FDI Location Choice in China: Agglomeration vs.Institutions

Julan Du, Yi Lu and Zhigang Tao 1

Abstract

Using an extensive data set on US invested enterprises in the Chinese mainland, we explore

the role of agglomeration economies (network externalities) and government institutions as

well as other more traditional factors in determining the (regional) locational choice of foreign

direct investments. Employing firm-level discrete choice model analysis, we find that provinces

with higher concentration of foreign investments and domestic enterprises with forward or

backward linkage to the potential foreign enterprises tend to attract more foreign investments.

More importantly, stronger regional government institutions reflected in such aspects as less

interventionist in the private sector, more protective of intellectual property rights and less

severe government corruption tend to boost the entry of foreign invested enterprises into those

provinces. Moreover, we obtain mixed evidence regarding the effects of the interplay between

agglomeration economies and institutional quality on FDI locational choices and the differential

impacts of institutions on FDI entry in the mode of joint ventures or fully-owned enterprises.

Keywords:

JEL Classification Number :

1>From the Chinese University of Hong Kong, University of Hong Kong and Univesity of Hong Kong respec-tively.

1

I. Introduction

In China’s economic transition and development, the state sector has kept dwindling while

the nonstate sector has been growing fast. As one striking feature of the Chinese economy, for-

eign direct investment (FDI) is a major component of the nonstate sector. China has emerged

as the largest FDI recipient country in the world, and FDI is widely regarded as making signifi-

cant contributions to China’s economic growth. For instance, World Bank (1997) credited FDI

as a main driving force behind China’s economic miracle. Since FDI boosts local economic de-

velopment, different regions in China are attracting FDI by various means. For instance, local

governments provided special tax incentives in numerous special investment zones to adjust the

relative attractiveness of investment locations in their favor (Boyreau-Debray and Wei, 2004).

Even the central government is also trying to lure FDI into the western hinterland by launch-

ing the massive "Go West" program. However, FDI still exhibits a highly uneven distribution

across regions in China with the east coast area taking the lion’s share. What are the location

determinants of FDI in China? The conventional view suggests that factors such as production

costs and infrastructure adequacy are the main determinants of FDI flows.

This paper investigates the importance of institutional quality and agglomeration economies

in addition to the conventional factors like production costs and infrastructure as determinants

of FDI locational choices. Institutional quality refers to rule of law, property rights protection

and government interference with business operations. Agglomeration economies catch the

network externality effects of regional concentration of business operations.

Using an extensive firm-level dataset on US-invested enterprises (UIEs) in China, we employ

discrete choice model developed by McFadden (1974) (see Anderson, Palma and Thisse (1992)

for an excellent summary) to examine the factors determining the locational choices of FDI from

the US. Our empirical analysis shows that regions (provinces) with stronger institutions, higher

horizontal agglomeration (higher agglomeration of UIEs and domestic enterprises engaged in

the same industry), and higher vertical agglomeration (higher concentration of downstream and

upstream enterprises to which the potential UIE has forward and backward linkages) are more

likely to attract US-based multinational enterprises (MNEs) to set up business operations in

those regions. It is noteworthy that these findings are obtained after controlling for conven-

tional factors shaping FDI locational choices such as regional wage costs and infrastructure

quality. In addition, we explore the interplay of agglomeration economies and institutions in

determining FDI locational choice and obtain mixed findings. On the one hand, the positive

network externalities may be more powerful in regions with stronger public institutions; on the

other hand, agglomeration, especially FIE clustering, may enhance the collective bargaining

power of FIEs against local governments so that agglomeration plays a more important role in

2

regions with weaker institutions.

We further investigate the impact of agglomeration and institutions on the choice of entry

modes between joint ventures (JVs) and fully owned enterprises (FOEs), i.e., the subsidiaries

of US-based MNEs. We find that stronger regional institutions and a higher level of horizontal

agglomeration promote the establishment of both JVs and FOEs. However, there is no clear-cut

pattern of the differential impact of regional institutions on the likelihood of setting up JVs

or FOEs. Sometimes regional institutions exert a positive impact of larger magnitude on the

likelihood of setting up FOEs than on that of establishing JVs; while other times the reverse

pattern appears. We interpret these results by emphasizing the dual role potentially played by

local business partners: they can either help or expropriate the foreign investor.

As the largest FDI recipient country, China has recently caught much attention in the

academic literature on FDI locational choice. For instance, Head and Ries (1996), Cheng

and Kwan (2000), He (2002), Chang and Park (2005), and Amiti and Javorcki (2005) address

the effects of agglomeration on FDI location determination in China. Belderbos and Carree

(2001), Fung, Iizaka and Parker (2002), Zhou, Delios and Yang (2002), and Fung, Iizaka and

Siu (2003) examine a host of FDI location determinants, but they did not touch upon the roles

of agglomeration and institutions. Limited by the unavailability of firm-level foreign invested

enterprises (FIEs) data, most of these studies only include city-level, region-level or industry-

level data. 2

The studies of the impact of institutions on FDI flows are rather limited. In a cross-country

study using aggregate data, Wei (2000a, 2000b) finds that corruption in a host country sub-

stantially deters inward FDI. Campos and Kinoshita (2003) use an aggregate panel data set for

transition economies to show that institutions, mainly reflected in rule of law and government

regulations of businesses, are an important determinant of FDI flows. As a matter of fact,

cross-country studies are likely to confound numerous factors. In contrast, our single-country

analysis focuses on the variation in institutional strength across regions within one country –

China. This allows us to hold constant many aspects such as political system, legal tradition,

de jure legal codes, culture and language, national tax policies, and trade policies that could

vary dramatically across countries. This helps us single out the aspects of institutional qual-

ity that are most closely related to the effectiveness of law enforcement and the efficiency of

government institutions. Furthermore, as far as we know, ours is the first study that examines

the impacts of institutional quality on FDI locational choice by using firm-level data. In so

2As an exception, Chang and Park (2005) employ the firm-level data to examine the determinants of FDIlocation choice of Korean firms in China. However, they mainly focus on the role of agglomeration effects withoutconsidering regional institution strength.

3

doing, we can virtually minimize the concern for endogeneity (including reverse causality) issue

in econometric analysis. At the same time, our study is also the first one that investigates

how the cross-region variation in institutional quality affects FDI geographical distribution in

China. China is a vast country with substantial regional disparity in institutional quality and

industry agglomeration as well as infrastructure, production costs and human capital endow-

ments. This rich variation across regions makes China an ideal platform to study the impact of

institutions and agglomeration in determining FDI locational choices. Our study also examines

the interplay of institutional quality and agglomeration economies, which, to the best of our

knowledge, is the first attempt to address this issue.

The earlier studies on FDI locational choice in China most often employ city-, region-

or industry-level FDI without identifying and classifying the source country. Because of the

different nature of the source countries, the total FDI flows may cover up many important

features of FDI. As we know, a large proportion of FDI into the Chinese mainland comes from

Hong Kong, Taiwan and Macao that can be called the ethnically Chinese economies (Huang,

2003). As these source areas share the same culture as the Chinese mainland, the FDI from

these areas typically shows different characteristics than those from other source countries. For

instance, FIEs from these areas are typically small and medium enterprises operating simple

and labor-intensive production and assembly processes. As a result of low technology content in

FDI, they may not be very sensitive to the regional differential in the protection of intellectual

property rights. Moreover, FIEs from these areas may not be so suscepitible to the inadequacy

of regional institutions as the cultural affinity may help them overcome the institutional barriers.

They know better how to deal with local governments through various connections.

Our study singles out FDI from one country – the US. US FDI is generally regarded as

more "truly foreign" in nature. Consequently our dataset on US FDI provides a good setting to

examine the impact of regional institutions on FDI locational choice. FDI from the US is also

particularly ideal for exploring the effects of local institutions on FDI. The 1977 Foreign Corrupt

Practices Act forbids American companies from making "grease payments" to foreign officials

to speed up transactions (Bardhan, 1997). Therefore, UIEs may be particularly vulnerable to

weak public institutions. Looking at US FDI locational choice could provide a clearer picture

of how sensitive FDI is to the disparity in regional institutions.

The rest of the paper is organized as follows. Section II discusses the importance of ag-

glomeration and institutions in FDI locational choice. The data and variables are described

in Section III. Section IV lays out the empirical estimation strategy. Results are discussed in

Section V. Section VI concludes the paper.

4

II. Agglomeration vs. Institutions

2.1. Agglomeration

Agglomeration comprises both horizontal and vertical agglomeration. Horizontal agglom-

eration has two distinct aspects. One is the agglomeration of FDI source country firms of the

same industry in the same region, i.e., the agglomeration of UIEs in our study. The other is

the agglomeration of domestic enterprises of the same industry in the same region.

The growing literature on new economic geography mainly lists the following potential

benefits (positive network externalities) of horizontal agglomeration – it creates knowledge

spillovers; it improves access to specialized labor and intermediate inputs; it improves access

to and sharing of information about markets and technology trends; it improves access to

infrastructure and public goods; and it increases competitive pressure among firms (Krugman,

1991; Porter, 1998).

On the other hand, agglomeration could also generate negative externalities. A firm’s own

knowledge and technologies can spill over to other firms when there is a concentration of firms.

