BRIEFING PAPER DISASTER RISK REDUCTION Spending where it should count AUTHORS: Jan Kellett & Dan Sparks WORKSTREAM: Global trends DATE: March 2012 VERSION: 1 INTRODUCTION The humanitarian system is under considerable strain. Needs are increasing, and commodity prices remain at near-record highs. There is pressure on donors either to reduce their humanitarian expenditures or at the very least, more than before, to justify the value in each dollar spent. This is in a context of mega-disasters on an almost unheard of scale and continual expenditures of vast sums in complex emergencies. There are considerable concerns about whether or not the current trend of year on year increases in humanitarian funding can be sustained. Disaster risk reduction (DRR) is seen by many as a means not only to reduce this continued pressure on humanitarian expenditures but also to protect development investments made by both the international community and national governments – and, of course, to reduce the effects that disasters have on families, communities and countries. These effects, and the scale of the damage caused by natural disasters, cannot be underestimated. Over the ten years from 2000 to 2009, more than 2.2 billion people worldwide were affected by 4,484 natural disasters. These disasters killed close to 840,000 people and cost at least US$891 billion in economic damage. In this report we examine the top 40 humanitarian recipient countries in the context of natural disasters and especially with regard to financing to reduce risk. We highlight how prevalent disasters are in these countries, and their particularly significant impact. Beyond this, we examine the current state of funding for DRR and, in the context of those countries most at risk of natural disaster, ask questions about the volume and type of funding, and its equity. Are the right choices being made? contents Introduction 1 Top humanitarian recipients: aid and disaster 3 Financing for disaster risk reduction 10 Disaster risk reduction financing in context 15 The inadaquacy and inequity of disaster risk reduction financing 23 Conclusion: Spending where it should count 30 Postscript: Disaster risk reduction in the context of conflict 31 Data and guides 33 Endnotes and references 35 Acknowledgements 36 1 THE BIG NUMBERS DISASTER RISK REDUCTION in THE TOP 40 HUMANITARIAN RECIPIENTS, 2000–2009 IMPACT 847.5 million Affected population 595,783 Population killed US$74 billion Economic damage 1,472 Number of disasters INTERNATIONAL ASSISTANCE INTERNATIONAL DRR FUNDING US$9.3 billion of US$10.1 billion of humanitarian aid to the top 40 recipients in 2009; US$9 out of every US$10 US$3.7 billion for DRR out of US$363 billion development aid, to the top 40 humanitarian recipients US$363 billion of US$1,229 billion of development aid to the top 40 humanitarian recipients; US$3 out of every US$10 Therefore 1% of all development aid is DRR; US$1 out of every US$100 spent on aid is for reducing disaster risk Four countries alone account for 75% of all DRR – US$2.8 billion of the US$3.7 billion In 2009, 68% of DRR financing came from humanitarian funds DRR AND EQUITY DRR IN CONTEXT Five countries received more than US$5 per person in DRR over the decade Fourteen countries ranked as mortality risk medium -high to severe received only US$351.1 million of DRR combined Twenty-three countries received less than US$1 per person in DRR over the decade Note: All figures are for 2000–2009 except where indicated. 2 Fourteen countries collected less than US$100 per person in government revenues in 2010; only two of these received more than US$50 million in DRR over the decade TOP HUMANITARIAN RECIPIENTS: AID AND DISASTER The top recipients of humanitarian aid are particularly relevant when it comes to examining risk reduction financing. Firstly, these countries account for the vast bulk of humanitarian expenditure each and every year; it is to them that nine out of every ten humanitarian dollars go annually, and therefore they account for the continually upward pressure on humanitarian financing. Secondly, these are precisely the countries that are susceptible to the risk of natural hazards becoming natural disasters, and where those natural disasters have particularly significant impacts. This suggests that these are the places where the bulk of risk reduction funding should be spent. HUMANITARIAN EXPENDITURE IS CONCENTRATED By 2009 the top 40 recipients accounted for 92% of all humanitarian funding, with a similar figure reported in each of the previous six years. In the earlier years of the decade humanitarian aid was much less concentrated, with the top 40 recipients accounting for 74% in 2000. The top five recipient countries account for a particularly remarkable proportion of humanitarian financing, peaking at nearly 50% in 2003 and 2004. In the past five years this proportion has remained relatively stable at around 40%. AID PROFILES OF TOP HUMANITARIAN RECIPIENTS DIFFER CONSIDERABLY Same aid, different countries The major recipient countries of humanitarian assistance are in many ways very different. Some cover millions of square miles in area and have large populations, such as Pakistan and Indonesia. Others are relatively small and landlocked, such as Chad, Burundi and Serbia. Some of the countries have suffered from major single disasters that that have killed tens of thousands of people in a matter of moments, such as Haiti, Pakistan and Iran. Others have suffered long-developing droughts that rarely make the headlines but affect millions of people year on year. Some countries are major commodity producers and have significant government revenues as a result – Angola, Iraq and Algeria are examples – while others are amongst the smallest economies in the world and have the least available government revenues. One feature that almost all the countries share, beyond considerable humanitarian financing, is conflict. Only 6 of the 40 top recipients have not suffered at least one year of conflict during the decade: Democratic People’s Republic of Korea (DPRK), Jordan, Mozambique, Tanzania, Zambia and Zimbabwe. India and Mozambique have received US$17.8 billion and US$16.1 billion of aid respectively over the decade, of which US$628 million and US$534 million has been humanitarian financing. These two countries are good examples of how aid flows can be very similar in countries that are contextually very different. India has the tenth largest economy in the world, with a gross domestic product (GDP) in 2011 of US$1.84 trillion, and a population of more than 1.2 billion. Mozambique is ranked 122nd in terms of GDP with a 2011 figure of US$12.1 billion. It has a population of 23 million. India’s economy is 150 times bigger than Mozambique’s, with a population only 51 times bigger.1 Per person, residents of Mozambique received US$781 in ODA over the decade. Indian residents received only US$16, the lowest of all top humanitarian recipients. Figure 1: Concentration of humanitarian aid in the top 40 recipients 100% 90% % of aid top 40 % of aid top 20 % of aid top 5 80% 70% 60% 50% 40% 30% 20% 10% 0% 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Source: Development Initiatives based on OECD DAC and FTS 3 Figure 2: Aid to top humanitarian recipients, ranked by volume of aid that is humanitarian2 FIGURE 2 Sudan Palestine/OPT Iraq Afghanistan Ethiopia DRC Somalia Pakistan Indonesia Lebanon Kenya Serbia Uganda Sri Lanka Zimbabwe Angola Jordan Burundi Chad Liberia Eritrea Tanzania DPRK Myanmar Sierra Leone Colombia Bangladesh Haiti India Syria Mozambique BosniaHerzegovina Iran Georgia Cote d'Ivôire Nepal Guinea Zambia Thailand Algeria -5 0 5 10 15 20 25 30 35 US$ BILLION (CONSTANT 2009 PRICES) Total development aid, 2000-2009 Humanitarian aid, 2000-2009 Source: Development Initiatives based on OECD DAC 4 40 Some countries – Sudan, Palestine/OPT, Democratic Republic of Congo (DRC), Somalia, Iraq, Ethiopia – have been major recipients of humanitarian assistance for the whole of the decade. Three have become major recipients since 2005: Pakistan, Lebanon and Chad. Others have received smaller but regular amounts of assistance on a near-yearly basis, such as Uganda, Zimbabwe, Burundi and Jordan. The relationship between humanitarian aid and overall development aid is not necessarily the same for all the top humanitarian recipients. Some countries have received a much higher proportion of overall aid as humanitarian assistance. For example, humanitarian assistance to Sudan, Somalia and Palestine/OPT makes up a particularly significant proportion of all aid (61%, 68% and 38% respectively), reflecting the seeming intractability of the crises in each of these countries and the challenges of fostering development. Meanwhile, some major humanitarian recipients have received far less aid in the form of humanitarian assistance than for development. This is partly a factor of timing. Countries such as Serbia, Angola and Mozambique were in conflict at the beginning of the decade but the need for humanitarian expenditure has diminished in favour of development aid, with priority being given to supporting reconstruction. Other countries are a particular focus for donors: for example, both Ethiopia and Pakistan border areas of conflict and are of strategic interest. Iraq and Afghanistan stand out; in both cases, conflict with US-led coalitions was followed by massive donor nation involvement in reconstruction and state-building. In terms of official development assistance (ODA) volumes, the reconstruction of Iraq and Afghanistan has resulted in these countries coming first and second over the decade. The US$33.7 billion and US$28.7 billion that they have received, respectively, accounts for 9.3% of all aid expenditure over the period, equating to just under one in every ten dollars spent. Ethiopia was fourth with US$22.2 billion and Pakistan was sixth with US$18.6 billion. India, Palestine/OPT and Mozambique also make it into the top ten over the decade. Some of the top recipients of humanitarian assistance have received very little development aid at all. The DPRK and Myanmar are examples, largely for reasons to do with the challenges of delivering development aid in these countries. There is actually significant overlap in the top 20 countries of the two aid categories, with 15 of the top recipients of humanitarian assistance also in the top 20 for ODA. Yet this is not the case for the next 20 highest recipients of ODA; major humanitarian recipients account for only five of these. Over the decade, the top 40 humanitarian recipients accounted for US$363 billion of all ODA, 30% of the total of US$1,229 billion for all countries. They, in effect, received nine out of every ten humanitarian dollars, but only three out of every ten ODA dollars. This balance of humanitarian and development funding is important. Risk reduction is recognised as a long-term national investment that needs to be mainstreamed through a country’s ministries and activities. Humanitarian funding, with its relatively short-term planning and engagement, is considered unsuitable for supporting these activities, though it should be noted that examples of this happening are not uncommon. Data on the split of DRR financing between humanitarian and development funding is made problematic by reporting issues. However, perhaps worryingly, for 2009, the year for which we have the most robust data, 68% of all DRR funding to the top humanitarian recipients came from humanitarian financing, not development. Given the pressure on humanitarian financing to ‘respond’ to more and larger crises, the question must be asked whether this is sustainable, notwithstanding the structural issues within humanitarian aid. Table 1: Top 40 