Export Performances in China and India - Ricerca sull'intelligenza artificiale specializzata al MIT

Export Performances in China and India
Export Performances in China and India

A Comparative Analysis of China’s and
India’s Recent Export Performances*

Kaliappa Kalirajan
Foundation for Advanced
Studies on International
Development and National
Graduate Institute for Policy
Studi
7-22-1 Roppongi
Minato-ku
Tokyo 106-8677
Japan
kpkalirajan@yahoo.com

Kanhaiya Singh
National Council of Applied
Economic Research
Parisila Bhawan
11, Indraprastha Estate
New Delhi 110 002
India

Astratto
Drawing on the convergence theory, one would expect that the
export performance of India (a latecomer to integrating with the
global economy) would be at least on par with that of China be-
cause China’s performance has happened as predicted by the the-
ory. This study, using performance measures based on the endog-
enous growth theory that internalizes the ability to export the
maximum possible exports under the determinants of exports in-
cluding the existing behind the border and beyond the border
constraints, shows that India’s export performance is still far be-
hind that of China. The implication of this study is that India’s re-
form measures need to be bolstered effectively to catch up and
to overtake China.

1. introduzione

In the ranking of the largest economies of the world mea-
sured by their gross domestic products in terms of 1995
constant US$, China and India stood at the 19th and 20th positions in 1980, but in 2005 the ranking places them at the 7th and 12th positions, rispettivamente. Such a quantum jump of these two economies, particularly China, over two and a half decades is remarkable.1 What is interesting is, measured in terms of per capita income in current interna- * This paper was presented at the Asian Economic Panel Meeting at the Brookings Institution, Washington, D.C., SU 10 April 2007. Comments and suggestions on an earlier version by discussants Lael Brainard, Brookings Institution, and Zhang Xiaojing, Chi- nese Academy of Social Sciences, and participants are gratefully acknowledged. A special thanks to Wing Thye Woo for commis- sioning this study for this Panel Meeting. 1 Woo (1998), Sachs and Woo (2000), and Lardy (2002) have pro- vided a comprehensive exposition about the factors behind China’s successful economic performance. Asian Economic Papers 7:1 © 2008 The Earth Institute at Columbia University and the Massachusetts Institute of Technology l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Export Performances in China and India tional dollars with purchasing power parity, China was lagging behind India by US$ 223 In 1980, but overtook India with a difference of US$ 1,450 In 2000. Based on the IMF data, the per capita income in current international dollars with purchasing power parity in 2005 worked out to be US$ 3,320 and US$ 7,150 for India and China, rispettivamente. The dynamic growth performance of China and the respectable growth performance of India raise several interesting questions.2 For example, is China’s growth miracle different from what we observed in other Asian countries? Although China has demonstrated its potential to grow faster con- sistently for several years, why doesn’t India exhibit the same kind of dynamism? As a latecomer, what can India learn from China’s growth process? These interesting and important questions have occupied the minds of development economists. There is now a rich literature on the economic developments of these two countries including their reform processes and their impacts on macroeconomic policies and overall economic growth. Though some of the conclusions in these studies are con- troversial, there is consensus that opening up the economies for export-led growth through trade liberalization is a crucial factor among others, which signiªcantly inºuenced the growth performance.3 Is China’s growth performance anything special? When China’s growth experience is examined against the growth patterns of other Asian countries, particularly Japan, it is noticeable that Japan’s growth rate fell 15 years after its catching-up process started in 1955, whereas China has continued its growth for more than 25 years.4 However, when China’s share of global GDP is compared with that of Japan’s, it is evident that the latter’s share of global GDP grew faster than that of the former dur- ing Japan’s catching-up process. Così, there do not seem to be any signiªcant mira- cles in the growth performances of China when compared with that of Japan.5 Nev- 2 In the eyes of many observers, by the end of the 1990s India had moved to being a “six per- cent growth” economy: not a “miracle” perhaps, but certainly respectable. 3 Per esempio, some authors found differences in the political system as the key instrument creating variations in the performance of the two countries. Sachs and Woo (2000) labeled the competing interpretations of China’s post 1978 economic growth process as institutional in- novations versus institutional convergence, which are in other words, the Experimentalist School and the Convergence School, rispettivamente. Important econometric studies of the link- age between trade reform and the rate of economic growth include Sachs and Warner (1995) and Frankel and Romer (1999). 4 The starting period of the catch-up process for a country is based on the IMF’s notion of hav- ing an annual rise in exports of more than 10 percent for 3 years continuously (IMF 2004, chapter II). 5 In this context, it is worth noting the publication Growth without Miracles by Garnaut and Huang (2001). 2 Asian Economic Papers l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Export Performances in China and India ertheless, China’s growth performance looks more impressive if its integration into the global economy in terms of international trade in goods is considered. Per esempio, China’s total merchandise trade increased from US$ 1,155 billion in
2004 to US$ 1,422 billion during 2005. The surge in China’s exports has drastically changed the structure of East Asia’s trade surplus with the United States and the European Union in favor of China from Japan. Drawing on the convergence theory, if, as a latecomer, China has been able to improve its export performance faster, why not India, which opened up its economy much later than China? It is in this context that this paper examines the merchandise export performances of China and India with the following three empirical questions: (1) If China’s exporting environment is emulated by India, what would be the latter’s export performance?; (2) If India’s ex- porting environment is duplicated by China, what would be the latter’s export per- formance?; E (3) How far are China and India from reaching their exports poten- tial with their trading partners given the existing “behind the border” constraints and “beyond the border” constraints to exports?6 The following section brieºy describes important trade policy reforms in China and India. Sezione 3 discusses the concept and measurements of potential exports and data, which is followed by empirical estimations of different measures of potential exports from China and India with their trading partners. This section also provides the simulation results of export performances in China and India with the assump- tion of China emulating the exporting environment of India and India duplicating the exporting environment of China respectively. A ªnal section discusses what In- dia can learn from the export performance of China to improve its trade policy re- forme. 2. Trade policy reforms of China and India 2.1 China Trade policy in China underwent a major change between 1979 E 1980, when the central government decided to establish four Special Economic Zones (SEZs) in two coastal provinces, Guangdong and Fujian, to attract foreign direct investment and 6 Behind the border constraints to export, within the home country, which mainly include regulatory policies that impede competition, restrictions on foreign trade and investment, tolerance of business cartels, monopoly privileges given to public enterprises, and the cost and performance of infrastructure services that are important to the functioning of busi- ness, services such as ports, customs and transport, generally affect the domestic costs of production. Beyond the border constraints mainly refer to non-tariff barriers and other insti- tutional rigidities of partner countries, which generally inºuence the shifting of the export frontier. 3 Asian Economic Papers l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Export Performances in China and India new technologies (National Statistical Bureau, various years; henceforth, NBS(UN)). This was the beginning of China’s Open-Door Policy. Initial success encouraged Chinese policymakers to adopt similar policies in 14 east coastal cities in 1984, which were further extended to a far wider area of China’s east coast region in 1985 and in the following years. It is worth noting that the 12 East Coast provinces, out of the total 30,7 contributed two-thirds of China’s total exports in 1990 (China Custom Statistical Bureau 2002). The openness of the Chinese economy was accelerated in the 1990s, after Deng Xiaoping’s push for faster economic reforms and openness in 1992. Twenty inland cities became “open cities” that could enjoy a series of preferen- tial policies in 1993. Border areas in North and West China, namely, Xinjiang, Inner Mongolia, Heilongjiang, Yuannan, and Guangxi, were also opened to border trade (Wang 2004). FDI, which was only US$ 1.7 billion in 1985, increased dramatically in the 1990s. In
1995, FDI increased to US$ 37.5 billion, and then to US$ 40.7 billion in 2000, and to
US$ 72.4 billion in 2005. Domestic and foreign trade sectors were opened to FDI in the late 1990s. Foreign enterprises, which include enterprises with investment from Chinese Hong Kong, Macao, and Taiwan, played more and more important roles in the manufacturing sector of China (Jiang 2002). Trade policy was not shifted immediately from import-substitution to export- orientation. For a long period during the reform era, it was a mix of both import- substitution and export-orientation, but gradually shifted toward the East Asian growth model of export-oriented growth. High import tariffs remained in China, al- though the real tariff rate was far lower, due to various preferential policies and smuggling. In 1995, Per esempio, the average nominal tariff rate on electronic prod- ucts was 40 per cento, but the actual rate (questo è, tariffs actually collected as a share of the value of imports) was only 11.8 per cento (National Statistical Bureau, various years; henceforth NBS(B)). In the 1980s and 1990s, there were also trade-related investment measures (TRIMs) such as the requirement of domestic components in production, and foreign ex- change balance requirements. Despite these measures, the foreign-invested indus- tries were not foreign-exchange earners in the 1980s and the early-to-mid 1990s be- cause their exports could not exceed their imports before 1998, though they did contribute to economic growth, employment generation, and an increase in foreign trade (Wang 2004). 7 This includes four Minority Autonomous Regions and three Central-Administrated Munici- palities. The total number became 31 Dopo (NBS(UN)). 4 Asian Economic Papers l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Export Performances in China and India There were more changes in the 1990s. In 1996, joint ventures with foreign invest- ment were allowed to deal with foreign trade. In 1998, private enterprises were also allowed to engage in foreign trade. The state monopoly in foreign trade was gradu- ally replaced by market competition. Deduction, or removal, of tariff and non-tariff barriers was also an important part of trade policy reform. From 1982–92, the nomi- nal tariff rate, as an average, reduced from 56 percent to 43 per cento. From 1992–03, it further reduced from 43 percent to 11 per cento (Wu 2003).8 The average tariff in 2005 era 9.9 per cento. Non-tariff barriers, Per esempio, import licensing and other require- ments for special import approvals, were reduced in the 1990s and eliminated in the early 2000s as the government’s commitment toward joining the WTO grew. There were major changes after the WTO accession in 2001, pure. Concerning TRIMs, mainly requirements on domestic components, export performance, and foreign ex- change balance of foreign enterprises, were removed. Upon China’s WTO accession in 2001, the banking/insurance and telecommunication sectors, which were not opened to FDI before, were opened.9 Not only were the trade policies relating to FDI changed, trade liberalization also occurred in the domestic sectors. More and more manufacturers that produced export goods were also permitted to directly purchase inputs and sell products overseas. Così, it is apparent that trade policy reforms signiªcantly contributed to economic growth in China, which was more or less on average at the two-digit level over more than 2 decades.10 Nevertheless, there is room for further improvement in China’s trade policies.11 Some analysts have suggested that the imbalance of policy treatment between FDI and domestic investment, which favors FDI, has resulted in rent-seeking behavior and inefªciencies. Inoltre, there are needs for further policy reform toward 8 As mentioned earlier, the actual tariff rate in the 1990s should be far below the ofªcially an- nounced rate because of various tariff exemptions and deductions, and smuggling. This should not be the case in the early 1980s, because the coverage of policy preferences on tariff deduction was only limited at the time, and smuggling was less serious. 9 Sachs and Woo (2003) argued that the Chinese leadership’s opinion has been that in the short-run, there could be signiªcant displacement of Chinese state banks by foreign banks, but in the long run, Chinese banks (most likely private ones) would rise in importance. 10 Literature indicates that countries that liberalized their trade (raising their trade-to-GDP ra- tio by an average of 5 percentage points) between 1950 E 1998 enjoyed on average 1.5 per- centage points higher GDP growth compared with their pre-reform growth rates (Greena- modo, Morgan, and Wright 2002; Baldwin 2003). 11 Drysdale, Huang, and Kalirajan (2000) argued for the need for more trade policy reforms to enhance China’s trade efªciency. Gang Fan and Xiaojing Zhang (2003) discussed how the further reform agenda can be designed to achieve another period of 2 decades of high growth. 5 Asian Economic Papers l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Export Performances in China and India Figure 1. Pattern of economic growth of India l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 transparency and better business environment (Sachs and Woo 2003; Huang and Khanna 2003). 2.2 India Figure 1 presents a simpliªed record of India’s aggregate growth (growth in real GDP at factor cost, 1993–94 prices) performance over the 52 years from 1951–52 to 2002–03. It also plots trend growth (TG) rates for each decade starting 1951–61 and some of the key events responsible for slowdown episodes and includes a summary table indicating the coefªcient of variation across decades and average growths after ignoring the drought and crisis periods. Sweeping policy changes were made in the trade sector during the 1990s in India, though at a pace slower than in China. Cus- toms tariffs are now lower and quantitative restrictions on imports have been elimi- nated. Export restrictions have been reduced along with the implementation of vari- ous export promotion measures. Tuttavia, the pace of tariff reforms slowed after 1996–97. Whereas the peak rate of duty has been reduced gradually, the average tar- iff rate remained broadly unchanged at about 30 percent during 1997–02, though the average tariff was about 18 percent in 2005, which is almost double that of China. This tariff rate is also high by the current world standards. Figura 2 shows plots of four indicators of tariff-related trade barriers, all-products simple mean, standard deviation of tariff lines, simple mean of tariff lines for manu- 6 Asian Economic Papers Export Performances in China and India Figure 2. Trade reforms in terms of tariff policy across selected countries Source: World Development Indicator, 2002. factured goods, and share of tariff lines with international peaks. When compared with countries, such as China, Brasile, South Korea, Sri Lanka, Malaysia, Japan, and the United States, India turns out to be an outlier in terms of all-products simple mean tariffs. What is most disturbing is the number of lines with world peak. It ap- pears that the Indian authorities simply look at the highest rates prevailing any- where in the world and adopt the same tariff without much analysis. There are also concerns about the institutional role in determining tariffs. At least four institutions are assigned the role of ªxing tariffs in one way or the other. Among them, the Tariff Commission is the most relevant. The commission has re- sources to determine tariffs with more techno-economic analysis, but it has never been involved in tariff determination or regulation since its inception in September 1997. Then, there is the Tariff Research Unit (TRU) (presumably the most effective in determining tariffs) in the revenue department of the Ministry of Finance, which ob- viously would be more concerned about short-term effects of changes in tariffs, par- ticularly on revenue, than long-term effects on trade and growth. The Ministry of Agriculture reportedly determines agricultural tariffs. Besides, there is an anti- dumping directorate in the Ministry of Commerce to look into complaints of dump- ing of agricultural products such as skimmed milk powder from the European Union. Così, lack of institutional coordination may not be overlooked. Though the medium-term exports strategy (MTES 2002–07), which was announced in January 2002, aimed to increase India’s share in world trade from about 0.7 per- 7 Asian Economic Papers l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Export Performances in China and India cent to 1 percent by 2006–07, the current target is to reach 1.5 percent of world trade by 2009.12 Latest trade ªgures in the World Trade Report 2006 reveal that in calendar year 2005, India’s merchandise exports were worth US$ 90 billion that is approxi-
mately 0.89 percent of total global exports worth US$ 10,121 billion. China’s share, on the other hand, increased from 6.67 percent in 2004 A 7.52 percent in 2005 with the country exporting goods worth US$ 762 billion during the year. Although In-
dia’s share in world total merchandise exports surged from 0.4 percent in 1992 A
0.8 percent in 2002, it took 3 long years for India to move another step farther. A
this rate, the target of reaching 1.5 percent of world trade by 2009 would not be that
easy to achieve. To keep pace with the growth in world trade and grab a larger share
of the world exports market, India has to aim higher.

