Jump to ContentJump to Main Navigation
Dull Disasters?How planning ahead will make a difference$

Daniel J. Clarke and Stefan Dercon

Print publication date: 2016

Print ISBN-13: 9780198785576

Published to Oxford Scholarship Online: June 2016

DOI: 10.1093/acprof:oso/9780198785576.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy).date: 24 November 2017

Defining the Problem

Defining the Problem

Begging bowls and benefactors

(p.13) 2 Defining the Problem
Dull Disasters?

Daniel J. Clarke

Stefan Dercon

Oxford University Press

Abstract and Keywords

The post-disaster relationships between national and subnational governments, government and farmers, government and homeowners, and government and the international humanitarian system often take the form of a begging bowl, although there are notable exceptions. Begging bowls arise because of benefactors—the people who retain discretion over how to allocate their budgets after a disaster strikes. For beneficiaries, begging-bowl financing of disaster risk is fraught with uncertainty—they do not know what help to expect and when help will arrive. It can also undermine their incentives to invest in disaster risk reduction and preparedness. All of these factors can increase the economic and human costs of catastrophes. Even if a budget can be mobilized, ambiguity over how money will be executed can lead to a slow response and poor targeting.

Keywords:   Samaritan’s Dilemma, charity hazard, begging-bowl financing, disaster risk management, disaster risk reduction, disaster risk finance, politics of disaster relief

There is something predictable about the way most of us learn about a natural disaster. It usually starts with a news item on the radio or television or a Tweet or news alert on a mobile phone. Fast-onset disasters, such as earthquakes or floods, always tend to make the headlines. Then reports of the estimated numbers of lives lost and of the damage caused begin to come in. Politicians and senior officials may strut before the cameras to demonstrate their leadership by embracing the three C’s of crisis management: concern about the situation and suffering, commitment to do something about it, and control of the situation. When disasters take place in relatively poorer countries, appeals for aid are quickly broadcast: they take the form of requests for help from local communities or national governments and formal appeals to richer countries for contributions to international agencies or to the public for contributions to non-governmental organizations (NGOs).

So it was for the Nepal earthquake in April 2015.1 Within a few days, extensive reports of damage and loss of life were followed by an appeal by the United Nations for US$415 million to cover the first three months of relief efforts. Across the world media, NGOs appealed for donations to their efforts on the ground. For slow-onset disasters such as a drought or a viral outbreak such as Ebola, it is typically not that much different except that the crisis may not reach the news headlines until it becomes visible, and even then substantial airtime and social-media discussions are devoted to debating whether it really is a disaster.

(p.14) Once a crisis is clearly imminent, expectations of leadership by politicians and international organizations, media attention, and appeals for support are front and centre. What happens next is also predictable: frustration, fallout, and blame games. Support often comes late, and when it finally arrives it is described as ineffectual and insufficient. Those in need expect the authorities to help them, and those delivering the help seem underprepared and underfunded, and they begin to spar publicly. Nepal was no exception. Within days, media reports noted the slow progress of the response, and the political fallout.2

All this makes good news copy, but something more profound is also at play. There is a clear expectation from those in need, as well as those seeing need, that assistance will be given. The moral obligation to help people undergoing hardship because of events clearly beyond their control is well accepted; indeed, even the most diehard opponents of international aid sense that humanitarian support is right and reasonable. This global concern is embedded in international declarations, including the global commitments in UN General Assembly Resolution 46/182, 1991, which offers humanitarian principles. The first of these principles, that of ‘humanity’, states: ‘Human suffering must be addressed wherever it is found. The purpose of humanitarian action is to protect life and health and ensure respect for human beings.’ There is, of course, nothing wrong with this principle—indeed, quite the contrary.

But how is this principle put into practice? Usually not well. Responses to natural disasters need to be timely and on the right scale. For example, how local and national governments and international agencies respond in the immediate aftermath of earthquakes is crucial to protect lives and livelihoods in the short term, but also to restrict the scale of the long-term consequences. An early response in slow-onset disasters is just as critical. As the spread of the Ebola virus in Africa revealed, containing an Ebola outbreak when there are just a few dozen or a few hundred cases—as was the case in Uganda in 2004 and in the Democratic Republic of Congo in 2014—is far less costly and far easier than when thousands of people are infected—which was (p.15) the situation by August 2014 in West Africa when the World Health Organization (WHO) declared the epidemic a Public Health Emergency of International Concern and a major international response was launched. According to data from Kenya and Ethiopia, substantially more lives and millions of dollars can be saved by an early response to a drought compared with a later response.3

Behind these delays is recurring ambiguity about who is in charge beforehand to plan and finance the consequences of these extreme events—that is, who should be taking on the risk and preparing for a response. This ambiguity leads, in turn, to poor plans and poor financing arrangements, despite many statements by all concerned that all human suffering will receive a response.