recipients of ODA, 2000-2009 ODA (US$ bn) Iraq 33.7 Afghanistan 28.7 Vietnam 23.6 Ethiopia 22.2 Tanzania 19.0 Pakistan 18.6 China 17.8 India 17.8 Palestine/OPT 16.6 Mozambique 16.1 Bangladesh 14.6 Uganda 14.2 Sudan 14.0 Indonesia 13.6 Serbia 12.4 DRC 11.9 Ghana 11.5 Egypt 11.4 Kenya 9.5 Zambia 9.5 Morocco 8.8 South Africa 8.4 Bolivia 8.0 Nicaragua 8.0 Senegal 7.9 Burkina Faso 7.8 Mali 7.6 Jordan 7.5 Colombia 7.5 Nigeria 7.3 Sri Lanka 6.9 Bosnia-Herzegovina 6.8 Madagascar 6.6 Malawi 6.6 Cambodia 6.4 Turkey 6.3 Rwanda 6.2 Philippines 6.0 Nepal 5.8 Honduras 5.8 Note: Figures in constant 2009 prices. Highlighted countries are also amongst the top 40 humanitarian recipients. Source: Development Initiatives based on OECD DAC 5 MAJOR HUMANITARIAN RECIPIENT COUNTRIES ARE DISPROPORTIONATELY AFFECTED BY DISASTER Major recipients of humanitarian assistance have accounted for substantial numbers of disasters, with many people affected and killed and substantial economic damage sustained. This is unsurprising since this impact is in many cases precisely why they needed the assistance. What is clear, however, is that the impact of disasters is significantly higher in such countries when compared with other disaster-affected nations. Over the decade 2000–2009 the top 40 humanitarian recipients accounted for 1,286 disasters, 32.1% of the total. The proportion of people affected was significantly higher than this, at 52.5%, and the proportion of people killed over the ten years was 78.7%. Essentially, although major humanitarian recipient countries suffer three in every ten disasters, they account for five out of every ten people affected and seven out of every ten people killed. There is no discernible trend for disasters within humanitarian recipient countries except that in general they follow the trend for all affected nations, with a rise in figures from 1997 to 2000. Peak years in terms of numbers of disasters in 2000, 2002 and 2005 were similarly high for humanitarian recipients. The year on year figures, however, reveal even more clearly how natural disasters in crisisaffected countries can have a particularly damaging impact. Table 2: Disaster impact on the top 40 humanitarian recipients against total disaster figures, 2000-2009 Top 40 humanitarian recipients All countries % top 40 recipients Total affected 248,506,824 473,460,284 52.5% Total mortality 536,321 681,313 78.7% Total number of disasters 1,286 4,012 32.1% Total estimated damages US$50 billion US$682 billion 7.3% Note: These figures exclude India inside the top humanitarian recipients and China in ‘all countries’. Source: EMDAT CRED The pattern of disaster numbers is remarkably stable, with between 29% and 38% of all disasters occurring in major humanitarian recipients over the period examined. The proportion of people affected is startlingly different, however, and much more erratic (this depends in part on the nature of different kinds of disaster and their impact. See box at base of page 7 on Haiti, “Disasters and their varying impacts”). 2001 was the year with the lowest figure, with fewer than 35% of people affected living in the top 40 humanitarian recipient countries. In four years, however – 1999, 2003, 2007 and 2010 – the proportion was 120,000,000 600 100,000,000 500 80,000,000 400 60,000,000 300 40,000,000 200 20,000,000 100 0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Number of disasters in the top 40 HA recipients (excl. India) Number of disasters outside of the top 40 HA recipients (excl. China) Source: EMDAT CRED 6 NUMBER OF DISASTERS NUMBER OF AFFECTED Figure 3: Total number of disasters and affected numbers for the top 40 humanitarian recipients over time in comparison with total figures 0 Total number affected (excl. China and India) Number affected in top 40 HA recipients (excl. India) Figure 4: Proportions of disasters, people affected and people killed, living in top humanitarian recipient countries, excluding India and China 100% 90% 80% % of total deaths % of total affected % number of disasters 70% 60% 50% 40% 30% 20% 10% 0% 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Source: Development Initiatives based on EMDAT CRED close to 80%. There is no easily identifiable reason for the peaks in 2003 and 2007. In 1999 a drought in Iran affected 37 million people, 31% of all those affected in that year (excluding India and China). 2010 was remarkable for two mega-disasters of very different kinds, the Pakistan floods and the Haiti earthquake, but it was Pakistan’s flood victims (20 million affected) who accounted for 20% of all people affected, while Haiti’s 3.5 million were only 3.5% of the total for that year. The proportion of people killed is also startlingly high for humanitarian recipients. In three years – 2004, 2005 and 2008 – the proportion of people killed living in these countries was more than 90% of the total. In these years, although accounting for just three in every ten disasters, major humanitarian recipients accounted for nine out of every ten deaths, with the 2004 Boxing Day tsumani, the Kashmir earthquake in 2005, and Cyclone Nagris in Myanmar in 2008 partly responsible. Undoubtedly this severity of impact is due to these countries’ relatively poor infrastructure, weaker government capacity for planning and response and their often large populations living on the fringes of habitable space, frequently Table 3: Haiti disaster impact, 2010 Disaster Type Occurrence Total Deaths on the edges of urban areas. These populations often lack basic facilities such as adequate housing, clean water and sanitation, roads and electricity. This, combined with their relatively limited means of coping with sudden crises, and compounded by weak infrastructure and government capacity, can easily turn a natural event into a disaster. Disasters and their Varying Impacts % of Total % of deaths Affected affected Total Number Damage affected (US$bn) for each death Earthquake 1 222,570 98.3% 3,500,000 92.5% 8 15.7 Epidemic 1 3,790 1.7% 185,012 4.9% n/a 48.8 Flood 3 44 0.0% 22,087 0.6% n/a 502.0 Storm 2 27 0.0% 78,142 2.1% n/a 2,894.1 Total 7 226,431 3,785,241 Not all disasters have the same impact. In some cases a single disaster can have the same effects as many others. Haiti is a good example. In 2010 seven natural disasters were reported in the country. One of these, the January earthquake, accounted for 98.3% of all deaths and 92.5% of all people affected. (In contrast, the far larger earthquake that hit Chile in 2010, causing US$30 billion in damages, had a much lower death toll of 562 people.) Source: Development Initiatives based on EMDAT CRED 7 Country FOCUS: India and China – eight out of every ten affected India and China are such large countries with huge populations that inclusion within the data on numbers of disasters and in particular numbers of people affected can tend to mask key data and trends within other countries affected by disaster. We have therefore removed both countries from the overall analysis (India from the top 40 humanitarian recipients and China from all other countries). However, the very fact that so many millions of people each and every year are affected in these countries demands attention. Over the 11-year period from 2000 to 2010, nearly eight out of every ten people affected by a natural disaster have been either Indian or Chinese. 2002 was remarkable, with more than 627 million people affected being either Chinese or Indian – 95% of the total in that single year. China suffered a wave of droughts, floods and storms, which affected more than 285 million people. In India one single disaster, a drought, affected 300 million people, 42% of the total that year worldwide. This drought, which occurred after one of the shortest monsoon seasons on record, affected ten states and 56% of the entire country. Agricultural GDP shrank by 3.1% and overall GDP by 1% as a result.3 Perhaps unsurprisingly, the size of both these countries means that when disasters occur they can affect much higher numbers of people than elsewhere. Taking 2002 as an example once more, the total number of disasters in China and India was 47 out of a total of 458, equivalent to just over 10%. So while the two countries accounted for more than nine out of every ten people affected in that year, it was with less than one out of every ten disasters. Over the decade the proportion of disasters occurring in India and China has been stable at between 8% and 13% each year – an average of 10% of all disasters over the ten years. This compares with the fact that 78% of all people affected over the decade were Chinese or Indian. The disaster profiles of both countries are in some ways similar, with drought and floods being particularly damaging. Through the decade China has experienced repeated and severe natural disasters, including significant earthquakes and storms. In each of the last four years more than 120 million people in China have been affected by natural disaster, a period in which an average of only 15.7 million Indians were affected. Figure 5: Populations affected by natural disaster: India, China and the rest of the world NUMBER OF AFFECTED (MILLIONS) 100% 89% 90% 600 500 73% 73% 61% 75% 68% 69% 80% 67% 70% 55% 400 60% 50% 300 40% 30% 200 20% 100 0 10% 2000 2001 2002 2003 2004 2005 Source: Development Initiatives based on EMDAT CRED 8 76% 2006 2007 2008 2009 2010 0% % OF AFFECTED 95% 700 All other countries affected India affected China affected % of total affected China and India Table 4: Disaster Profile of China and India China 2000-2010 Affected Mortality Drought 241,524,000 134 Earthquake (seismic activity) 52,807,377 Epidemic India 2000-2010 Number Affected Mortality 16 350,000,000 20 3 90,924 51 5,012,599 37,705 4 6,829 423 5 225,412 1,528 24 Extreme temperature 77,000,000 191 5 - 5,276 18 Flood 665,832,382 7,692 104 242,173,116 14,399 106 Mass movement 2,157,600 3,144 36 12,000 572 13 Storm 282,039,647 3,319 91 5,475,905 1,273 36 Wildfire 0 22 2 – – Total 1,321,367,835 105,849 602,899,032 60,773 311 Number – 204 Source: EMDAT CRED Context is key, however, both to how countries cope with natural hazards and their impact and to the place of international aid, as we discuss later in this report. Perhaps China is more able to cope with natural disasters than India. Despite similarities in population (China’s is the largest in the world and India’s the second) and in size of the country (third and seventh largest respectively), China is an upper middle-income nation while India is ranked in the lower category of lower middle-income countries. The poverty figures are particularly revealing. According to the Multidimensional Poverty Index (MPI), the countries are ranked first and second in terms of total numbers of people considered to be poor. However, India has more than 600 million poor people, four times the number of China.4 Figure 6: Number of disasters in India, China and the rest of the world and proportion of affected population 100% 600 90% NUMBER OF DISASTERS 500 80% 70% 400 60% 50% 300 40% 200 100 0 30% 10% 12% 9% 11% 2000 2001 2002 2003 8% 2004 13% 2005 12% 2006 9% 10% 11% 10% 2007 2008 2009 2010 20% 10% 0% Total number of disasters all other countries Total number of disasters China and India % of total disasters China and India Source: Development Initiatives based on EMDAT CRED 9 FINANCING FOR DISASTER RISK REDUCTION DRR IS A FRACTION OF AID At first glance, trends in the financing of DRR to these key countries seem positive. Since 2000, DRR investments to the value of US$3.7 billion have been made from all aid to the top recipients of humanitarian assistance. Funding grew from US$121 million in 2000 to a peak of US$809 million in 2007, before falling back to US$338 million in 2009. Yet at