Modern Economy, 2010, 1, 118-124
doi:10.4236/me.2010.12012 Published Online August 2010 (http://www. SciRP.org/journal/me)
Research on the Relationship between Foreign Trade and
the GDP Growth of East China—Empirical
Analysis Based on Causality
Yuhong Li, Zhongwen Chen*, Changjian San*
Business School, Jinggangshan University, Ji’an, China
Received April 19, 2010; revised June 8, 2010; accepted June 12, 2010
Abstract
In open economy, development of foreign trade greatly impacts GDP growth. Adopt modern testing methods
like unit root, time-series co-integration analysis and error correction model for researching the causalities
between foreign trade including total export and import with the collected 28-year statistical data of east
China from 1981 to 2008, including total export and total import and GDP growth of east China. The result
suggests that there exist long term or short termcausality between GDP and total export and import as well as
between GDP and export, foreign trade is the long term and short term reason of GDP growth, but no evi-
dence can prove that there exists long term stationary causality between import trade and GDP. This paper
finally provides with some instructive recommendations on how to develop the foreign trade of east China
under the new global economy environment.
Keywords: East China, GDP Growth, Foreign Trade, Causality Analysis
1. Introduction
Since the reform and opening up, China’s foreign trade,
which is playing a significance role in the world, has
become more and more important. But the proportion of
China's total import value in the GDP cannot match the
average level of developed countries. Obviously, the for-
eign trade is closely related to economic growth in China.
The importance of foreign trade for a country is increas-
ingly prominent, though many researchers like Xu Qifa
and Jiang Cuixia (2002) focused on the research related
to the contribution of foreign trade on GDP growth, re-
searchers particularly only focused on one region[1], a
province as an example, literature related to the research
on the economic development of east China is rare. Since
the reform and opening up, foreign trade in east has ex-
perienced rapid development. From 1981 to 2008, ex-
ports and imports in east increased from 8.564 billion
dollars to 2289.189 billion dollars. The increase of fore-
ign trade is faster than the increase of GDP, and the pro-
portion of foreign trade in GDP is increasing too. Howe-
ver, is there serious internal logical causality between
GDP and foreign trade? or, is there long term or short
termcausality between them? Thus, this paper will try to
research and discover it.
Due to the administrative division of China (2005) and
the statement of financial department about graduates go-
ing down to the grass-roots units (referring to the less
developed areas of east China and the region of western
China), now east China refers to the relevantly developed
areas including Beijing city, Tianjin city, Shandong pro-
vince, Liaoning province, Jiangsu province, Shanghai
city, Zhejiang province, Fujian province and Guangdong
province. However, some data of Beijing city are miss-
ing, so the data cited in this paper will not include that of
Beijing city. Although it will somewhat affect the rese-
arch, it will not so much impact the research of the rela-
tionship between GDP and foreign trade in east China in
essence.
According to the provincial yearbooks (1981-2008),
the relevant data of east China (shown as Table 1), inc-
luding (1981-2008) total foreign trade value (IE), the
total export value (EXP) and the total import value (IMP)
as well as GDP are collected and sorted out. From the
result, it could be seen that the mentioned indexes repre-
sent a trend to increase on the whole. In order to further
grope for the relationship between GDP and certain in-
dex, the concept of foreign trade dependency is cited to
describe it [2]. It is just the ratio of certain index of for-
eign trade and GDP, which can reflect the relationship of

Y. H. LI ET AL.
119
dependency between GDP and certain index. It can be
written in a formula as,
Dependence on foreign trade
= certain index of foreign trade / GDP
With the result calculated by the above formula, the de-
pendency of each index of foreign trade represents a tre-
nd to increase on the whole, while in some years, they
occasionally represented fluctuations. With the result, we
could see a trend of dependency (Figure 1). The ratio of
dependence on foreign trade in east increased from
12.01% in 1981 to 69.37% in 2008. Especially, in 2006,
it reached the top (80.88%); the ratio of dependence on
export increased from10.74% in 1981 to 43.17% in 2008,
it reached the top (48.60%) in 2006; the ratio of depend-
ence on import increased from 1.27% in 1981 to 26.2%
in 2008, it reached the top (32.27%) in 2006. These Fig-
ures intuitively indicate that each index of foreign trade
of east China contributes to GDP growth at different lev-
els. But, whether there exists internal logical causality
should be further tested. Therefore, the time-series data
and latest more stationary analysis methods are adopted
for testing the relationships of the indexes of foreign tra-
de and GDP of east China. We will use unit root, time-
series co-integration long term causality and short term-
causality analyses to expect more stationary results [3].
