The impact of FDI on trade: evidence from China's bilateral trade.

ABSTRACT

Within a gravity model framework, this paper investigates the impact of foreign direct investment (FDI) on the trade performance of China using trade and stock data on a bilateral basis between China and 75 partner countries/regions over the period 1989 to 2000. It has been found that outward FDI has a larger predicted impact on China's exports than does inward FDI. On the other hand, inward FDI is found having a larger predicted impact on China's imports than does outward FDI. The results from regional breakdown analysis show that the extent to which FDI is trade-enhancing appears to depend on FDI's motivation and region-specific characteristics.

1. INTRODUCTION

Studies (such as Lardy 1996, Zhang, et al. 1999, Liu et al. 2001) have shown that foreign-invested firms have contributed significantly to China impressive export expansion and economic growth. Using panel data at the provincial level in the period of 1986-97, Tse (1997) proved inward FDI positively affect provincial manufacturing export performance. But Sun (2001) argued the role of FDI changes across the regions in China. Although FDI shows a positive and significant impact on exports from coastal region to the central region, its impact on the western region is found to be insignificant. Earlier literature has mainly focused on the impact of inward investment on china's export, while the linkage between inward FDI and exports is quite well understand, there is still a paucity of systematic study on the impact of FDI on China's foreign trade. What is the role of FDI in China's trade expansion, and what is the impact of FDI on China's bilateral trade with its trade partner? Investigation of the linkage between FDI and trade growth would help us to reveal the major motives for investing in Chinese economy and Chinese firm investing abroad, which in turn might allow Chinese policy makers to take appropriate measures for stimulating further capital inflows and outflows. If a negative parameter on inward FDI is obtained in export (import) equation, FDI is considered displacing trade. China's Exports are partially replaced by multinational enterprises (MNEs)' local sales on domestic market, detrimental to the domestic industry's development. Reciprocally, home country's trade balance would benefit from this substitution effect, especially additional imports from the parent company are induced. Trade and outward FDI are complements if a positive parameter is obtained in export (import) equation. Investing abroad leads to an increase in exports (imports) of China towards (from) the host country. Accordingly, an impact on China's trade balance should be positive or negative depending on whether exports increase by more than imports or not.

My work improves upon former studies in three aspects. First, it relates to a set of countries, the contribution made by this paper is in more fully evaluating an important policy question regarding the effect of FDI. For example, it is broader in statistical terms than most other studies by using a panel data set covering 75 countries and 12 years (1989-2000). Second, it takes into account national changes both in inward FDI and outward FDI over a considerable period of time. Third, it is also explicitly based on the idea that the gravity model is a transactions cost model but that regional breakdowns are necessary to incorporate the different motivations for FDI.

The major empirical conclusions of this paper are: (1) Much of the measured trade effect is through FDI rather than cost, as the theory of FDI would indicate, and that studies which concentrate on cost as the channel significantly understate the extent of such expansion. (2) On the whole bilateral country level, outward FDI has a larger predicted impact on China's exports than does inward FDI. On the other hand, inward FDI is found having a larger predicted impact on China's imports than does outward FDI. (3) There is much cross-regional variation and differences in the patterns of FDI-trade links. Regarding to the impact of inward FDI on Chinese trade, FDI is found to boost both export and import growth in Asia, Europe and Oceania. As far as outward FDI is concerned, a unanimous complement link between FDI and trade exists only for Asia, and Africa.