The 5-year Export and Import (EXIM) Policy (2002–07) announced on 31 Marzo 2002
intended to remove all quantitative restrictions on exports except for a few sensitive
items reserved for exports through the state trading enterprises. It also outlined a
farm-to-port approach for exports of agricultural products with a special focus on
the cottage sector, handicrafts, and assistance to states for infrastructure develop-
ment for exports (ASIDE). New private sector-run SEZs were created to provide in-
vestors an export-friendly environment. The incentives offered under the SEZ
scheme included duty-free importation/domestic procurement of goods for the de-
velopment of the SEZ and setting up of units, 100 percent FDI in the manufacturing
sector under the automatic route, 100 percent income tax exemption for the ªrst
5 years, E 50 percent tax exemption for 2 years thereafter. Other incentives in-
cluded sub-contracting part of production abroad, reimbursement/exemption for
central sales tax on domestic purchases by the SEZ units and retention of 100 per-
cent foreign exchange earnings in the Exchange Earners Foreign Currency (EEFC)
Account. In terms of ªnancing SEZs, overseas banking units (OBUs) that were ex-
empt from cash reserve and statutory liquidity requirements (CRR and SLR, respec-
tively), were permitted to set up in SEZs. These OBUs have given access to SEZ
units and SEZ developers to international ªnance at international rates. SEZ units
were exempt from external commercial borrowing (ECB) restrictions and were al-
lowed to make overseas investments and carry out commodity hedging. SEZs were
exempt from central sales tax in respect of supplies from domestic tariff area (DTA)
and transactions from DTA to SEZs were treated as exports under the Indian Income
Tax and Customs Acts.

12 The MTES is a comprehensive exercise, which includes product and market identiªcation

for exports and indicates sector-wise strategies for identiªed potential sectors.

