1 preface

With the integration of the world economy, foreign direct investment (hereinafter referred to as FDI) has become an important driving force to promote the economy of developing countries. Since 1993, China has become the second largest foreign investor in the world after the United States. By 2002, the actual use of FDI in China totaled US$ 52.7 billion, surpassing the United States for the first time and becoming the largest foreign investor in the world. Due to the influx of foreign capital, foreign direct investment is playing an increasingly important role in China's economic development. It is conducive to providing scarce resources to China, improving the level of science and technology, creating employment opportunities and promoting economic development, which is a possible accelerating factor for China to realize modernization.

However, everything has two sides. Although the total amount of foreign direct investment flowing into China is huge, the regional distribution is seriously unbalanced. At present, most of the foreign investment attracted by China is concentrated in the coastal areas, especially in the southeast coastal areas such as Guangdong, Jiangsu, Fujian and Shanghai. In 2002, among the foreign direct investment actually utilized in various regions, the eastern region accounted for 87. 43%, the central region accounted for 9. 88%, the western region accounts for 2.65%, and the central and western regions only account for 12.57%[ 1]. A large number of foreign capital flows into coastal areas, which has played a very important role in promoting the rapid growth of coastal economy, but at the same time it has also aggravated the imbalance of China's economic development and further widened the gap between the eastern, central and western regions.

In order to promote the coordinated development of regional economy and gradually narrow the regional gap, based on the regression model established by the decisive factors affecting FDI, this paper weights the decisive factors of FDI in eastern provinces and cities, and takes this as a reference value, compares and analyzes the corresponding indicators of Jiangxi Province with the reference value, and calculates the difference index of attracting FDI between Jiangxi Province and the eastern region, thus providing policy suggestions for the future FDI decision of Jiangxi Provincial Government.

2 Literature review

Since 1960s, the research on the theory of foreign direct investment has been deepened. In 1960s, the theory of foreign direct investment focused on explaining international capital flows with the traditional principle of comparative advantage. After entering the 1990s, the research method of combining internalization theory with general equilibrium system has become a research hotspot. To sum up, in recent years, the discussion of FDI location in academic circles has basically started from two ideas: one is to explore and study the motivation and influencing factors of FDI location selection from different sides or angles in theory; Secondly, many scholars at home and abroad are increasingly using econometric models to conduct empirical research on the location choice of FDI.

2. 1 theoretical research on location selection of foreign direct investment

2. 1. 1 location theory

Theoretically, the initial location theory is the basis of FDI location selection theory, and many related international direct investment location selection theories are based on location theory, which more or less contains the budding idea of location theory. The development of location theory has mainly formed three schools: cost school, market school and behavior school, which explain the location choice of foreign direct investment from the aspects of enterprise pursuing cost minimization, profit maximization and enterprise managers' own needs. The analysis of location layout factors and location selection in this theory provides a theoretical basis and methodology for the location analysis of foreign direct investment, but rarely involves the study of economic activities of multinational enterprises.

2. 1.2 FDI location selection theory

Harmo's monopoly advantage theory holds that enterprises in the home country have more favorable monopoly advantages than similar enterprises in the host country, which is the motivation for enterprises to make foreign direct investment. Furong put forward the theory of product cycle, which holds that the place of production depends on different stages of product life cycle (Li Yunjing) [2]. Japanese scholar Kiyoshi Kojima's "Japanese-style" direct investment theory holds that foreign direct investment is different from general capital transfer, but a comprehensive transfer of capital, technology and management methods. Foreign direct investment refers to the transfer of industries with comparative disadvantages to industries with comparative advantages in the host country, or the investment in industries with comparative advantages in the host country, thus bringing about the expansion of trade and the increase of profits.

Deng Ning's eclectic theory of international production holds that the advantages of location choice affecting foreign direct investment include natural resources and man-made resources, as well as the spatial distribution of the market, the price, quality and productivity of inputs, the cost of international transportation and communication, investment incentives and obstacles, man-made obstacles in product trade, social and infrastructure conditions, and ideological, linguistic, cultural, commercial and political differences between countries, R&; The economy, economic system and government policy of centralized production and sales.