2. THE MODEL

This study examines the linkage between trade and FDI at an aggregate level using bilateral country data. I utilize the gravity model, which has been used to explain bilateral trade flows among many countries over long periods (Frankel et al. 1995; Hejazi and Trefler 1996; Fontagne L. and Pajot M. 2000). In fact, many factors may affect trade flows to and from each country. Trade between two countries should be positively related to their incomes. It can be justified by the modern theory of trade under imperfect competition. Furthermore, GDP per capita has a positive effect on trade: as countries become more developed, they tend to be specialized and trade more. Countries near to China geographically and those with similar languages (cultural similarity) should have lower transactions costs and correspondingly larger levels of bilateral trade. The larger difference in economic stage between investing and recipient countries should have a positive role on trade (Boiling and Somwaru 1999). Obviously, cheap labor or undeveloped internal market in China may attract more investors from developed countries but cannot attract those from the Africa or Latin America since investors from the transition countries have the same conditions in their own countries. Increases in the value of RMB are expected to increases China imports but reduce exports. During the period under study (1989-2000), the Chinese currency has depreciated significantly, and only in recent years (1995 to 1999) did the foreign exchange rate of the currency appreciated slightly. Therefore, foreign exchange rates EX are expected to play a positive role in stimulating exports from China and have a restrain effect on its import. Therefore, the gravity models are modeled as a function of GDP, GDP per capita (PGDP), Economic distance (EDIS), language (LANGU), exchange rate (EX), and outward and inward FDI. To measure the elasticity of changes in trades with regard to percentage changes in the independent variables, the logarithmic form of these variables is used. Let t denote years, c is China, and i is the trading partner country, the transactions cost function is expressed as follows:

(1) Log ([X.sub.cit]) = [[beta].sub.0] + [[beta].sub.1]log (GD[P.sub.ct] x GD[P.sub.it]) + [[beta].sub.2]log (PGD[P.sub.ct] x PGD[P.sub.it]) + [[beta].sub.3]log (E[X.sub.cit]) + [[beta].sub.4]log (EDI[S.sub.cit]) + [[beta].sub.5]LANG[U.sub.cit] + [[epsilon].sub.cit]

Where [X.sub.cit] stands for bilateral exports or imports between China and its partner countries. In order to measure how FDI has been influencing the trade, I add outward and inward FDI as a determinant of trade:

(2) Log ([X.sub.cit]) = [[beta].sub.0] + [[beta].sub.1]log([GDP.sub.ct] x [GDP.sub.it]) + [[beta].sub.2]log(PGD[P.sub.ct] x PGD[P.sub.it]) + [[beta].sub.3]log (E[X.sub.cit]) + [[beta].sub.4]log(EDI[S.sub.cit]) + [[beta].sub.5]LANG[U.sub.cit] + [[beta].sub.6]log(OUTFD[I.sub.cit]) + [[beta].sub.7]log(INFD[I.sub.cit]) + [[epsilon].sub.cit]

Here OUTFD[I.sub.cit] is the stock of Chinese FDI located in country i in year t, and INFD[I.sub.cit] is the stock of country i's FDI located in China. The presence of FDI reduces the costs associated with trade. It is unclear, however, whether exports or imports should increase more. The beta coefficients [[beta].sub.1] to [[beta].sub.7] are the elasticity of trade with respect to the product of GDP, the product of GDP per capita, exchange rates, economic distance, inward FDI and outward FDI. The values of [[beta].sub.6] and [[beta].sub.7] are of particular interest for this study because these two coefficients reflect the percentage change in trade in response to a given percent change in FDI.

3. DATA SOURCE

The data used in this study are panel data including Chinese 75 partner countries and which further are disaggregated into 6 partner regions (Asia, Africa, Europe, Latin America, North America and Oceania) over a 12-year period. Chinese inward and outward FDI stock and bilateral Chine's exports and imports were obtained from the Almanac of Foreign Economic Relations and Trade. Purchasing power parity-based exchange rates, GDP data were obtained from the World Development Database. The sample size of the tests is 900 (75 countries times 12 years per countries). To remove influences of inflation on real relationships between variables, all variables used in this study are deflated by the GDP index to convert to constant prices (1995 prices).