Asian Economic Papers

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Export Performances in China and India

The number of goods reserved for the small-scale sector is set to reduce further. IL
strategic sectors identiªed for providing special focus include electronics, electrical
goods, and engineering goods referred to as “3Es” (Chadha 2003). Policy on entry of
direct foreign investment has been eased greatly, but investors continue to face a
daunting regulatory framework beyond the foreign investment regime itself.

Although policy initiatives are yielding favorable results to some extent, the forego-
ing discussion indicates that there are several concerns and issues, which mainly in-
volve behind the border constraints issues, that need to be addressed if exports are
to grow faster.

How effective have these trade policy reforms been in improving the export perfor-
mances of China and India? Export performance can be measured in several ways.
A simple conventional method is to work out the growth rate of absolute values of
exports between two times and compare it with another time within the country, O
compare it with the growth rate of another country during the same period. Though
this kind of measure is useful in a way; what is more interesting is to measure the
country’s potential exports, given the determinants of exports and compare it with
its own actual exports. Such a measure provides a better understanding of the link
between trade policies and export performance, which is explained in the following.

3. Measuring export performances of China and India

3.1 Methodology I
A common feature of all performance measures is that performance is deªned with
respect to a benchmark. Though there are several methods to arrive at a benchmark,
the method of comparing one’s own potential to his or her own actual achievement
is more appealing because any performance improvements come from “within.”
The endogenous growth theory popularized by Romer (1986) and Lucas (1988) facil-
itates the assumption of internalization of the “within” aspect through policy mea-
sures that increase the incentive to innovate and to have an impact on the long-run
growth rate of an economy (Roberts and Setterªeld 2007). In line with these argu-
menti, potential exports can be measured by following either a general equilibrium
approach or a disequilibrium framework. In the former approach, a home country’s
exports to all its trading partners, which may be exhaustive and represent a general
equilibrium framework, would be estimated and added up to arrive at total values
of exports. Alternatively, drawing on Kalirajan (1999), in a disequilibrium frame-
work where a home country’s actual exports are assumed to differ from its potential
exports with respect to each trading partner and the partner-speciªc export gap is

Asian Economic Papers

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Export Performances in China and India

explicitly included in the model explaining export ºows and the speciªc estimation
method yield potential exports. Whereas there are several studies following the for-
mer approach, studies using the latter approach are sparse in the literature.13 The
gravity model has been established in both approaches as a popular methodology to
measure potential trade between countries.

The gravity model, which is deªned following Newton’s Law of Gravitation, ex-
plains trade ºows between two countries as directly proportional to the product of
each country’s “economic mass” that can be measured by GDP and inversely pro-
portional to the distance between the countries (Bergstrand 1985). It is one of the
most frequently estimated empirical relationships in economics. Earlier studies have
estimated the difference between observed values and the predicted values that are
calculated from OLS estimates of the gravity model as potential exports (Baldwin
1994; Nilsson 2000). A simple baseline gravity model can be written as equation (1).

B
Xij C Y Y D
io

G
j

D(cid:2)
ij

(1)

where C, (cid:2), (cid:3), E (cid:4) are positive coefªcients to be determined empirically. Xij refers
to exports of country i to country j. Yi and Yj are the national gross domestic prod-
ucts of countries i and j respectively; Dij is the distance between country i and coun-
try j relative to the average distance between country i and all its trading partners.
For simplicity of exposition, the time subscript is avoided. Taking the logarithm, IL
base line of equation (1) can be conveniently represented in log-linear form as equa-
zione (2).

lnXij

(cid:5) (cid:6) (cid:7) (cid:2)lnYi

(cid:7) (cid:4)lnYj

(cid:8) (cid:3)lnDij .

(2)

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

The real-world situation is too complex to be represented by a simple equation like
(2). The geographical size, population, trade policies, and openness to trade of the
importing country are also important factors affecting exports from any country. È
a bilateral relationship and representing such factors by a vector of variables Zij, E
an error term ((cid:9)
ij) representing other left out variables and the deviation of the se-
lected functional form from the actual relationship whose impact on export is con-
sidered to be on average negligible. Così, the gravity equation (2) can be written in
a more general form as equation (3). Così, equation (3) in general can be estimated
taking panel of data across time and across countries.

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

13 Drysdale, Huang, and Kalirajan (2000) used the disequilibrium framework to evaluate the
efªciency of China’s bilateral trade with its 57 trading partners for the period 1991–95, E
Kalirajan (2000) used it to examine Australia’s export efªciency with its trading partners in
IOR-ARC.

Asian Economic Papers

Export Performances in China and India

lnXij

(cid:5) (cid:6) (cid:7) (cid:2)lnYi

(cid:7) (cid:4)lnYj

(cid:8) (cid:3)lnDij

(cid:7) (cid:10)Zj

(cid:7) (cid:9)

ij .

(3)

Researchers have used a number of dummy variables in the set of Zij to augment the
modello. An important assumption in this model is that the exporting environment in
the home country does not impose any restrictions on the home country’s exports.
In other words, this model although admitting that there are behind the border con-
straints in the home country and also that the home country faces beyond the bor-
der constraints in partner countries, these constraints are not important and are ran-
domly distributed across observations. In other words, the assumption is equivalent
to saying that there are no signiªcant behind and beyond the border constraints for
exports of home country. Tuttavia, effects of regional trading arrangements, con-
nectivity by road/sea, language afªnities, historical relationships, and product pref-
erences shown through brand names have been included in the gravity of equation
(3). OLS methods or variants of OLS have been used to estimate models such as
equation (3).

3.2 Methodology II
In Methodology I, it was assumed that behind and beyond the border constraints to
export are not signiªcantly affecting export ºows from the home country (China
and India). This means that the impact of behind and beyond the border constraints
to export on export ºows from China and India are merged with the statistical error
term (cid:9) with “normal” characteristics in equation (3). Tuttavia, such an assumption
may be restrictive and may not be in line with reality. We would like to elaborate on
this by concentrating on important means to promote trade ºows between coun-
tries. One such means is trade liberalization. Trade liberalization, from a theoretical
viewpoint, promotes efªciency by re-allocating resources to productive uses, stimu-
lates competition, increases factor productivity, increases trade ºows, and thereby
promotes economic growth (Wacziarg 1997). Tuttavia, empirical facts on trade
ºows across countries do not always support this theoretical viewpoint. Export
ºows are constrained by three factors: (UN) natural constraints, which are geograph-
ical distance and transport cost; (B) behind the border constraints, which are institu-
tional and infrastructure rigidities that exist in exporting countries; E (C) beyond
the border constraints, which are institutional and infrastructure rigidities that exist
in importing countries. The impact of the latter constraints can be divided into two
groups, namely, “explicit beyond the border” constraints and “implicit beyond the
border” constraints. Beyond the border constraints, which are explicit, are mainly
tariffs and exchange rate. The impacts of these constraints on the home country’s ex-
ports may be measured from the coefªcients of variables such as average tariffs and
real exchange rate, which can be included directly into the gravity model. On the
other hand, identifying and measuring implicit beyond the border constraints that

Asian Economic Papers

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Export Performances in China and India

emanate from institutional and policy rigidities of importing countries are very
difªcult, and are considered as “given” for the present study.

Nevertheless, these implicit beyond the border constraints can be reduced or elimi-
nated through multilateral and bilateral negotiations to a considerable extent. Essere-
hind the border constraints in the home country could arise due to socio-economic,
institutional, and political factors in the home country. Per esempio, large govern-
ment size (Rodrik 1998), weak and inefªcient institutions in the home country in
terms of, Per esempio, custom and regulatory environments, port inefªciency, E
inadequate e-business (Bhagwati 1993; Rodrik 2000; Levchenko 2004), and political
inºuences through powerful lobbying by organized interest groups (Gawande and
Krishna 2001) have been found to affect export ºows, among other things. The com-
bined effects of behind the border constraints to export, Tuttavia, which may be in-
terpreted as an “economic distance” factor referred to by Anderson (1979) E
Roemer (1977) can be measured on export ºows. This requires that the error term of
the standard gravity model (3) needs to be decomposed into u, indicating the impact
of behind the border constraints, and v, indicating “normal” statistical errors and
implicit beyond the border constraints.

lnXij

(cid:5) (cid:6) (cid:7) (cid:2)lnYi

(cid:7) (cid:4)lnYj

(cid:8) (cid:3)lnDij

(cid:7) (cid:4)Zj

(cid:8) uij

(cid:7) vij .

