Professor Stephanie Lackner explains her research on the impact of natural disasters on economic growth in the long run. Despite different theories supporting both negative and positive impacts of natural disasters on the economy, she argues that the long-run productivity effects of natural disasters are not predetermined. She argues these effects can vary depending on factors such as vulnerability of infrastructure, disaster preparation and response, or access to financing and insurance for reconstruction.
Stephanie Lackner is an Assistant Professor in Economics at the IE Business School and the School of Global and Public Affairs.
Her research interests revolve around natural hazards, the socioeconomic impacts of disasters and geospatial data analysis. She was trained as an economist and she uses applied methods to address topics in environmental economics and development economics in her work. After finishing her PhD in Sustainable Development at Columbia University, she completed a postdoc at the Program in Science, Technology, and Environmental Policy (STEP) at the Woodrow Wilson School at Princeton University.
Natural disasters have often devastating negative impacts on the people affected by them, and they cause large immediate economic damages every year.
When it comes to the impacts of disasters on productivity and economic growth in the long-run however, it is actually much more ambiguous in which direction these impacts go.
We can visualize this issue by plotting GDP per capita over time. GDP, gross domestic product, per capita is a commonly used measure for the economic productivity of a country.
Let’s assume that our country would have been on a trajectory of economic growth that we can call its “baseline trend”. This trend here implies that GDP per capita would have grown continuously over time.
But instead of following this path, an unexpected natural disaster hits our country at some point in time and things change.
The big question now is: what will happen to the growth of GDP per capita in the long-run? To answer that question we need to think about the different direct and indirect impacts that natural disasters have. Their direct effect is obviously the destruction of capital and in the worst cases also human casualties.
Intuition might suggest that destroying capital would make it harder for our country to produce goods and services and growth would therefore temporarily be negatively affected. The economy would then start to recover and growth would return to its original level, but the GDP per capita itself would be permanently negatively impacted.
The new trajectory would be parallel, but below the baseline trend. The area between the two lines represents the size of the long-term productivity loss due to the natural disaster. We can think of this scenario in a way as moving the country back in time undoing the growth that happened during some period of time.
An even more pessimistic hypothesis suggests that not only the level of GDP per capita would be reduced in the long-run, but also the growth rate of GDP per capita. While there will always be some recovery, the country might be so much negatively affected that the growth rate will stabilize at a lower level than before. This is the worst case scenario.
But there are also other theories. According to neoclassical economic growth theory, the country would experience a temporary drop in GDP per capita due to the destruction of productive capital such as damaged roads or destroyed buildings. But the lower capital levels would also allow for higher productivity gains and the disaster would therefore be followed by higher growth rates than what the country experienced previously.
This would continue until the country reaches its original baseline trend. In this scenario there would be no long-run impacts of the natural disaster on GDP per capita nor its growth rate. Nevertheless, there would have still been temporary productivity losses, which can again be visualized by the area between the baseline trend and the return-to-trend trajectory. The smaller this area is, the better is the recovery process of the country.
There are also arguments for why a country’s productivity might actually be positively affected by a natural disaster. Shortly after a disaster strikes, destroyed capital will be replaced and reconstruction will begin to rebuild the destroyed infrastructure. This can cause a construction boom and through this also give the entire economy a temporary boost. We might therefore end up on a trajectory that is parallel but above our original baseline trend.
Finally, when natural disasters tend to destroy old and weak infrastructure, which might not have been the most productive in the first place. But when this infrastructure is rebuilt, it will often be built in a higher quality up to the current new standards and technologies. The disaster can also act as a catalyst for institutional or social change and thus allow for more growth than previously. These “Build back better” and “Creative destruction” effects can actually make it possible for a country to be in the long-run better off in terms of economic productivity than they would have been without the natural disaster.
But what happens in reality? Solomon Hsiang and Amir Jina, two professors of Public Policy have shown in a global empirical analysis that on average even after 2 decades cyclones still have a significant negative effect on the level of GDP per capita of an affected country.
In my own research, I find similar results for the impact of earthquakes after about one decade. On average growth seems to recover, but the level of GDP per capita is still negatively affected many years after the disaster. So, what about those “creative destruction” and “build back better” arguments?
Well, the answer is: it depends. When I split my analysis on earthquakes by country income levels into three groups, I found that the average negative effects are actually primarily coming from low and lower-middle income category countries, while high-income countries actually do seem to experience those positive productivity effects.
What we learn from this is that the long-run productivity effects of natural disasters are not predetermined. They can depend on many different factors: the vulnerability of infrastructure, disaster preparation and response, access to financing and insurance for reconstruction after an event, and effective policies and institutions.