4. CHINA'S BILATERAL FDI AND TRADE GROWTH PERFORMANCE

4.1 The Regional Pattern of FDI

An important characteristic of FDI in China is its uneven source. For China's inward FDI, Asia was the overwhelmingly dominant FDI source. It accounted for from 75% in 1979-89 up to 84% in 1990-95, then down to 70% in 1996-2000. Why FDI into China are mainly from developing Asian countries not from developed Western countries? There are two major factors underlying this unbalanced regional FDI source. First, Asia, especially Hong Kong, Macao and Taiwan have advantages over the other regions both in terms of the cultural environment and the economic development conditions. Asian MNEs' advantages such as its marketing skills that make investors specialize in delivering timely, uniform quality products to Western markets (Wells 1993), or the adaptation of mature technologies to more labor-intensive contexts and to local raw materials, make them have a strong desire to capture location advantages (cheap labor or growing market) to maximize their profits via direct investment in China, they use Mainland China as an export platform, relocated its labor-intensive and export-oriented industries to China, according to a survey in 2001, almost all Hong Kong's low-skilled industries have now moved to Mainland (Fung 2003). Second, The Asian FDI primarily is export oriented, which has been encouraged by China's cheap labor and incentive policies toward this type of FDI. The western countries' FDI have been motivated essentially by China's potentially huge market and trade barriers, not for cheap labor (Zhang 2000). The amount of FDI flows from North America and Europe has been limited largely due to China's restrictive policy toward the market-oriented FDI and troubled Sino-U.S.

In terms of outward FDI, the foreign investments by Chinese firms are concentrated geographically in Asian countries, Africa, Europe and North America, in order (see table 1). The largest investment was in Asia (which was largely pushed by Chinese capital into Hong Kong and Macao.), amounting to US$ 0.35 billion in 1979-89, US$ 2.53 billion in 1990-1995 and US$ 4.94 billion in 1996-2000, accounting for 60%, 78% and 71% of the total, respectively. The main reason why Hong Kong and North America have attracted a significant proportion of Chinese capital is that these countries and regions offer a favorable investment environment, including sound financial markets and consumer markets, highly developed technology, advanced management methods, superior infrastructure and so on. Using overseas investment to acquire the advanced technology it needs, and learn advanced management methods is one of the policy objectives of China's overseas investment, which led the Chinese government to encourage overseas investment in the first place (Ma and Li 1995).

Contrast to rapid rise in Asia, China's investment in Africa experienced a sharp decline in share from 25% in 1979-89 sharply down to 10% in 1990-95. This reflects the change of China's FDI policy objectives: during the earlier period of economic reform since 1979, the motivation behind overseas investment in Africa was intended only to expand the area of collaboration with foreign countries, establish international trade relationships and enhance China's international political and economic influence, rather than to maximize profit. Beginning in 1993, China strove to undertake more rigorous screening of overseas investment projects to improve the poor operational efficiency, which entailed a period of adjustment for overseas investment. Project reviews implemented by relevant government agencies were tightened up, and re-registration was required of those overseas enterprises already established (Wu and Chen 2001). As a result, the number of registered overseas investment enterprises, and those being set up in Africa, fell drastically. It is important to note that over time China's comparative advantage in the agricultural resource-intensive category has declined and that in labor- and skill-intensive goods it has tended to increase (Zhang and Hock 1996), at the same time, economic development is restricted by a shortage of resources, and necessary adjustments to the industrial structure meant that mature technologies and industries would have to be transferred to Africa, Latin America and less developed Asian countries which have comparative advantages. As a result, although Chinese investment in Africa did not achieve its original share (18% in 1996-2000), the absolute terms increased greatly.

4.2 The Regional Pattern of Trade

The regional pattern of bilateral trade of China is similar to the regional distribution of FDI. As shown in Table 2, the primary destination and source of China's trade has been Asia, although both exports from and imports to all regions increased over the period from 1979-2000. In particular, Asian countries depend most heavily on the Chinese market mainly for its exports of food, textile & clothing, machinery, electrical equipment, mineral products and resource based manufactures, ranging from 70.7% of its total imports in 1979-89, 64.9% in 1990-95, and gradually decreased to 54.4% in 1996-2000. The Chinese market also attracted 3% in 1979-83, 6.7% in 1990-95, and 9.52% of Asian total exports in 1996-2000, accounting for 51.8%, 59.1% and 61.7% of the Chinese total imports respectively. In value terms, Asian annual average imports from China grew from US$20.7 billion in 1979-89 to US$104.6 billion in 1996-2000, while annual average exports to China grew from US$17.4billion to US$100.4 billion over the same periods. This made Asia China's largest trade partner, followed by Europe, North America, Latin America, Oceania and Africa. It is interesting to note that the resource-rich Africa and Latin America market has become increasingly important for Chinese mature technology and industries, as an outlet for its exports. The share grew rapidly from 1.4% and 1.1% in 1979-89, to 1.9% and 2.6% in 1996-2000, respectively. This could be attributed to their strong comparative advantage in the resource-intensive industry relative to China (Zhang and Hock 1996).

