A natural disaster is regarded as the quintessential “act of God.” When a drought or a tsunami strikes a poor country and images of disaster victims are broadcast around the world, individuals, nonprofits, and governments elsewhere respond with generous relief. This picture seems to exemplify mankind at his finest, a rare field in which humanitarian needs trump political concerns. For the most part, this conception is true. Responding to a natural disaster is a rare locus of engagement for countries like the United States in countries like Burma or North Korea. 

Yet the notions that disasters are natural, and that disaster relief is apolitical, are ultimately flawed. Even though innocent victims are involved, the laws of political economy unfortunately apply. Those laws dictate that -- like individuals -- countries look out for their own self-interest, and governments and non-state actors respond to incentives. In the case of natural disasters, this means that disasters are sometimes more severe than they otherwise would be, and that disaster relief follows the incentives of the donors just as much as it does the needs of the beneficiaries.

How is it that natural disasters are, in fact, unnatural? Consider the recent earthquakes in Haiti and Chile. The earthquake that struck Haiti in January 2010 measured a bone-jarring 7.0 on the Richter scale. Over 250,000 Haitians were killed, with 1.5 million further losing their homes. Nature seemed to hold a particular vengeance against the small country, adding a horrendous earthquake to a country that in previous years had already been subjected to hurricanes, mudslides, floods, and droughts. Meanwhile, in Chile just over a month later, an earthquake measuring 8.9 on the Richter scale literally moved Chile's second-largest city, Concepcion, 10 feet to the west, according to GPS data. Though just separated by two points on the Richter scale from the Haitian quake, due to the logarithmic measurement the Chilean quake was 64 times stronger, yet fewer than 500 people were killed.

Natural hazards only become disasters when the particular patterns of human settlement leave a population vulnerable to the whims of nature. In Chile, a competent government enforced well-known building codes and ran rehearsed disaster response drills. In Haiti, self-reinforcing patterns of poverty, natural resource over-exploitation, and limited social cooperation meant the impact from any natural shock was amplified when it struck.

Meanwhile, any form of international engagement between rich countries and poor countries is subject to pulls from multiple forces. Not every foreign engagement is seen as a potential battle in the so-called war on terror, yet basic geopolitical forces are still at work whenever a dollar is dispersed through official channels. Even the large minority of disaster relief that comes from not-for-profits is not always spent on the victims who are most in need. Nongovernmental organizations (NGOs) cannot simply print their money, and their aid must go to those places where the agency's future funds can be raised. These are the politics of disaster relief: aid driven by donor priorities to calamities that shouldn't have occurred in the first place.

The role of disaster relief

According to research by Swedish economist David Strömberg, around US$5 billion is spent on natural disaster relief per year, whereas disasters inflict economic damage on the order of US$40 billion per year. Much of that economic damage occurs in rich countries, and much of that is privately insured. The bulk of aid flows from richer countries to poorer countries, and is naturally redistributive. While the US$5 billion figure may seem impressive, it must be put in proper perspective. It represents less than 0.02% of the combined GDP of the European Union, the United States, and Japan, and is smaller than the vending machine industry in the United States.

Unlike much foreign aid that is provided as a transfer between governments (or to support the functions of the recipient government), disaster relief is often provided in kind. This means that donors bring in actual goods, like blankets and shelter, or services, like helicopter transport. More often, donor governments support in-kind relief provided by international organizations like the International Committee of the Red Cross and the United Nations' World Food Programme (WFP), or contract specific relief tasks to international NGOs like Catholic Relief Services and CARE. These NGOs and others also raise funds privately, and the resources that they bring to bear on disaster relief are substantial, though dwarfed by official funds.

Disaster relief is often free from the political restrictions that more development-oriented foreign aid faces, and is one of the few acceptable channels of aid to pariah states. Countries like Burma and North Korea, otherwise cut off from the world, received generous offers of support from countries that would otherwise have nothing to do with them after being struck with a cyclone and a famine.

Disaster relief, then, functions as a sort of insurance to people who live in the places in which private insurance is either unavailable or unaffordable, and government mitigation and assistance is lacking. It relies on voluntary mechanisms, even as those mechanisms -- such as the web of government-sponsored disaster professionals -- behave in relatively straightforward and predictable fashion.

