2016-global-report-internal-displacement-IDMC

GRID Global Report ON2016 INTERNAL DISPLACEMENT

ACKNOWLEDGEMENTSProject director: Alexandra BilakProject coordinator: Elizabeth J. RushingAuthors: Alexandra Bilak, Gabriel Cardona-Fox, Justin Ginnetti, Elizabeth J. Rushing, Isabelle Scherer, Marita Swain,Nadine Walicki, Michelle YonetaniData analysis and methodological annex: Ali Anwar, Cheng Boon Ong, Gabriel Cardona-Fox, Justin Ginnetti, MayeulKauffmann, Luisa Meneghetti, Williams Ojo, Isabelle Scherer, Wilhelmina WelschWith special thanks to the International Organization for Migration (IOM): Nuno Nunes, Global CCCM Cluster Coor-dinator, and Anthony Sequeira at Geneva headquarters, as well as country staff in Afghanistan, Armenia, Burundi,Cameroon, China, Croatia, Dominican Republic, Ethiopia, Haiti, Honduras, India, Indonesia, Iran, Iraq, Kenya, Malawi,Mali, Mozambique, Namibia, Pakistan, Peru, Philippines, South Sudan, Sri Lanka, Timor Leste, Uganda, Vanuatu,Yemen, Zimbabwe for displacement data and analysis.IDMC also extends its thanks to those governments that provided up-to-date and detailed displacement information.Peer review: Ahmednur Abdi, Vlatko Avramovski, Biruktawit Belay, Michael Cernea, Jérôme Elie, Melissa Fernandez,Christopher Gaffney, David Garcia, Kelly Gilbride, Jacopo Giorgi, Dessalegn Gurmessa, Miloon Kothari, Geeta Kuttipa-rambilgeeta, Daniel Macguire, Erin Mooney, Cerianne Robertson, Eliana Ruedes, Victoria Stodart, Duncan Sullivan,Andy Tatem, Christian Visnes, Martin WyndhamEditor: Jeremy LennardDesign and layout: Rachel NataliCover photo: An older woman walks through the rubble of her neighbourhood in Kathmandu, Nepal. Two majorearthquakes in April and May 2015 killed 8,700 people, damaged or destroyed more than 712,000 houses anddisplaced more than 2.6 million people. Photo © www.gavingough.comWith thanksIDMC’s work would not be possible without the generous contributions of its funding partners. We would like tothank them for their continuous support in 2015, and we extend particular gratitude to the following contributors:Australia’s Department of Foreign Affairs, Liechtenstein’s Ministry of Foreign Affairs, Norway’s Ministry of ForeignAffairs, Sweden’s International Development Cooperation Agency, Switzerland’s Federal Department of ForeignAffairs, the UN Refugee Agency (UNHCR), the UK’s Department for International Development and the US Agency forInternational Development (USAID).

Global Report ONINTERNALDISPLACEMENTMay 2016

New displacements associated withConflict and violence (Total: 8.6 million)Disasters (Total: 19.2 million) Ukraine Conflict: 942,000 Egypt Conflict: 78,000 Libya Conflict: 100,000 United States SudanDisasters: 63,000 Conflict: 144,000 Mexico Dominican Rep. Niger Disasters: 91,000 Disasters: 28,000 Conflict: 47,000 Disasters: 38,000 Colombia Venezuela Chad Conflict: 224,000 Disasters: 45,000 Conflict: 36,000 Brazil Guinea Disasters: 59,000 Disasters: 34,000 Paraguay Nigeria Disasters: 171,000 Conflict: 737,000 Disasters: 100,000 Cameroon Conflict: 71,000 Central African Rep. Conflict: 210,000 Uruguay Dem. Rep. Congo Disasters: 24,000 Conflict: 621,000 Disasters: 106,000 Argentina Chile Disasters: 36,000 Disasters: 1m

conflict and disasters in 2015 Iraq Afghanistan Pakistan Conflict: 1.1m Conflict: 335,000 Disasters: 1m Disasters: 23,000 Disasters: 71,000 Bangladesh Disasters: 531,000 Syria China JapanConflict: 1.3m Disasters: 3.6m Disasters: 486,000 Nepal Taiwan Disasters: 2.6m Disasters: 27,000 Yemen Myanmar Philippines Malaysia India Disasters: 1.6m Conflict: 2.2m Conflict: 33,000 Disasters: 21,000 Disasters: 83,000 Disasters: 3.7m Disasters: 2.2m Vanuatu Somalia Disasters: 65,000 Conflict: 90,000 Disasters: 59,000 Sri Lanka Disasters: 66,000 Ethiopia Indonesia Conflict: 56,000 Disasters: 204,000 Disasters: 104,000 S. Sudan Conflict: 199,000 Kenya Disasters: 105,000 Madagascar Disasters: 87,000 Burundi MozambiqueConflict: 23,000 Disasters: 61,000 Malawi NOTE: For both types of displacement, the number is shown only when it exceeds 20,000. The size of the pie charts is fixed for estimates of 5,000 orDisasters: 343,000 less. In a few cases, the same person may be displaced more than once.