In general, the above analysis shows that in the last two decades China has achieved remarkable progress in attracting FDI, investing abroad, and improving its trade performance. And trade pattern has nearly the same geographical distributing structure as does FDI pattern. The question is: are FDI and trade complement in China? Forging ideal linkages between FDI and trade is a main challenge for Chinese policy-makers in order to benefit from FDI as much as possible. The relationship between trade and FDI is complicated and depends largely on the types of trade and FDI being considered. It is basically country-, industry-, and even firm-specific. This suggests the importance of empirical investigations in assessing true FDI-trade relationships. My interest here is to investigate the impact of FDI on China's bilateral trade growth. Therefore, in the next section we will examine the direct contribution of FDI to China's bilateral trade growth.

To examine the impact of FDI and other variables on China's trade, a number of regression analyses have been undertaken using the gravity model discussed in Section 2. The primary model (Model 1), as specified by Equation 1, tests the impact of transaction costs, including the product of GDP (GDPs), the product of GDP per capita (PGDPs), economic distance, language and exchange rates. The regression results are presented in Table 3.

To test the robustness of the impact of FDI on trade in various circumstances, a number of alterations to the primary model have been made. These include: 1) Adding Inward FDI to test for the possible influence of FDI on trade, this will indicate whether trade and FDI are substitutes or complements after controlling for comparative advantage. 2) Adding outward FDI. 3) Adding inward and outward FDI simultaneously. The results of these altered models are also presented in Table 3. Table 3 shows that much of the measured trade effect is through FDI rather than cost, as the theory of FDI would indicate, and that studies which concentrate on cost as the channel significantly understate the extent of such expansion. Therefore, FDI and trade are complementary at the whole bilateral country level. Table 3 also tells that Chinese outward FDI stimulates more exports than does inwards FDI (the coefficient of outward FDI in export equation is 1.263, compared to inward FDI coefficient 0.396). By contrast, inward FDI stimulates more imports than does outward FDI (the coefficient of inward FDI in import equation is 0.757, compared to outward FDI coefficient 0.122).

In order to precisely investigate whether the different motivations for FDI result in different linkage between trade and FDI, we further divided the whole sample countries into 6 groups, that is, Asia, Africa, Europe, Latin America, North America and Oceania. The results of the GLS tests are summarized in table 4:

1. For Asia, there exists a unanimous complement links between FDI and trade.

2. For Africa, the relationship between inward FDI and trade is found as statistically insignificant. This means that FDI from Africa does not explain the volatility of China's trade. Outward FDI is found to be a complement to trade.

3. For Latin America, both inward and outward FDI have insignificant influence on trade.

4. In the case of Europe and Oceania, inward FDI is found to be a complement for trade, outward FDI is a substitute for export and has insignificant relation to import.

5. in the case of North America, although there is no statistical significant link between inward FDI and export, outward FDI and import, a complement relationship is indicated between inward FDI and import, outward FDI and export.

How do we interpret the empirical results of links between FDI and trade? Complementary or substitute seems region (county)-specific and sensitive to FDI motivations.

The lack of general data on sectoral distribution of individual country's investments limits the assessment of how partner countries' direct investment is related to the changes in their comparative advantage and trade with China. In table 5, I compiled the available information on Sino-foreign joint ventures registered in 1979-96 from the Almanac of China's Foreign Economic Relations and trade to trace the sectoral distribution of regional investment. Although this may not generate a general picture of the overall distribution, we can at least get some idea of the sectoral allocation of foreign capital in the pattern of Sino-foreign joint ventures.