(4)

Così, apart from the geographical distance constraint, the behind the border con-
straints and explicit beyond the border constraints need to be included explicitly
into the standard gravity model. Unfortunately, most of the empirical trade models
do not consider this argument, as they do not incorporate these constraints into
their trade model.14

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

Tuttavia, OLS estimation of the gravity equation (4) leads to biased results. Draw-
ing on Kalirajan (2007), the procedures developed for estimating stochastic frontier
production functions (Aigner, Lovell, and Schmidt 1977; Meeusen and van den
Broeck 1977), which do not require the researchers to have information on the exact
components of u, can be used to estimate the modiªed gravity equation that in-
cludes the impact of behind the border constraints and explicit beyond the border
constraints to export for a given level of implicit beyond the border constraints.

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

14 Recentemente, Anderson and van Wincoop (2003) suggested an approach to tackle this problem,
which they name as “multilateral resistance.” However, their suggested method suffers
from a number of limitations. Per esempio, they assumed symmetric trade costs to solve
their model, which is an unrealistic assumption. Also, their modeling of multilateral resis-
tance, as a function of distance and tariffs only, ignores the presence and impact of variation
in behind the border trade resisting factors in home country, and the implicit beyond the
border constraints in respective importing countries.

Asian Economic Papers

Export Performances in China and India

The estimation procedure requires the assumption, which may be veriªed statisti-
cally, that u is a truncated (at zero) normal with mean (cid:11) and variance su
2 and takes
values either 0 or greater than 0. When u takes the value 0, this means that the im-
pact of behind the border constraints are not important and the actual exports and
potential exports are the same, assuming that the inºuence of v is not signiªcant
(cioè., v (cid:5) 0). When u takes the value other than zero, this means that the effects of be-
hind the border constraints are important and they reduce potential exports de-
pending on the value of u. Così, the term u represents the difference between poten-
tial and actual exports in logarithmic values, which is a function of the inefªciencies
that are within the exporting countries’ control. It is also assumed that error term v
captures the inºuence on trade ºows of other variables, including measurement er-
rors and implicit beyond the border constraints that are not under the control of the
exporting country and are randomly distributed across observations in the sample.

Maximum likelihood methods can be used to estimate the above modiªed gravity
model and the magnitude of u. Computer programs such as STATA and FRONTIER
4.1 can be used to estimate the modiªed gravity model.15

3.3 Data
The trade data are taken from the Direction of Trade Statistics of the IMF. Data on
real GDP, which is a proxy for the size of the economy; population (POP), area
(AREA), and tariff barriers are taken from the World Development Indicators (WDI)
2004 and WDI CD-ROM 2004. The most recent information on weighted average
tariff rate for the primary products (TBPR), manufactured products (TBMFG) and all
prodotti (TBALL) have been used.

Openness to trade is measured by trade in goods taken as a fraction of the gross do-
mestic product (TRDGZ). Perception about prevailing restrictions on imports pub-
lished in World Competitiveness Report 2004 of World Economic Forum (WEF)
(Sala-i-Martin 2004) has been used to proxy non-tariff barriers. The non-tariff barrier
is calculated as an index (NTBI) on a scale of 1–7 where lower values of index indi-
cate higher non-tariff barrier. Così, the expected sign of NTBI is positive. Factors
such as the macroeconomic environment, the quality of public institutions, and tech-
nology are also important determinants, and are likely to affect the intensity of im-
port across countries. WEF publishes a growth competitiveness index (GCI) on a
scale of 1–7 where a higher value indicates a higher level of competitiveness. IL
GCI is founded on the previous three factors and, interestingly, GCI and NTBI are
highly correlated (Sala-i-Martin 2004). Therefore, these variables are used selectively.
All variables are taken in logarithms or fractions.

15 Details of the estimation procedure of FRONTIER 4.1 are given in Coelli (1996).

Asian Economic Papers

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Export Performances in China and India

4. Empirical results and discussion

4.1 Absence of behind the border constraints
Both models estimated in this study for China and India separately were as follows:

lnXij

(cid:5) (cid:6) (cid:7) (cid:2)lnGDPj

(cid:7) (cid:4)lnDISij

(cid:7) (cid:3)lnPOPj

(cid:7) (cid:9)

ij ,

lnXij

(cid:5) (cid:6) (cid:7) (cid:2)lnGDPj
4TBPR (cid:7) (cid:3)

(cid:7) (cid:3)

(cid:7) (cid:4)lnDISij
5NTBI (cid:7) (cid:9)

ij .

(cid:7) (cid:3)

1lnPOPj

(cid:7) (cid:3)

2TRDGZ (cid:7) (cid:3)

3LAREA

(5)

(6)

The variables are as deªned earlier. Over a small span of time the relative size of the
trading partners and the exporting environment in the home country are not ex-
pected to change signiªcantly. Therefore, for the purpose of analyzing trading char-
acteristics of the countries concerned during the recent period, the average values of
exports during 2000–03 and average size of economies for 2000–02 are considered
appropriate.16 Data on trade restrictions and openness to trade are also taken for the
period 2000–02. Così, there is an inbuilt lag in the value of explanatory variables. In
the place of NTBI, the variable GCI was also used in the estimation for India. The se-
lected sample sizes of the partner countries, which are the same 77 countries for
both China and India, represent about 90 percent and 80 percent of exports from
China and India, rispettivamente, and therefore the estimated models can be considered
to be representative models for these economies in a general equilibrium frame-
work.17 All the equations were estimated by OLS and a complete diagnostic result is
provided in the respective tables. A series of estimations have been completed to de-
lineate the strengths and weaknesses of both countries. At the outset, the basic
modello (5) with GDP, distance, and population with respect to partner countries was
estimated for China and India and the results are reported in Table 1. The base
model was further expanded to include the proxies of openness and explicit beyond
the border constraints and the results are presented in Table 2.

Almost all the estimated equations are statistically consistent, and the R2 values are
reasonably high. Tuttavia, the magnitudes of the coefªcients are markedly different
between China and India. Whether the size and signiªcance of these variables are

16 Because 2001 is characterized by a number of political and terrorist disturbances, including
data from 2000 is expected to provide a better average, while considering the most recent
available consistent data for countries of interest. Further, there are statistical advantages in
taking average values as it reduces the problems of heteroskedasticity and functional forms
leading to more reliable interpretation of the relationships.

17 For the purpose of the present study of comparing the performances of China and India em-
ulating the exporting environment of each other, it is necessary to consider the same coun-
tries with which both China and India traded during the sample periods.

Asian Economic Papers

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Export Performances in China and India