Table of Contents || Foreword. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 || Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 Part 1 On the GRID: Internal displacement in 2015. . . . . . . . 7 || Key findings and messages . . . . . . . . . . . . . . . . . . . . . . . . . . .8 || New displacement: People who fled conflict and violence in 2015. . . . . . . .9 Displacement associated with conflict on the rise. . . . . . . . . . . . . . . . . . . 9 The Middle East: Home to half of all new IDPs . . . . . . . . . . . . . . . . . . . . 9 Displacement associated with conflict in the rest of the world. . . . . . . . . . . . 10 || Yemen: An overlooked crisis with no end in sight . . . . . . . . . . . . . . . . . 11 || New displacement: People who fled disasters in 2015. . . . . . . . . . . . . 14 Disasters caused twice as many new displacements as conflict. . . . . . . . . . . . 14 South and east Asia dominate the figures again. . . . . . . . . . . . . . . . . . . 15 || Nepal: Obstacles to protection and recovery . . . . . . . . . . . . . . . . . . . 17 Small countries, big impacts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Developing countries hardest hit . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Exposure to weather extremes. . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 || El Niño: Aggravated risk associated with extreme weather. . . . . . . . . . . . . 22 Large-scale geophysical hazards: few in number, many displaced. . . . . . . . . . 24 || Total headcount: People living in displacement as of end 2015 . . . . . . . . 27 Protracted crises swell unprecedented numbers . . . . . . . . . . . . . . . . . . . 27 || DRC: Multiple and chronic displacement. . . . . . . . . . . . . . . . . . . . . . 28 Sub-Saharan Africa: Chronic displacement keeps figures stubbornly high . . . . . . 29 Invisible IDPs: protracted displacement following disasters. . . . . . . . . . . . . . 29 || For richer or poorer: Protracted displacement in Japan and Haiti. . . . . . . . . . 30 Part 2 Inside the GRID: Caveats for the estimates. . . . . . . . 33 || Key findings and messages . . . . . . . . . . . . . . . . . . . . . . . . . . 34 || IDMC’s data model: Capturing the human toll of displacement . . . . . . . . 35 Conservative estimates for displacement associated with conflict . . . . . . . . . . 35 Data gaps for displacement associated with disasters. . . . . . . . . . . . . . . . 36 || Syria: Trapped in the country, and out of the picture . . . . . . . . . . . . . . . 37 || Working with decaying data. . . . . . . . . . . . . . . . . . . . . . . . . 39 || Decaying displacement data in Colombia. . . . . . . . . . . . . . . . . . . . . 412 G20R1ID6

PART 3Off The Grid: The world’s overlooked IDPs. . . . . . . . 43|| Key findings and messages . . . . . . . . . . . . . . . . . . . . . . . . . . 44|| Displaced by criminal violence. . . . . . . . . . . . . . . . . . . . . . . . . 45 Mexico and Central America: a million displaced by organised crime. . . . . . . . . 45 Unseen and in displacement limbo. . . . . . . . . . . . . . . . . . . . . . . . . . 45 Piecemeal data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 || Honduras: Unearthing a hidden displacement crisis . . . . . . . . . . . . . . . . 47 Falling through the cracks: data gaps and their consequences. . . . . . . . . . . . 49|| Displaced by drought-related disasters . . . . . . . . . . . . . . . . . . . . 50 Complex causality under drought conditions. . . . . . . . . . . . . . . . . . . . . 50 || Defining drought. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Linkages to other hazards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 The voluntary to forced migration continuum . . . . . . . . . . . . . . . . . . . . 52 Improving data collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 || Drought-related disasters in 2015/2016 . . . . . . . . . . . . . . . . . . . . . . 55 || Ethiopia: Extreme conditions, extreme measures. . . . . . . . . . . . . . . . . . 56|| Displaced by development projects . . . . . . . . . . . . . . . . . . . . . . 58 || Development projects that displace people. . . . . . . . . . . . . . . . . . . . 59 Overlooked and unprotected . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Startling estimates despite incomplete data . . . . . . . . . . . . . . . . . . . . . 60 Displacement undermines development gains. . . . . . . . . . . . . . . . . . . . 62 || Brazil: Olympic Games preparations displace thousands in Rio de Janeiro . . . . . 63|| Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65|| Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67|| Methodological annex. . . . . . . . . . . . . . . . . . . . . . . . . . . . 74|| Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Table 1: New displacement by country for disasters and conflict and total number of IDPs for conflict and violence. . . . . . . . . . . . . . . . . . 96 Table 2: Largest disaster-related displacement events of 2015 . . . . . . . . . . . 100 Table 3: Conflict-related displacement in 2015. . . . . . . . . . . . . . . . . . . 102 Global report on internal displacement 2016 3