A Flawed Funding Model

Working backwards from the way responses are financed sheds light on why this process goes wrong. When a disaster strikes, the first point of call is usually the government, at both the local and national levels. Across the world, governments are usually keen to respond, but they often do not have large contingency budgets—that is, budgets that are spent only if something exceptional occurs. Even if a government had such a budget, it may have already been spent. During a disaster-related crisis, cash needs to be mobilized, but either this requires new borrowing (essentially running an overdraft), or a government must reallocate funding from other budgets. This means it is not free money, and it will have a cost. For example, cash may have to be raised by cutting government services such as maintaining roads. In richer countries, borrowing or reallocating funds would be difficult and costly but politically important enough that it would go forward. For poor countries, its consequences would be worse because spending is typically already tight, borrowing is more costly, and reallocations may affect basic government functions. These countries, then, are likely to turn to global humanitarian agencies and development partners for support.

(p.16) The parties making up the humanitarian system also do not have large contingency budgets; they operate with something that resembles a cash budget (spending only the resources that are available). As of December 2015, the humanitarian need (the sum of all global UN appeals) stood at US$20 billion, but only about 50 per cent of that amount was funded.4 Behind this scale of underfunding is a deeply flawed funding model: appeals, linked to a particular crisis. All the main international agencies and NGOs involved are largely funded through appeals that solicit voluntary contributions from governments and the general public. The size of contributions is not pre-specified; in fact, no one is obliged to contribute. In the same way, many poorer countries appeal directly to the international community for support after disasters—an approach that leads to very uneven funding and recurring large funding gaps.

This ad-hoc, post-disaster model for financing disasters is hardly worthy of the twenty-first century. In fact, it feels distinctly medieval. It is a funding model based on begging bowls, whereby individuals, communities, local and national governments, international agencies, and NGOs are required to play the part of a beggar, as though they are pleading for alms, sitting in a row in front of a medieval cathedral or mosque. Benefactors may well be committed to digging in their pockets to share their coins with those who clearly are facing hardship, but their coins may run out midway through the row. Without more information, the benefactors do not know which beggars are the neediest, and so the neediest might not receive the most benefits. In short, begging has hardly ever been a stable source of resources to deal with the vagaries of life, not least when time is of the essence and benefactors need to make very quick decisions based on limited information.

The Consequences

The consequences of a financial model that encourages reasonable people and organizations to play the part of a beggar after each disaster can be dire.

(p.17) Ambiguities

The model creates ambiguity about who owns the risk: who will need to act and who will need to pay for it. Across the world, national and local governments and politicians will say with confidence that in the first instance they do. As for developing countries, the various agencies of the United Nations, such as the United Nations High Commissioner for Refugees (UNHCR), World Food Programme (WFP), World Health Organization (WHO), United Nations Children’s Fund (UNICEF), and United Nations Office of the Coordinator of Humanitarian Affairs (OCHA), as well as various big donors, will all claim they are there to work with governments so that the appropriate responses and financing materialize. Local and international NGOs and the International Federation of Red Cross and Red Crescent Societies (IFRC) will also claim to be responsible for responding to crises.

All this is reasonable, and in fact governments should have key responsibilities. But across all organizations, such statements are hollow because there is no guaranteed financing for fulfilling promises. Credibility requires a finance model that will deliver the funds when needed. Within countries, the lack of clarity between national and local governments over who should respond and who should finance the response leads to ineffectual relief efforts. Evaluations of the response to Hurricane Katrina in 2005 in the United States brought this home.5