Agglomeration may give rise to intensified competition in both product and factor markets

among adjacently located firms. It could also reduce or discourage innovation via groupthink

(Chang and Park, 2005). 3

Vertical agglomeration reflects the concentration of domestic firms with backward and for-

ward linkages to the UIEs, i.e., the domestic upstream and downstream firms and final goods

consumers in the same region. Most of the recent developments in the new economic geography

theories focused on the role of backward and forward linkages. In addition to the benefits and

costs of agglomeration listed above, vertical agglomeration could have one distinct advantage

– it promotes complementarities and cooperation among firms through backward and forward

linkages. The concentration of upstream firms indicates the accessibility to component suppliers

in the region, whereas the concentration of downstream firms and final goods consumers shows

the accessibility to market in the region. Producers typically like to choose locations that have

good access to large markets and to suppliers of intermediate inputs. Vertical agglomeration

in the region will increase the variety of intermediate inputs or final goods available for choice

and lower the average purchasing costs, will enhance the chances of matching and mitigate the

3A growing literature applies the new economic geography theory to test the importance of agglomeration inFDI locational choice. Head, Ries and Swenson (1995) show that there do exist agglomeration effects of Japanesemanufacturing firms in the United States. Afterwards, many other papers study the clustering of multinationalfirms using data from different countries, e.g. Portugal by Guimaraes, Figueiredo and Woodward (2000); Chinaby Head and Ries (1996), Cheng and Kwan (2000), He (2002), Chang and Park (2005), and Amiti and Javorcki(2005); France by Crozet, Mayer and Mucchielli (2004); and Hungury by Boudier-Bensebaa (2005).

5

holdup problem in contracting between the upstream producer and downstream client, and will

generate knowledge spillover through learning (Krugman and Venables, 1995; Venables, 1996;

Duranton and Puga, 2004).

It is often the case that the horizontal and vertical agglomeration are bundled together. A

place that for whatever reason already has a concentration of producers tends to offer a large

market because of the demand the producers and their workers generate, and tends to provide

a good supply of inputs and consumer goods made by the producers already there (Fujita,

Krugman and Venables, 2001).

Vertical agglomeration and thus access to suppliers and markets are expected to be partic-

ularly important determinants of FDI location choice in China. As pointed out by researchers

such as Young (2000), Bai, Du, Tao and Tong (2004), and Amiti and Javorcki (2005), there

exists prevalent regional protectionism in China, i.e., regional governments are erecting barriers

to entry of goods from other provinces. Regional economic fragmentation causes FDI fragmen-

tation. Case studies conducted by Huang (2003) demonstrate that FIEs often find it difficult

to obtain component supplies from or ship goods to regions other than the one the FIEs are

located. As a consequence, FIEs are often forced to invest in different regions as a way to obtain

intermediate goods or supply their products to their customers. Naturally, regions with higher

backward and forward agglomeration become ideal locations for FIEs because they offer larger

supplier and market access.

We expect that if the positive externalities dominate the negative ones, regional agglomer-

ation promotes FIE entry and vice versa.

2.2. Institutions

Regional institutions mainly refer to the state of rule of law, the government intervention in

the private sector, the government protection of property rights and government corruption in

a region. Regions with weak institutions are typically characterized by weak rule of law, heavy

government intervention, inadequate protection of property rights and severe corruption, which

may increase the expropriation risks to UIEs and harm US business interests. In recent years,

various cross-country and within-country studies such as, among others, Besley (1995), Knack

and Keefer (1995, 1997), Mauro (1995), Hall and Jones (1999), La Porta, Lopez-De-Silanes,

Shleifer and Vishny (1999), Acemoglu, Johnson, and Robinson (2001, 2002) (See Pande and

Udry (2005) for a brief review) have produced largely consistent results that a high quality of

public institutions contributes to a good economic performance. We thus expect that regions

with stronger institutions will be more appealing to UIEs.

China is a unitary state with uniform de jure laws across the country. However, public

institutions may exhibit wide variation in law enforcement effectiveness across regions. In this

6

sense, examining the variation in institutional strength across regions in China allows us to

conduct a natural experiment to focus on the de facto law enforcement after holding constant

the de jure legal codes. This certainly offers a better setting to distinguish between legal codes

and law enforcement than the cross-country analysis does.

2.3. Interplay of Agglomeration and Institutions

Furthermore, we investigate the interplay between the effects of agglomeration economies

and those of institutions. Institutions could affect the nature and effectiveness of network

externalities.

(1) Horizontal Agglomeration and Institutions

Theoretical consideration gives ambiguous predictions about the role of horizontal agglom-

eration in regions with different strength of institutions.

On the one hand, in regions with weaker institutions, it could be the case that the neg-

ative side of horizontal agglomeration dominates the positive side so that the importance of

agglomeration economies declines. For instance, in regions with weaker protection of intellec-

tual property rights, knowledge spillover among closely located companies could more likely be

transformed into the infringement of patent rights. In regions with heavy government interven-

tion and rampant corruption, the potential intensified competition and congestion of access to

infrasture and public goods stemming from agglomeration may deteriorate to firms competing

for preferential treatments and privileges through illegitimate means such as bribing government

officials. From this angle, weak institutions would cripple the positive effects of agglomeration

in attracting FDI.

However, on the other hand, agglomeration could under some circumstances alleviate the

negative effects of weak institutions, and thus the importance of agglomeration becomes larger

in regions with weaker institutions. For instance, with agglomeration of UIEs in a region, UIEs

could share information about local markets, which helps the new entrant know better the

local market environment and local governments and enhances the survival chances of the new

entrant. UIEs could also cooperate with each other in negotiating with the local governments,

which increases the bargaining power of the UIEs against the corrupt or interventionist govern-

ment. Thus cooperation among firms, particularly among UIEs, could also partially overcome

the weak institutional environment. In this sense, the positive impact of horizontal agglomera-

tion, especially that of the clustering of UIEs, on US FDI entry will be more salient in regions

with weaker institutions.

So, it is an empirical question what the net effect of agglomeration on UIE entry in regions

with strong or weak institutions is.

(2) Vertical Agglomeration and Institutions

7

Vertical agglomeration implies supplier and market access in each region. The public insti-

tutions in each region shape the contracting environment between the FIEs and the suppliers

or customers. When a region has better institutions, contracting between the FIEs and the

component suppliers or customers will go more smoothly, and thus the advantages of a larger

variety of suppliers and a larger market size associated with the backward and forward agglom-

eration will be more likely brought into full play. Consequently, the backward and forward

agglomeration effects are expected to be more striking in regions with better institutions.

2.4. FDI Entry Modes, Agglomeration and Institutions

In terms of entry modes, FDI can take various forms. In our dataset, we find that US

firms typically set up a JV with a local partner firm or establish an FOE. This difference in

organizational structure could have profound implications for FDI locational choice. We expect

that the determinants of the FDI locational choice, especially agglomeration economies and

regional institutional strength, would exhibit different patterns in influencing FDI entry as JVs

or FOEs. Theoretically speaking, the positive network externalities of US firm agglomeration

are expected to be stronger in the case of FOEs than in the case of JVs because local partners

in JVs may provide substantial help in coping with local markets and local institutions so that

the importance of the prior experience of other UIEs would decrease. In contrast, the positive

network externalities of domestic firm agglomeration could be more pronounced in the case of

JVs than in the case of FOEs because the local partners of joint ventures can better deal with

and cooperate with domestic firms of the same industry than a wholly-owned FIE and thus are

better able to take advantage of the concentration of domestic firms.

However, theory does not have clear predictions about whether backward and forward ag-

glomeration will have differential impacts on the entry of FIEs as JVs or FOEs. Generally

speaking, the markets for intermediate inputs and final goods are essential to both JVs and

FOEs. However, differentiation could take place under some circumstances. For example, if

local business partners play a more significant role in securing supplier access than market ac-

cess, then downward agglomeration casts a more salient impact on JVs (with local partners)

than FOEs (without local partners), and vice versa.

Regional institutional quality may also have differential effects on FIE entry in the form of

JVs and FOEs. However, the theoretical prediction is still somewhat ambiguous. On the one

hand, it is likely that local partners in JVs can help deal with local governments and overcome

the barriers posed by the inadequacy of local institutions so as to smooth business operations,

while FOEs have to cope with local governments on their own. In this scenario, we expect that

FOEs will be more sensitive to regional institutional quality than JVs do – better regional

institutions promote FOEs more than JVs.

8

On the other hand, it is possible that in JVs local partners may hold up the foreign partner

in business operations or expropriate the assets or technology of the foreign business partner,

a situation that is more likely and more severe under weaker regional institutions. Under this

circumstance, FOEs can rule out the local partner’s holdup problem or expropriation risks,

and thus display more resistance to weak institutions than JVs do. In contrast, JVs are more

sensitive to regional institutions and exhibit stronger tendency toward moving to regions with

stronger institutions than FOEs do.

So, it is again that only empirical tests can tell us which scenario fits better the reality.

III. Data and Variables

3.1. Data

Our data on US-invested enterprises in China come from a broad dataset of FIEs in China

compiled by China National Bureau of Statistics. This extensive dataset on FIEs contains

150,602 FIEs in 2001, accounting for 74.44% of the total 202,306 FIEs in China as reported

by China Statistical Yearbook 2002. Among them, our dataset has 141,668 enterprises engaged

in the manufacturing sector, covering 75.45% of the total number of foreign manufacturing

enterprises in China in 2001.

Our study focuses on FIEs from the US. US ranks as the third largest source of FDI to

China, after only Hong Kong and Taiwan. In our sample, UIEs account for 8.82% of the total

number of the FIEs, and UIEs engaged in the manufacturing sector constitute 8.64% of the

total number of manufacturing FIEs. As argued earlier, Hong Kong and Taiwan are ethnically

Chinese economies, and FDI from these areas may not be representative of FDI in general.