Foreword Much focus has been placed on the hundreds of displacement becoming more protracted and of thousands of refugees, asylum seekers and more of a development challenge. migrants who have put their lives at risk to reach European shores. Their bravery and despair has To take some of these considerations into drawn much attention to the phenomenon of account, we are presenting our estimates of displacement. In reality though, they represent internal displacement in 2015 in a radically new only the tip of an iceberg. Of those the Syrian way, with figures on people displaced by conflict, conflict has uprooted, around 6.6 million people by violence and by disasters in a single report. have been displaced internally. Away from the media glare and out of reach of humanitarian The Global Report on Internal Displacement agencies, many struggle to survive in subhuman (GRID) aims to provide a more holistic picture of conditions. the phenomenon, regardless of cause. In time for the World Humanitarian Summit in Istanbul, There are now twice as many internally displaced it also aims to highlight displacement as a multi- people (IDPs) as refugees worldwide. In some dimensional challenge that must involve humani- ways, the distinction between internal and cross- tarian, sustainable development, peace-building, border flight is unhelpful in a globalised world. disaster risk reduction and climate change adap- Large crises such as Syria should lead to a new tation work. and more holistic thinking about displacement. It also discusses types of displacement that A similar shift is needed in analysing the causes receive too little attention, such as that associ- and consequences of displacement. We tend to ated with generalised criminal violence, gradu- think in terms of single, isolated triggers, but ally-evolving crises such as drought, and develop- the reality is far more complex. Displacement ment projects. in Sudan ostensibly caused by conflict has been traced back to root causes such as drought and This year’s GRID is an important body of evidence, environmental degradation, and a food crisis but it is not the complete picture. We can only that became a famine because of government be as good as our data, so it also constitutes an neglect and changing regional demographics. In appeal for those who collect it to redouble their Haiti, the establishment of overcrowded informal efforts to provide comprehensive and up-to-date settlements and authorities’ inability to enforce information on all displaced populations. building and safety standards formed the back- drop to the mass displacement caused by the Behind our data lie millions of blighted human 2010 earthquake. lives. IDPs often lose everything when they flee, and the trauma and upheaval of displacement Such complexity has profound implications leave many with deep psychological and phys- when it comes to preventing, responding to ical scars. Our ultimate aim remains the same, and resolving displacement. Failure to conduct to improve understanding of their plight and a thorough assessment means responses will be ensure that efforts to protect, assist, consult and fragmented at best and ineffective at worst. empower the displaced are better resourced and targeted. A comprehensive approach must address political factors, and improve resilience to a range of risks Jan Egeland, so people do not have to flee in the first place. Secretary General of the This is development and governance work. When Norwegian Refugee Council displacement becomes inevitable, humanitarians attend to more immediate needs, but they must work with the development sector if sustainable solutions are to be achieved. There is a clear trend4 G20R1ID6