the same time this is a very small percentage of overall development aid spent in these same countries. Only in two years, 2006 and 2007, has DRR expenditure ever reached above 2% of ODA, and over the entire decade it was only 1%. Essentially less than one dollar for every 100 has been spent on reducing disaster risk. Given that the economic damages for top humanitarian recipients have been estimated as at least US$74 billion, this figure of US$3.7 billion seems insubstantial. There is further concern, looking beyond overall annual volumes of DRR financing, as those volumes hide not just variability but also very high concentrations of investments in just a few contexts. The aid trends over the decade do not show a sudden increase in expenditures from 2005 through to 2007, such as may be influenced by the lessons learned after the Indian Ocean tsunami of 2004. There is no sudden general increase in the amount of funding going to countries in need. Rather, the increases are much more about single large projects that dominate overall spending. In 2005, 63% of all DRR funding to top humanitarian recipients was for just two projects: US$116 million (32%) for a single World Bank reconstruction programme in Pakistan that focused on seismic-resistant house construction and a US$110 million (31%) project, also by the World Bank as part of its Gujarat recovery programme, that focused on “sustainable disaster management capacity”. In 2006, 71% (US$539 million) of all DRR funding for the top 40 humanitarian recipients was made up of two World Bank projects for Pakistan, one that linked DRR to reconstruction and another that mainstreamed DRR across multi-sectoral activities. In the following year, a single contribution from Japan of US$244 million to Indonesia to support its disaster management policy implementation accounted for 30% of funding to the top humanitarian recipients. The drops in funding for 2008 and 2009 are therefore much less about changes in donor priorities and more about the lack of single large projects for single recipient countries in those years. The Semantics of DRR, preparedness and the data connections The semantics of DRR are somewhat intimidating, with complex interconnected terms made more complicated by the loose use of those terms. This confusion is also reflected in the data available. The use of “DRR” in this report is taken from UN International Strategy for Disaster Reduction (UNISDR) terminology: “systematic efforts to analyse and manage the causal factors of disasters”. As there is no DRR code within the Organisation for Economic Co-operation and Development (OECD)’s Development Assistance Committee (DAC) database, a forensic method has been used to pull out investments made in reducing risk. This research has been particularly generous since many projects may actually include DRR as an element, perhaps as something cutting across other sectors; we have, without further information, added the entirety of this to our data. See methodology section for details. For the purposes of this report we have used preparedness to mean “disaster preparedness and prevention” (DPP), a humanitarian code that contains mostly activities designed as preparedness to respond to disasters. US$ MILLION (CONSTANT 2009 PRICES) Figure 7: Disaster Risk Reduction expenditure in top humanitarian recipients 900 800 756 4.0% 700 591 600 3.0% 500 400 357 300 200 100 0 121 2000 176 2001 152 2002 205 2003 338 184 2004 Source: Development Initaitives based on OECD DAC 10 5.0% 809 2.0% 1.0% 2005 2006 2007 2008 2009 0.0% Total DRR in ODA, top 40 HA recipients % of ODA as DRR A further complication, however, is that the available data does not show us the duration of projects. This is particularly significant in the case of longer-term loans and grants, such as those made by the World Bank, where a single project appearing in the available data as a DRR investment in a single year is actually a multi-year project. With this detailed information at hand (and it is not currently available in an easy-to-use form), we could expect a much smoother transition of DRR from 2006 through to 2009. Through the decade, the number of countries actually funded for DRR has grown. In 2000 there were 13 countries; this grew to 39 countries by 2009, with Iran the only top humanitarian recipient not to receive DRR financing in that year. There was a considerable jump from 18 to 28 countries from 2003 to 2004, and levels continued to rise for much of the rest of the decade. There is evidence to suggest that this sudden increase was less to do with changing donor priorities, however, and more to do with the introduction of the “disaster preparedness and prevention” code as part of humanitarian reporting within the OECD DAC database. Top humanitarian recipients receive less than other countries In 2009, the year in which we have investigated DRR expenditures for all countries, we find that the major humanitarian recipients combined received Table 5: Number of Top Humanitarian Recipients receiving DRR financing Number of top recipients recEIving DRR 2000 13 2001 16 2002 15 2003 18 2004 28 2005 31 2006 31 2007 29 2008 37 2009 39 Source: Development Initiatives based on OECD DAC less funding than other aid recipients, this despite the fact that they suffer particularly from natural disasters. In 2009, 37% (US$338 million) of DRR went to the top 40 recipients of humanitarian assistance, with US$414 million going to the remaining countries. Notable recipients outside of the top 40 were Vietnam (US$50 million), the Philippines (US$50 million) and China (US$36 million). Figure 8: Breakdown of DRR recipients 1,000 1,000 US$ MILLION (CONSTANT 2009 PRICES) US$ MILLION (CONSTANT 2009 PRICES) 900 900 800 800 414 414 700 700 600 600 500 500 400 400 300 300 162 162 338 338 200 200 Total DRR other recipients Total DRR allall other recipients Bilateral, unspecified Bilateral, unspecified Top humanitarian recipients DRR Top 4040 humanitarian recipients DRR 100 100 00 2009 2009 Source: Development Initiatives based on OECD DAC 11 Figure 9: Total volumes of disaster prevention and preparedness US$ MILLION (CONSTANT 2009 PRICES) 500 454.6 450 400 333.3 350 300 250 200 150 0 146.0 69.7 100 50 Remaining recipients (excl. bilateral unspecified) 98.6 5.6 3.3 2004 59.2 2005 Top 40 HA recipients 99.8 41.0 12.2 28.6 2006 2007 Total disaster prevention and preparedness 2008 2009 Source: OECD DAC MINIMAL INVESTMENTS IN DISASTER PREPAREDNESS AND PREVENTION We would imagine that even though top humanitarian recipients do not receive a huge proportion of their overall aid as DRR, they would at least receive a significant proportion of their humanitarian assistance as disaster prevention and preparedness, especially since there are specific reporting lines for donors to report their prioritisation for activities before a crisis strikes. Over the six years for which data is available, there is considerable growth from a very low figure of US$4.4 million to all countries in 2004 (a figure we consider to be largely driven by the fact that the reporting was new). In 2009 preparedness and prevention expenditures to all recipients rose to their highest ever figure of US$454.6 million. Of this, US$146 million was clearly given to the top 40 humanitarian recipients (see box on “Bilateral unspecified”, page 13). However, these sums are almost completely insignificant when compared with overall humanitarian financing in the same period to the top 40 humanitarian recipients. Table 6: Top 40 recipients’ preparedness/prevention expenditure, 2005-2009 Year Disaster prevention and preparedness All other bilateral HA to the top 40 recipients Total bilateral HA to the top 40 HA recipients 2004 3.3 4,501.2 4,504.5 0.1% 2005 59.2 7,057.9 7,117.1 0.8% 2006 12.2 6,365.3 6,377.6 0.2% 2007 28.6 5,734.4 5,763.0 0.5% 2008 99.8 7,796.7 7,896.5 1.3% 2009 146.3 7,939.9 8,086.2 1.8% Total 349.4 39,395.4 39,744.9 0.9% Note: Figures expressed in US$ million, constant 2009 prices. Source: OECD DAC 12 Bilateral, unspecified % HA as disaster prevention and preparedness Figure 10: Proportion of bilateral humanitarian sector funding to top 40 humanitarian recipients 100% 90% Disaster prevention and preparedness Emergency/distress relief Relief co-ordination; protection and support services Emergency food aid Reconstruction relief Total all 80% 70% 60% 50% 40% 30% 20% 10% 0% 2004 2005 2006 2007 2008 2009 Source: OECD DAC The 2009 figure of US$146.3 million for disaster prevention and preparedness is dwarfed by the US$8.1 billion of overall humanitarian expenditure allocated to the top 40 recipients of humanitarian aid, and is equivalent to just 1.8% of the total. Of the total US$39.7 billion of humanitarian aid spent specifically in these same countries over these five years, only US$349 million was on disaster prevention and preparedness, or 0.9%. For every US$100 spent on response to humanitarian need, therefore, less than 90 cents was spent on preparing for or preventing that need. One would imagine that, despite the poor overall volumes dedicated to disaster prevention and preparedness, the major humanitarian recipients should make up the major proportion of these investments before a crisis occurs. However, this is not the case. By 2009 humanitarian funding going straight to recipient countries (bilateral funding) reached US$10.7 billion. Of this funding, the top 40 countries made up a very stable percentage of close to 70% each year. For some of the specific humanitarian reporting lines they received a particularly large proportion; for example, in 2009, 91% (US$2.9 billion) of total food aid of US$3.2 billion went to the top 40. A similarly large proportion of the other line that received considerable volumes of funding, emergency distress/relief, also went to top humanitarian recipients – in 2009, US$3.6 billion of the total US$5.4 billion, equating to 68%. However, while the top 40 made up the largest proportion in four of the five codes, the share of DPP was remarkably low, hovering at only about 30% in each of the last four years. Essentially, more funding for disaster preparedness and prevention is going to countries that are not major humanitarian recipients than to those that are. What is bilateral unspecified? A number of members of the OECD DAC report portions of their bilateral ODA as “bilateral, unspecified”. As a result, much of their aid is not allocated geographically by recipient or region. Bilateral unspecified allocations are made for expenditures on administrative costs, global programmes and unearmarked contributions to implementing partners that cannot be allocated by recipient country. For practical purposes in this report, the bilateral unspecified DRR expenditure indicated in Figure 9 includes, for example, contributions to the UNDP Bureau of Crisis, Prevention and Recovery (BCPR) and the World Bank Global Fund for Disaster Risk Reduction (GFDRR). These organisations then disburse funds to recipient countries, some of which are also top humanitarian recipients. 