Thus US is the largest "truly foreign" source country of FDI to China.4 Though our dataset

covers only one year (2001), we follow the common practice in the literature by using the year

in which a UIE is registered as the year of its entry. This enables us to identify the entry year

of all UIEs.

Originally our sample contains 13,290 UIEs spanning the period 1983-2001. After deleting

those UIEs without registration dates, we have 13,270 firms. By restricting our analysis to

the manufacturing sector and deleting those UIEs involving individual US investors, we obtain

4Following the US, Japan and Korea are the fourth and fifth largest source countries of FDI in China. FDIfrom these two countries has been extensively studied in the literature. Japan and Korea are geographicallyclosely related to China and share some similar cultural heritage with China. In this sense, even FIEs fromJapan and Korea are not "truly foreign" enough. Moreover, we are unable to control for the impact of businessgroups (Keiretsu in Japan and Chaebol in Korea) in FDI, which have been shown to be very important in shapingthe location choice of Japanese and Korean firms.

9

a sample of 7521 US firms. We further focus on the period 1993-2001 because the data on

many of the independent variables in regression analysis are not available in the years before

1993 and the FDI flow (including FDI from the US) into China has increased dramatically

only since 1992. Finally we end up with 6,288 UIEs. Table 1 provides a breakdown of US

invested manufacturing firms across 30 different provinces in China. A brief look at this table

tells us that FDI from the US is mainly concentrated in the east coast areas: Jiangsu, Shanghai,

Shandong, Zhejiang, Guangdong, Beijing and Liaoning claim the largest numbers of UIEs.

3.2 Regression Variables

Agglomeration

In this study, we consider both horizontal and vertical agglomeration. Horizontal agglom-

eration has two distinct aspects: the agglomeration of UIEs of the same industry and the

agglomeration of domestic enterprises of the same industry in a region. We gauge the degree

of concentration of the same 4-digit US or Chinese domestic manufacturing enterprises in the

same region by the following two variables respectively.

(1a) Agglomeration_USirt =Number_USirtNumber_USit

(1b) Agglomeration_Domesticirt =Number_DomesticirtNumber_Domesticit

where i represents industry, r denotes region and t indicates year. Obviously, the two

variables measure the proportion of either US or domestic manufacturing firms of a certain

industry i in region r in the total number of US or domestic firms of a certain industry i in the

whole country in a particular year.

Vertical agglomeration reflects the concentration of the domestic upstream or downstream

firms in the same region. The backward and forward agglomerations are defined as

(2a) Backwardirt = Σjαij

Number_domesticjrtNumber_domesticjt

(2b) Forwardirt = Σjβij

Number_domesticjrtNumber_domesticjt

+ βiCGDPrtGDPt

where αij is the input-output ratio reflecting the inputs from the upstream industry j

required for one unit of output of industry i; βij is the input-output ratio showing the input

made by industry i required for one unit of output of downstream industry j; and βiCGDPrtGDPt

indicates proportion of the final output demand for industry i’s output by region r in the total

final output demand for industry i’s output by the whole country. Here we employ regional

GDP to proxy for market demand and use the ratio of regional GDP to national GDP to

indicate the share of final demand accounted for by some particular region.

To construct the vertical agglomeration measures, we select manufacturing firms from the

dataset collected and compiled by the National Bureau of Statistics of China in 2001. The

initial dataset contains 124,317 domestic firms. The input-output ratios are from the China

10

national input-output (I/O) table for 1997. (???)

Clearly, the backward agglomeration indicator shows the proportion of upstream (primary

and intermediate goods) suppliers in region r in total domestic suppliers in the country weighted

by the input-output ratio. It shows the potential network externality of domestic upstream

suppliers in one region for a potential US firm entering the region. The forward agglomeration

indicator reflects the proportion of downstream domestic firms and final demand customers

in region r in total domestic downstream firms and final demand customers in the country

weighted by the input-output ratio. It demonstrates the potential network externality of do-

mestic downstream firms or final output demand in one region for a potential US entrant.

Our backward and forward agglomeration indicators are similar in nature to the supplier

access and market access measures respectively adopted in Amiti and Javorcki (2005). In their

work, industry output instead of number of firms is used to gauge the market access and supplier

access. As in our dataset industry output is only available for year 2001 and as we need to

compute agglomeration measures for different years to match with the establishment years of

various UIEs, we have to employ the number of firms to construct the backward and forward

agglomeration measures. However, our compromised approach still keeps to the spirit of the

output-based measures.

Institutions – Legal Institutions and Law Enforcement

The regional institution indices mainly cover four broad categories of institutional strength.

The first category relates to legal institutions and law enforcement. In particular, it is concerned

with the effectiveness of courts in contract disputes resolution.

(3) Rule of law. Based on the dataset of the Chinese Private Enterprises Survey 1995-2002,

we construct one measure of the effectiveness and reliability of the courts in China’s different

regions. One question in the survey asks private entrepreneurs: When you encounter contract

disputes or breach of contract, will you go to the court to resolve the disputes? We use the

proportion of entrepreneurs choosing to go to court in case of contract disputes as an indicator of

the reliability and effectiveness of courts in providing a good contract environment for private

businesses. It also reflects the general strength of legal institutions and law enforcement in

different regions.

(4) We adopt some index about the legal institution development in China’s different re-

gions. Fan, Wang and Zhu (2003) have followed the process of institutional development and

marketization in China’s different regions over years and constructed the China Regional Marke-

tization Indices. We thus examine the impact of the one-year lagged regional institution indices

on UIE locational choice. Since the compilation of these indices started as late as 1997, we have

to restrict to the subsample of the period 1998-2001 when we use these indices. In the category

11

of legal institutions and contract enforcement, we have the index of market intermediary de-

velopment. This is the index that is most closely related to the legal institutions development

in Fan and Wang’s index system. It employs the proportion of lawyers in a region’s total pop-

ulation to reflect the development of market intermediary organizations. Market intermediary

development is in turn one central precondition for the establishment and the maintenance of

a well-functioning contracting environment.

Institutions – Government Intervention in Business Operations

The second aspect of institutions concerns the degree of government intervention in economic

activities. We employ several different types of measures.

(5) Government Intervention in Contracting. Based on the dataset of the Chinese Private

Enterprises Survey 1995-2002, we construct one measure of government intervention. One ques-

tion asks private entrepreurs: When you encounter contract disputes or breach of contract, will

you go and ask for government help? We use the proportion of entrepreneurs requesting gov-

ernment help in case of contract disputes as an indicator of government intervention in private

businesses. Resorting to government help in solving business disputes has ambiguous implica-

tions for institutional quality. On the one hand, government help may fill the void created by

the lack or weakness of the court system; it can be a substitute for the malfunctioning formal

court system. In other words, the government could provide a helping hand to private entre-

preneurs. If this is the case, UIEs may find government help in contract disputes an appealing

feature of regional governments. On the other hand, government help may involve rent seeking

and even corruption: entrepreneurs lobby or bribe government officials to seek favor in resolving

contract disputes. This becomes the grabbing hand of the government (Frye and Shleifer 1997;

Shleifer and Vishny, 1999). Faced with the hand-binding law against bribing local government

officials and various cultural barriers, the US investors may be particularly disadvantaged in

a region where rent seeking activities are prevalent. So, UIEs may be deterred from making

investments into the regions where governments often intervene in contract disputes resolution.

(6) We adopt some index about the regional government intervention in China from Fan,

Wang and Zhu (2003). In the category of government intervention, the index of government

interference with enterprises is probably most relevant. It specifically addresses the regional

government’s control of and intervention in enterprises’ business operations. It is based on a

questionaire about how large percentage of time the enterprise managers spend dealing with

government agencies and officials.

Institutions – Property Rights Protection

One of the government’s most important responsibilities to ensure a well functioning market

economy is to protect property rights and maintain a secure contracting environment. We focus

12

on the intellectual property rights protection. This has the following advantages. First, com-

pared with many other aspects of property rights protection, the measurement of intellectual

property rights protection is easier to be based on objective measures such as the quantity of

patents. Second, intellectual property rights protection captures the central concern of UIEs

and it has been a thorny issue in US-China trade friction. Unlike small and medium sized FIEs

from ethnically Chinese economies like Hong Kong, Taiwan and Macao, UIEs are typically

large companies with modern technology. This is consistent with the importance of intellec-

tual property in the American economy. According to Israel (2006), US intellectual property

industries account for over half of all US exports; intellectual property accounts for over 1/3

of the value of all US corporations and represents 40% of US economic growth. It is thus not

surprising that FIEs from the US are particularly concerned with intellectual property rights

protection. In recent years, the rising tide of counterfeiting and piracy in China has posed an

enormous threat to US business interests. In a 2005 survey of the US-China Business Council,

members put intellectual property rights protection enforcement at the very top on their list of

concerns. The serious intellectual property infringement in China reflects the lack of proactive

and deterrent intellectual property enforcement, especially at the local level (Israel, 2006; Strat-

ford, 2006). Depending on the difference in government coordination capacity, corruption, staff

training and legal enforcement power across regions, the degree of intellectual property rights

protection also exhibits large variation from region to region. We use the following indicators

to measure intellectual property rights protection.

(7) The number of approved patents per capita across regions. Though patent number could

be an outcome of research and development capacity and inputs, human capital endowment and

other factors in various regions, property rights protection provided by regional governments

no doubt plays an important role.

(8) The index of intellectual property rights protection constructed by Fan, Wang and

Zhu(2003). It is derived, after some statistical treatment, on the basis of the ratio of the

number of various types of patents applied to GDP and the ratio of the number of various

types of patents approved to GDP. We use it as an alternative measure of intellectual property

rights protection.

