From the Political Economy Newsletter: Earthquake Mortality and Damage

Jul 30 '11

Continuing “our relationship”:http://tmc.local/blog/2011/01/10/collaboration_with_the_apsa_po/ with the Political Economy Section of the American Political Science Association, here is an article from the Section’s most recent newsletter, preceded by a short note from the editors of the newsletter. Please note that I am happy to set up this kind of relationship with any section newsletter – and indeed, we should be starting a similar arrangement with the Comparative Democratization shortly – so please feel free to contact me directly if you are interested.


As editors of The Political Economist, the newsletter of APSA’s Section on Political Economy, we are happy to continue our relationship with the Monkey Cage to make select newsletter content available to readers who are not section members. The current issue of the Political Economist focuses on the political economy of natural disasters, with essays by Daniel Aldrich (“The March 2011 Earthquake, Tsunami, and Nuclear Crisis in Japan: A Political-Economy Perspective”) and Thomas Plümper and Eric Neumayer (“Earthquake Mortality and Damage”). David Victor adds a column on “What to Read on Natural Disasters: Some Insights from the Literature on Natural Resources.” Plümper and Neumayer’s essay is available at the jump. Members of the section may log in to “APSA Connect”: to download the full newsletter.

“Scott Gehlbach”:
“Lisa Martin”:


Earthquake Mortality and Damage

“Thomas Plümper”: and “Eric Neumayer”:

Since the turn of the century, at least four earthquakes have occurred that have entered the collective disaster memory: the March 2011 quake-cum-tsunami in Japan that overtook Hurricane Katrina as the costliest natural disaster ever, the February 2010 quake in Chile for its strength of 8.8 on the Richter scale (and possibly the absence of many fatalities), the December 2004 quake-cum-tsunami in Indonesia (for it was the first major tsunami of the media age and the high number of fatalities), and the January 2010 quake in Haiti (for the vast devastation of a major city and the very large number of fatalities). For lack of extensive film coverage, a fifth disaster is often ignored but it certainly qualifies as major disaster: the May 2008 Sichuan earthquake, which killed almost 90,000 people.

Most earthquakes go largely unnoticed, however. We know of them because we have installed sensitive instruments that capture the tremor emanating from earthquakes and other activity (e.g., the test of a nuclear bomb). These instruments report the occurrence of approximately 50 earthquakes per day, most of them far too weak to cause noticeable damage on the earth’s surface. However, about 250 earthquakes per year reach a magnitude of 6.0 or above on the Richter scale. These earthquakes are dangerous if they take place close to inhabited areas or if they trigger a tsunami. They may kill human beings by making buildings collapse, by triggering landslides and destroying dams, by bursting gas pipes and oil tanks and thus causing fire etc. On average, about 10 earthquakes per year take place that have fatalities, and even more cause significant economic damage. What explains the large variation in disaster fatality and disaster loss?

A naive, apolitical view looks for geological and simple economic determinants. Such a view is correct in that these determinants do matter, but flawed in that they overlook the political economy behind preventing (or failing to prevent) quake mortality and loss. First, and most obvious, the strength of an earthquake matters. An earthquake of magnitude 8.0 unleashes 32 times the energy of an earthquake of magnitude 7.0, 1024 times the energy of an earthquake of magnitude 6.0 and 32,768 times the energy of an earthquake of magnitude 5.0 on the Richter scale. The latter, a magnitude 5.0 quake, dispenses roughly the same amount of energy as the Hiroshima bomb. The March 2011 Tōhoku earthquake off the shore of Sendai in North-East Japan thus unleashed almost 35,000 times the energy of the Hiroshima bomb.

Second, the strength of an earthquake causes more damage if the quake is located directly under the earth’s surface. Accordingly, relatively minor quakes can cause significant damage if their focal depth is low. For example, the February 2011 Christchurch quake struck only 5 kilometers below the surface and though the quake had a magnitude of only 6.3 many buildings collapsed and 181 people died. Similarly, the May 2011 earthquake in the Spanish city of Lorca had a magnitude of only 5.1. Most quakes of this magnitude are hardly felt, but this particular quake had an extremely shallow focal depth of only 1.0 kilometer and thus killed nine people and caused damage estimated at 100 million US dollars. In contrast, the Chilean earthquake of 2010 had a magnitude of 8.8 but a focal depth of 35 kilometers. It killed 562 people while the economic damage has been estimated to lie between 15 and 30 billion US dollars.

