Analyzing the Effects of Foreign Aid on FDI - A Gravity-Equation Approach

RIETI Report: Your research builds on and extends prior studies on the impact of foreign aid on FDI, and provides new analyses as well. Could you elaborate on the background of this research?

Kimura: Capital inflows are extremely important for a country's economic growth, and both foreign aid and foreign direct investment constitute important capital inflows for less developed countries (LDCs). However, although foreign aid has been provided in huge amounts for decades, its direct impact on economic growth still remains an open question. There have been many econometric studies on this issue, but the results have shown no clear effects. Although the direct effect of foreign aid on growth is not clear, it may still promote the growth of the recipient country indirectly, for example, by facilitating domestic investment, physical infrastructure investment, and FDI. As we look at FDI, the other capital inflow, many empirical studies have shown its positive impact on economic growth, though it is dependent on factors such as the educational and technological level of the recipient country. Therefore, we presumed that foreign aid may have an indirect positive effect on growth if it is associated with a rise in FDI in the recipient country.

Previous studies based on the same idea have been conducted, but they aggregated positive and negative effects of aid on direct investment for each recipient country, and they did not demonstrate conclusively whether aid has a positive effect on FDI. However, our intuition told us differently. It is more natural to think that aid from a certain donor country is more likely to draw FDI from that country, rather than from a third country, through the transmission of information and the building of intergovernmental relations. To use a specific example of aid and FDI from Japan, Japanese aid may promote direct investment from Japan, but it may not necessarily attract FDI from Germany. Therefore, we inferred that the reason why existing studies could not show significant results might lie in their methodology for aggregating data. This led to our attempt to conduct a new study using decomposed data. The results of our econometric analyses show that our intuition did not apply to all countries, but was right only in the case of Japan.

RIETI Report: What is the most significant contribution of this paper? Also, the paper discusses how aid has a "vanguard effect," not observed in aid of other donor countries. As can be seen in Figure 2 [PDF:341KB] (Sectoral Share of Total Foreign Aid by Donor), the share of aid for economic infrastructure is greater in Japan than in other countries. How do such variations relate to differences in the aid effect among donor countries?

Kimura: Previous studies analyzed the effects of aid on FDI by breaking down the total aid amount from all donor countries into aid for infrastructure and aid for non-infrastructure, and distinguished between an infrastructure effect and rent-seeking effect. This paper examines the existence of a third, vanguard effect, by using data for each source-recipient country pair in order to estimate the effect of aid from a particular donor country on FDI from that donor. Thus, the paper's major contribution is the decomposition of the three effects of foreign aid on FDI.

We first estimated the effect of foreign aid on FDI based on the assumption that there is no variation across donor countries. To summarize our results: foreign aid does not promote FDI inflows either by supplying economic/social infrastructure in the host country (no infrastructure effect) or by providing information on the business environment in the host country (no vanguard effect), nor does it shrink FDI by encouraging unproductive rent-seeking activities (no rent-seeking effect).

However, because the objectives, methods, and modalities do in fact vary widely among donor countries, this assumption may not hold. Accordingly, the next step was to relax this assumption and examine whether foreign aid from any particular donor country has a distinct infrastructure effect, rent-seeking effect, or vanguard effect compared with other countries' foreign aid. As regards the infrastructure effect and rent-seeking effect, we found that aid from any donor country had no significant impact on FDI at the 5% level. If aid from Japan, much of which is for economic infrastructure, promotes private-sector FDI through the development of infrastructure, it should give rise to FDI not only from Japan, but also from other countries. However, such an effect could not be observed. We therefore narrowed the focus of our research, estimating the effect of foreign aid from five major donor countries in promoting bilateral FDI from those countries (the vanguard effect), and found a positive vanguard effect in the case of Japanese aid for infrastructure. That is to say, Japanese aid for infrastructure promoted FDI from Japan, although it had no effect on FDI from other countries. (Note: Aid for infrastructure includes social infrastructure, economic infrastructure, production, and multi-sector activities.)

RIETI Report: The next question concerns your methodology of using data on foreign aid and FDI for each source-recipient country pair. Between 1995 and 2002, the subject of your analysis, the main recipients of Japan's foreign aid and FDI were East Asian countries, whereas the recipients of foreign aid and FDI from Germany, France, the United Kingdom, and the United States include many Central and Latin American countries and African countries. Do you think the disparities in FDI growth among these regions affect your analysis?