The Ebola outbreak in West Africa in 2014 was another example of a situation in which a lack of clarity on who owned the risk and response was at the core of long delays. Most observers expected the national governments to own the risk, with WHO offering a second layer of risk protection in terms of expertise and finance if governments could not handle it. Indeed, this is how previous outbreaks of Ebola had been managed. This time, however, national governments were unable to contain the epidemic, and WHO did not have the resources to act at scale on this widespread expectation. Large bilateral donors and their agencies were not prepared to respond quickly once (p.18) this became clear because they did not realize they were carrying the risk, and therefore they did not have the plans and preparation in place to scale up to deliver on the response quickly. The result: valuable time was lost, and, as evaluation reports have shown, the long period of inaction led to more misery and higher costs.6 One study suggested that a response one month earlier could have averted more than half the cases in Sierra Leone.7

Procrastination and Delays

The lack of pre-commitment embedded in begging-bowl financing also leads to one of the enemies of effective decision making: procrastination among implementers as well as donors. Procrastination, the decision to delay or postpone something, is a well-studied phenomenon affecting the decision making of individuals or groups. According to the research, when they are faced with a decision on whether to act but the choice is not obvious, or when the action required is demanding and difficult, inaction is the common response.8 In many recent cases, including slow-onset disasters such as droughts and pandemics, this has been a real problem. Responding to a possible large-scale disaster early is harder than taking no action at all, especially because funding needs to be found at the right scale. The result is a tendency to procrastinate over acting and committing funding. This phenomenon was present in the national and international decision making around the Ebola response in West Africa, as well as in the responses to the 2010–11 drought in Somalia and the neighbouring regions9 and in the recent responses to likely extreme weather events linked to El Niño. Crises tend to be rather well developed before decisions are made to respond.

Crying Wolf

Actually, the consequences of begging-bowl financing can be worse. A funding model based on voluntary contributions and appeals does not just risk underfunding some causes, thereby leading to delays and more suffering. It also creates serious distortions and bad incentives (p.19) that make poor responses more likely. Because underfunding is common and little pre-committed funding is available, strong incentives surface among the implementing agencies to exaggerate crises and appeals. In June 2011, a UN press release suggested and was interpreted by the international media as meaning that in East Africa the worst drought in sixty years was under way.10 The truth, however, was that it was the worst drought in a relatively small number of specific pastoralist areas in Somalia, Kenya, and Ethiopia. No doubt, a bad drought was taking place, but the press release was somewhat parsimonious with the truth. Raising false alarms, ‘crying wolf’ as in Aesop’s famous fable, is a big risk here: because potential donors are aware of the incentives for overstatement, many appeals will hardly receive a response. As it turned out, this particular crisis did end up in a massive human disaster because the drought coincided with the raging conflict in Somalia, and responses were late and ineffective.11 It was not evident, however, that more finance would have avoided the crisis.

Such an exaggeration for effect is not uncommon, and rarely does one hear of retractions when the media overstate risks. For example, some agencies’ declarations and media reports that West Africa was at risk of famine during the Ebola outbreak in 2014 did not quite have a truthful ring to them based on the evidence available. If the information to be used for decision making is manipulated to overstate the need, it is difficult for those contributing to the system to make sensible trade-offs over where and when to contribute.

Fragmented Responses

The funding model also needs political or other leaders to be seen as doing the right thing. And when the cameras are rolling and after the crisis has already unfolded, this is the right time to fill the begging bowl and offer its proceeds to those in need. But for many disasters, this is too late. That said, let us be clear: there is nothing wrong with this media attention, and there is nothing wrong with opportunities for politicians and other leaders to show leadership. Indeed, there is (p.20) much research to suggest that political leadership ensures that natural hazards do not turn into disasters, and its absence can cause serious failures—such as in the case of Hurricane Katrina in 2005, in Haiti in 2010, and in the late responses to the Ebola virus in West Africa in 2014. An active press is also an effective mechanism for accountability and better responses. Nobel laureate Amartya Sen argued long ago that media attention ensured the disappearance of famines in India after independence.12 Subsequent research revealed how the press and better competitive elections made this true for the quality of response across India.13

Meanwhile, another consequence of lack of pre-commitment of financing to plans and implementers is the incentives it creates for flag planting and limelight hugging. ‘Everyone loves a good drought’, as the Indian journalist Palagummi Sainath famously wrote.14 A humanitarian crisis allows the flags of nations and organizations to be planted, showcasing their generosity and success. Internationally and locally, a crisis allows politicians to show leadership and seek to portray themselves as the saviours of those experiencing hardship. Everyone wants to hug the limelight and show off their effectiveness. But that kind of grandstanding only exacerbates the incentives for fragmentation rather than coordination: going it alone makes it easier to claim to be at the centre of the response, whether as a funder or an implementer. And it creates a class of benefactors: leaders in receiving and donor countries and in local and international organizations who can choose which begging bowl to fill or what to do with their available cash. Planning beforehand would have seemed futile since financing only follows a disaster. Therefore, these benefactors will play a direct role in deciding what the response will look like and how the subsequent recovery will roll out.