SummaryFor the first time, IDMC is publishing its estimates Disasters displaced around 19.2 million peopleand analysis of people internally displaced by across 113 countries in 2015, more than twice theconflict, generalised violence and disasters in a number who fled conflict and violence. Over thesingle report. This new publication, the Global past eight years, a total of 203.4 million, or anReport on Internal Displacement (GRID), presents average of 25.4 million displacements have beenour knowledge of the phenomenon more faith- recorded every year. As in previous years, southfully and constitutes a significant step in our and east Asia dominated in terms of absoluteefforts to paint as complete a picture as possible. figures, but no region of the world was unaf-Future iterations will go further still. fected. India, China and Nepal had the highest numbers, with 3.7 million, 3.6 million and 2.6Part 1 of the report covers displacement that is million respectively. The vast majority of displace-already “on the GRID”. During 2015, there were ment took place in developing countries, and the27.8 million new displacements associated with populations of small island countries were hitconflict, violence and disasters in 127 countries. hard relative to their size. The devastation cycloneThis is roughly equivalent to every man, woman Pam wrought on Vanuatu is a case in point.and child in New York City, London, Paris andCairo grabbing what they could carry and fleeing Part 2 of the report takes our readers “inside thetheir homes in search of safety. GRID” and IDMC’s generic displacement model. In this part we outline our efforts to improve theInternal displacement associated with conflict coverage and transparency of the global evidenceand violence has been on an upward trend since base on internal displacement. By providing the2003. There were 8.6 million new cases during breakdown of the age of our figures for the2015, or an average of 24,000 a day. Some first time in this report, we are appealing to the4.8 million people were newly displaced in the governments concerned and to our partners inMiddle East alone, significantly more than in the the field to contribute to this ongoing effort.rest of the world combined. Yemen, Syria andIraq accounted for over half of the total. Else- Part 3 of the report explores displacement whichwhere, Ukraine, Nigeria, Democratic Republic until now has been “off the GRID”. Global figuresof the Congo (DRC), Afghanistan, Colombia, do not capture other contexts in which peopleCentral African Republic and South Sudan had flee their homes, and this year we look at threethe highest numbers. often overlooked drivers – criminal violence, drought and development projects. We discussIn terms of total headcount, there were 40.8 IDMC’s initial efforts to estimate the numbermillion IDPs worldwide as a result of conflict of people they displace, and some of the issuesand violence at the end of 2015 – an increase of inherent in doing so.2.8 million on 2014, and the highest figure everrecorded. It is also twice the number of refu- As the global monitor of internal displacement,gees in the world. Just ten countries accounted we intend to expand our provision of knowledgefor over two-thirds of the total, or around 30 with the aim of advancing global commitmentsmillion people. Colombia, DRC, Iraq, Sudan and to reduce the risks and impacts of displacementSouth Sudan have featured in the list of the ten and find lasting solutions for the millions of IDPslargest internally displaced populations every year worldwide. Our ability to do so will depend onsince 2003. There were no total global figures for the breadth and strength of our partnerships,people still displaced by disasters, but a sample of and on states’ continued commitment to supportcases in 2015 identified hundreds of thousands these efforts.living in some form of protracted displacement. Global report on internal displacement 2016 5

Thousands of displaced people seek shelteroutside the UNAMID base in Um Baru, North Darfur. Photo: UNAMID/Hamid Abdulsalam, January 2015.

Part 1On the GRID Internal displacement in 2015There were 27.8 million new displacements in 127 Part 1 of this year’s report presents estimates forcountries during 2015, roughly the equivalent of new displacements associated with conflict andthe populations of New York City, London, Paris violence during 2015, and those associated withand Cairo combined. Of the total, 8.6 million disasters over the same period.were associated with conflict and violence in 28countries, and 19.2 million with disasters in 113 It also reports on the “total headcount”, or thecountries. overall number of people internally displaced as a result of conflict and violence as of the end of2015 was, tragically, another record year for the year, including those who fled in previousinternal displacement associated with conflict and years. Unfortunately such cumulative data is notviolence. The increases recorded during the year available for people displaced by disasters, so itwere driven primarily by the waves of violence is not possible to report a total headcount in thethat continued to spread across the Middle East same way.following the 2010 uprisings known as the Arabspring. New displacements were recorded in all Figure 1.1 Total number of people internally displaced byregions of the world, but by far the worst-affected conflict and disasterscountry was Yemen, where 2.2 million people – oreight per cent of the population – fled their homes New Total numberand sought refuge within the country’s borders. displacements of IDPs at Jan – Dec 2015 end 2015There were 19.2 million new displacements asso-ciated with disasters brought on by rapid-onset CONFLICT 8.6 40.8natural hazards in 2015, more than twice as many million millionas for conflict and violence. The vast majority ofthis displacement was caused by extreme weather DISASTERS 19.2 ?events such as storms and flooding, but the April millionand May earthquakes in Nepal, which forced 2.6million to flee their homes, were a stark reminderof the potential of geophysical hazards to precipi-tate mass displacements.As in previous years, south and east Asia and Note to our readersthe Pacific regions were worst-affected bydisplacement associated with disasters, and the In this section, we refer to “incidents” and “cases” ofvulnerable, coastal populations of small island displacement rather than “people displaced”, becausedeveloping states (SIDSs) were disproportionately some people will have been displaced more than once.affected again. Low and middle income countries When we use the term “people displaced” explicitly, itwere hardest hit across the world as a whole. should be taken to mean single incidents or cases affect- ing one person. ON THE GRID: Global internal displacement in 2015 7