The politics of relief allocation

If disaster relief were like insurance, then it would be allocated according to who had paid their premiums, and what coverage they had purchased. If it were run completely in the interests of potential victims, then it would go to those most in need and worst hit -- assuming that the victims were reachable without extraordinary expense. But the providers of relief are bound neither to a contract nor to a specific population of beneficiaries.

Governments, when they are conducting international activities, act in the interest of their taxpayers rather than in the interests of the other nation. (Of course, sometimes those interests coincide, as they usually do in disaster relief.) In addition, government agencies that want to survive another budgetary cycle find themselves catering to a complex web of other agencies, branches of the government, and voters. NGOs, while doing their best to meet the needs of their beneficiaries, must also meet their fund-raising requirements.

Just what is the national interest of a donor country like the United States when a natural disaster has occurred? A brief look at the political economy of foreign aid more generally may be instructive. Harvard economist Alberto Alesina, writing with the World Bank's David Dollar, examined the determinants of bilateral aid flows. They found that political variables explained “a large, but not exclusive extent” of cross-country differences. In particular, measures of former colonial status, voting in the United Nations, and being Israel or Egypt were far better predictors of foreign aid receipt than income per capita, trade policy, or democratization.

Princeton's Ilyana Kuziemko and I examined aid patterns to countries serving as non-permanent members of the UN Security Council. We noticed that when countries rotated onto the Security Council, they experienced a huge temporary boost in foreign aid, which in all likelihood was a play for their support. Far from being just a tool to promote economic development in poor countries, foreign aid is quite clearly an instrument of statecraft.

Can the same cold-hearted calculus apply when lives are at stake from a drought or a flood? Harvard Public Health economist Guenther Fink and World Bank economist Silvia Redaelli looked at emergency aid provided in response to rapid-onset disasters. Not surprisingly, they found that disaster-stricken countries received more aid when the disasters were more severe and affected more people. However, they found that two countries with different political characteristics experiencing the same intensity of disaster could receive very different amounts of international assistance. Donors were more generous to oil exporters than to countries without oil, and more generous to former colonies than to countries with whom they did not have a colonial presence. Former imperial powers were 25 to 30 percent more likely to respond to a disaster if it occurred in a former colony.

Interestingly, they found that donors were more likely to respond to a disaster in countries that were not traditional allies, measured by voting similarity at the United Nations' General Assembly. As with the American engagement with North Korea following their famine (or Venezuela’s Chavez offering heating oil assistance to needy New Englanders), a natural disaster could provide an opportunity for countries experiencing chilled relations to engage with one another, or to bypass unsympathetic governments and appeal directly to their people through aid.

The politics of disaster aid are not limited to geopolitics. After all, most disaster relief is simply to help disaster victims, which also happens to be a very easy way to earn goodwill internationally. But goodwill can also be earned domestically. When people sitting at home in rich countries see images of a disaster broadcast in their living rooms, they want their governments to be among those assisting the victims. It would have been unimaginable for the government of any major economy to sit back following the 2004 tsunami in the Indian Ocean while other nations stepped forward with doctors and supplies. Yet this introduces a political bias of its own: relief providers are more incentivized to respond to disasters that make news headlines.

David Strömberg and co-author Thomas Eisenee looked at media coverage and natural disaster relief. Not surprisingly, disasters that were "in the news" were more likely to receive relief than those that were not, even after controlling for the intensity of the disaster. Of course, this might simply be due to the fact that reporters covered more severe disasters, utilizing subjective information that might not have been captured by the quantitative indicators of disaster severity. To test whether news coverage itself brought about more relief, Strömberg and Eisenee scrutinized other news events at the time of a natural disaster. They found that when a major story, like the Olympics or the O.J. Simpson trial, was cluttering the headlines, natural disasters received less relief than when no other newsworthy events were occurring.

This effect is present for NGOs even more than it is for national governments. The so-called CNN effect brings NGOs to newsworthy disasters, as they help the NGOs connect with their donors and raise additional funds. But the CNN effect can result in a misallocation of aid. When well-meaning individuals contribute to a salient disaster, sometimes they leave the NGO with little choice but to try to shift the funds towards where it is needed more. The American Red Cross, for example, faced public scandal when it diverted funds raised after September 11, 2001 to underfunded emergencies as well as preventative expenditures.