With this funding model, no organization can guarantee before the onset of disasters that it can offer the right response. Because funding is not secure, the overall picture is one of incentives for extensive fragmentation when crises occur rather than for coordination. Each organization may have plans, but without guaranteed funding of their (p.21) plans none can act. The result is a dispersed, unwieldy system with much multiplication rather than economies of scale. Those who can mobilize resources respond, and so whenever a disaster develops, dozens of national aid agencies, international organizations, and NGOs all get involved. But each must be sustained, and each needs resources. Local governments, national leaders, international donors, and multilateral agencies also vie for their roles: all claim to recognize the importance of coordination, and yet none of them wants to be coordinated.

Disaster Risk Reduction and Preparedness

A funding model based on appeals and other begging bowls hardly encourages national or international organizations to undertake serious investments in reducing the risk of a disaster and preparedness. After all, if an organization has no idea how much funding might be available, how can it plan for the appropriate scale of response? Without a secure budget, how can it invest before a disaster in a post-disaster implementation capacity?

But it gets worse. Benefactors are anointed in full view, not from quiet graft in the shadows. A politician probably will not get credit for pushing good preparedness plans and investments. Plus, such planning demands more effort for limited immediate gain. Only lip service will be paid to investments in early warning systems, preventive infrastructure investments, building codes and zoning, and education and technical assistance in preparedness. Why should a politician invest in a sensible system to reduce risks and enable a quick response to a strong earthquake if the political benefits from such a system are likely to be reaped by that politician’s political successor? These facts of political life tend to lead to procrastination in setting up good response systems beforehand and in delays in making firm decisions about how to respond under various circumstances—after all, decision makers are under little pressure and the rewards are scarce. And risk-reduction investments will lose out.

(p.22) There is good evidence that electoral politics tends to encourage these behaviours: political leaders are rewarded by the voters for offering disaster relief, while disaster preparedness has no impact on election outcomes. For example, in India incumbent parties are apparently rewarded if they vigorously respond to disastrous weather events, but only in election years. Relief is much higher, then, in such years.15 These electoral behaviours are also observed in the United States. Presidential disaster declarations in affected states are rewarded by the electorate, but a president is punished if a governor’s request for relief is denied.16 These behaviours skew incentives in the choice between relief and preparedness: voters reward the delivery of disaster relief, but not investments in disaster preparedness.17 Improving disaster responses and preparedness will require balancing these incentives.

Moving Beyond a Medieval System of Finance

So, how can countries and their partners do better in handling disasters? They surely can go beyond a medieval system of finance that ignores centuries of progress in developing insurance and other financial protection instruments. Modern financial principles can ensure certainty in finance in a world of uncertainty.

But decision makers do not necessarily have to go to the City of London or to Wall Street for ideas and advice. Across the world, communities have long experimented with finding ways of protecting themselves against disasters. For example, slum dwellers in Dhaka, the capital of Bangladesh, can teach them a few things. The buildings in the slums are built of woven bamboo, and food is cooked inside over open fires. Highly combustible, the dwellings often catch fire, destroying dozens of homes and shops in one go. There is no fire insurance for such informally planned settlements or public compensation after fires. The solution? Residents have set up groups of a few dozen or more members. Each week, a cashier collects a fixed amount of money (p.23) from each member and banks it. In the event of a fire, the money is withdrawn and distributed to the members in proportion to their contribution. This is an extremely simple form of a reserve or contingency fund, pooled to ensure that it is only used for the designated purpose: paying for fire losses. When a fire occurs, the cash for compensation is readily available and disbursed using clear and simple rules.18