Institutions – Government Corruption

As US MNEs are constrained by legal prohibition against bribery in foreign countries, we

expect that they may be particularly vulnerable to government corruption in host countries.

We form the following indicator to explicitly measure the degree of government corruption in

China’s different regions.

(9) Government corruption indicator. In the dataset of the Chinese private enterprises

13

survey, there is one question: Do you think it is urgent to eliminate government corruption

in your region? The respondents are required to choose between Yes and No. We use the

proportion of entrepreneurs answering Yes in a region as the government corruption indicator.

Clearly, a higher value of the variable suggests more serious government corruption in the

region. Because of the constraint of data availability, we confine to the subsample of 1998-

2001 when we examine the impact of corruption on US-based FDI entry. Our subnational

corruption measure for China is clearly based on entrepreneurs’ perceptions of the severity of

corruption. As a matter of fact, corruption indices in cross-country comparison such as those

constructed by Transparency International or International Country Risk Guide are always

subjective survey-based indices. In this sense, our measure is consistent with the spirit of the

prevailing cross-national corruption indices.

Other Regional Characteristics as Control Variables

We follow the literature on FDI locational choice to control the following factors in regression

analysis.

(10) Wages. Low production costs mainly reflected in low wages are widely regarded as a

longstanding advantage of China in attracting foreign manufacturing firms. To see how the

regional differentiation in wage costs affects FDI distribution, we include in our analysis the

average manufacturing wages in each region.

(11) Education. The average human capital level of the workforce could be an important

determinant of FIEs from the US as UIEs tend to engage in technology-intensive industries.

We therefore use the proportion of the number of students enrolled in higher education in a

region to its population as a proxy for the average level of human capital in the region.

(12) Infrastructure. It is widely reported in the literature that regions with superior trans-

portation facilities are more appealing to FIEs. We use highway density, i.e., the length of

highway per square kilometer, as an indicator of infrastructure adequacy.

(13) The existence of US consulates. US has embassy in Beijing and consulates in Chengdu,

Guangzhou, Shanghai and Shenyang. US embassy and consulates could play an important

role in facilitating information transmission to UIEs regarding national and regional situations,

which could help UIEs overcome various informational and cultural barriers to making invest-

ments in China. We therefore include a dummy variable taking value one if there is US embassy

or consulate in a region and zero otherwise.

(14) Government promotion policies. The Chinese central government and the local gov-

ernments at various levels set up a large variety of promotion policies to attract FDI. One

important aspect of these promotion policies is establishing different types of special develop-

ment zones. At the national level, the central government set up four special economic zones

14

and fourteen open coastal cities in the 1980s. Later, the central government established various

national-level economic and technological development zones in many cities in various regions.

These areas are granted various types of preferential policies (like preferential tax policy) by the

central government and are allowed to deal flexibly with FIEs. At the same time, the provincial

and the municipal governments have also established numerous provincial- or local-level eco-

nomic and technological development zones and offered special tax incentives to attract FDI.

However, it is virtually impossible to have a clear picture of how many provincial- or local-level

development zones and what kinds of special tax incentives there are in different provinces

because there are no complete statistics from publicly available informational sources. We thus

focus on the national-level zones.

Following Fung, Iizaka and Parker (2002), we adopt two dummy variables. One (SEZD)

takes value one if a region has either special economic zone or open coastal city and zero

otherwise. The other one (ETDZD) takes value one if a region has national economic and

technological development zone and zero otherwise. By including these promotion policies, we

are able to control for the effects of government incentive policies on FDI locational choice and

at least partially distinguish between the effects of regional institutional strength and those of

government promotion policies.

In the data appendix, we present summary statistics for our major regression variables.

IV. Estimation Strategy

In this study, we use the discrete choice model developed by McFadden (1974) to investigate

the importance of various regional characteristics in determining the UIE locational choice. The

conditional logit model is a widely used method to analyze how agents choose from a large set of

alternatives. It focuses on the attributes of each region in the choice set. The model estimates

how each attribute increases or decreases the chances that a region will be chosen rather than

all other potential regions available for choice.

We define an underlying latent variable πij as the profits firm i derives from setting up

a manufacturing operation in region j at time t. Suppose πij is determined by the regional

attributes according to the following formula

πij = θ + β ·Xj + εij (1)

where πij is the profit of the i-th firm locating in region j, Xj is the vector of region j’s

chacteristics, and εij is a disturbance term.

In choosing the location of its establishment, the FIE compares its profits from each region in

15

the choice set and selects the region that generates the maximum profits. Thus the probability

of the i−th firm locating in region j is given by the optimal decision problem

Pi(j) = Pr ob{πij ≥ πik} for all k 6= j

= Pr ob{θ + β ·Xj + εij ≥ (θ + β ·Xk + εik)} for all k 6= j

= Pr ob {εij − εik ≥ β(Xj −Xk)} for all k 6= j (2)

According to McFadden (1974), if εij follows Type I extreme distribution, it can be shown

that

Pi(j) =eβ·XjPk∈K e

β·Xk(3)

where K is the set of locational choices faced by the i-th firm. The above equation is then

estimated by the conditional logit method.

By employing the conditional logit model for firm-level data analysis, we can largely avoid

the various potential endogeneity problems. This is particularly important when we examine

the impact of institutions on FDI locational choice. For instance, in cross-country analysis

using aggregate FDI data such as Wei (2000a, 2000b) and Campos and Kinoshita (2003), the

endogeneity of the institution strength is always a concern. One feature of this endogeneity is

the possible reverse causality: a large amount of total FDI inflow may generate demand for more

efficient governments and force local governments to improve local public institutions. In this

study, we are investigating how individual US firms make locational choice given the regional

institutional strength and agglomeration economies as well as other regional characteristics.

This can minimize the potential reverse causality as an individual US firm is unlikely to have

such a big impact on local institutional quality.

We have two types of regression model specifications. In one set of regression models, the

independent variables consist of four horizontal and vertical agglomeration indicators (variables

1a, 1b, 2a, 2b), one of the institutional strength indicators each time (variables 3-9), and all

the other regional characteristics control variables (variables 10-14). The other set of regression

models further explores the interaction of agglomeration and institutions where we add to

regressions the interaction terms between one institution indicator each time with the four

agglomeration measures respectively.

V. Results

5.1. Agglomeration, Institutions and FDI Locational Choice

Table 2 presents a set of benchmark regressions over the whole sample period 1993-2001

16

where in each regression model we put the rule of law indicator, the government intervention in

contracting indicator, the intellectual property rights protection indicator and the corruption

indicator separately. We find that all the four measures of agglomeration economies generate

positive and statistically significant impact on UIE entry. This suggests that US firms tend

to choose those regions where there are concentration of UIEs engaged in the same industry,

clustering of domestic enterprises of the same industry and wide market and supplier access. It

is interesting that the positive impact of domestic enterprise agglomeration is much larger than

that of UIE agglomeration: if the ratio of the domestic enterprise agglomeration increases

1%, it raises the probability of UIE entry by 3.01%, while a 1% rise in the ratio of UIE

agglomeration boosts the chances of UIE entry by 1.53%. Also interestingly, the effects of

backward agglomeration (supplier access) are much larger than those of forward agglomeration

(market access) on UIE locational choice. It can be inferred that a 1% increment in the ratio

of the backward agglomeration indicator will push up the chances of UIE entry by 11.37%,

whereas the same increment in the forward agglomeration indicator will raise the probability

of UIE entry by 6.43%. This suggests that supplier access is more important in attracting US

firms than market access does. 5

Turn to the institution variables. The rule of law indicator and the intellectual property

rights protection indicator demonstrate the expected positive effects on the likelihood of US

firm entry. This shows clearly that regional institutional strength does play an important role in

shaping FDI locational choices. Similarly, the government corruption indicator produces nega-

tive and statistically significant effects on FDI entry. Interestingly, the government intervention

in contracting indicator exerts negative effects on the FDI from the US. This suggests that

UIEs do regard regional government help in contract dispute resolution as a signal of two-way

rent-seeking activities between the government and the enterprises and shun those regions with

prevalent practice of government intervention in contract disputes resolution.

The other regional characteristics variables all exhibit results consistent with our theoretical

predictions. Regional average wage rate has a negative effect on US firm entry; this reveals

the labor cost effects in US firm locational choice. Highway density in a region promotes US

firm entry, suggesting that basic infrastructure is one essential factor in luring FDI. Human

capital endowment reflected in higher education enrollment also boosts US FDI. This is not

surprising because US FDI tends to involve high level of technology. This is also consistent

with the findings of Fung, Iizaka and Parker (2002) and Gao (2005) that regional labor quality

5The effects of agglomerations on UIE entry calculated in this paragraph are based on the average of theestimated coefficients of the relevant explanatory variable in regressions in Columns 1-3. Regression in Column4 has a much smaller sample size, which may affect the magnitude of the estimated coefficients.

17

significantly affects regional aggregate FDI flows from developed countries including the US.

As expected, the existence of US consulates facilitates US FDI as it provides more channels of

information sharing. The national government promotion policies produce the expected positive

and significant impact on US firm entry. This also helps us to separate out the effects of regional

institutions and those of regional endowment of central government promotion policies.

Table 3 introduces the interaction terms between institution indicators and the four ag-

glomeration economy indicators. The four agglomeration economy indicators and the institu-

tion indicators largely remain statistically significant with the same signs as in Table 2. We

see that the UIE agglomeration economies are stronger in regions with better rule of law and

intellectual property rights protection. Nonetheless, this is not the whole picture. The UIE

agglomeration economies are also stronger in regions with more serious government corruption.