It is evident from table 5 that Asian MNEs mainly engage in food-processing industry, followed by petroleum and chemicals, electronics, textiles and clothing, and heavy industry. These industries are in fact the ones in which Asian countries/regions (such as Hong Kong, Singapore and Korea) have been experiencing declining comparative advantage relative to China. It would be advantageous for those firms to invest accordingly in china instead of organizing production at home. So they use China as a cost-effective platform for servicing global export markets. The investment distribution in the five sectors as classified according to its factor-intensive uses shows this trend clearly. Of Asia's total investment in forming Sino-foreign joint ventures, US$58.63 billion (45%) was in agricultural resource-intensive industry, US$33.14 billion (25.4% of the total) in un-skilled labor-intensive sector, and finally US$13.18 billion (10.1%) and US$ 9.24 billion (7.1%) went to the mineral resource-intensive and technology-intensive industries, respectively. This investment pattern shows a complementarity that corresponds to revealed comparative advantage as described above.

On the other hand, encouraged by China's cheap labor and incentive policies toward export-oriented, Asian MNEs are mainly producing processing trade, where raw materials, components or unfinished goods are imported into the country for final assembly and re-export. In 1985, General trade exports took on 87% of the total export, processing trade only took on 13%, the two ratios were 51% and 47% in 1992, 48% and 50% in 1995, 41% and 56% in 1997, and 27% and 70% in 2000, respectively (data from various issues of Foreign Investment Statistics). Probably, a lot of these imports are of intra-firm character, especially if the intermediate products are highly firm-specific. The import of machines and equipment as well as other investment goods is another important base plank of FDI firms purchasing from the parent company. This finding is consistent with Helpman's vertical MNE theory that outward direct investment facilities structural adjustment in the investing country by transferring abroad the industries in which the country is losing its comparative advantage.

Unlike Asian MNEs' export-orientation, North America, European and Oceania MNEs are market-oriented (Zhang 2000). Table 5 shows that North America and European MNEs are mainly concentrated in high-tech industries like transport equipment (US$16.45 billion and 2.74 billion), electronics (US$1.92 billion and 1.03 billion) and electrical machinery (US$3.59 billion and 0.95 billion) industries. Oceania MNEs are mainly concentrated in petroleum & chemical and paper & wood industries. In these industries the production process often requires internal handling of firm-specific assets such as basic R&D. This can be achieved more easily within an organization than across markets (Nunnenkamp, Gundlach, and Agarwal 1994). So intra-firm trade becomes an important issue for these countries. This explains our empirical evident that the benefits from the operations of MNEs from these regions for China's trade are more import- than export-intensive (the coefficients of inward FDI in import equation for North America, Europe and Oceania are 0.526, 0.528 and 1.813, respectively, compared to 0.129, 0.399 and 0.329 in export equation, respectively). The insignificant effects of inward FDI on trade in Africa and Latin America may be explained by their small FDI stock relative to trade.

Sectoral distribution of China's investment abroad by industry and factor intensities is not available due to data limitation, in general, Chinese investments in Hong Kong, U.S. and other western countries aim at improving efficiency in production, maintaining the export market, securing raw materials and technical know how (acquiring know-how), and planning for a long-term marketing strategy. They are concentrated in the service sectors (UNCTAD 1995). Given that services are largely non-tradable, this kind of FDI would be predicted an increase in exports from China, especially in retail and wholesale trade, with little or no predicted impact on imports back to China. Chinese overseas investments development in Africa and Latin America are mainly mature technologies and industries which are transferred overseas to make use of comparative advantage, they are often small scale and labor intensive, producing low value added goods such as textiles, footwear, bicycles and motorcycles, electrical and electronic appliances, where firms take advantage of factor price differences, including low-cost natural resources, with at least part of the foreign production being exported back to China, whereas we can not get this empirical evidence for Latin America due to its very small outward FDI stock relative to trade.

6. CONCLUSIONS

This paper investigates the impact of FDI on trade performance for China and its partner countries within a gravity model framework. It is found that the complementary relationship between FDI and trade dominates. However, the patterns of relationship between FDI and trade flows depend on the stage of similarities between investing and recipient countries. Moreover, my research has revealed that different groups of countries obey different patterns of relationship between FDI and trade.