Table 1. 1981-2008 relevant statistical data of GDP and foreign trade of east China.
year GDP
(100 million dollars)
Foreign trade
(10000 dollars)
export
(10000 dollars)
import
(10000 dollars)
Dependence
On foreign trade
（%）
Dependence
on export trade
（%）
Dependence
on import trade
（%）
1981 1461.75 1755308 1569233 186075 12.01 10.74 1.27
1982 1455.68 1706645 1517628 189017 11.72 10.43 1.30
1983 1549.11 1765104 1540626 224478 11.39 9.95 1.45
1984 1603.69 2119018 1699111 419907 13.21 10.59 2.62
1985 1597.52 2558642 1793590 765052 16.02 11.23 4.79
1986 1571.12 2598819 1773466 825353 16.54 11.29 5.25
1987 1768.34 4303466 2675955 1627511 24.34 15.13 9.20
1988 2317.23 5901747 3339995 2561752 25.47 14.41 11.06
1989 2604.58 6692030 3878348 2813682 25.69 14.89 10.80
1990 2271.60 7339161 4593243 2745918 32.31 20.22 12.09
1991 2380.09 8907810 5342241 3565569 37.43 22.45 14.98
1992 3022.14 11334101 6424146 4909955 37.50 21.26 16.25
1993 4096.97 13560294 7136708 6423586 33.10 17.42 15.68
1994 3733.62 16810653 9362583 7448070 45.03 25.08 19.95
1995 4851.21 19671665 11297396 8374269 40.55 23.29 17.26
1996 5653.93 21668124 12217895 9450229 38.32 21.61 16.71
1997 6342.48 25125789 14851481 10274308 39.62 23.42 16.20
1998 6808.93 25955390 15194880 10760510 38.12 22.32 15.80
1999 7240.87 29076356 16413372 12662984 40.16 22.67 17.49
2000 8122.25 38067988 21040036 17027952 46.87 25.90 20.96
2001 8974.95 41122470 22756708 18365762 45.82 25.36 20.46
2002 9931.45 51292334 28085051 23207283 51.65 28.28 23.37
2003 11675.58 70691612 37877449 32814163 60.55 32.44 28.10
2004 14049.33 95750072 51540201 44209871 68.15 36.69 31.47
2005 16780.12 117542900 66064400 51478500 70.05 39.37 30.68
2006 20401.49 165002100 99160700 65841400 80.88 48.60 32.27
2007 26651.01 200954500 122842100 78112400 75.40 46.09 29.31
2008 33000.07 228918900 142452200 86466700 69.37 43.17 26.20
Data sources: 1981-2008 Provincial Yearbooks
Figure 1. the Trend of dependence of GDP on foreign trade.

Y. H. LI ET AL.
120
2. Literature Review
Economists who concern about foreign trade mainly fo-
cus on the effects of foreign trade on national economy,
and it has been a focus to discuss the theory about the
relationship between foreign trade and economic growth.
It is Adam Smith who first studies the relationship bet-
ween international trade and economic growth. In his
view, the development of division is the principal factor
to improve the long-time growth of productivity, and the
degree of division is constrained by the scope of market.
Expansion of market will naturally deepen the division
and improve the productivity, and then improve econo-
mic growth; David Ricardo and J. S. Mill as well as D. R.
Nurkse all possess the mentioned views shown in their
works; while the special dispute starts at Robertson’s pro-
position of that foreign trade is the engine of economic
growth (1973) he mainly focus on the theory, which is co-
mplemented and developed by Nurkse (the 1950s), that
lagging countries can improve their economic growth by
foreign trade ,especially, by export growth. He suggests
that foreign trade is the crucial factor of economic grow-
th engine.