Kimura: Although we confine our analysis to the period of 1995-2002, our dependent variable is the FDI stock, which is constructed by the perpetual inventory method using data on FDI flows from 1985 to 2002. The major independent variable of "aid" is the aid stock constructed by the same method using data on aid flows from 1973 to 2002. Looking at the geographical trends in FDI from 1985, FDI to East Asia grew, but so did FDI for Latin America and Central and Eastern Europe. Also, FDI increased at an enormous rate among industrialized countries, which shows that FDI growth in this period was a global trend, although it is not directly related to the analysis in this paper. In addition, FDI to Asia declined following the Asian crisis in 1997, but the trend was not confined to the Asian region, with FDI to other regions also declining. Also of note is the sharp rise in FDI to Sub-Saharan Africa from 2001. Given these circumstances, I do not think there were particularly large regional disparities. As a reference, this graph shows a regional breakdown of FDI in LDCs from seven major donor countries (Japan, the U.S., UK, Germany, France, the Netherlands, and Sweden).

Source: OECD Direct Investment Statistics, author's calculation

Regions classified by the World Bank: EAP (East Asia and Pacific), EUCA (Eastern Europe and Central Asia), LTA (Latin America and Caribbean), MDA (Middle East and North Africa), SA (South Asia), SSA (Sub-Saharan Africa)

For this estimate we used gravity equations frequently applied in quantitative analysis of bilateral trade. To be specific, the analysis incorporates variables considered to exert a strong influence on FDI, such as the economic size of the investing and host countries, relative technological levels, and their geographical distance. In addition, when using correlated variables such as aid and GDP as explanatory variables simultaneously, it is important to remove the endogeneity (which causes biases in the analysis), and so endogeneity is excluded through the use of the system generalized method of moments (GMM) method.

RIETI Report: How do you hope the results of this research will be reflected in Japan's foreign aid policy?

Kimura: Our empirical results underscore that Japanese foreign aid is highly distinctive compared to that of other donors. The remaining question is: Why is it distinctive? There is the thinking that it is in the interest of Japan, as a trading nation, to contribute to promoting the economic development of its trading partners, particularly other Asian countries; the argument that in Japan there is a close cooperative relationship between the government and the private sector, and the primary objective of Japanese foreign aid since the mid-1980s has been to promote Japanese FDI; and the collaboration between government economic agencies and the private sector in advocating "trinity (three-in-one) type" official development assistance (ODA) of a kind that links Japanese FDI, trade, and ODA in order to foster the development of the economies of aid recipient countries.

We believe that the vanguard effect of foreign aid arises mainly due to (1) transmission of the local business conditions of the recipient country through information exchange between the government and private sector, (2) the fact that the Japanese government is providing aid lowers the country risk of the recipient country that private sector firms form judgments at their own discretion, and thus acts as a "quasi government guarantee," and (3) the fact that the provision of aid brings with it Japanese standards and business rules, facilitating the implementation of direct investment by the private sector. This is not an exclusive list; there may well be other factors.

As I said at the outset, although foreign aid has been provided for decades on a massive scale, its direct effect on economic growth has been ambiguous. In addition, due to the development of globalization and the rapid growth achieved by China and India, major FDI recipient countries, recent debate has renewed emphasis on the importance of private capital inflow to LDCs. In this sense, we could argue that Japanese foreign aid has been "good aid," due to its capacity to attract private sector funds with a higher possibility of promoting economic growth. However, the possible drawback is that Japanese foreign aid may be interpreted as aid for the sake of Japanese private sector companies and the Japanese economy, and not for the reduction of poverty in LDCs. Such a view is not desirable for Japan. In this regard, it will be an important task for the Japanese government to simultaneously provide aid more directly linked to poverty reduction, as well as to disseminate such efforts. The simultaneous achievement of economic growth and poverty reduction is extremely important.

In this paper we were able to verify that Japanese foreign aid facilitates FDI from Japan, but unfortunately our analysis does not cover the aid-growth nexus through FDI. We expect that future studies will be able to fill the missing link between the aid-FDI nexus and the FDI-growth nexus.

Event Information