There are three principal findings in this paper. First, every China's bilateral region has different exposures to bilateral FDI and trade, with different degrees of economic development and regional investment setting. These region-specific factors not only determine different patterns of trade growth and FDI flows, but also shape the relations between FDI and trade. Second, modes of inward FDI across 6 regions are different. Asian FDI into China are largely export-oriented, taking advantage of seaport facilities and closure to overseas markets, while such investments from North America, Europe and Oceania are primarily targeted at the domestic market. The different market orientation is an important explanation for the differential impacts of inward FDI on trade across the six regions. Export-oriented FDI is a complement to both export and import, market-oriented FDI is a complement to imports but a complicated influence on exports. Finally, in terms of China's outward expansion, its investments in Asia and North America are largely motivated by the need to expand its overseas market share and secure a stable supply of resources, outward FDI has a complement relation with trade. FDI to Europe and Oceania are mainly resource-seeker, the effect is a substitute for export. China's investment to Africa and Latin America aims to attain comparative advantages by transferring its mature technology and industries, the evidence manifests outward FDI is a complement to trade for this case.

TABLE 1. AVERAGE INFLOWS AND OUTFLOWS OF CHINESE FDI BY REGION
(1979-2000) (USD10 000)

                     1979-1989        1990-1995         1996-2000

Partner regions      Value    % of    Value     % of    Value     % of
                              total             total             total

           Total     140863   100     1977791   100     4269577   100
           Asia      105253   74.72   1662411   84.05   3018465   70.70
           Africa    661      0.47    1046      0.05    14736     0.35
Inflows    Europe    8931     6.34    90773     4.59    426254    9.98
FDI        Latin     676      0.47    9837      0.50    300311    7.03
           America   12509    8.88    163459    8.27    425737    9.97
           Oceania   1845     1.31    10999     0.56    52126     1.22

           Total     59076    100     325151    100     693804    100
           Asia      35371    59.87   252643    77.70   493709    71.16
           Africa    14932    25.27   30921     9.5     123507    17.80
Outflows   Europe    1174     1.99    22176     6.82    30491     4.39
FDI        Latin     457      0.77    3539      1.09    10691     1.54
           America   5097     8.63    10846     3.34    23712     3.41
           Oceania   2045     3.46    5026      1.54    11694     1.68

Note: Data are on a realized basis.

Source: Almanac of Chins Foreign Economic Relations and
Trade, 1991-2002.

TABLE 2. BILATERAL TRADE PERFORMANCE (USD 100 Million)

                          Asia   Africa   Europe   Latin
                                                   America

        Exports   Value   207    4        49       3
                  %       70.7   1.4      16.6     1.1
1979-   Imports   Value   174    2        83       14
1989              %       51.8   0.7      24.9     4.1
        Export+   Value   381    6        132      16
        Imports   %       60.7   1.0      21.0     2.6

        Exports   Value   628    16       147      16
                  %       64.9   1.6      15.2     1.7
1990-   Imports   Value   542    7        159      20
1995              %       59.1   0.8      17.4     2.2
        Export+   Value   1170   23       344      37
        Imports   %       62.1   1.2      18.2     1.9

        Exports   Value   1046   38       334      51
                  %       54.4   1.9      17.4     2.6
1996-   Imports   Value   1004   26       306      37
2000              %       61.7   1.6      18.8     2.3
        Export+   Value   2070   64       641      209
        Imports   %       58.3   1.8      18.0     5.9

                          North     Oceania   Total
                          America

        Exports   Value   27        3         293
                  %       9.2       0.9       100
1979-   Imports   Value   51        10        335
1989              %       15.1      3.0       100
        Export+   Value   77        13        628
        Imports   %       12.3      2.0       100

        Exports   Value   148       11        967
                  %       15.3      1.1       100
1990-   Imports   Value   125       23        915
1995              %       13.6      2.4       100
        Export+   Value   273       34        1883
        Imports   %       14.5      1.8       100

        Exports   Value   405       28        1923
                  %       21.1      1.4       100
1996-   Imports   Value   208       42        1624
2000              %       12.8      2.5       100
        Export+   Value   614       69        3547
        Imports   %       17.2      1.9       100

Source: Almanac of Chins Foreign Economic Relations
and Trade, 1991-2001.