Of ownership, agency, and the Samaritan's dilemma

Because disaster relief is so political, there is often controversy about who gets credit for it. India, for example, initially refused disaster relief after the 2004 tsunami. By receiving outside assistance, the logic went, India's government could look weak to its population. This accords with research I have done with my Harvard colleague Sean Cole and Andrew Healy at Loyola Marymount. We looked at disaster relief from Indian state governments that occurred following droughts and floods. As it turned out, voters rewarded governments at the polls when they provided more relief. They were using the government's response to disasters as a way to judge their competence.

It should be no surprise, then, that when the Burmese or North Korean governments received disaster relief, they held out during tense negotiations to get as much control over the distribution of relief as possible. If bags of grain showed up deep in their hinterland marked with the US government's "a gift from the American people," officials in these despotic regimes were worried that they would lose legitimacy in the eyes of their population.

However, there are many countries around the world that do not purport to have the capacity to respond to natural disasters. In such places, governments are often judged not by the assistance that they provide themselves, but rather by the quantity of external resources they can marshal. When this is the case, foreign disaster assistance provides a very perverse set of incentives.

Consider the perspective of the donor. Whether driven by humanitarian impulse, the desires of home-country constituents, or even by the possibility to achieve a geopolitical goal, the international community will never find itself in the position to deny disaster relief to an afflicted country. This means that poor countries essentially receive "free" insurance for natural disasters. As with any moral hazard problem, subsidizing the cure reduces the incentive to invest in prevention. Why should a poor country, with so many needs, invest in disaster prevention when the rich world has more or less agreed to bail it out in an emergency. 

Ethiopia and northern Kenya are two areas that make regular appearances on the famine warning map, even though each has suffered these patterns for decades, and received humanitarian relief consistently. By continually providing disaster relief, the international community reduces the incentive for governments to come up with lasting solutions -- whether irrigation, market development, or private insurance -- and the vulnerable populations remain vulnerable through the next year when, hopefully, disaster relief can be procured again. This leaves the donor facing a "Samaritan's dilemma," that the act of providing aid may lead to a dependency on the aid itself.

At the extreme, cash-starved governments or rebel movements in resource-barren landscapes can enjoy disaster relief as one of the only sources of hard income. During the humanitarian crisis that occurred as a result of the 2001 war in Afghanistan, the Taliban imposed a tax of thirty-two dollars per metric ton brought in by UN WFP convoys. Somali warlords regularly taxed relief supplies and relief agencies.

The net effect of this, of course, is to reduce investments in disaster prevention and mitigation, such as building dikes, irrigation, enforcing building codes, creating early warning systems, and -- in the extreme cases listed above -- simply allowing food markets to function. These are the very types of investments that prevent natural shocks from becoming disasters. Relief agencies have begun to recognize this and have shifted some of their attention from disaster relief to disaster prevention. But the political economy of disaster relief makes this an uphill battle.

What is to be done?

The goal of the disaster relief industry should not be to simply respond vigorously to every natural disaster. Such a strategy would ignore the potentially higher gains that can be achieved from investing in the much less glamorous world of disaster prevention. At the same time, the goal should not be to invest so much in systems of prevention and mitigation that disasters never occur. After all, insurance exists for a reason, and sometimes it is more cost effective to provide millions of dollars in relief for rare events than it is to spend billions of dollars to make sure that they never occur. 

The ad-hoc system of international disaster relief is a relatively effective -- and progressive -- way of providing the insurance. Ironically, part of its financial sustainability is driven by the very political biases in its allocation. Disaster relief funds are much easier to raise when they provide not only assistance to the innocent victims of disaster, but also political benefits to the rich-country governments and warm, fuzzy feelings to individual donors. If this ends up allocating more money to some disasters over others, this may be regarded as a necessary distortion to keep the larger picture operating smoothly.

But just because the system works pretty well does not mean that there is not room for improvement. Donors should recognize their own biases as well as the incentive on potential beneficiaries that their biases create. Institutional donors should commit a large fraction of their expenditures on disaster prevention, or at the very least lobby development institutions like the World Bank to fund preventative activities. When journalists report from a natural disaster, they should not ask, "what are we doing to help?" but instead, "how did this happen in the first place?"

When there is a major disaster in the news, you as an individual should certainly reach for your pocketbook. But you should not contribute to the "tsunami" of the day; instead, you should search out the most forgotten, far-flung disaster you can find and donate to its relief activities. Your dollar will surely go farther.