)0.849***

((西德:8)22.105)

(西德:8)0.145***

((西德:8)10.053)

0.788***

(10.701)

0.350*
(1.905)

0.870
(1.521)
(西德:8)0.654***
(5.685)

0.945

26,620

0.865***

(94.985)
(西德:8)0.852***

((西德:8)22.109)

(西德:8)0.157***

((西德:8)10.856)

0.824***

(11.297)

0.378**
(2.093)

0.790***
(4.226)

0.527***
(6.413)

0.061
(0.459)

0.443***
(6.320)

0.547
(0.998)

0.322***
(0.007)

0.814
(1.451)
(西德:8)0.432***
((西德:8)3.714)

0.945

26,620

笔记: Figures in parentheses are t-statistics.

***Statistically significant at the 1 percent level.

**Statistically significant at the 5 percent level.

*Statistically significant at the 10 percent level.

trade less with the rest of world than would be predicted, so a regional trade agree-
ment among them might divert trade (Frankel and Wei 1996, 17).

The results are summarized in table 6. The introduction of dummy variables for the
openness of the blocs reduces the estimated effect of ASEAN(西德:1)3, but the coefªcient
remains highly signiªcant statistically. 此外, the variable representing bloc
openness of ASEAN(西德:1)3 shows signiªcantly positive values for the period 1990–2001.
相比之下, the coefªcient on the openness dummy for NAFTA is not signiªcant,
while that on the openness dummy for MERCOSUR is negative and signiªcant in
the same period. After inclusion of the openness dummy for the EU, the coefªcient
of the EU bloc dummy turns from statistically signiªcant into statistically

100

Asian Economic Papers

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Formation of Natural Trading Blocs in East Asia

insigniªcant, while the coefªcient of the openness dummy for the EU becomes posi-
tive and statistically signiªcant. These empirical results show little evidence of trade
diversion resulting from an ASEAN(西德:1)3 bloc.

4.3 Widening ASEAN 3
Frankel and Wei (1996) test the bloc effect of “Asia Paciªc,” which includes all the
countries with eastern coasts on the Paciªc, including Australia and New Zealand.
Their results show that the coefªcient and signiªcance level of this bloc are both
higher than those of the East Asia dummy. They also broaden the bloc search and
test for the formation of Asia-Paciªc Economic Cooperation (APEC), which includes
the United States and Canada and other Paciªc economies. Frankel and Wei’s results
如下面所述: the coefªcient of the APEC bloc dummy was highly signiªcant,
whereas that of the Asia Paciªc dummy completely disappeared and that of the
ASEAN(西德:1)3 dummy remained signiªcant. This implies that APEC would be the next
alternative trading bloc for East Asian countries and that ASEAN(西德:1)3 would con-
tinue to function as a regional trading bloc even with the existence of a wider APEC
bloc.7

5. Summary and conclusions

Using a modiªed gravity equation, we have examined whether any regional trading
blocs are forming in East Asia. We ªnd no evidence that any trading bloc is natu-
rally forming among the three Northeast Asian countries, but we ªnd that
ASEAN(西德:1)3 shows a signiªcant intra-regional trade bias in all tested cases. 这
coefªcient of the ASEAN(西德:1)3 dummy is signiªcant and remains so, even when tested
together with the bigger APEC group, and even though the magnitude of the APEC
effect is great (Frankel and Wei 1996).

We ªnd that the United States and Japan exert special effects on ASEAN(西德:1)3, promot-
ing intra-regional trade. 更重要的是, ASEAN(西德:1)3 is open to trade with the rest
世界的 (IE。, no trade diversion). The creation of a regional trading bloc in East
Asia is expected to contribute to a worldwide institutional framework of multi-
lateralism, given that trade relations with countries outside the region are unques-
tionably important to the development of the ASEAN(西德:1)3 trading bloc. 因此, 这

7 One might argue that the bigger the regional group, the higher the intra-regional bias among
its members. 然而, there is signiªcant evidence that this is not the case. Frankel (1997)
tests a potential TAFTA (Trans-Atlantic Free Trade Agreement) bloc, deªned as including
全部 15 EU members plus all 3 members of NAFTA, and shows that the coefªcient is always
negative.

101

Asian Economic Papers

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Formation of Natural Trading Blocs in East Asia

evolution of ASEAN(西德:1)3 into a formal trading agreement would further promote and
reinforce both intra-regional trade and inter-regional trade.

