Natural Disasters and Foreign Direct Investment
M onica Escaleras and Charles A. Register
The aim of this paper is to address the linkage between foreign direct investment (FDI) flows and the number of natural disasters. By using the data of 94 countries in the period of 1984 to 2004 and applying a variety of empirical tests, the result appears that natural hazards have significantly negative effects on FDI of countries. A. Economic Effects of Natural Disasters and The Determinants of Foreign Direct Investment
Economic Effects of Natural Disasters
There are three patterns that concern with the economic effects of natural hazard. The first two strands concentrates on the primary or short-term effects and long-term effects of hazards on economy. While the short-term effect strand achieves abundant evidences of negative disasters’ impacts on GDP, the long-term effect strand cannot reach a clear conclusion. The third strand focuses on the capacity to mitigate the destructive effects of natural risks. A brief conclusion is that the negative impacts of risks can be diminished by country’s institutions.
Determinant of Foreign Direct Investment
There are three types of foreign direct investment, namely:
(1) Operating new
(2) Moving an existing
(3) Moving a part of existing
The first type is considered as location decision and categorized in pull factor, the latter two types are relocation decision and belong to push factor. Following this logic, propositional pull factors to put in models are the level of openness and the size of the economy. Obviously, the push factor in models is natural risks. Other determinants which are mainly focused are institutions, such as government infrastructure, political freedom, corruption, etc. B. Data and Methods
The data for analyzing impacts of natural disasters on FDI flows are taken from the EMDAT, which provides by the institution Center for Research on the Epidemiology of Disasters (CRED) and World Bank. Some observations were dropped because of missing data, the data which is used in this research contains an unbalance panel with 1,822 country-year observations from 94 countries (29 in Africa, 17 in Asia, 22 in Europe and 26 in Americas) in the period 1984-2004. Table 2 presents descriptions of dependent and independent variables.
At this point, it is important to look again at two primary variables which devoted to results of empirical tests. The first key variable is FDI, which is measured by the total net inflows of foreign direct investment as a percentage of GDP. FDI is the dependent variable in all models. The second key variable relates to natural hazards. Since both recent and longerterm risks have its impacts on investors, the authors deliver four variables that are concerned with the number of natural risks happening in four time period: Total events in the prior year, total events in the prior 5 years, total events in the prior 10 years, total events in the prior 25 years. Table 3 shows the correlations between FDI/GDP and each of four variables referring to the measures of natural risks.
It is undoubtedly true that both the counted measure as number of natural hazards and the monetary measure as the estimation of “dollar value of damages” affect decision makers. While it can be argue that result as the dollar amount of damages may have substantial influence on investors’ decisions, it is obvious that estimating the consequence of natural disasters is complex and not as accurate as “counts of disasters”. For this reason, models will mainly focus on counts of disasters. Moreover, the research emphasizes on five types of natural hazards that severely devastate infrastructures, physical capital and labor forces. As such, these five types are earthquakes, floods, volcanoes, landslide and windstorms (include hurricanes).
The following two variables which refer to the degree of openness and incentive in...