Tavolo 1. Base gravity model with distance, aggregate GDP in terms of US$ at 1995 prices and population for China and India, 2000–03 Code: Model Number: Sample size: CONSTANT LDIST LGDP LPOP R2 S.E. Diagnostic test Serial correlation Function form Normality Heteroskedasticity China CH-9 77 (cid:8)7.071** (2.951) (cid:8)0.773* (0.180) (cid:8)0.882* (0.068) (cid:8)0.076 (0.097) 0.836 0.870 0.386 [0.53] 1.072 [0.30] 28.126 [0.0] 3.28 [0.07] India IN-9 77 (cid:8)4.398*** (2.437) (cid:8)1.021* (0.175) (cid:8)0.633* (0.059) (cid:8)0.149(cid:7) (0.095) 0.793 0.856 [0.08 [0.7] [0.791 [0.37] [0.252 [0.88] [0.109 [0.74] Note: When there is problem of heteroskedasticity, White heteroskedasticity adjusted standard errors are presented. Values in parenthe- ses () are standard errors and values in square brackets [] are p values. *Signiªcant at the 1 percent level; **Signiªcant at the 5 percent level; ***Signiªcant at the 10 percent level; Signiªcant at the 15 percent level. robust or not in the presence of other variables is an important issue and is dis- cussed later. The relative distance variables in both models of China have smaller coefªcients than those of India.18 It appears that the production process in China, which is characterized by large manufacturing volumes, is able to absorb the dis- tance effects much more efªciently than India. The production cost in China is com- paratively lower than that in India and the advantage derived from this is reºected in the size of the relative distance variable. It may be noted that the average distance of China from its trading partners is greater than that of India from its trading part- ners (Tavolo 3). Therefore, India has to be more efªcient in cost management in order to compete with China in the same product group or else it has to design alternative strategies related to product and market. Per esempio, empirical studies examining the costs of doing business in India often have cited that private ªrms have to have their own power generators in order to avoid the problem of a power shortage, which tends to increase production costs (Rajan 2006). Further, China is more con- cerned with other barriers to trade rather than distance. Per esempio, in Model CHN-14 (Tavolo 2), the relative distance variable becomes insigniªcant when a tariff barrier to primary sector products is introduced. Inoltre, as new variables are added, the coefªcient of the relative distance variable in China’s models continues to reduce. Therefore, it can be safely argued that China’s cost advantages are great instruments to boost their exports compared to India. 18 The results could have been better, had we disaggregated exports of China and India by commodity categories such as labor-intensive, agriculture-intensive, and resource-intensive. We thank the discussant, Lael Brainard, for pointing out this issue. 15 Asian Economic Papers l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Export Performances in China and India Table 2. Augmented gravity model with area, openness to trade, and other trade barriers for China and India, 2000–03 Code: Model Number: Sample size: CONSTANT LDIST LGDP LPOP TRDGZ LAREA TBPR NTBI GCI R2 S.E. Diagnostic test Serial correlation Function form Normality Heteroskedasticity China CH-14 77 (cid:8)13.858* (3.395) (cid:8)0.269 (0.202) (cid:8)0.641* (0.132) (cid:8)0.432*** (0.229) (cid:8)0.007* (0.0025) (cid:8)0.141 (0.093) (cid:8)0.032*** (0.018) (cid:8)0.313 (0.240) 0.870 0.790 0.006 [0.94] 0.910 [0.34] 56.90 [0.00] 5.280 [0.02] India IN-13 77 (cid:8)11.680* (2.616) (cid:8)0.567* (0.182) (cid:8)0.409* (0.086) (cid:8)0.666* (0.150) (cid:8)0.0060** (0.0027) (cid:8)0.145*** (0.078) (cid:8)0.355** (0.146) 0.846 0.750 0.319 [0.57] 0.093 [0.70] 3.011 [0.22] 0.355 [0.55] India IN-14 77 (cid:8)10.94* (2.540) (cid:8)0.542* (0.184) (cid:8)0.300** (0.123) (cid:8)0.742* (0.169) (cid:8)0.0056** (0.0027 (cid:8)0.160** (0.079) (cid:8)0.560** (0.228) 0.846 0.75 0.71 [0.40] 0.195 [0.66] 0.591 [0.74] 0.529 [0.47] Note: When there is the problem of heteroskedasticity, White heteroskedasticity adjusted standard errors are presented. Values in pa- rentheses () are standard errors and values in square brackets [] are p values. *Signiªcant at the 1 percent level; **Signiªcant at the 5 percent level; and ***Signiªcant at the 10 percent level. Tavolo 3. Summary matrix of distances (kilometer) being negotiated by China and India across the sample structures of trade partners Sample size: Mean Minimum Maximum China 77 9,931.5 956.2 19,286.0 India 77 8,490.4 678.6 16,937.4 l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . The coefªcient of size of the economy measured by GDP is consistently signiªcant in all formulations. The size of this coefªcient is larger for China than that for India in both models. Tuttavia, when variables such as openness to trade and growth competitiveness are added in the model, the size of coefªcient of GDP reduces for China and India (Vedi la tabella 2 in comparison with Table 1). Nevertheless, the coefªcient of GDP is larger for China than for India. This means that clearly India has to progress signiªcantly to manufacture and export premium products con- sumed in richer countries as compared to the manufacturing activities in China. f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Population is indirectly covered in the size of the economy, and it can be argued to have independent demand side effects also. Per esempio, subsistent economies also need basic livelihood amenities such as cheap clothing and food. Countries such as China and India, which have a high degree of mechanized production systems with 16 Asian Economic Papers Export Performances in China and India cheap labor, could be a potential source of imports provided the importing country has a conducive trade regime. This fact is revealed when the coefªcients of popula- tion variable across models are compared. The openness to trade variable (TRDGZ) is introduced in Models CH14, IN13, and IN14 along with the area variable (Tavolo 2). Clearly, exports ºow more from both countries to those countries, which trade a higher proportion of their GDP. The coefªcient of TRDGZ is almost equal for both China and India. In the case of China, GCI is not a signiªcant variable; instead, tariff barriers to primary sector products are more important in reducing its exports. Even non-tariff barriers are insigniªcant in affecting China’s exports. D'altra parte, in the case of India, non-tariff barri- ers and the growth competitiveness index act alike in affecting its exports growth. Recall that the expected sign of coefªcient of NTBI is positive because a higher value of NTBI means fewer problems in importing, whereas lower values mean the oppo- site. To calculate potential exports, it is important to estimate the equation in a general equilibrium framework so that as many trading partners as possible, indicating as much distance as possible, are covered. Nevertheless, such a general equilibrium framework may not take into account all country-speciªc characteristics of the home country that inºuence its exports. Therefore, in this exercise we put each country in the exporting environment of the other to simulate each other country’s potential exports. The key difference in export performance is expected to arise due to the change in the values of the relative distance variable, as all other variables remain more or less the same across trading countries. Models CHN14 and IND14 given in Table 2 were used for simulating the exports from China and India with the assump- tion that they switched their exporting environments between them. Simulations were carried out by applying the coefªcient of India, which proxies the exporting environment faced by India, on trade data concerning China and vice versa. The simulated gain/loss in exports is presented in Appendixes 1 E 2. As a summary, when the coefªcients of China are applied to calculate India’s simu- lated potential exports, it results in very high values for India (672.9 per cento), which implies that if India enjoys China’s exporting environment, it would increase its ex- ports drastically. D'altra parte, when India’s coefªcients are applied to China, it leads to lowering of exports from China by 91.7 per cento, clearly indicating that China has been operating at much higher efªciency levels than India. Così, there is much for India to learn from China to improve its export performance. This result also implies that there are signiªcant behind the border constraints to export more in India than in China, which is examined in the next section. 17 Asian Economic Papers l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Export Performances in China and India Table 4. Modiªed augmented gravity model with area, openness to trade, other trade barriers, and behind the border constraints to export for China and India, 2000–03 Code: Sample size: CONSTANT LDIST LGDP LPOP TRDGZ LAREA TBPR GCI T Sigma square Gamma Eta Mu Loglikelihood China 77 (cid:8)12.675* (3.262) (cid:8)0.258 (0.208) (cid:8)0.644* (0.138) (cid:8)0.429** (0.217) (cid:8)0.006* (0.0023) (cid:8)0.139 (0.096) (cid:8)0.036** (0.016) (cid:8)0.322 (0.254) (cid:8)0.228 (0.321) (cid:8)0.543*(0.115) (cid:8)0.834*(0.226) (cid:8)0.138**(0.068) (cid:8)0.43**(0.22) (cid:8)157.68 India 77 (cid:8)8.56* (2.228) (cid:8)0.549* (0.178) (cid:8)0.314** (0.118) (cid:8)0.728* (0.175) (cid:8)0.006** (0.003) (cid:8)0.147** (0.072) (cid:8)0.566** (0.232) (cid:8)0.186 (0.202) (cid:8)0.642*(0.221) (cid:8)0.875*(0.232) (cid:8)0.067(0.121) (cid:8)0.56**(0.272) (cid:8)120.67 Note: Values in parentheses () are standard errors. *Signiªcant at the 1 percent level; **Signiªcant at the 5 percent level. 4.2 Presence of behind the border constraints Drawing on Kalirajan (2007), the following modiªed augmented gravity model was estimated using panel data from 2000–03 and the results are presented in Table 4: lnXijt (cid:5) (cid:6) 1 (cid:7) (cid:3) (cid:7) (cid:2)lnGDPjt (cid:7) (cid:3) 4TBPRt (cid:7) (cid:4)lnDISijt (cid:7) (cid:3) 5NTBIt (cid:7) (cid:3) 6T (cid:7) vijt 1lnPOPjt (cid:8) uijt . (cid:7) (cid:3) 2TRDGZt (cid:7) (cid:3) 3LAREAt (7) ituij (cid:5) (cid:13) The variables are as deªned earlier and T refers to time, which takes values 1, 2, 3, E 4 rispettivamente, for data from 2000, 2001, 2002, E 2003. The variable uij is as- sumed to be non-negative truncations of the normal distribution with mean, (cid:11), E (cid:5) { esp[(cid:8)(cid:13)(T (cid:8) T )] }uij means variance, (cid:12)2. Further, the assumption that uijt that behind the border constraints to export have been varying over time. This as- sumption implies that if the estimate of (cid:13), which is provided by the computer pro- gram FRONTIER 4.1 simultaneously along with the parameters of equation (7), is positive then the behind the border constraints decline exponentially to its mini- mum value, uij, at the last period T of the panel. In questo caso, the gap between poten- tial and actual exports has been declining. The coefªcient estimates for constant, which is larger than the estimates of equation (6) as expected due to the speciªcation of equation (7), and most variables are signiªcant at least at the 5-percent level. Fur- ther, these coefªcient estimates have the signs that concur with the theory. The coefªcient (cid:4) presents a measure of the total variation that is due to country speciªc behind the border constraints to export. IL (cid:4) coefªcient is an average over the peri- od. Questo è, (cid:4) (cid:5) [((cid:14) ut one-sided error term at period t, (cid:12)2 vt is the variance of the random error term at pe- riod t, and T is the total number of periods. The estimate of (cid:4) is large and signiªcant vt)] / T, where is (cid:12)2 ut is the variance of the ut) / ((cid:14) (cid:7) (cid:12)2 (cid:12)2 (cid:12)2 t t 18 Asian Economic Papers l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Export Performances in China and India at the 1 percent level. This means that the decomposition of the error term into u and v in equation (7) is valid for the present data set and the deviation of actual ex- ports from potential exports is due to behind the border constraints and not by just random chances. It may be interesting to see how the (cid:4) coefªcients vary over time. This is equivalent to examining whether the inºuence of behind the border con- straints to export within the home country have been decreasing from one period to another or not. To put it differently, it investigates whether policy reforms toward promoting exports in China and India have been effective during the sample period. Information on the temporal behavior of (cid:4) can be obtained by examining the (cid:13) coefªcient. IL (cid:13) coefªcient considers whether the impact of country speciªc behind the border constraints on reaching potential exports have been decreasing from one period to another or not. If the (cid:13) coefªcient were positive, then the impact of country speciªc behind the border constraints to export would be decreasing over time. If, Tuttavia (cid:13) were zero or not signiªcant, then the impact of country speciªc behind the border constraints to export could be considered constant over time. In this model, IL (cid:13) coefªcient is positive and signiªcant for China, whereas it is positive but not sig- niªcant for India. This implies that policy reforms in India do not appear to be effec- tive in reducing behind the border constraints to export during the sample period, though policy reforms seem to be effective in China. Overall, from these results the following can be inferred. Behind the border con- straints (measured by u) contribute a large and signiªcant proportion to the varia- tion in the gaps between potential and actual exports in equation (7) for both China and India. This point is further emphasized by the signiªcance of (cid:4). In other words, country-speciªc factors including trade policy are important determinants of poten- tial and actual exports. The results given in Table 4 indicate that the impact of be- hind the border constraints to export has reduced over time during the sample peri- od for China but not for India. With the existing trade resistance between China and its trading partners, and India and its trading partners, China has been able to re- duce the gap between its potential and actual exports with a majority of the member countries more than India could do over time. The analysis shows that an average of approximately 86 percent of potential exports have been realized by China, whereas only about 68 percent of potential exports have been realized by India (Appendix 3). This clearly indicates that there is an urgent need to design and intensify trade pol- icy reforms to enhance its effectiveness toward reducing constraints to export in India and in this respect, India certainly can learn from China’s experience, which requires a detailed study. Tuttavia, India needs to study carefully the recently de- bated regional income inequality problems created by China’s surging export reve- 19 Asian Economic Papers l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Export Performances in China and India nues in order to avoid the occurrence of such social problems while increasing In- dia’s exports.19 5. Conclusion Thus, China’s export performance contrasted with that of India over the years indi- cates that an important determinant of the beneªts that developing countries can reap from globalization is whether behind the border constraints to export can be decreased consistently through appropriate policy measures. Tuttavia, although this study did not explore what kind of behind the border constraints need to be eliminated in India to facilitate the realization of its export potential, conjectures can be made from China’s experience. Drawing on Hayami (1997) who argued that poor countries could structure their institutions to bring about rapid development through the borrowing of technologies, the adoption of technology from abroad is important for India, and appears to be constrained mainly by a lack of infrastructure and proper institutions. “Catching up with China” is a worthwhile slogan for India’s new millennium, along with a national commitment to grow at 10 percent a year. Both goals may be feasible and attainable, and within India’s grasp, provided infrastructure and institutional reforms are intensiªed effectively. China has not only managed a high rate of invest- ment, but has kept the prime lending rate (PLR) at a relatively low 8 per cento; the in- terest rate spread between lending and deposit rates was conªned to 2.6 per cento. In India, the PLR is 12 per cento, and the interest rate spread is at 3.4 per cento. Clearly, China’s conªgurations are more conducive to high domestic investment. Even though the Indian stock markets were established much earlier than China’s, in terms of market capitalization, China is ahead at US$ 231.3 billion, che è 2.20
times that of India’s. Chinese banks extend credit, measured as a ratio of GDP, at a
rate of two-and-a-half times India’s. Even in ªscal decentralization, the Chinese
Central government transfers 51.4 percent of the tax revenue to the provinces,
whereas in India the ªgure is about 36.1 per cento.