Third, the location of an earthquake determines the population density and wealth concentration above the quake and along the fault line. Everything else equal, an earthquake kills more people and causes more economic damage if the ground above and around its epicenter is more densely populated and has more valuable buildings standing on it. Such a high population density in combination with a shallow depth and considerable quake strength seem to be the main determinants of relatively high mortality rates of typical quakes in China.

Fourth, there is also general agreement that a higher per capita income of the region affected by the earthquake directly reduces the mortality rates but increases the economic damage. In relatively wealthy regions, the buildings and infrastructures tend to be better built and thus better suited to withstand the forces of the earthquake. As a consequence, the share of buildings collapsing is smaller in wealthier regions. However, the total value of the destroyed buildings is often considerably higher, simply because of the much larger concentration of physical wealth. Thus, the earthquakes that killed the most people without exception occurred in relatively poor and economically underdeveloped regions. The earthquakes in Tangshan (China 1976), the quake that triggered the tsunami in the Indian Ocean (mainly Indonesia 2004), the Haiti quake (2010), and the quake in Sichuan (a major agricultural province in China 2008) were the four deadliest earthquakes after 1950. Three of the four costliest earthquakes, on the other hand, took place in Sendai (Japan 2011), Kobe (Japan 1995), and Northridge (USA 1994), with only one situated in the developing world, the quake in Sichuan (China 2008).

This naive, apolitical view ignores the fact that while no government can prevent an earthquake, governments could in principle reduce quake mortality to almost zero and they could certainly undertake measures to drastically reduce the economic toll from quakes. Before exploring why some governments blatantly fail to achieve such reductions in mortality and loss, we briefly discuss why private individuals do not prevent large-scale mortality and loss either.

The basic incentives for private individuals (as well as governments) to prevent earthquake mortality and damage seem clear and easy to understand: earthquakes can kill and they can destroy the most valuable property. Private individuals can refrain from settling in areas close to a high-risk fault line or they can construct buildings in a way as to minimize the probability that the buildings collapse when an earthquake strikes. Neither strategy is particularly popular, however. High-risk areas often provide large economic and amenity values to those settling or operating there. The Pacific coast in North America is an economically attractive area for trade and settlement, yet virtually all land that perimeters the Pacific sits on a ring of fire: one of the seismically most active parts of the earth. The second strategy is costly and thus unpopular, too. Earthquake-proof constructions significantly increase building costs. Moreover, it is difficult for private individuals to verify the quality of the building materials and of the construction process with a view to ascertain whether the construction is quake-proof, resulting in a market for lemons due to information asymmetry between buyers and sellers of allegedly earthquake-proof buildings. Mobility worsens the situation. For example, Americans – more so than, say, Europeans – live in houses they did not build for themselves. While it is possible (though costly) to supervise the construction process to guarantee earthquake-proof construction, it is impossible to verify that a finished building has been built to highest construction standards.

Still, in comparison to the opportunity costs of not settling in earthquake-prone areas, the costs of earthquake-proof construction appears minor so that most investors would face strong incentives to make this investment if they knew with certainty that an earthquake would strike their specific location within the foreseeable future. But they do not. The theory of plate tectonics, developed in the early 20th century by Alfred Wegener and others, provides a good understanding of the forces of nature that cause earthquakes. This theory of plate tectonics helps us distinguish between high- and low-risk areas and allows us to come up with some rough understanding of the propensity of certain areas to experience major earthquakes. Unfortunately, however, geologists and other natural scientists remain unable to predict the timing, the exact strength or exact location of earthquakes.