The engine theory causes controversial disputes [4],
many economists suggest that trade growth of develop-
ing countries is correlated to their own economic growth,
their export growths are constrained by the economic gr-
owth of developed countries[5]. William. Lewis is the
representative personage of such economists. Irving Kra-
vis (1970) puts forward new viewpoint, which is later
accepted by lots of western economists, that foreign trade
is a maid of economic growth rather than an engine. Cla-
ssical school, Marxian school and New classical school
all suggest that foreign trade has just indirect impact on
accumulation and economic growth, in fact, foreign trade
impacts on them through profit margin. Further, some La-
tin American economists like Prebisch and Singer have
completely negative attitude on engine theory, they sugg-
est that ,in modern global economy regime, developed
capitalism countries are the core which is regulating the
outer consisting of developing countries, the outer coun-
tries must comply with the core countries. This kind of
depending relation makes foreign trade the reason of we-
akening the economy of the developing countries rather
than the reason of improving the economy of the devel-
oping countries.
Difference of opinions on the relation between foreign
trade and economic growth has activated scholars to gro-
pe for answers through empirical studies. Such as Jeff-
rey.Sach and Andrew Warner find that the economic gro-
wth rate of those developing countries which carry out
opening economies reaches 4.5%, while that of those co-
untries which carry out closed economies only researches
0.7%; in the mean time, they find the economic growth
rate of those developed countries which carry out open
economies reaches 2.3%, while that of the countries whi-
ch carry out closed economy only reaches 0.7%.
Some of the scholars mainly focus on co-integration
analysis. For instance, Jung and Marshal (1985), Chow
(1987), Love (1994), Dhawan and Biswa (1999) have
done much of co-integration analyses [6]; Improved me-
thods with error correction model were adopted by some
Chinese scholars for doing a lot of empirical studies on
trade and economic growth. The empirical analyses done
by Shi Chuangyu (2003) and Wang Xianzhu (2007) sug-
gest that growth of export could greatly promote econ-
omic growth in short time, while that of import didn’t
impact economic significancely. Liu Xiaopeng (2001) as
well as Li Yuhong and Wang Xiaoyin (2009) do co-in-
tegration analyses with the data of import, export and
economic, and the results suggests that growth of import
greatly promotes economic growth of China, while that
of export performs an opposite one;
Some of the scholars mainly focus on contribution ra-
tio of foreign trade. Wei Weixian (1999) drew the concl-
usion through co-integration analysis and variance ana-
lysis that 31% of economic growth of China ascribes to
the export -oriented strategy, while the contribution of
economic growth to that of export is less than 10%, this
accounts for that the fast growth of China’s economic
growth doesn't realize the scale effect of the growth of
export. Lin Yifu and Li Zhengjun improved the tradi-
tional single equation model and built simultaneous equ-
ations to calculate the contribution of foreign trade to
economic growth. The result suggests that 10% of the
export growth can lead to 1% of economic growth.
Still, there are many scholars mainly focus on correla-
tion analysis. For instance, Dong Migang (2000) indica-
ted that, in 1978-1998, the correlation of China’s foreign
trade and economic growth was significant, the coeffici-
ent was r > 94%. Guo Xin (2004) drew the conclusion by
recursion model, which indicates there exists a signifi-
cant positive correlation between foreign trade and econ-
omic growth. That is to say, the contribution of foreign
trade to economic growth is considerable;
While some of the scholars mainly focus on regression
analysis to study foreign trade and economic growth. Su-
ch as, Yi Xiangshuo (1997) found that the pull effect of
export on non-export sectors or the whole economic
growth was not strong at all; Yang Quanfa (1998) took a
regression analysis with the data (1978-1995) of relevant
indexes through the model built by Balasa and Vedur,
and he drew a conclusion that growth of export didn’t
meet the expectation of promoting the economic growth.
Sun Lin, Wang Qifan (2003) researched, by the impro-
ved Vedur Model, and pointed out that the mechanism
and approaches of the effect of China’s foreign trade on
economic growth has strong time tag.