This discussion has revealed important ªndings, which can be helpful in making
strategies with respect to trade policy in India. The cost competitiveness of China
appears to help its exports in negotiating large distances. India needs to learn from
China. It has to develop cost advantage and product process so that high-value mar-
kets can be captured. Duties and taxes are still on the higher side as compared to
world standards, and they need to be reduced further, as higher duties and taxes

19 We are thankful to Zhang Xiaojing for pointing out this important issue to us.

Asian Economic Papers

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Export Performances in China and India

lead to higher domestic prices and reduced market size by reducing domestic con-
assunzione, and hence deprive the scale-of-economy effect and make Indian ªrms
less competitive. A larger consumption base will lead to an increase in labor produc-
tivity through competition and provide backstop to domestic producers against
external shocks. Duties merit reduction on several other grounds also. The proven
technological potential of the country can best be exploited and made robust by ex-
posing the economy to external competition by strategically reducing tariffs. Basso-
level tariffs have strong signaling effects, besides reducing inefªciencies in resource
allocation and operations. A relatively restrictive foreign investment regime in India
needs review. FDI ºows should be viewed as a vehicle of technology transfer, spill-
over effects in production processes, and of increasing exports.20 Continuation of
small-scale industry reservation in the case of many sectors of production deprives
the beneªts of scale economy and a strategic decision of de-reservation should be
taken for all the products where export potential exists. The poor quality of public
infrastructure including power and transport remains a key problem for business
enterprises (see Appendix 4). The sooner it is rectiªed the better and, Perciò, it is
argued that the government should continue its efforts in building infrastructure in-
stead of managing production units. Relatively sluggish clearing at ports and cus-
toms houses and rampant corruption are increasing costs to domestic manufactures
and they must be addressed through technological measures and a greater partici-
pation of the private sector. The state-owned port trust is extremely inefªcient, E
the government has rightly assigned some responsibilities to international operators
recently.

It is not that India has not proved its successful performance in the trade sector. As
argued by Rajan (2006), India has proven that it could compete in the services trade
sector despite the poor infrastructure in high-value-added, high-skill industries
where the output is relatively lightweight and relatively less dependent on ports
and electricity. Per esempio, during the 1990s, India’s service sector grew at an aver-
age annual rate of 9 per cento, contributing to nearly 60 percent of the overall growth
rate of the economy. Further, India’s exports of services grew annually on average at
17 percent per year in the 1990s, which is about two and a half times faster than the
domestically focused part of the services sector (Hoekman 2004).

Così, it is argued that India should nurture this comparative advantage effectively
by relaxing behind the border constraints rather than introducing new constraints

20 Unlike other studies, which are cross-country based, this study is country-speciªc (India vs.
its trading partners and China vs. its trading partners) and therefore FDI could not be used
as an explanatory variable in the gravity model estimation.

Asian Economic Papers

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Export Performances in China and India

such as over-regulation of the higher education system. Yet, in order to provide sus-
tained employment to several million people, India cannot underestimate the bene-
ªts of following the East Asian growth model of labor-intensive manufacturing,
which is also causally linked with the services sector.

Appendix 1. Simulated annual potential exports of China using coefªcients from the India
modello