Some scholars have studied the location choice of FDI from the perspective of agglomeration effect. Porter believes that a region attracts FDI because it "has developed infrastructure, can obtain specific service facilities and skilled labor, has a good regional image and a large number of industrial clusters". Kmgma, Dunning, Dermot and Davelin have theoretically studied the effect of aggregation. Luger and Shetty confirmed the important influence of agglomeration economy on the investment location choice of foreign companies through the study of three-digit industries (industrial classification standards). Xu Rodin and Tan Weihong also analyzed the role of agglomeration in attracting foreign investment in China (Wu Yao) [3].

2.2 empirical analysis of FDI location selection

Empirical analysis actually quantifies various influencing factors on the basis of theoretical research, and uses econometric analysis methods to test the correlation between these factors and FDI level. Various influencing factors are usually divided into several categories: cost factors, market factors, agglomeration economic factors and institutional factors. In recent years, many scholars at home and abroad have made empirical analysis of various influencing factors.

2.2. 1 Empirical Analysis on Location Choice of Foreign Direct Investment

The influence of market and cost factors on FDI, Rushmi's research, like that of Globerman and Shapiro, found that the basic economic variables have a significant impact on FDI. Specifically, these factors mainly include: market size, labor cost, high-tech level, foreign debt and power generation. However, from the current empirical literature on the impact of government policies on FDI, the conclusions on the impact of government policies on FDI inflows are inconsistent. Rashmi's research shows that some government financial incentive policies have a positive impact on the inflow of FDI, but the impact is not significant, while the cancellation of some restrictive measures has a significant positive impact on the inflow of FDI. Devereux, Griffith and Hines believe that fiscal policy does affect the regional distribution of FDI, especially export-oriented FDI, while other policies only play a secondary role (Hu Zaiyong) [4].

However, the UNCTAD report shows that the incentive measures implemented by the government play a less important role. Some scholars, such as Villella and Barre, also put forward different views. They believe that if the influence of economic factors on FDI is considered, the influence of government incentives on attracting FDI will be obliterated. Hoekman and Saggi also believe that although incentives have played a role in attracting a certain type of FDI, they will not play a role if they are considered in a broader economic factor.

2.2.2 domestic empirical analysis of FDI location selection

Domestic scholars usually use cross-sectional data or panel data to analyze domestic provincial or regional FDI, and mostly use correlation regression analysis or comparative analysis.

Lu Minghong used the foreign investment data of 29 provinces and regions1988-1995 to analyze the influence of investment environment on the location of foreign investment. The results show that the regional GDP, the proportion of tertiary industry output value, the proportion of urban population, the preferential degree of special economic policies and the extroversion of regional economy are positively correlated with foreign direct investment in various regions. At the same time, he also calculated the deviation between the foreign direct investment absorbed by each region and its investment environment, and thought that Guangxi, Shaanxi, Jiangsu, Hainan, Guizhou, Gansu, Tianjin and other provinces belonged to foreign over-investment zones, which attracted too much investment. Xinjiang, Fujian, Henan, Hebei, Inner Mongolia, Guangdong, Qinghai and Shanxi are regions with insufficient foreign investment, but they have great potential.

Wei, He Canfei and 35 foreign-invested enterprises in Qinhuangdao/KLOC-0 made an empirical analysis of their investment motives and location factors through questionnaires. The results show that the motivations of foreign investment in China are production input and market motivation, production service motivation, cultural connection and emotional motivation, preferential policies and risk reduction motivation, competition motivation and export motivation. The main location factors that affect foreign investment in Qinhuangdao can be summarized as urban economic and cultural environment factors, transaction cost factors, production input supply factors, market factors and input cost factors.

Ge Shunqi compared the performance index and potential index of China's 3 1 provinces and cities in utilizing FDI. From 65438 to 0995, the leading provinces and cities of the index were Beijing, Shanghai, Guangdong, Tianjin, Zhejiang, Fujian and Jiangsu. By 200 1, the ranking of index values has not changed, but the index values of Beijing have declined, and those of other provinces and cities have improved to varying degrees. In addition, many scholars have compared the impact of China's entry into WTO on attracting FDI, and have also drawn some valuable theories.

