Hurricanes can naturally be incredibly devastating events, killing many people and wrecking infrastructure in the space of a few hours or days.
The economic damage from a hurricane, however, can stay with a country for decades.
A National Bureau of Economic Research working paper by Berkeley public policy professor Solomon Hsiang and Columbia sustainable development PhD student Amir Jina explores the long term impact of tropical storms on countries’ economic growth.
The researchers combined models from atmospheric physics and economics to find this effect. Based on a comprehensive database of cyclone activity from 1950 to 2009, and a physical model of cyclone behaviour, they were able to estimate the top cyclone wind speeds in each country for each year. This gave a measure of how hard countries got hit by hurricanes each year.
They then put the hurricane data into an economic model designed to detect the effect of hurricane intensity on GDP growth, while controlling for other attributes of different countries and different years. They found that hurricanes had a long term negative impact on GDP.
Here’s a chart from the paper showing their main result. For each year after a hurricane hits a country, they plot the incremental effect of the hurricane’s intensity, measured in wind speeds of meters per second, on per capita GDP. They found that, even twenty years after a storm, there was still a drag on economic growth, relative to a baseline trend with no cyclone:
A full fifteen years after a hurricane or typhoon strikes a country, that country’s per capita GDP will be lower by 0.38% for each meter per second of top wind speed than it would have been without the cyclone. So, a storm whose top wind speed was 10 m/s, or about 22 miles per hour, would have its per capita GDP lowered by about 3.8% fifteen years later, compared to a stormless baseline.
The authors also note that cyclones have a much more severe effect on countries with multiple storms. Because each hurricane or typhoon negatively affects GDP in the long term, multiple storms over a period of time can have effects that add up.
They illustrated this with a fairly simple thought experiment: they plotted different countries’ per capita GDP growth from 1970 to 2009, then used their model for GDP growth, but with all cyclones removed, to estimate a counterfactual storm-free situation. They note the fairly big caveat that this is very much an informal thought experiment: taking a country that suffers from repeated cyclones and removing those storms would naturally involve many other changes to that country’s geography and climate that could have major economic effects.
For a country like the Philippines that gets hit by typhoons very regularly, this has had a major impact on GDP growth. The following chart from the paper shows this effect. The blue line represents GDP simulated by the authors’ full model including the tropical storm effects, the black line is the actual historical per capita GDP level, and the red line shows the results of running the simulation with typhoons removed. The grey bars in the background show the average cyclone wind speed for each year:
The Philippines, according to the authors’ models, have been deprived of quite a large amount of economic growth as a result of repeated typhoons. They write, “this effect of removing Filipino cyclones is one of the most extreme cases, equivalent to raising the average annual growth rate in the Philippines by roughly 7.3 percentage points, and would cause growth in the Philippines to match that of its near neighbour China.”
The authors conclude by pointing out that climate change could lead to an increase in the frequency and intensity of tropical storms, and since these have long term effects on economic growth, the warming of the planet may contribute to slower growth in the future through these storms.
For more details, check out the working paper at NBER here.