13 POOLED FUNDS FOR PREPAREDNESS AND DRR Funding mechanisms do exist that finance DRR and preparedness activities, although to significantly varying levels. Humanitarian funds focus more on preparedness. There are three main types of pooled humanitarian fund that finance preparedness: the Central Emergency Response Fund (CERF), common humanitarian funds (CHFs) and emergency response funds (ERFs).5 Despite the global CERF being limited by its emergency response mandate, it has funded some preparedness activities, but only in the last year of operation. In the second half of 2011 it channelled almost US$12 million to preparedness in three of the top 40 recipient countries – Ethiopia, Kenya and Sudan – but this was still a small fraction of the total US$426.2 million spent that year. The four country-level CHFs in operation continue to attract considerable donor support – US$354 million in 2011. Nevertheless, only 3% of the total channelled by the funds was spent on preparedness. Three of the funds operate in countries that are among the top 40 humanitarian recipients – DRC, Sudan and Somalia. In total these funds disbursed US$8.7 million to preparedness activities in 2011, an increase on 2010 and 2009 figures but still a very modest proportion, accounting for 3.9% of total CHF funding. Despite the emergency response mandate of the ERFs, several have disbursed funds to preparedness projects. In 2011 Haiti’s ERF spent 11.8% (US$0.8 million) on disaster and cholera preparedness. In 2010 the fund in Kenya spent 20% of its budget on flood preparedness, one of only four projects funded that year. In 2009 the ERFs in Somalia and Zimbabwe spent 5.9% and 2.9% respectively on preparedness out of the total money allocated in those countries. Unsurprisingly, each ERF is in a top 40 humanitarian recipient country. The World Bank-managed Global Facility for Disaster Reduction and Recovery (GFDRR) is the only operational fund that is solely focused on DRR and carries out both global and country-level projects. Around 70% of its funding comes from humanitarian aid budgets and it has received almost US$200 million since its inception in 2006. Track II of the fund allocates money to 20 priority countries deemed to be most at risk of disasters, plus 11 countries earmarked by donors. Only 8 of these 31 countries are among the top 40 recipient countries. In 2010 the fund allocated a larger proportion to countries that were top 40 recipients, 70% or US$7.2 million. In 2011 it spent a similar amount, US$6.3 million, but this was a substantially lower proportion of overall expenditure that year, only 35% (note these figures do not include GFDRR expenditures that are allocated to more than one country or are regional). The total three-year DRR funding by country to the top humanitarian recipients that we can identify is still only small relative to overall aid flows, just US$15.6 million out of a total of US$33.4 million. Figure 11: Money spent on preparedness and DRR from four different pooled funds 2009 GFDRR ERF CHF CERF 2010 GFDRR ERF CHF CERF 2011 GFDRR ERF Top 40 recipients of humanitarian aid, 2000-2009 All other recipients CHF CERF 0 2 4 6 8 10 12 US$ MILLION (CURRENT PRICES) Source: Development Initiatives based on UN OCHA FTS and World Bank data 14 14 16 18 20 DISASTER RISK REDUCTION FINANCING IN CONTEXT DRR IN THE CONTEXT OF RISK It is through looking at individual countries that we can begin to understand how adequate or otherwise DRR expenditure has been over the decade, especially when we examine the context, in terms of the different risks faced from natural disaster. The most obvious point to emerge from the data is the high level of concentration of DRR financing over the decade in just a few recipient countries. The top four recipients by volume (Pakistan, India, Indonesia and Bangladesh) received US$2.8 billion of the total US$3.7 billion received by the entire top 40 countries, or 75%. This meant that close to eight out of every ten DRR dollars going to the top humanitarian recipients went to just the top four countries. The top ten meanwhile accounted for 91.2%, leaving the other 30 to share the remaining US$325 million. Figure 12: Volumes of DRR over last ten years to top recipients of humanitarian assistance, ranked by mortality risk US$ MILLION (CONSTANT 2009 PRICES) 0 2.50 0.06 3.60 5.50 0.94 0.17 0.12 5.40 0.01 0.13 0.50 0.60 1.30 0.29 0.40 0.10 % of ODA that is DRR 0.11 0.27 0.02 0.28 0.20 0.06 0.50 0.10 0.16 1.50 0.06 0.04 0.02 0.28 1.60 0.12 0.04 -0.24 0.44 0.07 0.05 0.90 0.01 0.03 Bangladesh Colombia India Indonesia Myanmar Afghanistan Iran Pakistan Algeria DRC DPRK Ethiopia Haiti Mozambique Nepal Sierra Leone Sudan Uganda Bosnia-Herzegovina 200 400 600 800 1,000 1,200 371.4 4.4 639.4 754.6 19.5 49.8 1.8 1002.3 0.2 15.1 7.9 140.7 64.7 46.9 23.5 3.8 15.9 38.0 1.3 10.2 Burundi Chad Cote d'Ivôire Georgia Guinea Iraq Kenya Lebanon 7.3 1.2 23.0 2.4 53.7 143.4 3.1 1.2 1.9 Liberia Serbia Somalia Sri Lanka Syria Tanzania Thailand Zimbabwe Angola Eritrea Zambia Jordan Palestine/OPT 9.9 110.5 1.5 6.7 2.8 16.5 3.0 1.3 82.8 0.5 4.2 0 1 2 3 4 5 6 7 8 9 10 All DRR within ODA, 2000-2009 Multiple Mortality Risk Class (UNISDR 2009) MORTALITY RISK Sources: Development Initiatives based on OECD DAC, UNISDR 15 Secondly, funding received varied considerably between countries that are classified as having the same mortality risk. Relatively large amounts were spent in three major risk countries, Bangladesh, India and Indonesia, but two similarly ranked countries, Colombia and Myanmar, received hardly any funding at all. Similarly low levels of funding were received by Afghanistan and Iran, both ranked very high for mortality risk. These low amounts going to high-risk countries seem even more surprising when we see how three countries much further down the rankings (Kenya and Sri Lanka ranked medium and Zambia medium-low) have received relatively high levels of funding. Zambia received US$82.8 million of DRR funding over the decade, far more than all other higher-ranked sub-Saharan countries (with the exceptions of Kenya and Ethiopia). DRC, ranked as high for mortality risk and suffering from multiple epidemics, droughts, floods and seismic activity, received only US$15.1 million over the decade. Even the relatively large amounts going to some of those high-risk countries suddenly seem relatively insignificant when compared with overall ODA levels. The top three countries by proportion of ODA that was used for DRR over the decade (Indonesia, Pakistan and India) received only 5.5%, 5.4% and 3.6% respectively for this purpose. Only 6 of the top 40 countries received more than 1% of ODA as DRR. Sub-Saharan African countries, which are in many cases affected by a mix of droughts and floods (often compounded by conflict) and where a very large proportion of overall humanitarian spending goes each year, have received particularly low levels. The proportion over the decade for both Somalia and Uganda was 0.3%; for Sudan and DRC it was only 0.1%. There are many countries that have high levels of risk but low levels of funds for tackling risk. What are the mortality rankings? Mortality risk is taken from the UN Global Assessment report on DRR produced by the United Nations International Strategy for Disaster Reduction (UNISDR), with a rating of ten denoting the most at-risk countries and one the least. Of the top 40 humanitarian recipients over the decade, five countries are classified as being at major risk (level nine), three as very high (level eight) and two as high (level seven). The risk level of medium (five) applies to the largest number (17) of the top humanitarian recipients. DRR IN THE CONTEXT OF ECONOMY All countries can be affected by natural disaster Natural hazards are not confined to poorer countries or to those without systems or infrastructure and with particularly vulnerable populations. Rich and seemingly well-prepared nations can also be severely affected. The list of top ten countries by number of people affected includes five countries that are not major humanitarian recipients: China, the Philippines, Vietnam, South Africa and the United States. These top ten countries accounted for 27% of all disasters over the decade and 90% of people affected, a percentage largely made up of Indians and Chinese. The US, which ranked ninth over the period in terms of numbers affected, suffered the largest estimated economic impact, at US$353.4 billion. Table 7: Top 10 countries affected by natural disasters, 2000 –2010 Top 10 countries affected by disasters Numbers Affected (million people) Number of disasters Number of deaths Ratio Disasters to Deaths Economic costs (US$bn) China 1,321.4 311 105,849 340 205,654,128 India 602.9 204 60,879 298 25,888,285 Bangladesh 73.2 90 9,696 108 5,884,000 Philippines 52.9 160 10,531 66 2,543,118 Thailand 43.6 57 9,750 171 2,433,613 Pakistan 32.8 74 79,325 1072 17,134,648 Ethiopia 29.2 48 2,926 61 9,400 Vietnam 21.8 89 3,754 42 5,759,905 United States 20.7 257 4,357 17 353,414,290 South Africa 15.3 42 708 17 866,305 % of total 90% 27% 25% – 61% Source: EMDAT CRED 16 COuntry focus: Japan, a rich nation prone to major natural disasters Japan has the third largest economy in the world and in the past has spent up to 5% of its annual general budget on DRR.Yet even the richest, best-prepared nations in the world cannot completely prevent natural hazards from becoming natural disasters. a century in 2000 and of course the huge tsunami in March 2011. It ranks as having the third highest number of disasters of all countries over the years 2000–2011. Its ranking for deaths due to natural disasters would have been only 49th were it not for the 2011 tsunami, which pushed it up to 10th place. The total estimated economic damages (US$288 billion) put Japan only one place behind the US; even without the massive impact of the tsunami, Japan still ranks very high, third in total estimated damages, behind the US and China. Japan has a multi-hazard profile. The country has suffered from regular seismic, flooding and storm disasters in particular, with notable events being the Kobe earthquake in 1995, the worst flooding in Table 8: The impact of disasters in Japan Japan disasters Total 2000-2011 Rank Number of disasters 72 13/201 Number affected 1,991,840 46/201 Number of deaths 21,365 10/201 Economic damages US$288 bn 2/201 Source: EMDAT CRED Figure 13: affected numbers against disaster type, Japan 1995 2011 2000 KOBE EARTHQUAKE WORST FLOODS IN A CENTURY SENDAI TSUNAMI 700,000 600,000 576,090 NUMBER OF AFFECTED 500,000 410,846 400,000 400,000 300,000 276,427 269,855 200,000 100,792 100,000 67,460 73,130 64,089 44,000 20,746 5,400 22,587 22,593 50,270 32,000 15,000 0 1995 1996 Volcano Storm 1997 Flood Wildfire 1998 1999 2000 2001 2002 2003 2004 2005 Earthquake (seismic activity) Mass movement Extreme temperature Epidemic 2006 2007 2008 2009 2010 2011 Total Source: EMDAT CRED 17 Figure 14: Number of disasters in developing countries NUMBER OF DISASTERS 350 300 250 50.0% 49.4% 41.6% 46.0% 42.3% 50.8% 52.3% 48.2% 60% 45.3% 45.9% 44.1% 40% 200 30% 150 20% 100 10% 50 0 50% 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 0% Upper middle-income country Lower middle-income countries (excl. China and India) Low-income country Least developed country % disasters in low-income and least developed countries Sources: EMDAT CRED and OECD (income groups are determined by the OECD for 2009 and 2010) Classification of Countries by Income Group Poorest countries suffer from higher proportions of affected people The poorer a country is, the more significant the impact of natural disaster. Over the 11 years from 2000 to 2010, the proportion of natural disasters in developing countries occurring in those with the lowest incomes (low-income and least developed countries) has been relatively constant at 40–50%. However, the proportion of people affected in these same two poorest country categories was higher in all but two of the 11 years. In some years, such as 2003, 2005 and 2007, the proportion of people affected was 25% higher than the proportion of disasters. It is therefore not just major humanitarian recipients that are disproportionately affected by natural disaster, but also poor countries in general. Unsurprisingly, there is plenty of crossover between top humanitarian recipients and poorest countries. Most of the top 40 are in the poorest groups, though not all. Of the top 40 humanitarian recipients over