The above methods all have separate advantages and
defects, especially, when single cross-set (time series)

Y. H. LI ET AL.
121
data analyses are adopted for testing. These methods all
have the problem of potential on the low side, and they
might lead to errors while calculating long term causali-
ties and the instabilities. In addition, simply dealing with
the data is just equal to considering them to be coessen-
tial, so their heterogeneity was ignored.
But in this paper, the authors will adopt improved co-
integration analysis with error correction model for test-
ing time series data of foreign trade and GDP to research
the relationship between foreign trade and GDP growth
of east China.
3. Models and Methods
Co-integration analysis, which is mentioned above, with
time-seriesis is adopted in this paper for testing whether
there exists long term or short term stationary causality
between foreign trade and GDP growth, and for testing
the unit root of each variable to confirm their stationari-
ties. The following would be the desired time-sequence
data model [7],
yit =ρi yi,t - 1 + Xitδi +εit
where i = 1,…, N represent the number of time-series
data; t = 1,…,T represents time span; xit are the exoge-
nous variables in the model including fixed effect or time
trend of each time-series unit; ρi is autoregressive coeffi-
cient, suppose that disturbance terms εit are mutual inde-
pendence. If | ρi | < 1, yit represents the stationary process;
if | ρi | = 1, yit represents the process of unit root.
Take the logarithms of the gross domestic products
(GDP), total import and export value (IE), total export
value (EXP) and total import value (IMP) separately, and
they are LnGDP, LnIE, LnIMP, LnEXP. Then test their
logarithm values and first difference values through time-
series unit root. Logarithm cited here is for convenience
to get stationarity more easily, and is helpful to eliminate
the heteroscedasticity of time series and the characterist-
ics of time series and relationships would not be changed.
The relationships between relevant indexes would be
tested in this paper by three steps. First of all, test throu-
gh unit root using time-series data[8]; then, use two-step
method put forward by Engle and Granger (1987) to test
the mutual long term causalities of relevant indexes; if
the long term causality exists, then further test their short
term causalities.
3.1. Test of Time-Series Data by Unit Root
In order to overcome the deviation brought out by only
one method, LLC test, B test and IPS test are all used in
this paper to test the relationships between GDP and
relevant indexes (IE, EXP and IMP) of foreign trade of
east China by unit root.
3.2. Co-Integration Analysis of Time-Series Data
and Long Term Causality Test
In order to test the long term cassations between variab-
les, two-step test method put forward by Engle and Gra-
nger (1987) is used. When measuring the long term cau-
salities between GDP and relevant indexes of foreign
trade, the measured variables are mutually simple inte-
grated, and then the regression through the following
time-series Equation (1) can be processed. Further, re-
sidual errors Eit comes out and it’s tested through unit
root to determine their stabilities. If Eit is stationary, the
mutual long term causalities are proved to exist.
Ln (*) = α + β Ln (**) + εit (1)
where (*) and (**) separately represent GDP, IE, EXP
and IMP.
3.3. Time-Series Data Error Correcting Model
and Short Term Causality Test
Co-integration relationships just reflect the long term
balanced relations between relevant variables. In order to
cover the shortage, correcting mechanism of short term-
deviation from long term balance could be cited. At the
same time, as the limited number of years, the above test
result may cause disputes (Christpoulos and Tsionas,
2004). Therefore, under the circumstance of long term
causalities, short termcausalities should be further tested
as well. The error correcting models could be built as,
d LnGDPit = ηi + Σα1d Ln GDPi, t – 1
+ Σβ1d LnIEi, t-1 + λ ECMit + εit (2)
d LnGDPit = ηi + Σα1d LnGDPi, t – 1
+ Σβ1d LnEXPi, t-1 + λ ECMit + εit (3)
d LnGDPit = ηi + Σα1d LnGDPi, t – 1
+ Σβ1d LnIMPi, t-1 + λ ECMit + εit (4)
d LnIEit = ηi + Σα1d LnIEi, t-1
+ Σβ1d LnGDPi, t-1 + λ ECMit + εit (5)
d LnEXPit = ηi + Σα1d LnEXPi, t-1
+ Σβ1d LnGDPi, t-1 + λ ECMit + εit (6)
d LnIMPit =ηi + Σα1d LnIMP i, t-1
+ Σβ1d LnGDPi, t-1 + λ ECMit + εit (7)
where t represents year, d rerepresents first difference
calculation, ECMit represents the errors of long term
balance. If λ = 0 is rejected, error correcting mechanism
happens, and the tested long term causality is reliable, it

Y. H. LI ET AL.
122
could be unreliable. If β1 = 0 is rejected, and then the
short termcausality is not proved to exist.