Key findings and messages ||There were 27.8 million new displacements ||Weather-related hazards triggered 14.7 in 127 countries during 2015, more than the million displacements in 2015. Despite the total populations of New York City, London, onset of an El Niño episode, this was lower than Paris and Cairo combined. the annual average of 21.5 million since 2008. New displacement: conflict ||4.5 million displacements were brought on by large-scale geophysical hazards. This was higher ||There were 8.6 million new displacements than the annual average, driven up by the Nepal associated with conflict and violence in 28 earthquake disaster. countries in 2015. Total headcount ||This type of displacement is on the rise, with an average of 5.2 million incidents a year since ||There were 40.8 million people internally 2003. This equates to 14,000 people forced to displaced worldwide as a result of conflict and flee every day. violence as of the end of 2015. This represents an increase of 2.8 million from our 2014 esti- ||Displacement in the Middle East and north mate and the highest figure ever recorded. It is Africa has snowballed since the Arab spring twice the number of refugees in the world. uprisings in 2010 and the rise of the Islamic State (also known as ISIL or ISIS). There were ||As of December 2015, three-quarters of the nearly 4.8 million new displacements in 2015, world’s internally displaced people (IDPs), or 30 significantly more than in all other regions of million people, were located in ten countries. the world combined. Yemen, Syria and Iraq Five of them – Colombia, DRC, Iraq, Sudan accounted for over half of the global total. and South Sudan – have featured in the list of the ten largest displaced populations every ||Outside the Middle East, the countries with year since 2003. the highest numbers of people fleeing were Ukraine, Nigeria, Democratic Republic of the ||Of the other five countries, displacement in Congo (DRC), Afghanistan, Colombia, Central Nigeria, Syria, Ukraine and Yemen is largely African Republic (CAR) and South Sudan. the result of conflicts that erupted or escalated within the last five years. There is a risk these New displacement: disasters conflicts will become intractable, leaving the people they have displaced, who make up more ||There were 19.2 million new displacements than 30 per cent of the overall global esti- associated with disasters in 113 countries, mate, to face a future of protracted and more than twice as many as for conflict and possibly multiple displacement. violence. Over the past eight years, 203.4 million displacements have been recorded, an ||At nearly 12 million, the number of people average of 25.4 million each year. displaced by conflict in sub-Saharan Africa has flat-lined over the last decade, underlining the ||South and east Asian regions, countries and chronic nature of displacement in the region. events again dominated in terms of the highest absolute figures, but no region of the world ||There are no figures for the total number of was unaffected. India, China and Nepal people displaced by disasters as of the end accounted for the highest numbers, with 3.7 of 2015. However, among a sample of 34 million, 3.6 million and 2.6 million respectively. ongoing cases documented in 2015, there were hundreds of thousands of people identi- ||Most displacement associated with disasters in fied as living in protracted displacement 2015 took place in developing countries. for periods ranging between one and 26 years.8 G20R1ID6

New displacementPeople who fled conflict and violence in 2015There were 8.6 million new displacements associ- The Middle East: Home to halfated with conflict and violence in 28 countries of all new IDPsduring 2015 (see global map, cover page fold-out). Displacement in the Middle East and north Africa has snowballed since the wave of social upris-Displacement associated with ings known as the Arab spring in late 2010 andconflict on the rise the rise of the Islamic State (also known as ISIL or ISIS). The region accounted for the highestThe average number of people displaced each numbers of people fleeing violence in 2015 byyear by conflict and violence has risen over the a wide margin. There were nearly 4.8 millionlast 13 years (see figure 1.2). Our figures show an new displacements during the year, significantlyaverage of 5.2 million displacements a year since more than in all of the other regions of the world2003, which equates to 14,000 people forced to combined (see figure 1.3).flee their homes every day. Figure 1.3: New displacements associated with conflict andCalculated over the last five years, the average violence by World Bank-defined region in 2015rises to 7.6 million a year, or more than 20,000people a day. This increase correlates with find- Latin America and East Asia and Pacificings that conflict and violence intensified world- the Caribbean 49,000wide between 2008 and 2015.1 231,000 South Asia 336,000 Europe and central Asia 942,000 Sub-Saharan Middle East and Africa north Africa 2.2m 4.8mFigure 1.2: New displacements associated with conflict andviolence, 2003 to 2015 12Millions 10 8 6 4 2 - 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 ON THE GRID: Global internal displacement in 2015 9

