The Effect of Natural Disasters on Mexico’s Regional Economic Growth:Growing Disparity or Creative Destruction?

Over the last decade, Mexico has experienced a sharp increase in the economic costs associated with hydrometeorological disasters, most notably floods, hurricanes and droughts. This is attributed to the combination of an increasing population and expanding economic activities along Mexico’s coastal areas and arid zones with the mismanagement of its urban growth. To make it worse, risks would only grow in the future under the current basic Climate Change predictions for Mexico. While immediate and short-term costs of disasters are clear, little is known about its longer-run effects on (regional) growth. To address these questions this paper uses an endogenous growth model with panel data for all 31 Mexican states for the period 2002-2007, which includes as explanatory variables all major and medium disasters in this period. Our results show that major disasters, of a magnitude enough to be declared as such by the National Disaster Prevention Center and thus eligible to receive federal relief funds, do have a negative impact on economic growth in the short term, but that this effect gradually fades away and becomes negligible for medium term growth. Thus the opportunity cost of relief funds is not as high as expected as compared to other public spending. However, when analyzing “smaller” disasters, we find a surprisingly counterintuitive effect: we find evidence of a positive effect of disasters on growth, both in the short-run and medium-run terms. This can only be explained by arguments akin to Schumpeter’s creative destruction hypothesis, where capital replacement with embodied technological change is responsible for boosts to productivity after a disaster’s shock. If this holds true in further test we would have new arguments, not to disregard climate resilience, but to favor policies that mimic a disaster’s reconstruction effects on capital replacement, essentially forcing economic agents to forego capital they would not otherwise replace. New questions emerge from these results regarding the interaction of the magnitude of the disaster, the amount of relief support and growth?