This suggests that FDI concentration could help share information among FIEs and enhance

collective bargaining power with a corrupt local government.

The interplay between domestic enterprise horizontal, backward and forward agglomera-

tions and regional institution only produces statistically significant effect for one institutional

strength measure – intellectual property rights protection. The clustering of domestic enter-

prises of the same industry has weaker externality in regions with better intellectual property

rights protection. This could be the case that stronger regional intellectual property rights pro-

tection promotes the innovation capability of domestic firms, which in turn leads to stronger

competition between UIEs and domestic firms in the same industry. Moreover, the backward

agglomeration effects are stronger in regions with better intellectual property rights protection,

while the forward agglomeration effects are weaker in regions with better property rights pro-

tection. We interpret this result in this way: in regions with better intellectual property rights

protection, the increase in upstream firms’ innovation capacity would enlarge the available va-

riety of raw materials or intermediate inputs for the UIEs to choose so as to encourage US

FDI entry, whereas the enhancement in downstream firms’ innovation capacity could cause the

demand for UIE products to upgrade and change too fast, putting too much pressure on UIEs

and thus deterring US FDI entry to some extent.

Table 4 presents regressions with similar specifications as in Table 2 but uses the regional

institution indices constructed by Fan, Wang and Zhu (2003). Sample period is restricted to

1998-2001. We find that all these institutional indices show statistically significant positive

effects on promoting FDI inflow. In terms of agglomeration economies, the domestic horizon-

tal agglomeration no longer exhibits larger impact than UIE horizontal agglomeration does as

shown in Tables 2 and 3. Actually the magnitude of the former effect is a bit smaller than the

latter one. However, the backward agglomeration still exerts a positive impact of larger mag-

18

nitude than the forward agglomeration does. Other regional characteristics variables present

qualitatively equivalent results as in Tables 2 and 3.

Table 5 lays out regressions introducing interaction terms between agglomeration indicators

and institutions into Table 4. There is no significant impact of the interaction terms between

institutions and US firm horizontal agglomeration. Somewhat similar to the findings in Table

3, the domestic firm horizontal agglomeration effects are also weaker in regions with less severe

government intervention. This could be the case that the competitiveness of domestic firms is

higher in regions with less government intervention, and the concentration of domestic firms

in the same region could pose a threat to UIEs. Similar to what we found in Table 3, the

backward and the forward agglomeration effects are stronger and weaker in regions with better

institutions respectively.

5.2. Agglomeration, Institutions and Entry Modes

We turn to the comparison of location determinants of JVs and FOEs in Table 6. The US

firm horizontal agglomeration effects do not exhibit much difference in statistical significance

between JVs and FOEs, though the magnitude of the estimated coefficients for FOEs are a little

larger than those for JVs. The other three types of agglomeration economies show no difference

in statistical significance for JVs and FOEs either, but the economic significance (magnitude)

clearly differs.

The domestic firm horizontal agglomeration effects are consistently larger in magnitude for

JVs than for FOEs. A 1% increment in the ratio of the domestic agglomeration indicator raises

the chances of US firms entering as FOE by 1.61% but those as JV by 3.57%. 6 This probably

suggests that with a high domestic horizontal agglomeration, the competition pressure from

domestic firms in the same industry in the same region encourages US firms to form JVs so

that local partners may provide some help in coping with the regional market competition.

The positive backward agglomeration effects are smaller in magnitude for FOEs than for

JVs, while the positive forward agglomeration effects are larger in magnitude for FOEs than

for JVs. This suggests that the concentration of upstream suppliers benefit JVs more than

FOEs, while the concentration of downstream firms in the same region benefit FOEs more than

JVs. This could be the case that local partners are particularly needed to deal with component

suppliers. As a result, when UIEs want to enjoy the low-cost supply, they tend to form JVs

with local partners, and when UIEs put more emphasis on serving the local market, they prefer

to establish FOEs.

6This is calculated on the basis of estimated coefficients of domestic agglomeration in regressions in Columns1-3. Regressions in Column 4 have a much smaller sample size that may affect the magnitude of the estimatedcoefficients.

19

In terms of institution indicators, the effects of the government intervention in contracting

indicator and those of the intellectual property rights protection indicator show no difference

in statistical significance, but the magnitude of the effects is larger for FOEs than for JVs. For

instance, an increment of one point in the ratio of the government intervention in contracting

indicator diminishes the probability of US FDI in the form of FOE by 4.18% but that of JV

only by 2.04%.

The rule of law index shows more striking difference. It has statistically significant positive

effects on FOEs but insignificant effects on JVs. This pattern of difference suggests that FOEs

are more sensitive to the local institutional environment because they lack local partners’ help

in dealing with local institutions.

However, the corruption index turns out somewhat surprising results. Though corruption

deters US FDI entry in both JV and FOE modes, the magnitude of the deterrent effect is

larger for JVs than for FOEs. This suggests that the risks of local parters holding up and

expropriating US MNE could be a real concern under corrupt governments so that FOEs are

preferred to JVs.

Other regional characteristics variables also exhibit some interesting patterns of difference

between the two modes of entry. Regional average wage rate has much stronger impact on JVs

than on FOEs. This is probably because JVs are likely to be engaged in more labor-intensive and

less technology-intensive industries than FOEs do. Infrastructure (highway density), in most

cases, casts a larger impact on FOEs than on JVs. It is likely that local partners in JVs can help

US firms overcome infrastructure deficiency. It could also be the case that FOEs are targeting

the national or international market for their products more than JVs do so that transport

facility looms larger in FOE location decision-making. Human capital (education) produces a

larger impact on FOEs than on JVs in most cases. It is reasonable to expect that FOEs set up

by US-based MNEs are more likely to have proprietary technology than JVs do so that FOEs

may set higher standards for their employees’ skills. Surprisingly, the existence of US consulates

generates much more statistically and economically significant effects on JVs than on FOEs.

This seems at odds with our prediction that the information transmission function played by

the US consulates should be more beneficial to FOEs than to JVs. Finally, the government

promotion policies (SEZD and ETDZD) do not show much difference in determining US FDI

entry in the form of JVs or FOEs.

In Table 7, we examine how the institution indices constructed by Fan andWang (1997, 1998,

1999 and 2000) and agglomeration economies as well as other factors determine the locational

choices of US FDI with the two different modes. The four agglomeration indicators produce a

pattern of effects qualitatively equivalent to that in Table 6. The institution indices, however,

20

produce more mixed results than in Table 6. On the one hand, FOEs are more sensitive to

the regional intellectual property rights protection than JVs do. On the other hand, both the

market intermediary development index and the government interference with enterprise index

produce statistically significant positive impacts on US FDI entry as JVs but no significant

effects on UIE entry in the mode of FOEs. This evidence seems to illustrate again the potential

risks that local partners could expropriate US MNEs in JVs so that US firms are willing to

form JVs only in regions with better institutions and thus lower risks of expropriation.

In terms of other regional characteristics, the wage rate, the existence of US consulates,

and government promotion policies reflected in SEZD produce results similar to those in Table

6. Infrastructure (highway density) is no longer more imoprtant for FOEs as in Table 6. On

the contrary, it often produces larger impacts on JVs. The regional average human capital

level (education) also exhibits different results than those in Table 6: it often produces a larger

effect for JVs than for FOEs. The government promotion policies reflected in economic and

technological development zones most of the time no longer produce statistically significant

impacts on both JVs and FOEs.

The difference in the qualitative conclusions drawn from Tables 6 and 7 tells us that the

differential in the impact of institutions as well as some other regional characteristics on FDI

entry modes is quite sensitive to the variation in institution indices adopted and the sample

period and sample size covered. We obtain mixed results regarding the differential impacts of

institutional quality on FOEs and JVs. This probably signifies that local partners potentially

play a dual role in FDI – they may either help or expropriate US-based MNEs when the latter

penetrate China’s markets.

VI. Conclusion

Huge amounts of FDI inflow have been a driving force for China’s rapid economic growth

in recent decades. In a tournament for promoting GDP growth, provincial and municipal

governments in China have made various efforts to attract FIEs to set up business operations

under their jurisdiction. Our study shows that regional institutional quality has been a major

determinant of FDI locational choice. Regions with stronger institutions reflected in better

rule of law, more adequate protection of property rights, less government interference with

businesses and less government corruption will have stronger appeal to foreign investment. We

further explore the effects of the interplay between institutions and agglomeration on FDI entry

and the effects of institutions on FDI entry in the modes of JVs and FOEs. We obtain mixed

results. Both strong and weak institutions can help industry agglomeration economies play a

more significant role in attracting new FDI. This reflects the dual role that industry clustering

21

could play in coping with public institutions. Similarly, weak institutions deter FOEs more

than JVs in some cases and the other way round in other cases. However, our study still

contains clear policy implications for national and regional governments in China as well as in

many other developing countries: strengthening public institutions is one central precondition

for attracting FDI inflow.

References

Acemoglu, Daron, Simon Johnson and James A. Robinson (2001) "The Colonial Origins

of Comparative Development: An Empirical Investigation", American Economic Review 91:

1369-1401

Acemoglu, Daron, Simon Johnson and James A. Robinson (2002) "Reversal of Fortune:

Goegraphy and Institutions in the Making of the Modern World Income Distribution", Quar-

terly Journal of Economics 117: 1231-1294

Amiti, Mary and Beata Smarzynska Javorcki (2005) "Trade costs and location of foreign

firms in China", IMF working paper.