Poor people’s funeral insurance systems across the world also offer up some lessons. Among the fisherfolk in Cochin in Kerala, India, early death is all too common. However, funerals can be very expensive, and cash is not readily available when someone dies. Members of the annual burial fund pay in a fixed sum each week for a year. If the member or a close relative dies, a payout is offered from the fund, no strings attached. At the end of the year, if any funds remain they are distributed to the members. If the fund runs out of cash during the year, all members are asked to make up the shortfalls in equal shares.19

The funeral group operated by Muungano, the village union of women’s groups in the small village of Nyakatoke in Western Tanzania, provides insurance that supplements what the traditional village Bujuni (or ‘mutual help’) association offers at the time of mourning. The additional protection involves only a small contribution beforehand, but when someone dies a strictly enforced commitment goes into effect, and a particular sum of cash is paid by each member to the deceased family, plus a fixed contribution in kind or in labour.20 In recent years, these groups have expanded their coverage and now also offer a payout when someone needs to be hospitalized.

Ethiopia has its own version of these funeral societies, usually called iddirs (see Chapter 4). They are based on a simple model, similar to the one found in Cochin.

All these schemes are group-based versions of insurance, with fixed regular contributions and fixed payouts when a fire, death, or serious illness occurs. Such groups are at the root of many of the largest insurance companies in the world. They operate as a mutual (p.24) fund, a ‘risk pool’ of savings, which is used to pay for pre-specified risks.21

These examples from around the world point to the kind of finance model for disaster responses and recovery that would overcome many of the disincentives and failures of the current system of voluntary contributions after disasters have unfolded. It uses insurance principles to finance need and states clearly exactly who holds the risk: for example, the funeral society’s fund based on members’ contributions pays for the costs of funerals and possibly other pre-specified risks. Contributions are collected beforehand and predictable—no begging bowls required, or at least liabilities are well defined. The ways in which to raise extra emergency cash are specified. There is no ambiguity, no gaming, no procrastination.

These, then, are the elements of a sensible model to deal with the response to calamities internationally and within countries. The system is based on clear-cut decision rules, and there is no ambiguity about who owns the risk, who needs to respond, and how it is financed. The incentives to prepare a response and to reduce risks are in place. And, finally, a financing model offers resources when they are needed, turning risk and uncertainty into the certainty of support. No more begging bowls or benefactors.

This is a big ask for political leaders for whom discretion is the default setting. But in the chapters that follow, we outline how this can be done. Planners will begin by putting together and packaging politically sellable alternatives to begging-bowl financing, including thinking through who will be protected and against what, who will pay, and what the conditions for protection will be (Chapter 3). Then, in moving from discretion to rules, they will think carefully about what data will trigger action, and how it will be collected and protected from fraud and political opportunism (Chapter 4). Finally, planners will help benefactors to pre-commit their funds before disasters strike in ways that encourage coordination and proper incentives for risk reduction (Chapter 5).

(p.25) Recapping…

  1. 1. The post-disaster relationships between national and subnational governments, governments and farmers, governments and homeowners, and governments and the international humanitarian system often take the form of a begging bowl, although there are notable exceptions.

  2. 2. Begging bowls arise because of benefactors—the people who retain discretion over how to allocate their budgets after a disaster strikes.

  3. 3. For beneficiaries, begging-bowl financing of disaster risk is fraught with uncertainty—they do not know what help to expect and when help will arrive. It can also undermine their incentives to invest in disaster risk reduction and preparedness. All of these factors can increase the economic and human costs of catastrophes.

A Snapshot of the Literature

The effect of an altruistic benefactor on the risk-reduction investments made by a vulnerable beneficiary has been considered in depth in economic research. Situations in which the benefactor can provide post-disaster relief but cannot condition that relief on the pre-disaster behaviour of the beneficiary are a special case of a much more general situation known as hidden action, or moral hazard (Arrow 1971; Hölmstrom 1979). In the context of disaster risk, the beneficiary has fewer incentives to invest in self-protection, such as strengthening buildings against natural hazards such as earthquakes, limiting development to low-risk areas, shifting to economic activities that are more resilient, or purchasing insurance themselves (Kaplow 1991). This finding is sometimes referred to as a type of Samaritan’s Dilemma (Buchanan 1975; Lindbeck and Weibull 1988; Coate 1995) or more recently as a Charity Hazard (Browne and Hoyt 2000; Raschky and (p.26) Weck-Hannemann 2007). Cohen and Werker (2008) offer a political-economy version, applying it to case studies of disaster relief.