Figure 1.4: Countries with most new displacements associated with conflict and violence in 2015 Yemen 1.3 2.2 Syria Iraq 1.1 2.5 0.9 Millions Ukraine Nigeria 0.7 0.6 DRC Afghanistan 0.3 0.2 Colombia 0.2 CAR 0.2 South Sudan 0.0 0.5 1.0 1.5 2.0 Yemen, Syria and Iraq accounted for the bulk of such as Ramadi. Three governorates – Anbar, new displacement in the region, and more than Baghdad and Dohuk – host nearly half of the half of the global total. They were also the top country’s IDPs.2 As the crisis deepens, the likeli- three countries worldwide in terms of the number hood of IDPs being able to return to their homes of new displacements in 2015 (see figure 1.4). has diminished and they are increasingly unable to meet their own needs. Many have become Yemen had the highest number, largely the result reliant on dwindling public and others’ resources, of Saudi-led airstrikes and an economic blockade a point of growing tension between IDPs and imposed on the civilian population (see Yemen their host communities.3 spotlight). Nearly 2.2 million men, women and children were forced to flee their homes during Displacement associated with the year, a 20-fold increase on the 2014 esti- conflict in the rest of the world mate of new displacements. It was the worst year to date for displacement in the country, and Outside the Middle East and north Africa, the the needs of those affected and the rest of the countries with highest numbers of people fleeing civilian population are acute. the effects of conflict and violence in 2015 were Ukraine, Afghanistan, Colombia and a number In Syria, there were 1.3 million incidents of of sub-Saharan African nations. new displacement during 2015. The figure is an increase of 18 per cent from the 2014 estimate, Now in its second year, the conflict that broke and the humanitarian situation in the country out in eastern Ukraine continued to cause signifi- deteriorated significantly during the year. The civil cant loss of life, suffering and mass displacement. war is now in its sixth year, with four of the five Insecurity continued in a number of areas along permanent UN Security Council members actively the contact line between government forces and engaged in the hostilities, and it has caused one pro-Russian separatist groups, despite the signing of the largest displacement crises since World of the Minsk II agreements in February 2015. War Two. A renewed ceasefire in September to enable the Neighbouring countries have increasingly start of the school year has largely held. There restricted the flow of people trying to leave Syria has been a marked reduction in clashes and or have closed their borders altogether, leaving shelling, bringing partial relief to people who hundreds of thousands of families trapped in the had been living with the threat of violence for country and living as IDPs (see Syria spotlight). many months.19 These positive political and secu- rity developments have done little, however, to More than 1.1 million people were displaced convince people that they are safe in their homes. in 2015 by Iraq’s most recent wave of violence, There were more than 942,000 incidents of new which began in early 2014 with the rise of the displacement during the year, up nearly 50 per Islamic State (also known as ISIL or ISIS) and cent on our 2014 estimate. other non-state armed groups, and has included intense counter-insurgency operations in cities10 G20R1ID6