Anderson, Simon P., André de Palma, and Jacques-Francois Thisse (1992) Discrete Choice

Theory of Product Differentiation, Cambridge, Mass.: MIT Press

Bai, Chong-En, Yingjuan Du, Zhigang Tao and Sarah Y. Tong (2004) "Local protection-

ism and regional specialization: evidence from China’s industries", Journal of International

Economics, 63:397-417.

Bardhan, Pranab (1997) "Corruption and development: a review of issues", Journal of

Economic Literature, 35 (3): 1320-46.

Belderbos, René and Martin Carree (2002) "The location of Japanese investments in China:

agglomeration effects, keiretsu and firm heterogeneity", Journal of the Japanese and Interna-

tional Economies, 16: 194-211.

Besley, Timothy (1995) "Property Rights and Investment Incentives: Theory and Evidence

from Ghana", Journal of Political Economy 103: 903-937.

Boudier-Bensebaa, Fabienne (2005) "Agglomeration economies and location choice: FDI in

Hungary", Economics of Transition, 13: 605—628.

Boyreau-Debray, Genevieve and Shang-Jin Wei (2004) "Pitfalls of a state-dominated finan-

cial system: the case of China", CEPR discussion paper No. 4471.

Campos, Nauro F. and Yuko Kinoshita (2003) "Why Does FDI Go Where it Goes? New

Evidence from the Transition Economies", IMF Working Paper.

Chang, Sea-Jin and Sekeun Park (2005) "Types of firms generating network externalities

and MNCs’ co-location decisions", Strategic Management Journal, 26: 595-615.

22

Cheng, Leonard K. and Yum K. Kwan (2000) "What are the determinants of the location

foreign direct investment? the Chinese experience", Journal of International Economics, 51:

379-400.

Coughlin, Cletus C. , Joseph V. Terza, and Vachira Arromdee (1991) "State characteristics

and the location of forign direct investment within the United States", Review of Economics

and Statistics 73, 675-683.

Crozet, Mattieu, Mayer Thierry, and Jean-Louis Muccheilli (2004) "How do firms agglom-

erate? a study of FDI in France", Regional Science and Urban Economics 34, 27-54.

Duranton, Gilles, and Diego Puga (2004) "Micro-foundations of Urban Agglomeration

Economies", in J. Vernon Henderson and Jacques-Francois Thisse (ed.), Handbook of Regional

and Urban Economics, Vol. 4: 2063-2118

Fan, Gang, Xiaolu Wang and Henpeng Zhu (2003) Reports on the Relative Progress of

Marketization in Different Regions in China, Economic Science Publishing House, China.

Friedman, Joseph, Daniel Gerlovski, and Johnathan Silberman (1992) "What attacts for-

eign multinational corporations? Evidences from branch plant location in the United States",

Journal of Regional Science, 32: 403-418.

Frye, Timothy and Andrei Shleifer (1997) "The Invisible Hand and the Grabbing Hand",

American Economic Review Papers and Proceedings 87: 354-358.

Fujita, masahisa, Paul Krugman, and Anthony J. Venables (2001) The Spatial Economy,

MIT Press.

Fung, K.C., Hitomi Iizaka, and Stephen Parker (2002) "Determinats of U.S. and Japanese

direct investment in China", Journal of Comparative Economics 30, 567-578.

Fung, K.C., Hitomi Iizaka, and Alan Siu (2003) "Japanese direct investment in China",

China Economic Review 14, 304-315.

Gao, Ting (2005) "Labor quality and the location of foreign direct investment: evidence

from China", China Economic Review, 16:274-292.

Guimarães, Paulo, Octávio Figueiredo, and Douglas Woodward (2000) "Agglomeration and

the location of foreign direct investment in Portugal", Journal of Urban Economics, 47: 115-135.

Hall, Robert E. and Charles I. Jones (1999) "Why Do Some Countries Produce So Much

More Output per Worker than Others?", Quarterly Journal of Economics 114: 83-116

He, C. (2002) "Information costs, agglomeration economies and the location of foreign direct

investment in China", Regional Studies, 36(9), pp. 1029-36.

Head, Keith, and John Ries (1996) "Inter-city competition for foreign investment: static

and dynamic effects of China’s incentive areas", Journal of Urban Economics, 40: 38-40.

Head, Keith, John Ries and Deborah Swenson (1995) "Agglomeration benefits and location

23

choice: evidence from Japanese manufacturing investments in the United States", Journal of

International Economics, 38: 223-247.

Huang, Yasheng (2003) Selling China: Foreign Direct Investment During the Reform Era,

Cambridge University Press.

Israel, Chris (2006) "Piracy and Counterfeiting in China", Testimony of U.S. Coordinator for

International Intellectual Property Enforcement Before the US-China Economic and Security

Review Commission, June 7.

Knack, Stephen and Philip Keefer (1995) "Institutions and Economic Performance: Cross-

Country Tests Using Alternative Institutional Measures", Economics and Politics, 7: 207-227

Knack, Stephen and Philip Keefer (1997) "Dose Social Capital Have an Economic Payoff?

A Cross-Country Investigation", Quarterly Journal of Economics 112: 1251-1288

Krugman, Paul (1991) "Increasing returns and economic geography", Journal of Political

Economy, 99:483-99.

Krugman, Paul and Anthony J. Venables (1995) "Globalization and the Inequality of Na-

tions", Quarterly Journal of Economics, 110: 857-880

La Porta, Raael, Florencio Lopez-de-Silanes, Andrei Shleifer and Robert Vishny (1999)

"The Quality of Government", Journal of Law, Economics and Organization 15: 222-279

Luger, Michael I. and Sudhir Shetty (1985) "Determinats of foreign plant start-ups in the

United States: lessions for policymakers in the southeast", Vanderbilt Journal of Transnational

Law, 18: 223-245

Mauro, Paolo (1995) "Corruption and Growth", Quarterly Journal of Economics 110: 681-

712

McFadden, Daniel (1974) "Conditional Logit Analysis of Qualitative Choice Behavior", in

Paul Zarembka (ed.), Frontiers in Econometrics, New York: Academic Press, 105-142

Pande, Rohini and Christopher Udry (2005) "Institutions and Development: A View from

Below", working paper, Yale University

Porter, Michael (1998) "Clustering and the new economics of competition", Harvard Busi-

ness Review, 76(6): 77-90.

Shleifer, Andrei and Robert W. Vishny (1999) Grabbing Hand: Government Pathologies and

Their Cures, Harvard University Press.

Stratford, Timothy P. (2006) "China’s Enforcement of Intellectual Property Rights and the

Dangers of the Movement of Counterfeit and Pirated Goods into the United States", Prepared

Statement of Assistant U.S. Trade Representative Before the U.S.-China Economic and Security

Review Commission, June 7.

24

Wei, Shang-Jin (2000a) "How taxing is corruption on international investors?", Review of

Economics and Statistics, 82(1): 1-11.

Wei, Shang-Jin (2000b) "Local corruption and global capital flows", Brookings Papers on

Economic Activity, 2: 303-54.

Woodward, Douglas P. (1992) "Locational determinants of Japanese manufacturing start-

ups in the United States", Southern Economic Journal, 58: 690-708.

World Bank (1997) China Engaged: Integration with the Global Economy, Washington,

D.C.: World Bank.

Venables, Anthony J. ""Equilibrium Locations of Vertically Linked Industries", Interna-

tional Economic Review, 37: 341-359

Young, Alwyn (2000) "The Razor’s Edge: Distortions and Incremental Reform in the Peo-

ple’s Republic of China", Quarterly Journal of Economics, 115: 1091-1135.

Zhou, Changhui, Andrew Delios, and Jing Yu Yang (2002) "Locational determinants of

Japanese foreign direct investment in China", Asia Pacific Journal of Management 19: 63-86.

25

Table 1 Regional Distribution of US FDI in China

Province COUNT PERCENT Province COUNT PERCENT Anhui 129 2.051527 Jilin 82 1.304071 Beijing 455 7.236005 Liaonin 439 6.981552 Fujian 121 1.9243 Neimenggu 50 0.795165 Gansu 20 0.318066 Ningxia 9 0.14313 Guangdong 516 8.206107 Qinghai 4 0.063613 Guangxi 47 0.747455 Shaanxi 84 1.335878 Guizhou 19 0.302163 Shandong 683 10.86196 Hainan 10 0.159033 Shanghai 835 13.27926 Heilongjiang 89 1.415394 Shanxi 61 0.970102 Henan 121 1.9243 Sichuan 147 2.337786 Herbei 245 3.89631 Tianjin 338 5.375318 Hubei 108 1.717557 Xinjiang 22 0.349873 Hunan 54 0.858779 Yunnan 35 0.556616 Jiangsu 935 14.86959 Zhejiang 587 9.335242 Jiangxi 43 0.683842

Distribution of US manufacturing firms across regions in China

0

200

400

600

800

1000

Anh

uiB

eijin

g

Fujia

nG

ansu

Gua

ngdo

ng

Gua

ngxi

Gui

zhou

Hai

nan

Hei

long

jiang

Hen

anH

erbe

i

Hub

eiH

unan

Jian

gsu

Jian

gxi

Jilin

Liao

nin

Nei

men

ggu

Nin

gxia

Qin

ghai

Shaa

nxi

Shan

dong

Shan

ghai

Shan

xi

Sich

uan

Tian

jinX

injia

ng

Yun

nan

Zhej

iang

Table 2 Agglomeration, Institutions and FDI Location Choice

(Full Sample 1993-2001)a

1 2 3 4

Agglomeration

Agglomeration_US 1.582***

(0.092)

1.588***

(0.092)

1.598***

(0.091)

2.736***

(0.187)

Agglomeration_domestic 3.121***

(0.220)

3.097***

(0.220)