Simulated potential exports
(US$ million) Percentage difference of simulated poten- tial exports over actual average exports Algeria Argentina Australia Austria Bangladesh Bolivia Brazil Cameroon Canada Chad Chile Colombia Costa Rica Denmark Dominican Republic Ecuador Egypt El Salvador Ethiopia Finland France Germany Ghana Greece Guatemala Honduras Hungary Indonesia Italy Jamaica Japan Jordan Kenya Korea RP (S) Madagascar Malawi Malaysia Mali Mauritius Mexico Morocco Mozambique Netherlands New Zealand Nicaragua Nigeria Norway Pakistan Panama Paraguay Peru Philippines Poland Portugal Romania CHN as IND 63.6 63.4 193.7 213.3 342.4 9.1 346.7 20.3 321.0 4.7 80.2 92.0 22.0 207.0 36.6 23.2 170.3 32.2 52.1 186.4 898.8 1,825.9 40.9 119.2 23.4 10.6 155.9 599.2 639.3 11.1 5,197.4 41.9 41.9 2,947.7 15.3 16.1 651.3 9.3 15.9 363.7 88.4 15.5 663.2 54.1 7.7 149.2 105.3 378.0 10.3 6.8 53.9 548.8 245.7 144.7 79.2 CHN as IND (cid:8)82.31 (cid:8)90.11 (cid:8)97.21 (cid:8)76.11 (cid:8)77.45 (cid:8)16.65 (cid:8)84.04 (cid:8)66.43 (cid:8)95.49 313.41 (cid:8)94.08 (cid:8)73.70 (cid:8)80.32 (cid:8)87.59 (cid:8)71.85 (cid:8)88.02 (cid:8)83.61 (cid:8)82.25 (cid:8)46.23 (cid:8)88.15 (cid:8)88.36 (cid:8)90.36 (cid:8)80.69 (cid:8)87.97 (cid:8)91.83 (cid:8)90.21 (cid:8)91.47 (cid:8)86.29 (cid:8)90.91 (cid:8)88.00 (cid:8)91.33 (cid:8)89.43 (cid:8)80.25 (cid:8)84.29 (cid:8)89.25 53.63 (cid:8)89.27 (cid:8)79.37 (cid:8)91.89 (cid:8)89.75 (cid:8)81.06 (cid:8)55.61 (cid:8)94.68 (cid:8)93.84 (cid:8)89.03 (cid:8)87.94 (cid:8)88.28 (cid:8)68.84 (cid:8)99.51 (cid:8)96.83 (cid:8)80.98 (cid:8)86.85 (cid:8)82.36 (cid:8)66.64 (cid:8)77.56 22 Asian Economic Papers l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Export Performances in China and India Appendix 1. (continued) Simulated potential exports (US$ million)

Percentage difference of simulated poten-
tial exports over actual average exports

Russia
Senegal
Singapore
South Africa
Spain
Sri Lanka
Sweden
Svizzera
Tanzania
Thailand
Trinidad And Tobago
Tunisia
Turkey
Uganda
United Kingdom
stati Uniti
Uruguay
Venezuela
Zambia
Zimbabwe

CHN as IND

370.0
17.6
1,212.1
190.4
501.2
89.1
272.5
276.8
39.3
921.5
12.6
72.5
267.8
37.2
1,101.0
3,611.7
11.7
46.5
12.4
15.4

Fonte: Author’s estimation from the results of Table 3.

CHN as IND
(cid:8)90.37
(cid:8)78.49
(cid:8)89.12
(cid:8)89.68
(cid:8)88.20
(cid:8)88.81
(cid:8)84.38
(cid:8)84.13
(cid:8)71.01
(cid:8)81.27
(cid:8)75.40
(cid:8)50.66
(cid:8)82.85
13.15
(cid:8)92.84
(cid:8)96.74
(cid:8)94.77
(cid:8)89.64
(cid:8)66.47
(cid:8)61.37

Appendix 2. Simulated annual exports of India using coefªcients from the China model

Simulated potential exports
(US$ million) Percentage difference of simulated poten- tial exports over actual average exports IND as CHN IND as CHN Algeria Argentina Australia Austria Bangladesh Bolivia Brazil Cameroon Canada Chad Chile Colombia Costa Rica Denmark Dominican Republic Ecuador Egypt El Salvador Ethiopia Finland France Germany Ghana Greece Guatemala Honduras Hong Kong Hungary Indonesia Italy Jamaica Japan Jordan Kenya 401.8 917.0 2,417.0 3,158.2 968.9 58.9 4,155.2 89.4 4,682.1 13.4 711.6 596.5 171.3 2,602.5 254.9 166.4 1,030.4 180.4 193.1 1,979.9 14,365.2 27,411.3 104.0 1,462.5 162.3 54.9 15,387.4 1,259.7 3,828.3 10,234.4 70.2 40,031.6 189.8 151.0 574.4 834.8 412.9 3,286.5 (cid:8)14.2 1,635.3 911.0 462.0 530.2 350.4 790.5 754.8 1,529.2 1,257.1 1,640.3 1,617.1 239.2 3,586.1 166.7 2,850.7 1,209.8 1,181.0 15.7 904.4 674.2 249.9 511.9 1,874.0 522.9 632.0 816.2 2,055.9 113.9 (cid:8)9.2 23 Asian Economic Papers l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a s e p a r t i c e – p d / l f / / / / / 7 1 1 1 6 8 2 3 9 3 a s e p 2 0 0 8 7 1 1 p d . . . . . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Export Performances in China and India Appendix 2. (continued) Simulated potential exports (US$ million)

Percentage difference of simulated poten-
tial exports over actual average exports

IND as CHN

Korea RP (S)
Madagascar
Malawi
Malaysia
Mali
Mauritius
Mexico
Morocco
Mozambique
Netherlands
New Zealand
Nicaragua
Nigeria
Norway
Pakistan
Panama
Paraguay
Peru
Philippines
Poland
Portugal
Romania
Russia
Senegal
Singapore
South Africa
Spain
Sri Lanka
Sweden
Svizzera
Tanzania
Thailand
Trinidad And Tobago
Tunisia
Turkey
Uganda
United Kingdom
stati Uniti
Uruguay
Venezuela
Zambia
Zimbabwe

9,952.2
82.3
48.5
5,286.0
40.5
113.3
3,057.8
295.0
60.2
9,688.7
611.2
43.0
469.9
1,401.1
1,962.5
82.1
61.5
338.1
2,308.9
2,004.6
1,655.4
477.3
3,162.9
92.0
15,646.4
1,996.9
6,681.6
425.4
3,210.8
3,121.1
117.9
5,243.0
114.8
245.6
2,353.7
160.7
14,061.0
50,080.9
139.0
377.5
54.9
93.2

1,095.2
611.1
97.0
680.6
67.9
(cid:8)32.8
833.6
331.5
47.4
925.4
721.1
1,725.0
28.2
1,759.7
885.1
134.3
704.7
861.5
695.0
1,434.3
965.9
1,343.1
392.0
227.4
1,193.6
424.8
785.7
(cid:8)45.6
1,626.6
753.4
8.7
698.9
843.6
382.2
471.8
156.8
453.9
388.1
479.7
944.5
95.9
510.7

Fonte: Author’s estimation from the results of Table 3.

Appendix 3. Realization of potential exports (%) of China and India with partner countries

Number of partner countries

Realization of potential exports (%)

40–49
50–59
60–69
70–79
80–89
90–100

Mean level of realization of potential exports, %

Fonte: Author’s estimation from the results of Table 4.

China

6
9
12
16
30
4

China

India

10
13
31
15
6
2

India

Asian Economic Papers

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Export Performances in China and India

e
R
tu
T
C
tu
R
T
S
UN
R
F
N
io

sì
T
io
C
io
R
T
C
e
l
E

e
R
tu
T
C
tu
R
T
S
UN
R
F
N
io

sì
T
io
l
UN
tu
q

sì
T
io
l
UN
tu
q
sì
l
P
P
tu
S

sì
T
io
l
UN
tu
q

l
UN
T
S
o
P

e
N
o
H
P
e
l
e
T

T
R
o
P

D
UN
o
R
l
io
UN
R

T
R
o
P
S
N
UN
R
T

R
io
UN

e
R
tu
T
C
tu
R
T
S
UN
R
F
N
io

sì
T
io
l
UN
tu
q

e
R
tu
T
C
tu
R
T
S
UN
R
F
N
io

l
l
UN
R
e
v
O

sì
T
io
l
UN
tu
q

k
N
UN
R

e
R
o
C
S

k
N
UN
R

e
R
o
C
S

k
N
UN
R

e
R
o
C
S

k
N
UN
R

e
R
o
C
S

k
N
UN
R

e
R
o
C
S

k
N
UN
R

e
R
o
C
S

k
N
UN
R

e
R
o
C
S

1
5

7
4

3
4

7
4

8
6

5
4

7
5

0
.
6

4
.
5

8
.
6

5
8

0
6

0
.
3

2
.
4

8
.
6

9
6

4
5

6
1

2
.
3

7
.
3

6
.
5

0
2

7
3

7
.
4

7
.
3

7
.
6

7
4

8
6

1
3

8
.
4

9
.
3

3
.
5

9
6

4
5

6
1

2
3

7
3

6
5

UN
io
D
N
IO

UN
N
io
H
C

N
UN
P
UN
J

S
e
io
R
T
N
tu
o
C
D
e
T
C
e
l
e
S

S
S
o
R
C
UN
sì
T
io
l
UN
tu
q
e
R
tu
T
C
tu
R
T
S
UN
R
F
N
io

F
o
X
e
D
N
io

S
S
e
N
e
v
io
T
io
T
e
P
M
o
C

e
v
io
T
UN
l
e
R

.
4

X
io
D
N
e
P
P
UN

.
T
S
e
B

S
’
D
l
R
o
w
S
UN

T
N
e
io
C
ª
F
e

D
N
UN

e
v
io
S
N
e
T
X
e

S
UN

7
D
N
UN
D
e
P
o
l
e
v
e
D
R
e
D
N
tu

F
o

e
R
o
C
S

.
2
0
1

D
e
R
e
D
io
S
N
o
C

S
e
io
R
T
N
tu
o
C

F
o

R
e
B
M
tu
N

l
UN
T
o
T

:
e
T
o
N

M
tu
R
o
F
C
io
M
o
N
o
C
E
D
l
R
o

,
4
0
0
2
–
3
0
0
2

T
R
o
P
e
R
S
S
e
N
e
v
io
T
io
T
e
P
M
o
C

l
UN
B
o
l
G

:
e
C
R
tu
o
S

Asian Economic Papers

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Export Performances in China and India

Riferimenti

Aigner, Dennis, Lovell, Knox, and Peter Schmidt. 1977. Formulation and Estimation of Stochas-
tic Frontier Production Function Models. Journal of Econometrics 6:21–37.