A large empirical literature confirms this prediction more rigorously for both the general case and national disaster-relief programs. For example, Kousky et al. (2013) consider floods in the United States and estimate that for every US$1 increase in average relief investment, self-protection in the form of insurance decreases by approximately US$6. Van Asseldonk et al. (2002) found that Dutch farmers who believed government assistance would be available in the event of a disaster had a significantly lower demand for crop insurance.

Meanwhile, political scientists have offered a variety of explanations for underspending on disaster preparedness and overspending on disaster relief by national governments. First, voters may show a preference for private goods, such as direct relief payments, over public goods, such as investments in early warning systems or better building codes and zoning. Second, voters may simply be less aware of the benefits of disaster preparedness expenditures relative to disaster relief expenditures because of their lower relevance to them and the reduced media coverage. Third, voters may make attribution errors by attributing investments in preparedness expenditures to future government administrations, whereas they are more likely to attribute relief expenditures to the current administration. This attribution error reduces politicians’ incentives to make long-term investments in risk reduction. Fourth, voters may find it difficult to quantify the benefits from investments in risk reduction, and in particular to construct counterfactuals for how much larger the loss would have been after a disaster without risk-reduction investments. Without such a comparison, they may underappreciate disaster preparedness expenditures. Finally, voters may be shortsighted and underappreciate long-term investments in risk reduction that do not yield large short-term payoffs. Healy and Malhotra (2009) analysed these five potential mechanisms using a data set on natural disasters, U.S. government spending, and election returns, and found that the first mechanism, the desire for individually targeted goods over public goods, went a (p.27) long way towards providing a useful explanation for the behaviour of US politicians. They also estimated that federal investments in preparedness are extremely cost-effective, with every US$1 spent on preparedness leading to a reduction in damage of approximately $15.

Beyond Healy and Malhotra (2009), there is a wide range of evidence that disaster relief buys votes. For example, using data on rainfall, public relief spending, and elections from India, Cole et al. (2012) found that voters punish the incumbent party for weather events beyond its control, but they punish the incumbent party less when it vigorously responds to an event. This effect, however, is limited to election years. Correspondingly, the authors find that the government spends more on relief in election years. Reeves (2011) demonstrates that in the United States between 1981 and 2004 a single presidential disaster declaration to specific constituencies translated on average into a one-point increase in votes for the presidential party in a state-wide contest. Furthermore, when the president rejects a governor’s request for federal assistance, the president is punished in voting, whereas the governor is rewarded (Gasper and Reeves 2011). Fuchs and Rodriguez-Chamussy (2014) analysed the impact of insurance payouts on voter behaviour in the 2006 Mexican presidential election. In this study, the incumbent party was estimated to have garnered 8 per cent more votes where payments were made prior to the election. Eisensee and Strömberg (2007) explain US disaster relief payments by media coverage, showing that when a disaster occurs simultaneously with other newsworthy events such as the Olympic Games, post-disaster aid is reduced because media coverage of the disaster is crowded out by the other events.

Although there are probably a large variety of reasons behind delays in the provision of disaster relief, one reason may be the strategic behaviour of benefactors. It is well understood that strategic situations in which multiple parties could voluntarily contribute to achieving a common goal—in this case the provision of relief to disaster victims by multiple potential benefactors—often lead to the observance of strategic delays (Osborne and Rubinstein 1990). Some (p.28) convincing evidence of the cost of such delays has been uncovered. For example, looking at extreme droughts, Alderman et al. (2003) investigated the impact of preschool malnutrition on subsequent human-capital formation in rural Zimbabwe. The authors estimated that reduced nutrition in children under 2 may lead to a loss of 14 per cent of lifetime earnings. Using Ethiopian data, Dercon (2004) estimated the effects of rainfall on consumption growth. He found that, as a result of reduced consumption and increased distress sales, household income at the end of the nine-year study period was 16 per cent lower than that of households that had not suffered to the same degree. Clarke and Hill (2013) used these two studies to estimate that responding early in an extreme slow-onset drought would have been approximately three times more cost-effective than responding late.