the decade, 25 fall into the lowest two categories of income, with 20 classified as least developed and five as low-income. Thirteen are classified as lower middleincome and two are upper middle-income. The OECD classifies countries by income group: • Least developed countries (LDCs) are defined by the UN based on an assessment of economic vulnerability, human resource weakness (nutrition, health, education, adult literacy) and where gross national income (GNI) per capita, based on a three-year average, is under $750. LDCs are a subset of low income countries. • Low-income countries (LICs) are those with a per capita GNI of less than US$935 in 2007. • Lower middle-income countries (LMICs) are those with a per capita GNI of between US$936 and US$3,705 in 2007. • Upper middle-income countries (UMICs) are those with a per capita GNI of between US$3,706 and US$11,455 in 2007. Figure 15: Population in developing countries affected by natural disasters 90% NUMBER OF AFFECTED (MILLIONS) 80 79.0% 70 60 70.6% 80% 72.6% 62.8% 50 49.5% 40 70% 59.1% 60.5% 60% 55.9% 50% 49.7% 45.5% 30 40% 35.8% 30% 20 20% 10 10% 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Sources: EMDAT CRED and OECD (income groups are determined by the OECD for 2009 and 2010) 18 0% Upper middle-income countries Lower middle-income countries (excl. China and India) Low-income countries Least developed countries % affected in low-income and least developed countries IN FOCUS: DAMAGES AND LOSSES AND THE ECONOMIC IMPACT OF DISASTERS While the Centre for Research on the Epidemiology of Disasters (CRED) database provides the most comprehensive coverage of disasters, the damage and loss assessment (DaLA) methodology assesses their economic impact far more rigorously, though only for major natural disasters. It usually also identifies priorities for recovery. The United Nations, World Bank and other stakeholders often use a DaLA methodology to assess the economic impact of major natural disasters, as well as to identify the needs of recovery. Since 2003 this methodology has been used 16 times by the UN, World Bank and national partners to gauge the economic impact of disasters in top humanitarian recipients, including five times in Indonesia, after the Haiti earthquake and twice in India. Damage and losses can be significant: for example, Haiti suffered US$7.8 billion in losses due to the January 2010 earthquake, while Cyclone Nargis caused US$4.1 billion of damage in Myanmar. The 2004 Indian Ocean tsunami caused US$10.8 billion of damage and losses across the four countries assessed. The five disasters assessed in Indonesia incurred damages and losses amounting to US$9.4 billion in total. Table 9: The Global Facility for Disaster Reduction and Recovery (GFDRR) Global Disaster Damage and Loss Database Year Country Name All terms Damage (US$m) Losses (US$m) Total (US$m) 2010 Haiti Haiti earthquake 2010 Earthquake 4,245.3 3,517.3 7,762.6 2004 Indonesia Indonesia tsunami 2004 Tsunami 3,371.5 1,765.4 5,136.9 2008 Myanmar Cyclone Nargis Myanmar 2008 Hurricane/cyclone/storm 1,773.1 2,328.2 4,101.3 2001 India Gujarat (India) earthquake 2001 Earthquake 2,609.9 802.1 3,412.0 2006 Indonesia Indonesia earthquake 2006 Earthquake 2,637.1 739.3 3,376.4 2005 Pakistan Pakistan earthquake 2005 Earthquake 2,549.6 645.0 3,194.6 2004 Thailand Thailand tsunami 2004 Tsunami 587.3 1,950.0 2,537.3 2007 Bangladesh Cyclone Sidr Bangaldesh 2007 Hurricane/cyclone/storm 1,211.3 540.6 1,752.0 2004 Sri Lanka Sri Lanka tsunami 2004 Tsunami 1,320.0 357.7 1,677.7 2004 India India tsunami 2004 Tsunami 663.5 748.8 1,412.3 2008 Haiti Hurricane Gustave Haiti Hurricane/cyclone/storm 458.6 435.4 894.0 2009 Indonesia West Java quake 2007 Earthquake 760.8 20.3 781.1 2000 Mozambique Floods Mozambique 2000 Flood 304.6 357.9 662.5 2010 Indonesia Indonesia volcanic eruption 2010 Volcanic eruption 126.5 343.3 469.8 2004 Haiti Hurricane Jeanne Haiti 2004 Hurricane/cyclone/storm 210.3 91.1 301.4 2006 Indonesia Aceh floods 2006 Flood 195.9 30.4 226.2 2007 Indonesia West Sumatra quake 2007 Earthquake 172.9 2.8 175.6 2010 Sri Lanka Floods Sri Lanka 2010 Flood 23.4 27.2 50.6 Note: Table contains outcomes of post-disaster damage and loss assessments carried out by various international organisations. Source: GFDRR 19 GOVERNMENT REVENUES AND DRR Data on national capacity to reduce the risk of natural disasters is at present very limited. Despite the work done through the Hyogo Framework for Action (HFA), there is only partial, piecemeal information available on just a few developing countries. Government revenues do provide us with a proxy of national capacity, however, and indicate which countries could have finances available to fund DRR. Differences between countries are again evident. Some have significant revenues based on very large economies, such as Indonesia, Iran and India. Some are major commodity producers, while others are at the other end of the scale – such as Liberia, Afghanistan and Burundi, each of which features in the top five for aid dependency on a yearly basis. Government revenues range from the US$267.6 billion of India in 2010 to the US$253 million of Sierra Leone. The figures for revenues per person tell a different story than simple volumes. Lebanon generates nearly US$2,000 from each of its citizens, followed closely by Bosnia-Herzegovina, Angola and Algeria. Some of the top 40 humanitarian recipients generate very little revenue at all per person. Fourteen countries produced less than US$100 per person in 2010. On the positive side, growth in government revenue has been remarkably robust, even during the worst of the financial crisis. Of those 14 countries that generated less than US$100 per person in government revenue, only four did not see growth in revenues year on year from 2006 to 2010. Only in 2009 were there marked drops in revenues in many countries, and most of those were major commodity producers – Algeria, Iraq, Sudan and Angola – which suffered due to the sharp dip in prices from 2008 peaks. Thirteen countries suffered no decline at all over the five years. 20 The relationship between government revenues and mortality risk is revealing. Of those same 14 countries that generated less than US$100 of revenue per person in 2010, two are ranked as major risk (level nine) for mortality, one is ranked as very high (eight) and five are ranked as medium-high (six). All these countries have significant risk of mortality from natural disaster, but have very little revenue available to reduce risk. The level of funding available for DRR is obviously key. It would be expected that the greater the government revenue in general, the less DRR financing would have to come from external sources such as international aid. The picture for DRR funding when compared against government revenues is very mixed, however. We do not see, for example, that those 14 countries with less than US$100 per person in revenue are the ones that have received the bulk of DRR funding over the decade. In fact, only two of them received over US$100 million in that period (Ethiopia and Bangladesh). The rest received less than US$50 million. Some of these countries, those in sub-Saharan Africa in particular, have received almost no DRR funding at all, despite the risks and their lack of revenue. For example, Sierra Leone received just US$3.8 million over the decade, Eritrea US$1.3 million, Côte d’Ivôire US$1.2 million and Guinea US$2.4 million. Table 10: Government revenues in top 40 humanitarian recipients Country Lebanon 2005 (US$bn) 4.5 2006 (US$bn) 5.0 2007 (US$bn) 5.7 2008 (US$bn) 6.9 2009 (US$bn) 8.4 2010 (US$bn) 8.3 Mortality Risk 5 Govt Rev per person 2010 DRR (US$m) 1,955 3.1 Bosnia-Herzegovina 4.7 5.6 6.8 8.1 7.3 7.2 5 1,925 1.3 Angola 12.4 21.0 27.7 42.8 23.3 36.0 4 1,896 3.0 Algeria 38.2 50.1 53.2 80.3 50.7 62.3 7 1,760 0.2 Iraq 0.0 33.6 45.0 66.4 44.8 55.3 5 1,757 53.7 Serbia 9.9 11.4 16.4 19.5 16.4 15.0 5 1,523 1.9 Colombia 37.0 42.8 56.5 61.1 61.0 69.7 9 1,505 4.4 Iran 45.7 66.4 87.8 89.3 82.9 97.5 8 1,299 1.8 Thailand 39.6 45.8 52.9 58.0 54.7 65.9 5 968 2.8 Jordan 3.2 4.4 5.0 5.6 5.9 6.0 3 923 0.5 Georgia 1.2 2.0 2.9 3.5 2.9 3.0 5 716 23.0 Syria 7.9 8.5 9.2 10.6 12.9 13.1 5 580 1.5 Indonesia 53.9 72.6 81.9 107.4 87.4 119.1 9 512 754.6 Sri Lanka 3.4 4.4 4.8 5.6 5.5 6.8 5 332 110.5 Sudan 4.2 7.5 9.3 12.4 8.3 10.3 6 239 15.9 147.4 174.9 240.6 249.0 226.6 267.6 9 220 639.4 Zambia 1.3 1.8 2.1 2.7 2.0 2.9 4 217 82.8 Haiti 0.6 0.6 0.9 1.0 1.2 1.9 6 191 64.7 India Kenya 4.0 4.7 6.0 6.6 6.7 7.8 5 190 143.4 Zimbabwe 0.9 0.5 0.2 0.1 0.9 2.2 5 174 16.5 Chad 0.4 0.9 1.6 2.2 1.1 1.9 5 169 7.3 Pakistan 15.2 18.0 21.5 23.9 23.5 24.8 8 134 1,002.3 DRC 0.8 1.1 1.5 2.1 1.9 2.5 7 115 15.1 Mozambique 0.9 1.1 1.3 1.6 1.8 1.9 6 82 46.9 Liberia 0.1 0.1 0.2 0.2 0.2 0.3 5 80 1.2 Tanzania 1.6 1.8 2.1 3.0 3.4 3.6 5 80 6.7 Guinea 0.4 0.4 0.6 0.7 0.8 0.7 5 70 2.4 Bangladesh 5.6 5.9 6.4 7.2 8.8 11.5 9 70 371.4 Cote d’Ivôire 2.8 3.2 3.8 4.4 4.3 4.5 5 67 1.2 Nepal 0.7 0.7 0.9 1.1 1.3 2.0 6 66 23.5 Uganda 1.1 1.2 1.5 1.8 2.0 2.1 6 62 38.0 Afghanistan 0.0 0.6 0.7 0.8 1.3 1.7 8 60 49.8 Eritrea 0.3 0.3 0.3 0.3 0.2 0.3 4 54 1.3 Ethiopia 1.8 2.2 2.5 3.2 3.9 4.2 6 49 140.7 Myanmar 0.7 1.0 1.3 1.7 1.8 2.5 9 49 19.5 Sierra Leone 0.2 0.2 0.2 0.2 0.2 0.3 6 43 3.8 Burundi 0.2 0.2 0.2 0.2 0.2 0.3 5 34 10.2 More than 20% growth 0 to 20% growth 0 to -10% growth -10 to -20% growth Less than -20% growth Notes: 1) This table does not include DPRK, Palestine/OPT or Somalia due to lack of data on government resources. 2) Government revenues are made by a tax component and a non-tax one (revenue coming for example from sovereign wealth funds and state-owned enterprises and corporations). They include also fees, fines and mineral and resource rights. Sources: IMF Regional Outlooks (2011), IMF World Economic Outlook (2011), OECD DAC and UNISDR6 21 IN FOCUS: UNDERSTANDING NATIONAL SPENDING ON DRR Table 11: Reported national resources set aside for DRR by top 40 humanitarian recipients % allocated from national budget Mozambique Allocated from overseas development assistance fund (US$m) Allocated to hazard proofing sectoral development investments (US$m) Allocated to stand alone DRR investments (e.g. DRR institutions, risk assessments, early warning systems) (US$m) Disaster proofing post-disaster reconstruction (US$m) 5.2% 317.2 538.1 28.2 14.8 5% 394.5 1.6 0.1 – – – 17.0 1.1 Zambia Colombia 0.1% Bangladesh 4.5% Pakistan – Sri Lanka 2.6% – 1.5 – – 3.6 – 10.0 0.6 102.2 63.8 6.7 67 Source: Indicator 1.2: The Hyogo Framework for Action dedicated and adequate resources available to implement disaster risk reduction plans and activities at all administrative levels7 Obtaining comparable data from national governments on how much they are investing in DRR is at present considerably challenging. Those governments that have signed up to the Hyogo Framework for Action are supposed to declare the financing they have committed for DRR, as dictated by one of their indicators. However, of the top 40 recipients of humanitarian financing, only a handful have reported funding in 2009–2011. Whether this is a reporting issue or is due to lack of financing is difficult to gauge. Indonesia, which unlike the other countries above, reported no financial data to the HFA during the three years, has a well-developed DRR programme and a detailed breakdown of finances. Its planned DRR expenditure of US$7.5 billion over five years comes from a variety of sources, both government and private sector, domestic and international. Table 12: Budget for Indonesia’s National Disaster Management Plan, 2010-2014 Indonesia’s National Disaster Management Plan Budget (US$bn) Enhancement of regulatory framework and institutional capacity 3.6 Integrated disaster management