Economic Consequences of Catastrophes Triggered by Natural Hazards

Several important questions related to catastrophes are addressed in this dissertation. How closely are catastrophes and developmental process related? How is the post event economic growth related to the losses from catastrophic events? How important and how long lasting are the various effects likely to be? What is the record of past catastrophes and what regularities can be inferred from them? Can theoretical models explain some of these regularities? How will a regional economy behave after a catastrophic event? How is the effect of a catastrophe propagated to an interacting region? What pre-event conditions are crucial in explaining the fact that some economies do better after an event? Can the models explain other post-event behaviors too?

The purpose of this dissertation is twofold. On the one hand, it seeks to detect empirical regularities in the behavior of economies affected by catastrophes. On the other hand, it develops various models to study the effect of catastrophes on an economy, which explain some of the empirical regularities.

Appropriate economic models are developed to explain the observed phenomena. Theoretical simulations start by perturbing the Ramsey’s model to study the effect of sudden changes in capital and the post event changes in the productivity. Two extensions of this model are examined. The first of these studies the effect of efficiency of post-event reconstruction on subsequent behavior. The second extension studies the effect of a catastrophe on interacting economies. The behavior of the models from numerical simulations is corroborated with empirical regression results.

A cross-country study with data from countries from various income groups affected by different types of natural hazards (earthquakes, floods, hurricanes, and droughts) is presented. Results based on an econometric model imply that direct losses as a result of catastrophes are negatively correlated with the post event growth. It is seen that only by modeling the fact that after a catastrophe reconstructed capital takes time to become productive can one explain the negative correlation of the direct loss with post-event growth. Evidence point to the fact that catastrophes increase the external debt, budget deficit and inflation. However, these effects are only temporary. Two years after the event the effect of the catastrophe on the economic growth is statistically insignificant.

A standard regional economic model was used to simulate post event economic behavior for three historical events - the 1989 Loma Prieta earthquake, 1992 Hurricane Andrew, and 1994 Northridge earthquake. The model was then used to study the effects of scenario earthquakes in the Bay Area and the Silicon Valley. The gross regional product, consumption, investment, and local government spending show declines during the first two years and then recover depending on the external aid and the efficiency of the reconstruction process.

One of the main messages of this dissertation is that catastrophes cause myriad problems in the short term after an event. But efficient reconstruction policies should help the affected communities to emerge as less vulnerable, more productive and hence economically stronger regions in the long run. To achieve this efficiency catastrophe management has to be intimately linked with development policies.