2.974***

(0.219)

2.679***

(0.355)

Backward Agglomeration 11.778***

(0.845)

11.901***

(0.843)

11.061***

(0.845)

9.042***

(1.425)

Forward Agglomeration 6.657***

(0.752)

6.495***

(0.752)

5.398***

(0.763)

6.180***

(1.202)

Institution

Rule of Law 0.784**

(0.337)

Government Intervention in

Contracting -2.865***

(0.555)

Intellectual Property

Right Protection 0.344***

(0.036)

Corruption -1.691***

(0.244)

Controlled Variables

Wage -1.066***

(0.056)

-1.027***

(0.086)

-1.115***

(0.086)

-1.260***

(0.145)

Highway 0.630***

(0.040)

0.587***

(0.040)

0.474***

(0.042)

0.563***

(0.065)

Education 0.482**

(0.029)

0.448***

(0.029)

0.201***

(0.041)

0.367***

(0.063)

Consulate 0.379***

(0.038)

0.399***

(0.038)

0.329***

(0.038)

0.627***

(0.067)

Sezd 0.540***

(0.044)

0.508***

(0.045)

0.538***

(0.043)

0.590***

(0.070)

Etdzd 0.260***

(0.050)

0.277***

(0.050)

0.223***

(0.050)

0.296***

(0.087)

No. of choosers 6,288 6,288 6,288 2,259

No. of choices 29 29 29 29

Pseudo R2 0.1860 0.1865 0.1881 0.2075

LR chi2(11) 7876.84 7898.89 7965.75 3156.72

Standard errors are reported in parentheses. *, **, *** represent the significance at 10%, 5%, 1% level, respectively. a The number of years covered in the corruption index regression is reduced to 1998-2001 due to the availability of the index.

Table 3 Interplay of Agglomeration and Institution and FDI

Location Choice

(Full Sample: 1993-2001) a

1 2 3 4

Agglomeration

Agglomeration_US 0.851***

(0.191)

1.430***

(0.144)

1.669***

(0.103)

0.536

(0.991)

Agglomeration_US*Institution 7.675***

(1.720)

5.225

(3.723)

0.198*

(0.115)

4.256**

(1.878)

Agglomeration_domestic 3.534***

(0.415)

3.148***

(0.348)

2.624***

(0.262)

1.253

(2.289)

Agglomeration_domestic*Institution -4.546

(3.378)

-1.579

(7.302)

-0.725**

(0.295)

2.915

(4.512)

Backward Agglomeration 10.362***

(1.503)

12.017***

(1.407)

12.058***

(1.251)

15.655**

(7.676)

Backward_back*Institution 15.889

(12.732)

-3.189

(26.306)

2.186*

(1.200)

-14.043

(15.884)

Forward Agglomeration 4.839***

(1.357)

7.432***

(1.220)

3.715***

(0.944)

8.182

(5.779)

Forward Agglomeration*Institution16.562

(11.006)

-22.158

(22.486)

-2.609***

(0.893)

-4.234

(11.368)

Institution

Rule of Law -0.349

(0.403)

Government Intervention in

Contracting -2.312***

(0.682)

Intellectual Property Rights

Protection 0.411***

(0.042)

Corruption -1.774***

(0.357)

Controlled Variables

Wage -1.009***

(0.087)

-1.025***

(0.086)

-1.068***

(0.088)

-1.227***

(0.147)

Highway 0.602***

(0.041)

0.584***

(0.040)

0.454***

(0.043)

0.561***

(0.066)

Education 0.476***

(0.029)

0.454***

(0.030)

0.174***

(0.041)

0.344***

(0.067)

Consulate 0.375***

(0.038)

0.398***

(0.038)

0.336***

(0.038)

0.654***

(0.078)

Sezd 0.532***

(0.044)

0.510***

(0.045)

0.550***

(0.044)

0.607***

(0.070)

Etdzd 0.256***

(0.050)

0.274***

(0.050)

0.206***

(0.050)

0.276***

(0.087)

No. of choosers 6,288 6,288 6,288 2,259

No. of choices 29 29 29 29

Pseudo R2 0.1870 0.1866 0.1886 0.2079

LR chi2(15) 7919.77 7903.23 7988.29 3163.05

Standard errors are reported in parentheses. *, **, *** represent the significance at 10%, 5%, 1% level, respectively. a The number of years covered in the corruption index regression is reduced to 1998-2001 due to the availability of the index.

Table 4 Agglomeration, Institutions and FDI Location Choice

7 8 9

Agglomeration_US

2.758***

(0.189)

2.894***

(0.186)

2.925***

(0.186)

Agglomeration_domestic

2.617***

(0.356)

2.618***

(0.356)

2.524***

(0.356)

Backward Agglomeration

10.654***

(1.396)

10.258***

(1.409)

10.836***

(1.389)

Forward Agglomeration

7.536***

(1.199)

6.534***

(1.197)

6.578***

(1.198)

Market Intermediary

Development Index

0.094***

(0.020)

Gov’t Interference with

Enterprise Index 0.026**

(0.012)

Intellectual Property

Protection Index 0.016*

(0.010)

Wage -1.331***

(0.157)

-1.001***

(0.142)

-1.052***

(0.142)

Highway 0.413***

(0.067)

0.505***

(0.065)

0.509***

(0.064)

Education 0.157*

(0.083)

0.424***

(0.062)

0.430***

(0.063)

Consulate 0.546***

(0.064)

0.443***

(0.069)

0.470***

(0.065)

Sezd 0.800***

(0.075)

0.671***

(0.070)

0.671***

(0.070)

Etdzd 0.197**

(0.086)

0.145***

(0.086)

0.141***

(0.087)

No. of choosers 2,259

2,259 2,259

No. of choices 29

29 29

Pseudo R2 0.2057 0.2046 0.2045

LR chi2(11) 3128.81 3112.46 3110.93 The period covered is 1998-2001. Standard errors are reported in parentheses. *, **, *** represent the significance at 10%, 5%, 1% level, respectively.

Table 5 Interplay of Agglomeration and Institution and FDI

Location Choice (1998-2001)

7 8 9

Agglomeration_US 2.563***

(0.876)

2.888***

(0.296) 2.691***

(0.3198)

Agglomeration_US*Institution -0.061

(0.070)

0.037

(0.109)

0.048

(0.074)

Agglomeration_domestic 2.973***

(0.565)

4.816***

(1.319)

3.213***

(0.626)

Agglomeration_domestic*Institution-0.225

(0.176)

-0.302*

(0.166)

-0.160

(0.119)

Backward Agglomeration 5.703***

(2.012)

3.874

(6.150)

7.786***

(1.993)

Backward Agglomeration*Institution3.055***

(0.806)

0.703

(0.730)

0.428

(0.406)

Forward Agglomeration 8.026***

(1.612)

20.962***

(4.970)

10.501***

(1.676)

Forward Agglomeration*Institution-0.373

(0.584)

-1.647***

(0.569)

-0.982***

(0.282)

Market Intermediary Development

Index 0.060**

(0.025)

Government Interference with

Enterprise Index

0.072***

(0.018)

Intellectual Property Rights

Protection Index

0.065***

(0.017)

Wage -1.405***

(0.160)

-0.928*** -0.975*** (0.142) (0.144)

Highway 0.435***

(0.068)

0.444*** 0.478***

(0.067) (0.065)

Education 0.165**

(0.083) 0.410*** 0.369***

(0.063) (0.065)

Consulate 0.539***

(0.064)

0.510*** 0.502*** (0.070) (0.066)

Sezd 0.741***

(0.076)

0.694*** 0.706*** (0.069) (0.071)

Etdzd 0.242***

(0.087) 0.097***

(0.087)

0.102

(0.088) No. of choosers 2,259 2,259 2,259

No. of choices 29 29 29

Pseudo R2 0.2069 0.2061 0.2059 LR chi2(15) 3134.76 3147.09 3132.75 Standard errors are reported in parentheses. *, **, *** represent the significance at 10%, 5%, 1% level, respectively.

Table 6 Institutions and Choice between JV vs. FOE (Full sample: 1993-2001) a

1 2 3 4

JV FOE JV FOE JV FOE JV FOE

Agglomeration

Agglomeration_US 1.566***

(0.105)

1.577***

(0.192)

1.567***

(0.105)

1.596***

(0.191)

1.568***

(0.104)

1.642***

(0.189)

2.699***

(0.248)

2.723***

(0.290)

Agglomeration_Domestic 3.745***

(0.265)

1.747***

(0.400)

3.730***

(0.265)

1.703***

(0.399)

3.617***

(0.264)

1.550***

(0.399)

3.632***

(0.494)

1.640***

(0.519)

Backward Agglomeration 12.531***

(1.006)

9.672***

(1.577)

12.650***

(1.004)

9.772***

(1.574)

11.969***

(1.007)

8.564***

(1.579)

10.629***

(1.921)

6.722***

(2.163)

Forward Agglomeration 4.855***

(0.900)

11.287***

(1.400)

4.773***

(0.900)

10.888***

(1.401)

3.808***

(0.912)

9.445***

(1.420)

3.542**

(1.652)

8.842***

(1.780)

Institution

Rule of Law 0.415

(0.399)

1.433**

(0.639)

Government Intervention in

Contracting -2.112***

(0.656)

-4.331***

(1.041)

Intellectual Property Right

Protection 0.306***

(0.043)

0.452***

(0.068)

Corruption -1.917***

(0.340)

-1.352***

(0.357)

Controlled Variables

Wage -1.424***

(0.101)

-0.302*

(0.169)

-1.398***

(0.101)

-0.225

(0.170)

