Monday, March 14, 2011

The Economic Costs of Natural Catastrophes: History’s Lesson for Japan

The earthquake that struck Northern Japan on Friday is a human disaster of frightening dimensions. Early estimates place the loss of human life at ten thousand, but such a figure will regrettably prove a gross underestimate. Our first thoughts go to those victims, killed, injured, or suffering the hardship of little food and no electricity.

The Japanese markets opened as usual on Monday morning. Not unexpectedly Japanese stocks fell five percent, and the discussion of the earthquake’s economic cost to Japan has already begun. Talk has turned to how much Japanese GDP will be lost, thereby compounding the problems of Japan’s feeble recovery.

In such cases, economic historians are naturally drawn to past natural catastrophes and their effects on economic output. Some of the worst natural disasters occurred in a past, so remote that we do not have the necessary statistics. Also, we must recognize that historical statistics are much less inaccurate than the current ones, especially when it comes to estimating annual figures. With these reservations in mind, my methodology is to take the country’s GDP the year prior to the natural disaster and then calculate the loss of output until GDP returned to or exceeded its pre-disaster level. This is a primitive method. The correct method would be a counterfactual analysis of the loss of output as measured by what output would have been in the absence of the disaster.

Here is what the statistics show:

1. Of the four major earthquakes of the twentieth century (Italy, 1908, Japan, 1923, Peru, 1970, Iran 1990), I found no case in which GDP declined as a consequence of the earthquake, even though the loss of life was horrendous (ranging from 50,000 for Iran and 200,000 for Japan 1923). Clearly, output would have been higher without the earthquake, but the earthquake’s effect was not strong enough to cause output to actually decline.
2. Major famines caused a much greater loss of output than earthquakes. The greatest recorded loss was a twenty-one percent decline in output from the Russian famine of 1891-92 (500,000 lives), followed by a three percent loss from the India-Bengal famine of 1943, which cost two million lives. China’s 1976 famine cost two hundred thousand lives and a one percent loss of output. We still do not have accurate statistics on Mao’s Great Leap famine of the late 1950s, but it claimed the most lives of any famine of modern history.
3. Floods have also had a minimal impact on economic output. China’s 1935 flood which claimed 3.5 million lives was accompanied by a robust seven percent rate of growth.
4. Epidemics can have a substantial effect on output. The French-German smallpox epidemic of 1870 cost half a million lives and a drop of five percent in output.
5. Political upheavals, such as war, civil war, and revolution can eclipse the economic effects of natural disasters. The Russian civil war caused a drop in output of fifty percent, and the “small” 1905 revolution caused a staggering fourteen percent decline in GDP.

Does history provide any guidance to estimating the effect of Japan’s March 2011 earthquake on Japan’s GDP? Judging from past natural disasters, the impact on GDP depends on the affected area. Floods and earthquakes can kills hundreds of thousands, but they tend to be confined to a particular geographic region. This limits their impact. Famines can cause huge drops in output in relatively poor agricultural countries (like Russia and India) because agriculture is the economy’s major output, and most of the population is in rural areas. Epidemics that transcend national boundaries such as the French-German smallpox epidemic of 1870 or the Spanish Flu epidemic of 1918 (or the Black Death of the Middle Ages) cover large swaths of territory and population and can have large impacts on economic output.

If history is a guide, the March 2011 earthquake will not have a large effect on Japan’s GDP. There is no denying it could not have come at a worse time for the Japanese economy, which, like the United States, was in the process of emerging from the great recession of 2007- present.

The greatest long-run effect will be to slow down the development of nuclear power plants worldwide.


  1. Hi paul, my thoughts exactly.
    Except that BOJ has expanded the QE program to 10T Yen to aid the recovery. However it needs to be signficantly bigger to have a meaningful impact.

    Intrinsic Value