4. Result and Analysis
4.1. Test of Time-Series Data by Unit Root
Software Eview 5.0 is herewith used and the four variab-
les LnGDP, LnIE, LnEXP and LnIMP by LLC test, B
test, IPS test are calculated and processed separately(see
Table 2). The result indicates that LnGDP, LnIE, LnEXP
and LnIMP all perform non-stationary state through the
tests by the mentioned method. However, after first order
difference, through the same methods, it’s found that all
of them passed the significance test by 1%. So we can
say GDP growth, total foreign trade value, total export va-
lue and total import value are all integrated of order one.
4.2. Co-Integration Analysis of Time-Series Data
and Long Term Causality Test
Through the test by unit root, GDP, IE, EXP and IMP all
perform one-order simple-integration I (1), there may
exist mutual co-integration between relevant variables.
The results of their long term causalities and correspond-
ding residual errors Eit can be shown as Table 3.
From Table 3, it’s found that there exists mutual
long term causality between LnGDP and LnEXP, two
of the three tests (LLC test, B test and IPS test) passed
by 90% significance level. Nothing can prove that there
exists long term mutual causality between GDP and
LnIMP. Corresponding co-integration equations are the
following:
LnGDP =6.836721 + 0.672795LnEXP
(28.49411) (45.37549)
[0.0000] [0.0000]
F = 2058.395, R2 = 0.987530
LnEXP = –9.933901 +1.467801LnGDP
(–17.16705) (45.37549)
[0.0000] [0.0000]
F = 2058.935 R2 = 0.987530
(Notation: Figures in ( ) and [ ] are separately t-test values and p
values of t-test)
The above equations all passed t-test and F-test by
95% level. From the co-integration equations, it’s clear
that, in the long term, LnGDP and Ln EXP are positive.
And the elasticity between them is 0.672795; this means
one unit of LnIE increment will lead to 0.672795 units of
LnGDP increment. Similarly, one unit of LnGDP incre-
ment will lead to 1.467801 units of LnEXP increment.
4.3. Time-Series Data Error Correcting Model
and Short Term Causality Test
It’s found, according to the co-integration test of time-se-
ries data, that there exist mutual long term causality be-
tween LnGDP and LnEXP. For the limited number of ye-
ars, short causality between them should be used to test
their stationarities (see Table 4).
The result in Table 4 states the following clear: ECM
in model 1, of which test equation is Equation (6), is
positive and passes the test by 0.05 level, indicating cor-
recting error mechanism happens, and long term pull
effect of export trade on GDP has been proved. ECM in
model 6 is also positive and passes the test by 0.05 levels,
further indicating that export always promote GDP to
Table 2. Result of Time-series data test by unit root.
LnGDP LnEXP LnIMP LnIE
Time span:
28 Level test first difference Level test first differenceLevel testfirst difference Level test first difference
ADF test 2.1999
(0.999)
–4.2547**
(0.0028)
2.2935
(0.999)
–4.1016**
(0.0040)
–2.902*
(0.059)
–3.7241**
(0.0085)
1.0197
(0.996)
–4.5716**
(0.0013)
PP test 7.4563
(1.000)
–4.2076**
(0.0031)
2.2935
(0.999)
–4.0796**
(0.0042)
–1.701
(0.418)
–4.0464**
(0.0045)
–1.001
(0.995)
–4.5902**
(0.0012)
Notation: 1) the Figures in the brackets are p values; 2) *indicates panel data pass of the significance test by 95% level, **indicates panel
data pass of the significance test by 99% level; 3) testing form is only intercept, lagging exponent number is chosen as Schwarz rules.