-1.461***

(0.101)

-0.392**

(0.169)

-2.095***

(0.187)

-0.028

(0.243)

Highway 0.577***

(0.046)

0.786***

(0.085)

0.550***

(0.046)

0.708***

(0.083)

0.453***

(0.048)

0.540***

(0.089)

0.528***

(0.082)

0.474***

(0.117)

Education 0.443***

(0.034)

0.598***

(0.056)

0.417***

(0.035)

0.548***

(0.057)

0.183***

(0.049)

0.270**

(0.074)

0.464***

(0.085)

0.226**

(0.097)

Consulate 0.491***

(0.044)

0.065***

(0.074)

0.506***

(0.044)

0.089

(0.075)

0.442***

(0.044)

0.020

(0.073)

0.640***

(0.092)

0.527***

(0.101)

Sezd 0.537*** 0.609*** 0.513*** 0.549*** 0.515*** 0.673*** 0.661*** 0.520***

(0.052) (0.083) (0.052) (0.086) (0.051) (0.082) (0.092) (0.108)

Etdzd 0.265***

(0.058)

0.233***

(0.101)

0.279***

(0.058)

0.274***

(0.102)

0.247***

(0.057)

0.137

(0.100)

0.323***

(0.113)

0.361***

(0.138)

No. of choosers 4,445 1,843 4,445 1,843 4,445 1,843 1,244 1,015

No. of choices 29 29 29 29 29 29 29 28

Pseudo R2 0.1751 0.2252 0.1754 0.2262 0.1768 0.2284 0.2012 0.2235

LR chi2(11) 5240.51 2795.19 5250.05 2808.01 5291.87 2835.49 1685.50 1511.72

Standard errors are reported in parentheses. *, **, *** represent the significance at 10%, 5%, 1% level, respectively. a The number of choosers on the corruption index regression is reduced to 1998-2001 due to the availability of the index

Table 7 Institutions and Choice between JV vs. FOE (Subsample: 1998-2001)

1 2 3

JV FOE JV FOE JV FOE

Agglomeration

Agglomeration_US 2.771***

(0.232)

2.583***

(0.331)

2.971***

(0.229)

2.655***

(0.326)

2.988***

(0.229)

2.712***

(0.326)

Agglomeration_domestic 3.630***

(0.471)

1.277**

(0.557)

3.612***

(0.469

1.270**

(0.559)

3.518***

(0.471)

1.192**

(0.559)

Backward Agglomeration 11.663***

(1.790)

6.806***

(2.310)

11.048***

(1.794)

6.795***

(2.352)

11.770***

(1.779)

7.034***

(2.294)

Forward Agglomeration 4.879**

(1.554)

10.529**

(1.939)

3.420**

(1.549)

10.104***

(1.936)

3.743**

(1.551)

9.791***

(1.935)

Institution

Market Intermediary

Development Index 0.146***

(0.027)

0.046

(0.034)

Gov’t Interference with

Enterprise Index 0.039**

(0.016)

0.008

(0.020)

Intellectual Property

Protection Index

0.011

(0.013)

0.025*

(0.015)

Controlled Variables

Wage -2.077***

(0.196)

0.314

(0.284)

-1.550***

(0.172)

0.450*

(0.269)

-1.591***

(0.173)

0.383

(0.268)

Highway 0.295***

(0.079)

0.292**

(0.140)

0.441***

(0.076)

0.346***

(0.134)

0.461***

(0.076)

0.313**

(0.134)

Education 0.089

(0.103)

0.062

(0.151)

0.486***

(0.078)

0.205*

(0.108)

0.475***

(0.078)

0.255**

(0.112)

Consulate 0.608***

(0.081)

0.345***

(0.107)

0.462***

(0.086)

0.299**

(0.117)

0.528***

(0.082)

0.263**

(0.110)

Sezd 0.902***

(0.093)

0.808***

(0.140)

0.710***

(0.086)

0.730***

(0.127)

0.698***

(0.086)

0.755***

(0.129)

Etdzd 0.248**

(0.106)

-0.040

(0.155)

0.143

(0.106)

-0.040

(0.155)

0.168

(0.106)

-0.089

(0.160)

No. of choosers 1,411 848 1,411 848 1,411 848

No. of choices 29 28 29 28 29 28

Pseudo R2 0.1966 0.2010 0.1941 0.2007 0.1935 0.2012

LR chi2(11) 1868.44 1083.51 1844.28 1081.87 1838.74 1084.49

Standard errors are reported in parentheses. *, **, *** represent the significance at 10%, 5%, 1% level, respectively.

Data Appendix

Variable Mean Std. Dev. Min Max

Agglomeration

Agglomeration_US 0.0332 0.0833 0 1

Agglomeration_domestic 0.0343 0.0508 0 1

Backward Agglomeration 0.0188 0.0200 0.0000602 0.271

Forward Agglomeration 0.0255 0.0247 0.0000206 0.364

Institution Environment

Rule of Law 0.118 0.087 0.022 0.500 Government Intervention in

Contracting 0.0461 0.0511 -0.0273 0.424

Intellectual Property Right

Protection -1.232 0.895 -2.905 1.653

Corruption 0.582 0.119 0.367 0.879 Market Intermediary Development

Index 4.447 2.189 0.98 9.97

Gov’t Interference with

Enterprise Index 3.107 2.967 -0.71 11.39

Intellectual Property Protection

Index 3.358 2.822 0 11.77

Controlled Variables

Wage 8.456 0.430 7.681 9.752

Highway -1.595 0.845 -4.159 -0.212

Education 1.001 0.660 -0.223 3.135

Consulate 0.172 0.378 0 1

Sezd 0.375 0.484 0 1

Etdzd 0.529 0.499 0 1

Agglo_us4 Agglo_dm4 backward forward government contract log_patent log_wage education highway consulate sezd etdzd

Agglo_us4 1

Agglo_dm4 0.4054 1

Backward 0.3735 0.6849 1

Forward 0.3753 0.6658 0.7711 1

government -0.0781 -0.0713 -0.0798 -0.0748 1

contract -0.0584 -0.1213 -0.1571 -0.1621 0.1417 1

log_patent 0.3314 0.3100 0.2878 0.3248 -0.0982 -0.1927 1

Wage 0.1586 0.1520 0.1158 0.1512 0.3059 0.0122 0.4978 1

education 0.1848 0.0342 -0.0524 -0.0381 -0.0192 0.1530 0.7308 0.5388 1

Highway 0.2563 0.3163 0.3523 0.3878 -0.0838 -0.2673 0.4910 0.2892 0.1748 1

Consulate 0.2447 0.1910 0.1477 0.2489 -0.0821 -0.2105 0.5494 0.2835 0.4540 0.5249 1

Sezd 0.2864 0.3789 0.4522 0.4539 -0.1576 -0.2836 0.364 0.2152 0.3806 0.456 0.1875 1

Etdzd 0.2586 0.3169 0.3569 0.3942 -0.071 -0.4994 0.4729 0.2875 0.3207 0.4345 0.3171 0.5885 1

Variable Description Resources Wage the logarithm of the average manufacturing wage (Unit: Yuan) Chinese Statistical Yearbook 1994-2001 Highway the logarithm of the highway density (Unit: km/sq. km) Chinese Statistical Yearbook 1994-2001 Education the logarithm of the ratio of employment with college or above degree) Chinese Statistical Yearbook 1994-2001

Consulate the dummy variable which equals one when the region has the US consulates History of the diplomatic establishment between China and United States 1786-1994, Huang Gang, 1995 Taiwan

Sezd the dummy variable which equals one when the region has a special economic zones* or open coastal cities**

Comprehensive statistical data and materials on 50 years of new china 1949-1999

Etdzd

the dummy variable which equals one when the region has a economic and technological development zones*** Ministry of Commerce of China

* Special economic zones: Zhuhai, Shenzhen, and Shantou in the Guangdong Province, Xiamen in the Fujian Province, and the Hainan Province

**Open coastal cities: Tianjin Municipality, Qinghuangdao in the Hebei Province, Dalian in the Liaoning Province, Shanghai Municipality, Lianyungang and Nantong in the Jiangsu Province, Ninbo and Wenzhou in the Zhejiang Province, Fuzhou in the Fujian Province, Yantai and Tsingtao in the Shandong Province, Guangzhou and Zhanjiang in the Guangdong Province, and Beihai in the Guangxi Province

***Economic and Technological Development Zones: Beijing Municipality, Tianjin Municipality, Qinghuangdao in the Hebei Province, Danlian, Yinkou and Shenyang in the Liaoning Province, Changchun in the Jilin Province, Harbin in the Heilongjiang Province, Shanghai Municipality, Lianyungang, Nantong, and Kunshan in the Jiangsu Province, Ninbo, Wenzhou, Hanzhong, Ninbo Daxie, and Xiaoshan in the Zhejiang Province, Wuhu and Hefei in the Anhui Province, Fuzhou, Xiamen Haicang, Fuqing Rongqiao, and Dongshan in the Fujian Province, Nanchang in the Jiangxi Province, Tsingtao, Yantai and Weihai in the Shandong Province, Zhengzhou in the Henan Province, Wuhan in the Hubei Province, Changsha in the Hunan Province, Guangzhou, Zhanjiang, Guangzhou Nansha and HuiZhou Dayawan in the Guangdong Province, Hainan Yanpu in the Hainan Province, Chongqin and Chengdu in the Sichuan Province, Guiyang in the Guizhou Province, Kuming in the Yunan Province, Xi’an in the Shaanxi Province, Xining in the Qinghai Province, Urumuqi and Shihezi in the Xinjiang Province