(1.) The BBC website <http://www.bbc.com> stated on 26 April 2015 that ‘a major earthquake has struck Nepal’, ‘hundreds of people are feared dead’, and ‘the government has declared a state of emergency in the affected areas’. (p.115) Soon the first appeals for assistance were heard. ‘We need support from the various international agencies which are more knowledgeable and equipped to handle the kind of emergency we face now’, said the Nepali information minister. Time magazine ran an article on 27 April entitled ‘6 Ways You Can Give to Nepal Earthquake Relief’, giving details on six international charities. By 30 April, the BBC website was reporting more detailed estimates of damage and loss of lives: ‘Officials say Saturday’s quake killed more than 5,500 people, and injured at least 11,000. The UN says more than eight million people have been affected by Saturday’s 7.8-magnitude quake and some 70,000 houses have been destroyed.’ And the main appeal was reported: ‘The UN has appealed for $415m (£270m) to help provide emergency relief over the next three months.’

(2.) On the 26 April, the BBC website reported that ‘the [Nepali] government says it has been overwhelmed by the disaster’. By 28 April, the first fallout had begun: Time magazine reported under the headline ‘Why Nepal Wasn’t Ready for the Earthquake’. Documenting poor regulation and poor politics, it stated, ‘People have been trying for a long time to improve preparedness and resilience, but they’re resource-strapped.’ By 30 April, the BBC website was reporting that ‘survivors in some areas told the BBC that they were angry that neither food nor medicine has reached them’.

(3.) Clarke and Hill (2013); Cabot Venton et al. (2012).

(4.) Data from the Financial Tracking Service of the UN’s Office for Coordination of Humanitarian Affairs (OCHA) suggested that in December 2015 there were twenty-four live appeals for Humanitarian Response Plans, as well as six Refugee Response Plans and a number of flash appeals. The total appeals were for just under US$20 billion, of which just 50 per cent were either paid or committed via pledges by donors. Several of the largest involved conflict-related crises (such as Syria or the Democratic Republic of Congo), but large appeals also included the Nepal earthquake ($420 million, of which $150 million has been unmet), and more than $1.4 billion in total for Burkina Faso, Mali, Chad, and Niger, much of it related to drought, floods, and other extreme events. Of that $1.4 billion, $850 million was unmet.

(5.) See Congressional Research Service (2006) or Department of Homeland Security (2006).

(6.) WHO (2015).

(7.) Kucharski et al. (2015).

(8.) Procrastination could be related to different cognitive biases commonly found in people (Samuelson and Zeckhauser 1988). Ritov and Baron (1992) distinguish between status quo bias (explained by loss aversion) and (p.116) omission bias (explained by a failure to act, which arises through a failure to see the possibility of action). See <https://fts.unocha.org/reports/daily/ocha_R21_Y2015_asof___4_December_2015_(02_31).pdf>. O’Donoghue and Rabin (1999) discuss explanations linked to self-control problems.

(9.) Bailey (2013).

(10.) For example, on 28 June 2011, Reuters reported, ‘Worst drought in 60 years hitting Horn of Africa: U.N.’. The UK’s Channel 4 news website stated, ‘East African drought worst in 60 years’. On 22 October 2011, the Independent in the UK wrote, ‘Africa’s worst drought in 60 years threatens famine…’.

(11.) Bailey (2013).

(12.) See Sen (1983) for an early articulation or Sen (2009) for a recent discussion. Sen suggested that the 1958–61 famine during the Great Leap Forward in China happened because news was repressed, and so news of the famine did not even reach the top party leadership. India has been able to avoid famines since independence because a free press and the pressures of elections have meant that politicians have been held accountable for responding to disasters.

(13.) Besley and Burgess (2002).

(14.) Sainath (2000).

(15.) Cole et al. (2012).

(16.) Gasper and Reeves (2011); Reeves (2011).

(17.) Healy and Malhotra (2009).

(18.) Rutherford (2001).

(19.) Rutherford (2001).

(20.) Dercon et al. (2006).

(21.) Historically, fire insurance was also the first commercial product targeting households to emerge. A few years after the catastrophic Great Fire of London in 1666, the Insurance Office for Houses, probably the first fire-insurance company, was founded. It was set up as a mutual fund, pooling the contributions of its members, with payouts further guaranteed by a property-investment portfolio. Recognizing the benefits of preparedness, this company and its fast-emerging competitors ended up setting up the first fire brigades, offering protection to the properties they insured and only those.