TABLE 3. TESTING THE LINK BETWEEN FDI AND EXPORTS AND IMPORTS

Independent          Dependent Variable
Variables
                     Bilateral exports

                     A         B         C         D

LOG(GDPs?)           0.755     0.623     0.691     0.595
                     (55.28)   (40.68)   (45.50)   (42.63)
LOG(PGDPs?)          0.071     0.007     0.028     0.047
                     (2.96)    (0.34)    (1.20)    (2.01)
LOG(EX?)             0.028     0.033     0.029     0.031
                     (3.32)    (4.27)    (3.46)    (4.26)
LOG(EDIS?)           0.329     0.118     0.267     0.126
                     (9.99)    (4.72)    (9.78)    (5.39)
LANGU?               1.290     0.769     0.813     0.497
                     (28.98)   (16.21)   (15.20)   (9.01)
LOG(OUTFDI?)                   0.188               1.263
                               (19.11)             (17.09)
LOG(INFDI?)                              0.129     0.396
                                         (13.61)   (11.23)
Adjusted [R.sup.2]   0.659     0.731     0.705     0.750
Observations         879       877       859       849

Independent          Dependent Variable
Variables
                     Bilateral Imports

                     A         B         C         D

LOG(GDPs?)           1.109     1.119     1.076     1.083
                     (63.35)   (55.46)   (51.23)   (47.86)
LOG(PGDPs?)          0.085     0.019     0.045     0.133
                     (3.05)    (0.82)    (2.21)    (4.38)
LOG(EX?)             0.012     0.047     0.020     0.044
                     (0.91)    (2.98)    (1.41)    (2.79)
LOG(EDIS?)           -0.041    -0.091    -0.052    -0.071
                     (-1.46)   (-0.66)   (-1.81)   (-1.08)
LANGU?               2.811     2.620     2.633     2.483
                     (44.51)   (31.58)   (34.00)   (27.46)
LOG(OUTFDI?)                   0.050               0.122
                               (3.74)              (2.50)
LOG(INFDI?)                              0.046     0.757
                                         (4.10)    (3.92)

Adjusted [R.sup.2]   0.646     0.662     0.649     0.668
Observations         879       877       859       849

Note: The t-statistics are in parentheses and are constructed
using standard errors that are autocorrelation and heteroskedastic
consistent. The 5 percent critical value for the t-statistic is
1.96. Column A includes only standard gravity variables for trade;
column B adds outward FDI stocks; column C adds inward FDI stocks;
and column D adds both outward and inward FDI stocks.

TABLE 4. RESULTS OF TEST FOR COMPLEMENT OR SUBSTITUTE BETWEEN
FDI AND BILATERAL TRADE BY REGION

Regions         Export Equation              Import Equation
                Inward FDI     Outward FDI   Inward FDI   Outward FDI

Asia            4.215          6.915         5.074        3.692
                (31.42)        (18.52)       (34.15)      (21.70)
Africa          0.335          6.832         0.069        0.759
                (1.28)         (17.18)       (1.50)       (5.36)
Europe          0.399          -0.795        0.528        1.044
                (2.69)         (-9.34)       (4.20)       (1.61)
Latin America   0.086          0.673         0.329        0.193
                (0.82)         (1.42)        (1.46)       (0.46)
North America   0.129          0.833         0.526        0.026
                (0.99)         (4.92)        (9.78)       (0.39)
Oceania         0.329          -0.979        1.813        0.149
                (2.89)         (-7.61)       (11.20)      (1.01)

Note: The t-statistics are in parentheses and are constructed using
standard errors that are autocorrelation and heteroskedastic
consistent. The 5 percent critical value for the t-statistic is 1.96.

TABLE 5. SECTORAL DISTRIBUTION OF SINO-FOREIGN JOINT
VENTURES BY INDUSTRY AND FACTOR INTENSITIES, 1979-96
(in Billion US Dollars)

                        Asia    Africa   Europe

Food                    42.02   0.42     0.61
Textiles & clothing     20.37   0.58     0.98
Paper & wood            4.38    --       0.64
Products
Petroleum &             26.75   0.41     1.00
  chemicals
Electronics products    18.36   --       1.03
Electrical machinery    4.16    --       0.95
Transport equipment     14.38   --       2.78
Agricultural resource   58.63   0.42     0.61
Mineral resource        13.18   --       --
Unskilled labor         33.14   0.58     1.37
Human capital           16.22   --       3.21
Technology              9.24    --       1.95
  intensive