When we regard regionalization as a process of dynamic development of trade pat-
terns by way of building gravity centers (following Lorenz 1992), it seems clear that
a regional trade agreement among ASEAN(西德:1)3 countries would be the ªrst step for
regionalization in East Asia. The provision of a formal institutional framework
would help to stabilize the existing unidentiªed ties among ASEAN(西德:1)3 国家,
such as production networks and sociocultural linkages, and accelerate further
growth and efªciency in this region. 而且, the creation of an ASEAN(西德:1)3 formal
trading bloc would likely be deªned by market-driven integration (Drysdale and
Garnaut 1993). For a more successful and rapid development of the intra-regional
market potential in East Asia, the creation of a formal regional trading bloc to ad-
vance the process of regionalization should now be considered.

参考

安德森, James E., and Eric van Wincoop. 2003. Gravity with Gravitas: A Solution to the Bor-
der Puzzle. American Economic Review 93 (1):170–92.

Baier, Scott L., and Jeffrey H. Bergstrand. 2001. The Growth of World Trade: Tariffs, Transport
成本, and Income Similarity. Journal of International Economics 53 (1):1–27.

Bhagwati, Jagdish. 1992. Regionalism vs. 多边主义. World Economy 15:535–56.

Choudhry, Saud, Abu Suªan Abu-Bakar, and Peter Wylie. 2000. A Yen Bloc in Paciªc Asia:
Natural Economic Symbiosis or Overblown Rhetoric? Applied Economic Letters 7 (4):215–18.

Deardorff, Alan V. 1998. The Determinants of Bilateral Trade: Does Gravity Work in a Neoclas-
sical World? In The Regionalization of the World Economy, edited by Jeffrey A. Frankel, PP. 7–28.
芝加哥: 芝加哥大学.

Deardorff, Alan V., and Robert M. Stern. 1990. Computational Analysis of Global Trading Arrange-
蒙特. 安娜堡: University of Michigan Press.

Drysdale, 彼得, and Ross Garnaut. 1993. The Paciªc: An Application of a General Theory of
Economic Integration. In Paciªc Dynamism and the International Economic System, edited by Fred
Bergsten and Marcus Noland, PP. 183–223. 华盛顿, 华盛顿特区: Institute for International Eco-
经济学.

Frankel, Jeffrey A. 1992. Is Japan Creating a Yen Bloc in East Asia and the Paciªc? NBER
Working Paper no. 4050. 剑桥, 大量的。: 国家经济研究局.

———. 1997. Regional Trading Blocs in the World Economic System. 华盛顿, 华盛顿特区: 研究所
for International Economics.

Frankel, Jeffrey A., Ernesto Stein, and Shang-Jin Wei. 1997. Continental Trading Blocs: Are
They Natural or Supernatural? In The Regionalization of the World Economy, edited by Jeffrey A.
Frankel, PP. 91–113. 芝加哥: 芝加哥大学出版社.

102

Asian Economic Papers

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

Formation of Natural Trading Blocs in East Asia

Frankel, Jeffrey A., and Shang-Jin Wei. 1996. ASEAN in a Regional Perspective. UC Berkeley
Working Paper no. C96-074. 十一月. 伯克利: 加州大学.

International Monetary Fund. Various years. Direction of Trade Statistics. CD-ROM.

Krugman, 保罗. 1991. The Move toward Free Trade Zones. In Policy Implications of Trade and
Currency Zones, PP. 7–41. Kansas City: Federal Reserve Bank of Kansas City.

Lorenz, Detlef. 1992. Economic Geography and the Political Economy of Regionalization: 这
Example of Western Europe. American Economic Review 82 (2):88–91.

———. 1993. Europe and East Asia in the Context of Regionalization: Theory and Economic
政策. Journal of Asian Economics 4 (2):255–70.

Panagariya, Arvind. 1995. The Free Trade Area of the Americas: Good for Latin America?
Working Paper no. 12. 学院公园, 马里兰州。: Center for International Economics, 大学
Maryland.

Prewo, Wilfried. 1978. Determinants of the Trade Pattern among OECD Countries from 1958 到
1974. Jahrbucher fur Nationalokonomie und Statistick 193 (4):341–58.

Summers, 劳伦斯·H. 1991. Regionalism and the World Trading System. In Policy Implications
of Trade Zones, PP. 295–302. Kansas City: Federal Reserve Bank of Kansas City.

Wonnacott, 保罗, and Mark Lutz. 1989. Is There a Case for Free Trade Areas? In Free Trade Areas
和美国. Trade Policy, edited by Jeffrey J. Schott, PP. 59–84. 华盛顿, 华盛顿特区: Institute for Inter-
national Economics.

World Bank. Various years. World Development Indicators. CD-ROM.

World Trade Organization (WTO). Various years. Integrated Data Base (IDB). CD-ROM.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你
A
s
e
p
A
r
t
我
C
e
–
p
d

我

F
/

4
1
9
0
1
6
8
2
1
4
7
A
s
e
p
2
0
0
5
4
1
9
0
p
d

乙
y
G
你
e
s
t

哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3

103

Asian Economic Papers
下载pdf