To sum up, the empirical analysis of FDI location choice by domestic and foreign scholars shows that the basic economic variables, namely market, cost and agglomeration economic factors, have a significant impact on FDI, while the research results of institutional factors are controversial, so the formulation of policies should be analyzed according to specific conditions.

3 Analysis of the differences in attracting FDI between Jiangxi and the eastern region

The empirical analysis of this paper is divided into two steps: the first step is to choose regression model. According to the need of writing purpose and the limitation of space, it is necessary to choose a model that can fully reflect the influencing factors of foreign direct investment in China as the basis of analysis. The model needs comprehensive data and strong representative conclusions. The second step, on the basis of the model, compares and analyzes the influencing factors of Jiangxi Province with the corresponding indicators of the eastern region, and then calculates the difference index of attracting FDI between Jiangxi Province and the eastern region.

3. 1 model selection

This paper draws lessons from a regression model established by Wu Yao, a graduate student of capital university of economics and business University of Economics. According to the influencing factors of FDI described above, the author selects nine variables as explanatory variables of the equation:

In (FDI) = ao+a1ln (GDP)+a2ln (ggdp)+a3ter+a4cap+a5ln (salary)

+a 6 1n(TRA)+a7 infra+a8ln(FDI- 1)+a9 pol+C(3. 1)

FDI (unit: ten thousand dollars) is the explanatory variable; GDP (unit: 100 million yuan): gross domestic product; GGDP (unit: yuan): per capita GDP; TER (unit:%): the proportion of tertiary industry (financial, information, transportation and other industries in various regions) in GDP, which measures the degree of marketization development in a region; HCAP (unit:%): human capital stock in each region; Wage (unit: yuan): the average wage level of labor force in various regions, reflecting the labor cost level of foreign direct investment in various regions; TRA (unit: US$ 100 million): the total import and export volume, which is used to measure the degree of opening up of a region; INFRA (unit: km/km2): the comprehensive density of traffic lines, which is used to measure the infrastructure level of a province and city; FDI- 1 (unit: ten thousand dollars): the amount of foreign direct investment in the previous year; POL: Preferential policies for foreign investors. The area enjoying preferential policies is assigned as 1, otherwise it is 0[5].

These nine variables comprehensively consider factors such as agglomeration effect, economic scale and market capacity, economic efficiency, degree of marketization, human capital stock, degree of opening to the outside world, infrastructure level, labor cost and preferential policies, and the indicators are comprehensive and reasonable. At the same time, the author uses the panel data of 365438+ 1997-2003 in China, and uses econometric analysis methods to analyze the influence of various factors on FDI from both static and dynamic levels.

Stepwise regression method was used to analyze the model. Results Five explanatory variables passed the test and entered the equation, namely:

Ln (foreign direct investment) = 1.985+0.435ln (gross domestic product) +0.6 10ln(GGDP)-0.634ln (salary)+1.023infra.

+0.508 ln(FDI- 1)+0.467 pol+C(3.2)

The regression results show that the explanatory variables are 1n(FDI- 1), INFRA, ln(GDP), 1n (salary), 1n(GGDP),

POL all passed the significance test, which was significant at the level of 1%, and the r square of the overall model reached 9 1. 38%.

It has a good fitting degree. The f value is 35 1.69 17, which is significant at the level of 1%, indicating that the model is significant as a whole. The value of D-W is 1. 476, indicating that there is no serious sequence autocorrelation, and VIF values are all below 5, indicating that there is no serious multiple * * * linearity. In addition, no obvious heteroscedasticity was found by observing the residual diagram.

According to this model, agglomeration effect (FDI- 1), infrastructure level (infra), economic development level (GDP, GGDP), labor cost (wages) and policy factors (POI) all have important influences on FDI. The other three explanatory variables failed to pass the significance test, among which the tertiary industry's proportion to GDP (TER) may reflect the infrastructure situation at the same level as the infrastructure level (infra), so it is necessary to exclude multiple * * * linearities. The variable (TRA) representing a region's opening level may be that foreign direct investment in China pays more attention to China's local market, so the level of import and export is not significantly related to FDI. The human capital stock (HCAP) reflects the supply of human capital in a region, because it uses relative figures. In the analysis, HCAP did not enter the equation, which shows that foreign investors consider the demand of human capital more than the supply. This model considers the influence of macroeconomic development on FDI, and explains the location choice of FDI in China.