Yemen SPOT LIGHTAn overlooked crisis with no end in sightThe political and security situation in Yemen dete- tions, however, is predominantly the result ofriorated dramatically in 2015, and the ensuing sea, land and air blockades of commercial andhumanitarian crisis shows few, if any, signs of humanitarian imports.10abating. Violence displaced eight per cent ofthe country’s population, or 2.2 million people, Taizz, Amran and Hajjah governorates hadduring the year – more than in any other country the highest number of IDPs, between themin the world – and people fled their homes in all accounting for 900,000.11 Many more peoplebut one of its 22 governorates.4 Humanitarian may have wanted to flee, but found a range ofand protection needs among IDPs and the rest of physical, economic and social obstacles preventedthe civilian population are acute, and neither the them from doing so.national nor international response have gone farenough in addressing them. Flagrant disregard for international humanitarian and human rights law, and indiscriminate warfareThe only unaffected governorate, Socotra, was that has targeted civilians and civilian infrastruc-hit by two freak tropical cyclones that traversed ture, were the main triggers of displacement. Thethe Gulf of Yemen in November. Between them, destruction of infrastructure including hospitals,cyclones Megh and Chapala forced 56,000 schools, markets, shops and water supplies haspeople to flee their homes across three gover- left 82 per cent of Yemen’s population in neednorates, and by the end of the month around of humanitarian assistance, including 14.4 million23,000 people remained displaced.5 people who suffer from food insecurity.12 Calls for an independent international human rightsBy the end of the year, people were living in monitoring and reporting mechanism have gonedisplacement in every region of Yemen, an unheeded.13unprecedented situation in the country’s history. The majority of IDPs live in overcrowded rentedHumanitarian needs were already acute before accommodation, schools and other public spaces,the conflict escalated in March 2015. Yemen is or tents and other forms of makeshift shelter.14one of the poorest countries in the Arab world. They face a wide range of protection needs andIt has few natural resources, weak governance vulnerabilities including lack of shelter options,and social services, high youth unemployment lack of safety and security, harassment, lack ofand almost 50 per cent of its population of 26.8 livelihood options, gender-based violence, loss ofmillion lived below the poverty line in 2014.6 Half documentation, food insecurity and limited accessof the population, of whom 70 per cent live in to healthcare, education, water and sanitation.rural areas, had no access to safe drinking waterand three-quarters no access to safe sanitation.7 Displacement has also forced many families toGender inequality is widespread. Yemen has separate, and there are large numbers of unac-ranked last in the World Economic Forum’s annual companied minors.15 IDPs have few livelihoodglobal gender gap report in each of the ten years options, and most are dependent on charity orit has been published.8 humanitarian assistance for survival.16The escalation of violence made human suffering Cyclones Megh and Chapala brought the equiv-and the country’s displacement crisis significantly alent of five years’ of rainfall to Hadramaut,worse during the year. As of the end of 2014, Socotra and Shabwa governorates in just twothere were around 334,000 people displaced.9 days, leading to flash floods and widespreadBy the end of 2015, the figure had increased devastation.17 More than half of those who fledmore than seven-fold to more than 2.5 million. their homes returned within a month, and theThe upsurge in violence has largely been attrib- majority of the 22,970 people still displaced wereuted to the Saudi-led military intervention in the living with host families or in rental accommoda-conflict. The sharp deterioration in living condi- tion.18 ON THE GRID: Global internal displacement in 2015 11

Internally displaced The warring parties have enforced import and Given that there are no prospects for return orchildren play in front movement restrictions, which have led to scarce other durable solutions in sight, there is high riskof a school in Sana’a commodities, fuel shortages and price hikes, and that displacement will become protracted andhosting some 200 IDPs, left the country with little or no capacity to rebuild IDPs’ resilience to future shocks, including envi-mostly from Saada or repair damaged homes and infrastructure. ronmental hazards, will be further compromised.governorate. The damage the cyclones inflicted is relativelyPhoto: Charlotte Cans/ small compared with destruction the conflict hasOCHA, June 2015 caused, but people in Yemen are highly vulner- able to such shocks, which compounds the risks they already face. The international media and political discourse have widely overlooked the human narrative and widespread suffering in Yemen, and there has been little political resolve to stop the violence and improve humanitarian access. As a result, and because of competing crises in the region such as Syria, the response is seriously under- funded. Even if a stalled peace process bears fruit in 2016, economic and political recovery will take many years.12 G20R1ID6

In Nigeria, Boko Haram continued to launch Colombia’s long-running armed conflict, andattacks and commit atrocities throughout 2015. violence perpetrated by groups the governmentIn the six years since the onset of its insurgency, identifies as criminal gangs, triggered more thanthe group and military operations against it have 224,000 cases of new displacement in 2015.24forced more than a million people to flee their Most of the violence takes place in rural areas,homes and fuelled an unprecedented humani- but IDPs tend to flee to the country’s large andtarian crisis in the north-east of the country and medium-sized cities, where they take refuge inthe wider Lake Chad region. slums and shanty towns. Here, they and the rest of the urban poor face sporadic bomb attacks,Displacement trends show that as the military threats, killings, forced recruitment, gender-pushed Boko Haram back during the year, people based violence, abductions and other forms ofpreviously trapped by the militants moved to harassment and intimidation. These cause furtherurban centres in search of humanitarian assis- displacement, but the number of people affectedtance.20 Inter-communal clashes fuelled by ethno- is difficult if not impossible to gauge.25religious feuds, criminality, cattle rustling, landdisputes and tensions between pastoralists and Improved security in some areas of CAR duringfarmers also continued to flare across the Middle the first half of the year allowed many IDPs toBelt region, but were largely overlooked. There return to their homes. The situation remainedwere 737,000 incidents of new displacement in volatile, however, with violence in September andthe country during the year. October sparking nearly 210,000 new displace- ments and forcing some people to seek shelter inOngoing conflict and extreme violence also sites that were in the process of closing.26 Manyplagued the DRC, where there was an average IDPs were trapped in enclaves, some controlledof more than 50,000 cases of new displacement by warlords, where they were beyond the reachevery month in 2015. The quarterly average has of state authorities, humanitarians and Frenchdeclined since 2013, but population movements and UN peacekeepers.27could rise significantly again in 2016 if the polit-ical and security situation were to deteriorate.21 Despite the signing of a peace agreement in South Sudan in August 2015, civilians in allIn Afghanistan, the withdrawal of most of the ten of the country’s states continued to sufferinternational troops present in the country coin- violence, which caused the displacement of morecided with a rise in displacement, with more than 199,000 people during the year. Fightingthan 335,000 new incidents reported. This was erupted in new areas in the second half of thedriven by an increase in violence by non-state year, including Western and Central Equatoria.armed groups and counter-insurgency opera- Inter-communal violence, some sparked by cattletions by national and the remaining international rustling, caused further displacement.28security forces. Such operations have increas-ingly involved the use of mortars, rockets andgrenades in populated areas.22 Significant newdisplacement in Kunduz province and continuinginstability in Helmand and central areas of thecountry have aggravated a displacement crisisthat has been ongoing since 2001.23 ON THE GRID: Global internal displacement in 2015 13