planning 0.0 Research, education and training 0.0 Capacity building and improvement of people’s and stakeholders’ participation in DRR 0.3 Disaster prevention and mitigation 0.8 Early warning system 0.1 Preparedness 0.9 Emergency response 0.1 Rehabilitation and reconstruction 1.7 Total 7.5 Note: Presentation given by Dr. Suprayoga Hadi from the Indonesian Ministry of National Development Planning at a workshop on the Tracking of DRR and Recovery Investment Data with International Aid in 2011; sources of funding include the government, foreign loans, foreign grants and the private sector. Source: Republic of Indonesia, National Disaster Management Plan, 2010-20148 22 THE INADeQUACY AND INEQUITY OF DRR FINANCING INEQUITIES ACROSS SIMILAR CONTEXTS Unequal spending across countries is not necessarily indicative of inequity, especially given the range of countries amongst the top recipients of humanitarian assistance. Further analysis against population figures reveals, however, that considerable inequity does exist. Table 13: Comparisons based on DRR funding per person in top 40 humanitarian recipients DRR per person (average 2000-2009 population) Zambia Haiti Pakistan Sri Lanka Georgia Kenya Indonesia Bangladesh Mozambique Afghanistan Iraq Ethiopia Burundi Uganda Zimbabwe Somalia Palestine/OPT Nepal Lebanon Chad Sierra Leone India Sudan Myanmar Bosnia and Herzegovina Liberia DPRK Eritrea Guinea DRC Serbia Angola Tanzania Colombia Jordan Syrian Arab Republic Côte d’Ivôire Thailand Iran Algeria Average proportion People Affected each year Number of Disasters Mortality Risk 7.10 6.94 6.14 5.68 5.13 4.04 3.44 2.45 2.28 2.05 1.92 1.90 1.39 1.34 1.32 1.20 1.14 0.87 0.77 0.74 0.74 0.57 0.42 0.41 0.34 3.6% 1.2% 0.8% 3.0% 1.6% 3.1% 0.5% 4.8% 4.8% 2.4% 0.0% 3.9% 4.3% 1.0% 6.8% 6.7% 0.0% 1.1% 0.0% 3.6% 0.0% 5.3% 2.0% 0.6% 0.9% 21 40 67 24 8 53 152 84 45 83 9 45 31 37 21 40 1 27 3 22 9 186 36 14 14 4 6 8 5 5 5 9 9 6 8 5 6 5 6 5 5 2 6 5 5 6 9 6 9 5 0.34 0.34 0.29 0.26 0.26 0.19 0.18 0.17 0.10 0.09 0.08 0.06 0.04 0.03 0.01 1.6% 0.7% 3.9% 0.3% 0.1% 0.0% 0.6% 1.5% 1.0% 0.3% 0.7% 0.0% 4.3% 0.3% 0.1% 11 13 3 16 64 4 35 39 50 4 5 13 54 53 36 5 6 4 5 7 5 4 5 9 3 5 5 5 8 7 Note: Colour-coded cells indicate where those countries are in the top ten for the various elements. Sources: Development Initiatives based on OECD DAC, CRED and ISDR 23 Again, analysis reveals a concentration of DRR financing within a few countries. Over the decade the top five recipients of DRR per person received as much as the other 35 countries combined. This inequity is particularly worrying given how unrelated it appears to be to various proxies of investment need. Only one of the top ten countries for DRR financing – Bangladesh – made it into the top ten for number of people affected, number of disasters and mortality risk. Meanwhile, only two of the top ten countries ranked by people affected are also top ten recipients of DRR financing. Only five of the top ten ranked by number of disasters and four of the top ten ranked by mortality risk are recipients of DRR financing. There is a question to be asked about whether Zambia, Georgia and Sri Lanka are appropriate priorities for funding, given that they do not appear in any of the top ten proxies of need. This may be somewhat misleading, however, given that overall funding levels are so low. The US$7.1 per person spent on DRR per person in Zambia over the decade is hardly an indicator of significant priority being given to the reduction of risk. A question that should be asked, but that appears to have no clear answer at present, is how much is an appropriate level of investment in DRR, and on what should that investment be based? Zimbabwe and Somalia have had a yearly average of 6.8% and 6.9% respectively of their populations affected by natural disasters. Burundi has had 4.3%. Yet none of these countries are in the top ten for DRR funding. Mortality risk would appear to drive more DRR financing, with Pakistan, Indonesia, Bangladesh and Afghanistan, all ranked high-risk, in the top ten. However, Myanmar (ranked as a major mortality risk), and DRC (high) have received very little funding at all, just 41 cents and 26 cents per person respectively over the decade. Humanitarian inequity Development Aid inequity Table 14: Average amount of humanitarian assistance per person, 2000-2009 Table 15: Average amount of development assistance per person, 2000-2009 Country Average HA per person Country Average ODA per person Palestine/OPT 1,680.0 Palestine/OPT 4,466.1 Lebanon 431.5 Bosnia and Herzegovina 1,766.5 Somalia 296.6 Jordan 1,367.8 Liberia 269.1 Lebanon 1,301.5 Jordan 229.3 Serbia 1,247.0 Sudan 223.6 Iraq 1,206.0 Afghanistan 202.2 Afghanistan 1,182.7 Eritrea 192.1 Georgia 1,021.9 Iraq 180.2 Liberia 846.1 Burundi 157.8 Zambia 812.1 Sources: OECD DAC and UN population figures (downloaded November 2010) 24 Sources: OECD DAC and UN population figures (downloaded November 2010) General Inequity of Funding Similar inequities exist elsewhere within aid financing. The people of Palestine/OPT received US$1,680 per person over the decade, four times as much as in Lebanon, the country ranked second in terms of average humanitarian assistance per person. The disparity is similarly evident for development aid, where the Palestinian population is once more ranked first for financing per person: US$4,466 was received, more than 2.5 times higher than second-placed Bosnia-Herzegovina. OVERALL FIGURES MASK INADEQUACIES: BY COUNTRY Table 16: Number of people affected by natural disasters in Ethiopia by disaster type The generally inadequate level of investment in DRR, with year on year volumes dominated by single projects, is further revealed through an analysis of funding trends in individual countries. ETHIOPIA DRR PROFILE YEAR Drought Epidemic Flood Volcano 2000 - 7,033 30,000 - 2001 - 8,166 39,500 - 2002 - - 4,000 - 2003 12,600,000 - 110,000 - 2005 2,600,000 964 242,418 9,000 2006 - 32,848 434,050 - 2007 - - 245,386 2,000 2008 6,400,000 3,134 115,595 - 2009 6,200,000 13,652 - - 2010 - 967 80,700 - 2011 4,805,679 - - - Total 32,605,679 66,764 1,301,649 11,000 Ethiopia has been in and out of the disaster headlines ever since the famine of the mid-1980s. A massive drought in 2003 affected more than 12 million people and the 2011 famine was a reminder that underlying issues have still not been resolved. More regular though less severe natural disasters include the yearly flooding, which usually affects more than 100,000 people, and volcanic eruptions. The country has 65 volcanoes, 25 of which have 100,000 people living within a 30km radius.9 Mount Erta Ale erupted in 2005, displacing thousands of people. Source: EMDAT CRED Funding for Ethiopia was US$140.7 million over the decade (making it the 6th largest recipient), but its large population made it only 12th highest in terms of per person funding. Despite the major droughts, and the country being a major recipient of development aid, the proportion of ODA for DRR has been only 0.6%. The 2007 peak in funding is explained by a World Bank “productive safety net” programme of US$51 million, which was designed to help Ethiopians be more resilient to drought and to respond to it more effectively. This accounted for 86% of total DRR spending in that year. The 2009 peak was largely made up of a single US$16 million donation from Canada through the World Food Programme (WFP), for a mix of food provision and resilience. A low overall figure therefore is compounded by a DRR financing trend that starts off as nonexistent and then varies considerably, but this may be in part due to the availability of data. If the World Bank project was stretched over several years, the variability in DRR trends may be reduced. Figure 16: DRR funding and natural disasters in Ethiopia MAJOR DROUGHT US$ MILLION (CONSTANT 2009 PRICES) 2008/2009 2005 DROUGHT, FLOODING AND MOUNT ERTA ALE ERUPTION DROUGHT 70 14,000,000 60 12,000,000 50 10,000,000 40 8,000,000 30 6,000,000 20 4,000,000 10 2,000,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 DRR spending 2009 NUMBER OF AFFECTED 2002/2003 0 Number of affected Source: Development Initiatives based on OEDCD DAC and EMDAT CRED 25 HAITI DRR PROFILE Table 17: Disaster numbers in Haiti Haiti suffers from multiple natural hazards that almost always become disasters. Though the earthquake of 2010 had one of the greatest impacts of any disaster anywhere over the last decade, typhoons and flooding are more regular threats to Haitians. The earthquake appears very obviously in Haiti’s overall disaster profile. In terms of number of people affected, the country ranked 8th highest in 2010, pushing it into 30th position over the 11 years. The earthquake, the disaster that killed the most people of any single event over all 11 years, pushed it into first place for number of deaths (prior to 2010 it ranked 17th). Though Haiti received only the ninth highest volume of DRR funding over the decade, this was equivalent to the second highest (US$6.94) per person, although once more this was a very poor proportion of aid – just 1.3% over the decade. Like Ethiopia, Haiti received little or no funding at all before a major disaster struck, with the 2004 storms prompting a very small investment of US$1.5 million in 2005. Funding in 2007, 2008 and 2009 suggested a general increase, even before the massive earthquake of 2010. However, even funding of US$30 million 2000-2010 Rank Number of disasters 47 26/201 Average number of disasters per year 4 – Number affected 4,775,265 30/201 Average affected 434,115 – Number of deaths 234,846 1/201 Average number of deaths 21,350 – Source: EMDAT CRED in 2009 was hardly enough preparation for a country like Haiti with such a high mortality risk and such low government capacity and revenues. Despite the threat of an earthquake being high in and around the capital Port-au-Prince, and the risk of death and damage being massive, there was little evidence of any significant investment in reducing the risk in urban areas. Figure 17: DRR funding to Haiti against disaster impact US$ MILLION (CONSTANT 2009 PRICES) 2010 2008 HURRICANE JEANNE SERIES OF POWERFUL HURRICANES MASS DESTRUCTION IN PORT-AU-PRINCE FOLLOWING RICHTER 7 EARTHQUAKE 35 4,000,000 30 3,500,000 3,000,000 25 2,500,000 20 2,000,000 15 1,500,000 10 1,000,000 5 0 500,000 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 DRR spending Note: 2010 DRR data not available. Source: Development Initiatives based on OECD DAC and EMDAT CRED 26 NUMBER OF AFFECTED 2004 2010 0 Number of affected ZAMBIA DRR PROFILE Zambia is regularly affected by flooding in particular, with major disasters in 2001, 2007 and 2009. There was also a major drought in 2005. Despite not being ranked high either for the proportion of its population affected, the number of disasters or its mortality risk, Zambia received the eighth highest volume of DRR over the decade and was placed first in terms of per person funding. However, Zambia’s financing trend over the decade is worrying, perhaps even more so than the funding to Ethiopia and Haiti, which is generally rising. There was a sudden and initial high figure of more than US$30 million in 2003, followed by another high year with close to US$25 million in 2004. This was almost totally accounted for by a World Bank emergency drought recovery project that incorporated resilience and disaster management components. However, there were much lower DRR figures in 2005 and 2006. For the remainder of the decade there was hardly any DRR financing at all, despite repeated natural disasters. Table 18: Number of people affected by disaster type in Zambia Drought Epidemic Flood 2000 - 1,224 12,000 2001 - 425 617,900 2003 - 3,835 - 2004 - - 196,385 2005 1,200,000 7,615 4,000 