Factors affecting mitigation of natural disaster fatalities in selected countries

The main objective of this study is to investigate the relationship between fatalities due to natural disaster and its determinant, specifically the linkages between macroeconomic variables, institutional factor and fatalities due to natural disaster. The study is conducted in order to have a better understanding of how to mitigate the fatalities. The study covers 79 countries with data spanning for the period 1980 to 2005. To achieve the goal, we employed Generalized Method of Moment (GMM) estimation developed for dynamic models in panel data, popularized by Arellano and Bond (1991) and Blundell and Bond (1998). As for the dependent variable, we used total killed/death, total affected per capita and total economic damages as a proxy for the impact of natural disaster. To test for the independent variables, ranging from macroeconomic and institutional factors, we used per capita income, square of per capita income (measure of non-linear relationship), total population, population density, investment, government consumption,education, openness and corruption as the explanatory variables. The findings suggest that the levels of economic development are inversely related, which means that an increase in economic development will reduce the consequences or impact of natural disaster, in other words mitigates the fatalities. Another important finding is that there is a non-linear relationship between the stage of economic development and the fatalities due to natural disaster. Countries which have more risk exposure have an inverse U-shaped relationship between total killed, total economic loses and wealth and there is also a U-shaped relationship between disaster death, loses and wealth for countries which is more disaster-prone. The results also support the notion of that the positive urbanization coefficient and its negative interaction with GDP per capita indicating that low income highly urbanized countries will increase disaster fatalities than similarly urbanized high income countries. As for the population, we obtained positive relationship; both total economic losses and total affected are positively related with population, an increase in population would literally mean the dragnet would become greater. In contrast, the relationship between population densities indicates that it is positively and significantly related with fatalities due to natural disaster. The results for unemployment is also mixed, while we found that unemployment rate is significantly and inversely related with total economic losses due to natural disasters. In contrary, we also found that unemployment rate has positive relationship with total death and total affected due to natural disaster. As for investment, we found that the investment is positively related with fatalities (all three, total affected, total death and total economic loss) due to natural disaster. As for the government consumption, our finding suggest that it has a positive impact on natural disaster fatalities (total death and total economic loss), which means that if there is an increase in government consumption, it will increase consequences of human fatalities and total economic losses due to natural hazard. Meanwhile, the behavior of openness shows both positive and negative relationship with fatalities due to natural disaster, while it has an inverse relationship with death due to natural disaster; the same could not be said for total affected whereby it has a positive relationship. Our results suggest that there is negative relationship between total economic loss due to natural disaster and education attainment which strengthened the notion of higher education attainment reducing natural disaster fatalities. Finally, the result for corruption indicates that it is positively related with fatalities due to natural disaster (total affected, total death and total economic loss), which suggest that higher level of corruption will increase the consequences of natural disaster fatalities. The study shows that while it is almost impossible to predict the happenings or occurrence of natural disaster, it is possible to mitigate the fatalities if we could formulate proper policies to impact the macro variables as well as the institutional factors