The purpose of this paper is to test the significance of the variables selected in the model to FDI. The variables entering the regression equation show that these variables have a significant impact on attracting FDI. Only when this condition is met, it is of practical significance to calculate their difference index, otherwise it is of no practical significance. For example, human capital stock (HCAP) failed the test and entered the regression equation. The conclusion obtained by calculating the difference index of this variable can only show that there is a difference in the human capital stock between Jiangxi Province and the eastern region (the degree of difference is determined by the value of the difference index). However, because this variable does not enter the equation, this difference is not the factor that causes the two regions to attract different FDI, and the difference index of the variable has no practical significance.

3.2 Comparative analysis

3.2. 1 cluster analysis

In order to build a class of provinces and cities that effectively attract FDI, as a reference area compared with Jiangxi Province, this paper uses the FDI data of provinces and cities from 2004 to 1998 to cluster 3 1 provinces, cities and autonomous regions. The following are the original data and analysis results:

Table 3.11foreign direct investment in 998-2004 3 1 units of provinces, municipalities and autonomous regions: millions of dollars.

region

In 2004

In 2003

In 2002

200 1 year

In 2000,

1999

1998

Beijing

255974

2 19 126

172464

1768 18

168368

197525

2 16800

Tianjin

17209 1

153473

158 195

2 13348

1 1660 1

176399

2 1 136 1

Hebei Province

69954

96405

7827 1

66989

67923

104202

142868

Shanxi

9022

2 136 1

2 1 164

23393

22472

39 129

2445 1

Inner Mongolia

34297

8854

1770 1

10703

10568

6456

9082

Liaoning province

540677

2824 10

34 1 168

25 16 12

204446

106 173

2 19045

Jilin province

19237

19059

24468

33766

3370 1

30 120

409 17

Heilongjiang province

339 17

32 180

355 1 1

34 1 14

30086

3 1828

52639

Shanghai

63 1087

546849

427229

429 159

3 160 14

283665

360 150

Jiangsu Province

894830

1056365

10 18960

69 1482

642550

607756

663 179

Zhejiang Province

573256

498055

3076 10

22 1 162

16 1266

123262

13 1802

Anhui province

42850

36720

38375

33672

3 1847

26 13 1

27673

Fujian Province

192384

259903

383837

39 1804

343 19 1

402403

42 12 1 1

Jiangxi

204487

16 1202

108 197

39575

22724

32080

46496

Shandong (province)

866423

60 16 17

473404

352093

297 1 19

225878

220274

Henan Province

422 1 1

53903

40463

45729

56403

52 135

6 1654

Hubei province

17444 1

156886

142665

1 18860

94368

9 1488

97294

Hunan

14 1803

10 1835

90022

8 10 1 1

67833

65374

8 18 16

Guangdong

100 1 158

782294

1 133400

1 193203

1 12809 1

1 165750

120 1994

Guangxi

29579

4 1856

4 1726

384 16

52466

635 12

886 13

Hainan

1 1926

42 125

5 1 196

4669 1

43080

48449

7 17 15

Chongqing

25 196

26083

19576

25649

24436

23893

43 107

Sichuan Province

36503

4 123 1

55583

58 188

43694

34 10 1

37248

Guizhou (province)

627 1

452 1

382 1

2829

250 1

4090

4535

Yunnan Province

14 153

8384

1 1 169

6457

128 12

15385

14568

Xizang

-

-

-

-

-

-

-

Shanxi province

14 132

33 190

36005

35 174

28842

24 197

300 10

Gansu

3539

2342

6 12 1

7439

6235

4 104

3864

Qinghai

-

2522

4726

3649

-

459

-

Ningxia

6704

1743

2200

1680

174 1

5 134

1856

Xinjiang

3996

1534

1899

2035

19 1 1

2404

2 167

Source: China Statistical Yearbook, 2005, China Economic Net.

The clustering results are as follows:

Tree diagram using average linkage (within group)