Anderson, James. 1979. A Theoretical Foundation for the Gravity Equation. American Economic
Review 69:106–116.

Anderson, James, and Eric van Wincoop. 2003. Gravity with Gravitas: A Solution to the Border
Puzzle. American Economic Review 93 (1):170–192.

Baldwin, Richard. 1994. Towards an Integrated Europe. London: CEPR.

Baldwin, Richard. 2003. Openness and Growth: What’s the Empirical Relationship? NBER
Working Paper No. 9578. Cambridge, MA: National Bureau of Economic Research.

Bergstrand, Jeffrey H. 1985. The Gravity Equation in International Trade: Some Microeconomic
Foundations and Empirical Evidence. The Review of Economics and Statistics 67:478–481.

Bhagwati, Jagdish. 1993. India in Transition: Freeing the Economy. Oxford: Oxford University
Press.

Chadha, Rajesh. 2003. External Sector. In: Mid-Year Review of the Indian Economy 2002–2003, ed-
ited by N. N. Vohra, pag. 78–82. Delhi: Shipra Publishing.

China Custom Statistical Bureau. 2002. China Custom Statistical Yearbook. Beijing: China Custom
Statistical Bureau.

Coelli, Tim. 1996. A Guide to FONTIER Version 4.1: A Computer Program for Stochastic Fron-
tier Production and Cost Function Estimation. Center for Efªciency and Productivity Analysis
Working Paper No. 96/07. Armidale, Australia: Center for Efªciency and Productivity Analy-
sis, University of New England.

Drysdale, Peter, Yiping Huang, and Kaliappa Kalirajan. 2000. China’s Trade Efªciency: Meas-
urement and Determinants. In: APEC and Liberalisation of the Chinese Economy, edited by Peter
Drysdale, Zhang Yunling, and Liagang Song, pag. 259–271. Canberra, Australia: Asia Paciªc
Press.

Fan, Gang, and Xiaojing Zhang. 2003. The Chinese Reform Agenda. In: China’s Role in Asia and
the World Economy—Fostering Stability and Growth, edited by J. J. Teunissen, pag. 214–228. IL
Hague: Fondad.

Frankel, Jeffrey A., and David Romer. 1999. Does Trade Cause Growth? American Economic Re-
view 89 (3):379–399.

Garnaut, Ross, and Yiping Huang. 2001. Growth Without Miracles: Readings on the Chinese Econ-
omy in the Era of Reform. New York: Oxford University Press.

Gawande, Kishore, and Pravin Krishna. 2001. The Political Economy of Trade Policy: Empirical
Approaches. Working paper, Economics Department. Providence, RI: Brown University.

Greenaway, David, Wyn Morgan, and Peter Wright. 2002. Trade Liberalization and Growth in
Developing Countries. Journal of Development Economics 67:229–244.

Hayami, Yujiro. 1997. Development Economics: From the Poverty to the Wealth of Nations. Oxford:
Clarendon Press.

Asian Economic Papers

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Export Performances in China and India

Hoekman, Bernard. 2004. The World Bank Trade Research Program: Summary and Synthesis.
Development Economics Research Group (DECRG). Washington, D.C.: World Bank.

Huang, Yasheng, and Tarun Khanna. 2003. Can India Overtake China? Foreign Policy Magazine
July–August. Available at www.foreignpolicy.com/story/

IMF. 2004. World Economic Outlook. Washington, D.C.: International Monetary Fund.

Jiang Xiaojuan. 2002. The Foreign Invested Sector in China: Its Contribution to Economic Growth,
Structure Upgrade and Power of Competitiveness. Beijing: People’s University Press.

Kalirajan, Kaleeswaran. 1999. Stochastic Varying Coefªcients Gravity Model: An Approach in
Trade Analysis. Journal of Applied Statistics 26 (2):185–193.

Kalirajan, Kaliappa. 2000. Indian Ocean Rim Association for Regional Cooperation (IOR-ARC):
Impact on Australia’s Trade. Journal of Economic Integration 15:533–547.

Kalirajan, Kaliappa. 2007. Regional Cooperation and Bilateral Trade Flows: An Empirical
Measurement of Resistance. The International Trade Journal 21:85–107.

Lardy, Nicholas. 2002. Integrating China into the Global Economy. Washington, D.C.: Brookings
Institution Press.

Levchenko, Andrei A. 2004. Institutional Quality and International Trade. IMF Working Paper
No. WP/04/23. Washington, D.C.: IMF.

Lucas, Robert. 1988. On the Mechanics of Economic Development. Journal of Monetary Econom-
ics 22:3–42.

Meeusen, William, and Julian van den Broeck. 1977. Efªciency Estimation from Cobb–Douglas
Production Function with Composed Error. International Economic Review 18:435–444.

NBS(UN) (National Statistical Bureau). various years. China Statistical Yearbook. Beijing: China Sta-
tistics Press.

NBS(B) (National Statistical Bureau). various years. China Foreign Economic Statistical Yearbook.
Beijing: China Statistics Press.

Nilsson, l. 2000. Trade Integration and the EU Economic Membership Criteria. European Jour-
nal of Political Economy 16 (4):807–827.

Rajan, Raghuram. 2006. India: The Past and its Future. Asian Development Review 23 (2):36–52.

Roberts, Segno, and Mark Setterªeld. 2007. What Is Endogenous Growth Theory? In: Economic
Growth: New Directions in Theory and Policy, edited by P. Arestis, Michelle Baddeley, and J. S. l.
McCombie, pag. 14–31. Cheltenham, UK: Edward Elgar.

Rodrik, Dani. 1998. Why Do More Open Countries Have Bigger Governments? Journal of Politi-
cal Economy 106 (5):997–1032.

Rodrik, Dani. 2000. Trade Policy as Institutional Reform. Cambridge, MA: Harvard University,
Department of Economics.

Roemer, John E. 1977. The Effect of Sphere of Inºuence and Economic Distance on the Com-
modity Composition of Trade in Manufactures. The Review of Economics and Statistics 59:318–
327.

Romer, Paul. 1986. Increasing Returns and Long-run Growth. Journal of Political Economy
94:1002–1037.

Asian Economic Papers

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Export Performances in China and India

Sachs, Jeffrey, and Andrew Warner. 1995. Economic Reform and the Process of Global Integra-
zione. Brookings Papers on Economic Activity 1:1–95.

Sachs, Jeffrey, and Wing Thye Woo. 2000. Understanding China’s Economic Performance. Jour-
nal of Policy Reform 4 (1):31–48.

Sachs, Jeffrey, and Wing Thye Woo. 2003. China’s Economic Growth After WTO Membership.
Journal of Chinese Economic and Business Studies 1:1–31.

Sala-i-Martin, Xavier. 2004. The Global Competitiveness Report 2003–2004. New York: Oxford Uni-
versity Press.

Wacziarg, Robert. 1997. Trade, Competition and Market Size. Cambridge, MA: Harvard Univer-
sity Press.

Wang, Xiaolu. 2004. FDI in People’s Republic of China. In: Managing FDI in a Globalizing Econ-
omy: Asian Experiences, edited by David H. Brooks and Hal Hill, pag. 79–117. London: Palgrave/
MacMillan.

Woo, Wing Thye. 1998. Chinese Economic Growth: Sources and Prospects. In: The Chinese Econ-
omy, edited by Michel Fouquin and Françoise Lemoine, pag. 134–152. London: Economica.

Wu, Jing-Long. 2003. Economic Reform in Contemporary China. Shanghai: Shanghai Far East
Press.

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
S
e
P
UN
R
T
io
C
e
–
P
D

F
/

7
1
1
1
6
8
2
3
9
3
UN
S
e
P
2
0
0
8
7
1
1
P
D

B
sì
G
tu
e
S
T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Asian Economic Papers
Scarica il pdf