                        Latin     North     Oceania
                        America   America

Food                    0.21      3.67      0.21
Textiles & clothing     0.24      1.69      0.24
Paper & wood            1.06      0.35      1.06
Products
Petroleum &             1.45      1.78      1.45
  chemicals
Electronics products    0.46      1.92      0.46
Electrical machinery    --        3.59      --
Transport equipment     --        16.45     --
Agricultural resource   2.19      7.39      2.19
Mineral resource        --        1.35      --
Unskilled labor         0.70      3.45      0.70
Human capital           0.42      12.94     --
Technology              --        4.22      1.45
  intensive

Source: The Almanac of China's Foreign Economic Relations
and Trade, 1991-1997/1998, with authors' compilation.

REFERENCES

Bolling, Christine, and Agapi Somwaru, "The Role of Foreign Direct Investment in NAFTA and Mercosur Countries," selected paper, Northeastern Agricultural and Resource Economics meetings, Morgantown, WV, 1999.

Fontagne L., and Pajot M., "Foreign Trade and FDI Stocks in British, US and French Industries: Complements or Substitutes?," in N.PAIN ed Inward Investment, Technological Change and Growth, The Impact of Multinational Corporations on the UK Economy, Palgrave, December, 2000, 240-263.

Fung, K. C., Trade and Investment: Taiwan, China and Hong Kong, January 2003; Http://groups.northwestern.edu/tasc/events/kcfung.ppt.

Frankel, J. A., Wei, S., and Stein, E., APEC and Regional Trading Arrangements in the Pacific Pacific Trade and Investment: Options for the 1990s, Kingston, Canada: John Deutsch Institute, 1995, 289-321.

Lardy, N., "The role of Foreign Trade and Investment in China's Economic Transformation," In A. Walder, ed. China's Transitional Economy. Oxford: Oxford University Press, 1996.

Liu x., Wang C. and Wei Y., "Causal Links Between FDI and Trade in China," China Economic Review Vol. 12, 2001, 190-202.

Ma, S. and Li, J., "Location Decision of Chinese Outward FDI," Touzi Yanjiu (Investment Rresearch), Vol. 6, 1995, 47-49.

Nunnenkamp, P., Gundlach, E. and Agarwal, JP. "Globalisation of Production and Market," Die Weltwirtschaft, Vol. (1), 1994, 67-92.

Sun, Haishun, "Exports, Inward Foreign Direct Investment (FDI) and Regional Economic Growth in China," Regional Studies, Vol. 35, No 3, 2001, 187-196, Carfax Publishing Limited, United Kingdom.

Tse, D. K., Pan, Y. and An, K. Y., "How MNCs Choose Entry modes and Form Alliances: The China Experience," Journal of International Business Studies, Vol 28 (4), 1997, 779-805.

UNCTAD, "World Investment Report 1995: Transnational Corporations and Competitiveness," United Nations Publication, United Nations, New York and Geneva, 1995.

Wells, Louis, Mobile Exporters New Foreign Investors in East Asia, in Foreign Direct Investment K. A. Froot, ed., University of Chicago Press, Chicago, 1993.

Wu, H. and Chen, C., "An Assessment of Outward Foreign Direct Investment from China's Transitional Economy," Europe-Asia Studies, Vol. 53 (8), 2001, 1235-1254.

Zhang, K. H., "Why Is U.S. Direct Investment in China So Small?," Contemporary economic Policy, Vol. 18 (1), 2000, 82-94.

Zhang, Z. and Hock, O. C., "Trade Interdependence and Direct Foreign Investment between Asian and China," World Development, Vol. 24 (1) 1996, 155-170.

Zhang H., and Van den Bulcke D., "The restructuring of the Chinese automotive industry: the Role of Foreign Direct Investment and Impact of European Multinational Enterprises,"--Antwerp, CIMDA discussion paper/University of Antwerp, Center for International Management and Development, 38, 1999.

Yong Li is a Ph.D. is a Ph.D. candidate at University of Science & Technology of China and an associate professor of economics at Shandong Finance University. Her research activities have included finance theory, emerging markets, and China's economy and capital markets.