Table 3. Result of Time-series Co-integration test (test of residual errors by unit root).
Variables ADF test PP test
LnGDP is the induced variable of LnIE
LnIE is the induced variable of LnGDP
–1.6497(0.4444)
–1.8836(0.3345)
–1.5748(0.4813)
–1.8023(0.3714)
LnGDP is the induced variable of LnIMP
LnIMP is the induced variable of LnGDP
–0.9914(0.7416)
–1.9525(0.3047)
–1.3153(0.6075)
–2.0177(0.2779)
LnGDP is the induced variable of LnEXP
LnEXPis the induced variable of LnGDP
–2.9299*(0.0550)
–2.9830**(0.0493)
–2.8616*(0.0632)
–2.9036*(0.0581)
Notation: *indicates panel data pass of the significance test by 95% level, **indicates panel data pass of the significance test by 99% level.

Y. H. LI ET AL.
123
Table 4. Short term causality test of Time-series data.
variable C D(LnGDP(–1)) D(LnIEXP(–1)) ECM R2 Fvalue
Model 3 D(LnGDP) 0.03429
(0.2906)
–0.149232
(0.3971)
0.621686**
(0.0000)
0.3342*
(0.056) 0.451 60022** (0.03726)
Model 6 D(LnEXP) 0.130995**
(0.0011)
0.196343
(0.3233)
0.188206
(0.3287)
0.3755**
(0.0053) 0.341 3.80176** (0.02457)
Notation: *indicates panel data pass of the significance test by 95% level, **indicates panel data pass of the significance test by 99% level.
grow, and it’s passed F-test, so it’s concluded that export
trade and GDP are mutually causal.
5. Conclusions and Recommendations
The result of the tests indicates foreign trade is the long
term and short termsource of GDP growth of east China.
Total export has positive relationship with GDP growth,
and they are mutually causal. It has proved the intuitional
dependence relationship mentioned in part 1 (INTRO-
DUCTION). Developing foreign trade is good for prom-
oting GDP growth of east China, and GDP growth, in re-
verse, is also good for promoting the development of open
economy. No evidence can prove that there exists long
term stationary causality between import and GDP gro-
wth as well as that between total foreign trade and GDP;
it’s not necessary that import can directly contribute to
GDP growth. As the crowding-out effect of imported pr-
oducts from foreign countries and the indirect promoting
effect couldn’t be measured, it cannot be proved either
there exists long term causality between import and GDP,
or there exists long term stationary internal logical cau-
sality .
According to the empirical study results, strong devel-
opment of foreign trade greatly benefits the economic de-
velopment in east China. To overcome the problems ex-
isted in foreign trade, for the current financial crisis and
hard retrieve of global economy, and in order to reduce
its corresponding economic loss as much as possible, ke-
eping scale of exports is necessary. Thus, governments
of all levels in east China should do as the following:
1) Stable exchange rate must be remained. At present,
RMB can not be continuously upvalued, and in the long
run, it could be devalued. Then it can be helpful to keep
or even improve the competitiveness of the products th-
ere. Furthermore, it will be good to enrich the commodi-
ties to increase export supply.
2) Active industrial policy must be carried out. First,
develop hi-tech industries, improve the comparative adv-
antages of the products as well as the competitiveness;
Second, develop specialty industries to amplify the com-
parative advantages to increase exports; Third, cultivate
emerging industries, through importing FDI and high
technologies, to improve own productivity. That is to say,
governments should make great efforts to the construc-
tion of export by virtue of a series of industrial restructu-
ring and revitalization, and to the plans continuously in-
troduced by the state in textile, steel, automobile, equip-
ment manufacturing industries. Develop high-tech and
echo-friendly products, promote the exports of branded
products, and large machinery, complete sets of equipm-
ent as well as edge and labor-intensive products to supp-
ort tech innovation in small and medium enterprises so as
to increase competitiveness.