2006 - 105 - 2007 - 115 1,553,521 2008 - 8,312 5,000 2009 - 5,198 614,814 2010 - - 1,200 Total 1,200,000 26,829 3,014,820 Source: EMDAT CRED 2004 2008 2010 HURRICANE JEANNE SERIES OF POWERFUL HURRICANES MASS DESTRUCTION IN PORT-AU-PRINCE FOLLOWING RICHTER 7 EARTHQUAKE Figure 18: DRR funding and disaster impact in Zambia 4,000,000 2001 OVER ONE MILLION SUFFER FOOD SHORTAGES OWING TO DROUGHT FLOODING 35 1,800,000 1,600,000 30 1,400,000 25 1,200,000 20 1,000,000 15 800,000 600,000 10 400,000 5 0 NUMBER OF AFFECTED US$ MILLION (CONSTANT 2009 PRICES) 2007 2005 FLOODS AND DROUGHT 200,000 2000 2001 2002 2003 2004 2005 2006 2007 2008 DRR spending Source:2004 Development Initiatives based on OECD DAC and EMDAT CRED HURRICANE JEANNE 2009 0 Number of affected 2010 27 OVERALL FIGURES MASK INADEQUACIES: BY DISASTER TYPE Flooding: More deaths but much less financing Flooding is one of the most regular and damaging natural disasters. The top 40 humanitarian recipients include many countries that suffer regular annual flooding, such as Bangladesh, Mozambique, Sudan and Pakistan. Unsurprisingly then, only four of the top 40 humanitarian recipients make it into the top ten for overall flood-related funding, even though they include seven of the top ten by numbers of people affected by floods. Five countries make it into the top ten for flood prevention control even though they are not in the top ten for floodaffected people (Sri Lanka, Cambodia, Guyana, Yemen, Indonesia). The data on the effects of flooding in these countries compared with the flooding-related DRR expenditures received is revealing. Although the top 40 recipients regularly account for the highest proportion of deaths due to flooding (even before the massive Pakistan flood of 2010) and in many years for a sizeable proportion of the people affected, funding in general goes to countries outside the top 40. Only in 2000 and 2001 have the top 40 received more than 50% of all flood-related DRR funding. For the years 2003 to 2009, the figure never crept above 38%. In 2005 it was only 6.7%. Table 19: The top 10 countries affected by flooding and funding for flood prevention and control in total ODA, 2000-200910 Numbers affected by flooding Flood prevention and control funding (US$m) China 525,638,382 383.3 India 238,290,253 46.2 Bangladesh 58,601,793 33.6 Thailand 13,026,034 1.3 United States 11,317,806 0 Vietnam 10,961,710 136.8 Pakistan 9,562,092 13.1 Cambodia 6,644,182 29.9 Mozambique 6,212,111 27.3 Colombia 4,488,063 0.8 Note: Figures are in US$ million constant 2009 prices. Sources: OECD DAC and EMDAT CRED US$ MILLION (CONSTANT 2009 PRICES) Figure 19: Comparing funding for flood prevention and control and flooding impact between the top 40 HA recipients and all other recipients 300 250 85% 76% 75% 67% 80% 73% 63% 63% 39% 150 63% 36% 34% 50% 41% 40% 30% 29% 50 0 70% 60% 52% 100 20% 7% 11% 2000 2001 2002 Sources: OECD DAC and EMDAT 28 90% 62% 64% 200 73% 80% 2003 2004 2005 2006 2007 2008 2009 10% 0% Flood prevention funding to top 40 recipients Flood prevention funding to all other recipients Proportion of deaths in top 40 countries Proportion of affected in top 40 countries 100% Flooding in Bangladesh and Pakistan US$227m 90% Figure 21: Breakdown of DRR funding for Bangladesh and Pakistan US$969m 80% Between 2000 and 2011 Bangladesh and Pakistan had the highest numbers of people affected by flooding of all countries (excluding China and India). They accounted for 41% of all those affected by floods over this period, or four out of every ten people. 70% 60% 100% 90% Pakistan had only 9.6 million people affected up until 2010, but the massive floods in that year pushed its total to 35.7 million for the period as a whole. In Bangladesh over this period a total of 58.6 million people were affected. The 2004 flooding in Bangladesh was particularly severe, worse even than the 2010 flooding in Pakistan, with more than 60% of the country covered with water and 37 million people affected. The need for DRR investments connected to flooding is therefore quite obvious. However, the funding response from international donors has not been at all substantial. Of Bangladesh’s US$371.4 million of DRR funding over the decade, only US$91 million, or just 25%, was directly related to flooding. For Pakistan the figure is much worse. Despite receiving over US$1 billion for DRR, the highest of any of the top 40 recipients over the decade, only 2% of this, just US$16 million, was directly related to flooding. US$227m 50% US$969m 40% 80% 30% 70% 20% 60% 10% 50% 0% 40% 30% 0% US$53m Bangladesh US$17m US$16m Pakistan US$91m 20% 10% US$91m US$53m Bangladesh US$17m US$16m Total DRR in all other ODA, 2000-2009 Funding to flood prevention, 2000-2009 Total DRR funding within HA, 2004-2009 Pakistan Total DRR in all other ODA, 2000-2009 Funding to flood prevention, 2000-2009 Note: Figures are in constant 2009 prices. Source: Development Initatives based on OECD DAC Total DRR funding within HA, 2004-2009 Figure 20: Number of people affected by flooding in Pakistan and Bangladesh NUMBER OF AFFECTED 40,000,000 35,000,000 Bangladesh number affected by flooding Pakistan number affected by flooding 30,000,000 25,000,000 20,000,000 15,000,000 10,000,000 5,000,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Source: EMDAT CRED 29 CONCLUSION: SPENDING WHERE IT SHOULD COUNT A number of key lessons can be drawn from a detailed examination of financing for DRR in relation to the contexts that need it most. The first and most obvious one is that the data itself, although revealing of trends, needs considerable improvement in order to give a true picture of DRR financing. Secondly, the general balance of assistance appears to be wrong. Only 19 of the top 40 humanitarian recipients are also in the top 40 for overall development aid. This suggests that too many countries are missing out on the long-term development financing that is considered a prerequisite for DRR in our bifurcated model of aid. Worryingly, in the year with the best data available (2009), seven out of every ten dollars spent on DRR came from humanitarian funds. This would suggest that the pressure on humanitarian financing is unlikely to diminish. Major humanitarian recipients are particularly affected by natural disasters. Many of them are poor countries and many of them are affected by conflict. They are extremely vulnerable to shocks. Despite this, funding for DRR is very weak indeed, with only US$3.7 billion out of a total US$363 billion of ODA to the top humanitarian recipients going to reduce disaster risk. This low level of funding is compounded by concentration in a few recipient countries, and detailed country investigation reveals at best variable, and at worst worryingly sporadic, trends over time. Placing DRR in context reveals considerable inequities across similar country contexts, with funding seemingly going towards countries that have substantial resources of their own; arguably these are the very countries that do not need international DRR assistance. Funding for DRR does not appear to be directed logically to those countries that need it most; it is not based on the number of disasters, the risk of mortality or the proportion of people affected each year. Neither does funding for DRR seem to be obviously related to country revenues, with some of the richer countries receiving considerable funding and many of the poorest ones receiving almost nothing over an entire decade. Though these country comparisons are important, as clearly there are questions to be asked about DRR priorities, it is, however, the almost negligible investment in DRR overall that is the core issue. The first priority is surely to generate more financing in general, while the second should be the creation and implementation of a financing model based on a proper and cross-country assessment of need. A final note must be made on the quality of data. That which is available is revealing about the priority, or rather lack thereof, given to DRR. Considerable improvements must be made in the quality of reporting, making sure that this reflects the complexity of DRR programming, in order for a full picture to be drawn and, crucially, for the right decisions to be made. A key element of this is to ensure that country context remains paramount and that international funding for DRR is allocated with that context in mind, and seen as only one part of the overall solution to reducing risk. 30 POSTSCRIPT: DRR IN THE CONTEXT OF CONFLICT PEOPLE AFFECTED BY DISASTERS OFTEN LIVE IN CONFLICT-AFFECTED COUNTRIES Conflict plays a major role in the need for humanitarian assistance, and unsurprisingly the top 40 recipients of such assistance have suffered or continue to suffer from conflict of one kind or another. From 2000 to 2009, 28 of these 40 countries suffered at least seven years of conflict. Four of them suffered from nine years of conflict and ten experienced conflict for the entire decade. Only six countries had no year of conflict at all. This is particularly significant because of the way in which natural disasters have a particular impact in countries affected by conflict. In each year from 2005 to 2009, conflict-affected countries accounted for more than 50% of people affected. In some years, such as 2006 and 2008, the proportion was around 80%. Increasingly, attention is being given to the interplay between natural disasters and unnatural conflicts. Whether there is a systematic attempt to reduce conflict risk in the same way as DRR is open to question. Certainly there is no equivalent in conflict risk reduction of the Hyogo Framework for Action for DRR that binds coordinating actors, implementing agencies and national governments together to a framework of commitments. Funding for conflict risk reduction is largely limited to specific post-conflict scenarios that involve donor Table 20: Top 10 recipients of conflict prevention funding11 Conflict prevention (US$m) Number of years in conflict over the last 10 Mortality Risk Afghanistan 2,266.4 10 8 Iraq 2,146.0 10 5 Serbia 1,071.8 7 5 Sudan 803.5 10 6 Timor-Leste 557.6 6 6 Bosnia-Herzegovina 541.6 6 5 DRC 456.3 10 7 Liberia 307.1 10 5 Angola 262.6 7 4 Colombia 254.7 10 9 Note: Figures are in US$ million, constant 2009 prices. Sources: Development Initiatives’ methodology for conflict-affected, OECD DAC and UNISDR governments. For example, Afghanistan and Iraq accounted for 25% of all conflict prevention expenditure to the top 40 recipients of humanitarian assistance over the decade. What is particularly significant is again the number of countries that have roughly similar profiles, and that suffer from both natural disasters and conflict, that appear to have very different levels of funding for DRR and conflict prevention. There seems to be no obvious pattern, no clear indication of funding related to need and no clear setting of priorities. Figure 22: Location of people affected by natural disasters 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 18.6 8.8 56.5 28.3 40.2 12.3 32.2 58.1 49.8 37.6 Non conflict-affected (excl. China) Conflict-affected 2005 2006 2007 2008 2009 Sources: Development Initiatives’ methodology for conflict-affected and EMDAT CRED for disaster-affected 31 10.0% 2,500 8.0% 2,000 6.0% 1,500 4.0% 1,000 2.0% 500 0.0% -2.0% Bosnia-Herzegovina Eritrea Cote d'Ivôire Liberia Jordan Algeria 0 Pakistan Indonesia India Bangladesh Kenya Ethiopia Sri Lanka Zambia Haiti Iraq Afghanistan Mozambique Uganda Nepal Georgia Myanmar Zimbabwe Sudan DRC Burundi Somalia DPRK Chad Tanzania Colombia Palestine/OPT Sierra Leone Lebanon Angola Thailand Guinea Serbia Iran Syria US$ MILLION (CONSTANT 2009 PRICES) Figure 23: Investments in conflict prevention and DRR from the top humanitarian