This severe lack of predictability provides an incentive for investors to abstain from making buildings fully earthquake-proof. Decision-psychological experiments have shown that individuals often ignore even potentially large impacts that come with very small probability, unknown size and unknown timing. In many cases, private investors prefer to under-invest in earthquake-proof construction and rather purchase insurance that protects them against the economic losses of a disaster or simply ignore the low-probability high-damage event.

Governments could in principle overcome the information asymmetry problems and the myopic behavior that private individuals face. By implementing and enforcing strict construction regulation standards, they can render private buildings quake-proof. But governments not only influence disaster mortality and loss indirectly by influencing the decision-making of private investors via the setting and enforcing of standards, they also influence the destruction caused by earthquakes directly, as investors. Many buildings and the vast majority of a country’s infrastructure such as roads, ports, airports, power lines etc. are built for public ownership, in full or in part. Governments can directly influence the quality of these constructions.

Unfortunately, governments face very similar incentives as their citizens to under-invest in disaster-proofing public buildings and infrastructures. Since earthquake-proofness is costly and has – with the possible exception of prestige constructions like the Bay Bridge connecting San Francisco and Oakland – typically no immediate effect on political support, governments are better off in the short run if they propel resources to projects which promise to increase short-term political support. In other words, governments are confronted with a time-inconsistency problem in which they may perfectly know that a certain amount of long-term investments in disaster-proofness are in the public interest, but the short-term incentive not to invest in projects with uncertain long-term and negative short-term payoffs may dominate the decision. This incentive to neglect earthquake hazards extends beyond governments’ investment decisions. Governments may also abstain from implementing or, if they do so, from enforcing earthquake-proof construction regulation. Citizens who would opt against earthquake-proofness in the absence of regulation are not very likely to support a government that coerces them to adhere to strict construction regulation.

Yet, some governments appear to invest much more heavily in disaster-proofing public buildings and infrastructure and in passing and enforcing strict building codes than others. In two papers we discuss how variation in the political incentive structure determines the extent to which governments proactively seek to prevent earthquake fatality and damage (Keefer et al. 2011; Plümper et al. 2011). We argue that a strong influence on these incentives stems from the likelihood with which a disaster strikes and its probable magnitude. Such earthquake propensity varies largely across countries, as already mentioned in our brief review of the geology of earthquake prediction. This propensity can be approximately known by governments either via receiving expert advice from geologists or simply by inference from a country’s past history of disaster events. A high earthquake propensity raises the political costs to governments of under-investing in disaster prevention and mitigation. In turn, a low earthquake propensity raises the relative political costs of strict and strictly enforced construction standards as more voters find these measures unnecessary. As a consequence, a quake of the same magnitude will cause many more fatalities and much more damage in low-propensity Spain, where few policy makers have given even a thought to earthquakes, than in high-propensity Japan, where most infrastructure and buildings are quake-proof.

However, the effect of earthquake propensity on governments’ willingness to implement and enforce tight construction standards is conditioned by economic opportunity cost and political institutions. Poor countries often have other more pressing needs on the top of their priority agenda, such that governments in rich countries react more strongly to the incentives stemming from a high quake propensity than governments in poor countries. Governments in autocracies are mainly concerned with the well-being of a small and rich ruling elite that can either well look after itself and usually lives in buildings that can withstand the tremor unleashed by tectonic plates or can be compensated for any damage experienced via private transfers. Democratic governments need to be more accountable to a wider section of the population and thus have a stronger incentive to provide public goods. They will thus react more strongly to higher quake propensity. In addition, governments’ ability and willingness to fight corruption will condition the effect of quake propensity on mortality and damage. In corrupt countries, strict construction standards may merely represent an additional source of income for agencies that ought to enforce these standards, but allow private investors to “buy” their way out of having to obey these standards. Our data analyses provide evidence backing these predictions and thus support a political-economic rather than naïve and apolitical theory of quake mortality and damage.


Keefer, P., E. Neumayer, and T. Plümper (2011), Earthquake Propensity and the Politics of Mortality Prevention, World Development (forthcoming).

Plümper, T., E. Neumayer, and F. Barthel (2011), The Political Economy of Natural Disaster Loss. Working Paper. London School of Economics and Political Science and University of Essex.