3) Strategic trade policy must be performed. First, ch-
oose special industries in east China to protect or provide
with subsidies to possess bigger share of global market
[9]. Some labor-intensive products, like textile products,
have less profit for their lower prices, so subsidies can
keep their irreversed benefits to some extent, and labor-
intensive industries can improve employments; Second,
protect domestic market to protect and cultivate emerg-
ing industries, and finally increase exports; Third, take
advantages of the increased export rebate rate of some
products introduced by the states, and improve the upg-
rade and transformation of the trade in processing.
4) Pro-active fiscal policy should be executed in east
China. Governments of all levels should provide the ent-
erprises with more capital supports to improve the finan-
cial environment for exportation and financing for the en-
terprises. Meanwhile, stepping up their supports for entr-
epreneurs’ credit and improving secured financing cond-
itions are of great importance. However, preferential pol-
icies, such as lower income tax and sales tax, also could
be provided to support and promote the border trade as
well as international trade. And the governments should
increase investment in port construction, market develop-
pment, project declaration, utilization of funds and joint
inspection service to help exports, and increase expendit-
ure to bring in and cultivate talents in foreign trade by
training the employers’ ideas, practical abilities, knowle-
dge and negotiation skills etc, in order to improve their
capability of service and creation.
5) Trade protection must be always aware of under cu-
rrent hard retrieve of global economy. With the wide spr-
ead of the international financial crisis, all kinds of trade
protectionism, in the form of technique protection, green
products standard, anti-dumping, countervailing etc, in
countries across the world are getting rampant. In a word,
trade barriers are set for many excuses to reduce import,
which has become a prescription for some countries to

Y. H. LI ET AL.
124
get rid of crisis. Therefore, governments in east China
must be highly aware of it and ready to face provocation
of protectionism in foreign trade.
6. Acknowledgements
This paper is assisted by the projects: Jiangxi provincial
Co-operated Social Science Projects，A Study on the Dev-
elopment of Service Industry and Trade in Service in
Jiangxi Province (Project ID: 09YJ249)and A Research
on the Evolution of the Spatial Economy in Jiangxi and
Agglomeration of Industry (Project ID: 09YJ245 ); A St-
udy on the Development of Logistics in Ji’an City Based
on the Theory of Industry Cluster (Project ID: JR0816).
7. References
[1] Y. H. Li and X. Y. Wang, “Research on the Relation-
ship between Foreign Trade and the GDP Growth of So-
uthwest Minority Region of China-Empirical Analysis
Based on the Panel Causality,” IEEE International Con-
ference on Industrial Engineering and Engineering
Management, Hongkong, December 2009, pp. 798-802.
[2] Y. H. Li and K. Y. Deng, “An Empirical Study on the
Contribution of Foreign Trade to the Economic Growth
of Southwest Minority Region of China,” International
Conference on Computational Intelligence and Software
Engineering, Wuhan, December 2009, pp. 1-4.
[3] A. Levin, C.-F. Lin and C. Chu, “Unit Root Tests in
Panel Data: Asymptotic and Finite Sample Properties,”
Journal of Econometrics, Vol. 108, No. 1, 2002, pp. 1-24.
[4] J. Love, “Engines of Growth: the Exports and Govern-
ment Sectors,” World Economy, Vol. 17, No. 2, 1994, pp.
203-218.
[5] W. S. Jung and P. J. Marshall, “Export, Growth and Cau-
sality in Developing Countries,” Journal of Development
Economics, Vol. 18, No. 2, 1985, pp. 1-12.
[6] P. C. Y. Chow, “Causality between Export and Industrial
Development: Empirical Evidence from NICs,” Journal
of Development Economics, Vol. 26, No. 1, 1987, pp. 55-
63.
[7] C. Kao, “Spurious Regression and ResiduaL2based Tests
f or Co-integration in Panel Data,” Journal of Economet-
rics, Vol. 90, No. 1, 1999, pp. 1-44.
[8] K. S. Im, H. Pesaran and Y. Shin, “Testing f or Unit
Roots in Heterogeneous Panels,” Journal of Economet-
rics, Vol. 115, No. 1, 2003, pp. 53-74.
[9] E. Helpman and P. R. Krugman, “Market Structure and
Foreign Trade,” MIT Press, Cambridge, 1985.