recipients % Conflict ODA incl. DPP % DRR ODA Conflict prevention All DRR within ODA Source: Development Initiatives based on OECD DAC Disaster-Affected/Conflict-Affected countries in sub-Saharan africa The 2011 Horn of Africa drought and famine provide an example of the interplay between natural disaster and conflict. Each of the five countries most affected suffers from yearly natural disasters, and each has suffered or is currently suffering from conflict. Humanitarian funding has been significant for all countries and overall aid has been particularly high for Ethiopia and Uganda. Funding to reduce risk has been minimal, however, with only US$288.6 million spent on DRR and US$418.4 million on conflict prevention. Table 21: Total aid, DRR and conflict prevention expenditure against disaster impact and risk in Horn of Africa countries RISK Number of years conflictaffected Multiple Mortality Risk Class Eritrea 9 Ethiopia AID DRR expenditure Humanitarian Assistance (US$m) Total ODA (US$m) All DRR within ODA (US$m) 4 170,700 841.3 2,768.7 0.9 0% 71.7 3% 10 6 2,909,775 4,730.8 22,209.2 123.4 1% 70.2 0% Kenya 5 5 1,092,237 1,641.1 9,515.2 122.9 1% 46.7 0% Somalia 10 5 556,457 2,451.9 3,590.3 9.2 0% 119.1 3% Uganda 10 6 285,839 1,503.9 14,154.0 32.2 0% 110.7 1% 11,169.0 52,237.4 288.6 418.4 1% Totals % ODA to DRR Notes: Financial figures are in constant 2009 prices. Sources: Development Initiatives based on OECD DAC, CRED, UNISDR 32 Conflict prevention Average no. affected Conflict prevention (US$m) % ODA to conflict prevention ANNEX: DATA AND GUIDES METHODOLOGICAL NOTES The semantics of DRR Figure 24: How DRR works • Capacity building for response • Emergency management including contingency planning Response • Risk assessments • Legislative frameworks and policies • Support to national governments Emergency preparedness Recovery and reconstruction • Build back better Dis a Disaster risk reduction r ste Prevention ri s k m an ag Re em sil e nt ie n Climate change adaptation • Mitigation ce Note: The concentric circles denote connections between the various elements of DRR, resilience, emergency preparedness, disaster risk management, response, recovery and reconstruction, prevention and climate change adaption. Source: Development Initiatives report on emergency preparedness prepared for FAO on behalf of the IASC Methodology for tracking DRR in the OECD DAC CRS The OECD DAC Creditor Reporting System (CRS) provides probably the most accurate and easiest way of tracking funding for DRR, as it reports both development and humanitarian aid. The project-level data in the CRS allows particular issues to be examined through sector codes and project descriptions, albeit only through a forensic examination. In the CRS, funding to a sector relevant to each individual aid activity is recorded using a five-digit purpose code. The coding is intended to identify the specific areas of the recipient’s economic or social development that the transfer intends to foster. A purpose code for an element of DRR has existed since 2004; this falls within humanitarian aid under “disaster prevention and preparedness” (DPP). The data reported under the DPP code (74010) can be easily accessed. However, to search for DRR activities within development and humanitarian programmes (not coded 74010), the short and long project descriptions were used in this report. As searching each individual project description is very time consuming, 29 key terms (in four languages) were used to find projects with elements of DRR. A coding macro using visual basics was developed to search both long and short descriptions with the identified terms. 33 The search terms were selected from recent literature on DRR and the websites of key DRR-focused organisations (e.g. UNISDR). After running the macro, project descriptions were scanned and those not related to DRR were removed. Column filters on the descriptions were then searched using the same terms as the coded macro to identify any potential DRRrelated activities that had not been picked up. All funding reported to the flooding prevention/control development purpose code (41050) was included in the final estimate of DRR. All OECD DAC data used in this report is expressed in constant 2009 prices and was downloaded in April 2011. Data used for disaster statistics (numbers of natural disasters, affected numbers, deaths and estimated damages) The Centre for Research on the Epidemiology of Disasters (CRED) maintains the International Disaster Database. For a disaster to be entered into the database, at least one of the following criteria must be fulfilled: ten or more people reported killed; 100 people reported affected; a declaration made of a state of emergency; or a call for international assistance. The main sources for events listed are UN agencies, but information also comes from national governments, insurance organisations and the media. The total number of people affected by an event includes those who suffered physical injuries or illness, as well as those made homeless or who otherwise required immediate assistance during a period of emergency. The economic damage figures for disasters are considered to be underestimates, as only a third of reported disasters estimate economic losses, often due to problems relating to damage assessments. Conflict Prevention in DAC reporting Conflict prevention and resolution, peace and security within ODA includes: security system management and reform; civilian peace-building; conflict prevention and resolution; post-conflict peace-building (UN); reintegration and small arms and light weapons (SALW) control; land mine clearance and child soldiers. BASIC CONCEPTS AND DEFINITIONS Constant prices Gross domestic product (GDP) Constant (real terms) figures show how expenditure has changed over time, after removing the effects of exchange rates and inflation. DAC deflators, along with annualised exchange rates, are available at: www.oecd.org/dac. The base rate year used by the DAC during 2011 was 2009. The total market value of goods and services produced by workers and capital within a nation’s borders. Development Assistance Committee (DAC) The DAC is the Development Assistance Committee of the OECD. Its members are: Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Japan, Korea, Luxembourg, the Netherlands, New Zealand, Norway, Portugal, Republic of Korea, Spain, Sweden, Switzerland, the United Kingdom, the United States and the European Commission. These members have “agreed to secure an expansion of aggregate volume of resources made available to developing countries and to improve their effectiveness”. 34 Official development assistance (ODA) For the purposes of this report, development aid refers to ODA reported to the OECD DAC. ODA is a grant or loan from an ‘official’ (government) source to a developing country or multilateral agency for the promotion of economic development and welfare. It is reported by members of the OECD DAC according to strict criteria each year and by a small number of donors outside of the OECD DAC group, who typically report a less comprehensive dataset. It includes sustainable and povertyreducing development assistance (for sectors such as governance, growth, social services, education, health, and water and sanitation) as well as funding for humanitarian crises. ENDNOTES AND REFERENCES 1 I nternational Monetary Fund, GDP figures come from the Economic Outlooks database of the IMF. http://www.imf.org/external/pubs/ft/weo/2011/02/weodata/index.aspx 2 ugoslavia has been excluded from the list of top 40 humanitarian recipients for the Y purposes of this report. Thailand has an overall negative figure of ODA over the tenyear period due to debt repayments. 3 oorest Areas Civil Society Programme (PACS), Drought in India: Challenges and P Initiatives 2001-2008. http://www.empowerpoor.org/downloads/drought1.pdf 4 ultidimensional Poverty Index, Oxford Poverty and Human Development Initiative. M http://www.ophi.org.uk/policy/multidimensional-poverty-index/ 5 he CHFs fund planned humanitarian response for the following year based on T strategic planning and identification of needs that include an element of risk analysis and the likely event of an emergency occurring, and therefore address the need to prepare for that emergency. The ERFs are much smaller in scale compared with CHFs and support short-term projects of up to six months’ duration. 6 or countries with no data on government revenues excluding grants, a different F methodology has been applied. Grants (data from the OECD DAC) have been subtracted from general government revenues. Data for Haiti refers to government revenues including grants. 7 yogo Framework for Action (HFA) priority for action 1.2 (2009-2011), dedicated and H adequate resources are available to implement disaster risk reduction plans and activities at all administrative levels. http://www.preventionweb.net/english/hyogo/ progress/documents/hfa-report-priority1-2(2009-2011).pdf 8 I ndonesian Ministry of National Development Planning, presentation given by Dr. Suprayoga Hadi from the Indonesian Ministry of National Development Planning at a workshop on the Tracking of DRR and Recovery Investment Data with International Aid in 2011. Sources of funding include the government, foreign loans, foreign grants and the private sector. http://www.preventionweb.net/english/professional/trainingsevents/events/v.php?id=19655 9 spinall, W., Auker, M., Crosweller, S., Hincks, T.K., Mahony, S., Nadim, F., J. Pooley A and Sparks, R.S.J., Syre, E. (2011), Volcano Hazard and Exposure in Track II Countries and Risk Mitigation Measures - GFDRR Volcano Risk Study, Bristol University Cabot Institute and NGI Norway for the World Bank. 10 he OEC DAC defines flood prevention and control activities as ‘Floods from rivers or T the sea; including sea water intrusion control and sea level rise-related activities’. http://www.oecd.org/dataoecd/28/62/38429349.pdf 11 onflict prevention and resolution, peace and security as defined by the C OECD DAC; see DAC statistical reporting directive. http://www.oecd.org/ dataoecd/28/62/38429349.pdf 35 ACKNOWLEDGEMENTS This report would not have been possible without the input and support of a number of people. We would like to thank Daniel Kull of the World Bank for his review of the report. Thanks also to Diane Broadley for her work on design and David Wilson for proofreading. Internal contributors included Georgina Brereton (coordinating production), Hannah Sweeney, Daniele Malerba and Tim Strawson. 36 Disaster Risk Reduction Spending where it should count In this report we examine the top 40 humanitarian recipient countries in the context of natural disasters and especially with regard to financing to reduce risk. We highlight how prevalent disasters are in these countries, and their particularly significant impact. Beyond this, we examine the current state of funding for disaster risk reduction and, in the context of those countries most at risk of natural disaster, ask questions about the volume and type of funding, and its equity. Are the right choices being made? Global Humanitarian Assistance (GHA) is a Development Initiative which aims to improve the efficiency, effectiveness and coherence of humanitarian response by further increasing access to reliable, transparent and understandable information on the aid provided to people living in humanitarian crises. Global Humanitarian Assistance Development Initiatives, Keward Court, Jocelyn Drive, Wells, Somerset BA5 1DB, UK Tel: +44 (0)845 618 2939 Fax: +44 (0)1749 676721 Email: gha@devinit.org www: globalhumanitarianassistance.org Development Initiatives is a group of people committed to eliminating poverty. Follow us on Twitter @devinitorg