Empirical Results: Is there a PRC Effect?

6.1 Results for East and Southeast Asia

6.1.1 Does PRC Reduce FDI inflows to the East and Southeast Asian Economies?

Table 3 [PDF 84KB | 2 pages] shows the results from the first set of panel simultaneous regressions using the absolute level of FDI inflows as the dependent variables. To avoid the multicollinearity problem, variables that are highly correlated are not included simultaneously. That generates various specifications of our regressions. For our Asian regressions, the years considered are from 1985 to 2002.

Our main variable of interest CFDI is positive and significant in all specifications. A 10 percent increase in the FDI inflows to PRC would raise the level of FDI inflows to the East and Southeast Asian countries by about 1 to 3 percent, depending on the specifications. Despite considerable concerns in policy circles that an increase in FDI flow to PRC is at the expense of other regional economies, this study shows that those economies can actually benefit from it. This may be linked to the production-networking activities among Asian countries as well as the increased resource demand by a growing PRC. The evidence of production-networking among PRC and other Asian economies can be found in the substantial two-way trade of intermediate and final goods in the same industries among those countries.⁵

Many of the countries examined are heavily involved in vertical specialization, particularly in electrical equipment and electronics industries, which can be seen in the share of two-way trade in the same industry in the total volume of trade among the nations (Table 4 [PDF 94KB | 1 page]). The economic ties of mutual dependence among them have been deepening rapidly since 1990s. The significance of the PRC Effect in the level of FDI inflows to our group of Asian countries may reflect such interdependence. Thus our empirical study shows that an increase in PRC's FDI is positively and significantly related to FDI inflows in other Asian economies. Our central result here is then as follows: up to now the investment-enhancing effect dominates the investment-diversion effect, so that overall PRC is a positive force for FDI inflows into other Asian economies.

The effect of openness, denoted by the variable OPEN, has an expected positive sign and is always significant. Openness captures the degree of both tariff and nontariff measures including various trade costs. In contrast to the effect of tariff barriers proxied by DUTY, which is another significant variable, the impact of openness to trade on the inflow of FDI is substantial. The results in Table 3 suggest that, all else being equal, the marginal effect of trade liberalization of the Asian countries on the inflow of FDI can be more than twice as large as that of the PRC Effect. Trade impediments can take various forms such as local content requirements, technology transfer requirements, domestic sales and export requirements, and so on. Our results imply that reductions in the various types of trade barriers can play a vital role in promoting FDI to those countries.

Corporate tax is another variable that is found to exert a large influence on the level of the inflows of FDI in this analysis. Although many countries offer various forms of tax incentives for foreign investors, corporate tax rates can be considered as one of the most influential tools to promote investment, since it has a direct impact on the profitability of their investment projects. The effects of corporate tax rates are in most cases larger than the PRC Effect.

For the East and Southeast Asian economies, the GDP variable is significant but seems to have the wrong sign. However, its significance disappears once DUTY is added into the regressions. This seems to indicate that the GDP variable is not very robust. The degree of government stability, the index of corruption and the index for the rule of law, GOV, CORRUPT and LAW, are all insignificant. The OUTFLOW variables are positive and significant. They signify the impact of an overall "supply" effect on the inflows of FDI to these Asian economies. The proxy for infrastructure is also significant, even though it has a very small coefficient.

Overall, factors that affect the FDI inflows into East and Southeast Asia are the positive PRC Effect, policy variables such as the degree of openness to trade and the quality of infrastructure and the world supply of the FDI.⁶

6.1.2. Does PRC Reduce the East and Southeast Asian Economies' Shares of Total FDI inflows to Developing Economies?

In this empirical exercise, we change the dependent variable from the level of FDI to the country's share of the total FDI flowing into all developing countries (Table 5 [PDF 84KB | 2 pages]).

The idea is to capture the notion that some government officials may be concerned about their shares and not just the levels of their FDI. Here we found that the PRC Effect is negative and significant. This means that PRC does reduce the shares of these economies out of the total FDI inflows to all developing countries. Furthermore, the PRC Effect is large. OPEN and DUTY are as in the regressions with levels, significant. Corporate tax rates have the expected negative signs. The index of government stability has a small coefficient, but it is significant. Infrastructure is also positive and significant. But labor market variables including the wage rates and the degree of illiteracy seem to have the wrong signs.⁷ Overall, the dominant determinants of the Asian economies' shares of FDI into all developing countries are the negative PRC Effect, policy variables such as openness to trade, corporate tax rates and infrastructure, as well as the institutional factor of government stability.

6.2 Empirical Results for Latin America: Is There a PRC Effect?

6.2.1 Does PRC Reduce FDI inflows into Latin America?

In the next table we present results for the levels of FDI inflows into various Latin American economies (Table 6 [PDF 75KB | 2 pages]). For the Latin American regressions, the years we examine are from 1990-2002. In contrast to the corresponding regressions for East and Southeast Asia, the PRC Effect variable is in most cases insignificant. Even when they are significant (columns (3), (5) and (8)), the magnitudes of the coefficients are quite small, generally smaller than those in the regressions for Asia. This is consistent with the fact that the similarity of exports between PRC and the Latin American economies is still rather modest (Lall and Weiss 2004). Except for Mexico, multinational firms in general do not view PRC and most of the Latin American countries as competing sites for processing their products. We thus do not find a systematic negative PRC Effect.

On the other hand, unlike PRC and the rest of Asia, there is no comparable network of production-sharing in place between PRC and Latin America. There are indications that in electronics, a production fragmentation network may be forming between PRC and Mexico (Lall and Weiss 2004). At the same time, PRC's appetite for commodities may also spur FDI in the primary sectors of selective Latin American economies. This may explain the occasional positive signs of the PRC Effect. In sum, for Latin America, the PRC Effect is either insignificant or very mildly positive.

Levels of FDI in Latin America are mostly explained by their market sizes and their growth rates, the global supply of FDI and import barriers. In the Latin American regressions, higher trade barriers are correlated with more FDI, indicating the motive for tariff-jumping FDI. The positive sign of DUTY also indicates the lack of a production network, since with production and trade of intermediate goods, FDI will be correlated with lower trade barriers in general. This is in contrast with the results from the Asian regressions, where DUTY is negative and significant, which tends to be consistent with the existence of an East and Southeast Asian production network. A thriving business and production network in East and Southeast Asia (including PRC) in contrast to the relative lack of such clusters of production in Latin America may explain the different estimated results for Asia and Latin America.⁸

6.2.2. Does PRC Reduce Latin American Economies' Shares of Total FDI inflows into Developing Countries?

In the next table, we present our panel regression results using the Latin American economies' shares of FDI flows going to all developing countries as the dependent variable (Table 7 [PDF 74KB | 2 pages]). The PRC Effect in this case is negative and significant. As in the regressions with levels, other variables that are significant include the size of the markets, growth of per capita income and the extent of trade restrictions. Even though the PRC Effect is negative and significant here, its effect is much smaller compared to the market size variables. DUTY as an explanatory variable also has a larger coefficient. Thus, even if policymakers are concerned with their countries' FDI shares, the dominant influence here does not seem to be the emergence of PRC.

Download this Discussion Paper [ PDF 213.2KB| 32 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.