New displacement People who fled disasters in 2015 There were 19.2 million new displacements population of Brazil. The 2015 figure is lower associated with disasters in 113 countries across than the average recorded since 2008 of 25.4 all regions of the world in 2015, brought on million displacements per year, but similar to the by events such as floods, storms, earthquakes, reported global totals for the previous year (see volcanic eruptions, wildfires, landslides and figure 1.5).29 extreme temperatures (see global map, cover page fold-out). The scale and frequency of large-scale disasters account for much of the variance in the total Disasters caused twice as figures from year to year. Such events are less many new displacements as predictable because they are brought on by the conflict most extreme hazards. As in 2014 and 2009, there were no mega-events (defined here as more than three million displacements) in 2015. Disasters triggered by natural hazards caused twice as many new displacements in 2015 as conflict and violence. Over the past eight years, there have been 203.4 million displacements by disasters, a figure comparable to the entire Figure 1.5: New displacements associated with disasters by scale of events, 2008 to 2015 Mega events (> 3 million displaced) Total, 2008 to 2015 Very large events (1 – 3 million displaced) 18.7 Large events (100,000 – 999,999 displaced) (9%) Very small to medium events (fewer than 100,000 displaced) 61.8 45 42.3 (30%) 60.8 Millions 203.4m (30%) 40 62.2 (31%) 36.6 32.3 35 30 15.0 26.2 25 16.7 15.0 13.0 22.2 Average, 25.4m/year 20 6.4 3.5 10.8 4.1 19.1 19.2 15 14.4 1.5 10 8.3 6.4 6.3 5.9 9.3 2.0 8.2 2.1 5 4.9 2009 7.5 2012 9.3 10.0 7.3 0 2.3 1.8 6.4 2011 2.5 3.2 2.6 2008 2.2 2013 2014 2015 201014 G20R1ID6

South and east Asia The earthquakes in Nepal in April and May, thedominate the figures again thousands of aftershocks that followed and the landslides they triggered left 712,000 homesAs in previous years, south and east Asian regions, and much infrastructure damaged or destroyed.countries and events dominated in terms of the The disaster took a heavy toll on the developinghighest absolute figures in 2015 (see figure 1.6). nation, affecting almost a third of the popula- tion and killing 8,700 people. Many of the 2.6Figure 1.6: New displacements associated with disasters by million who were displaced have been unable toWorld Bank-defined region, 2015 return to their homes, and recovery and recon- struction will take many years to complete (seeLatin America Sub-Saharan Africa Nepal spotlight).and theCaribbean 1.1m (6%) Unlike China and India, the number of people displaced in Nepal was also high relative to its1.5m (8%) Other population size (see figure 1.8, p.19). It recorded the third highest level of new displacement 0.3m (1%) worldwide in both relative and absolute terms.South Asia East Asia and As in previous years, multiple typhoons struck the the Pacific Philippines in 2015, with three of the strongest 7.9m storms displacing two million people. Typhoon (41%) 8.4m Koppu (local name Lando) was the most severe. (44%) It made landfall on Luzon, the country’s largest and most populous island, in October, killing 54India, China and Nepal accounted for the highest people, displacing around 938,000 and causingnumbers of people displaced, with totals of 3.7 severe crop damage. Typhoon Melor (localmillion, 3.6 million and 2.6 million respectively name Nona) forced 743,000 people to flee their(see figure 1.7, p.16). homes in the central regions of Bicol Peninsula and Romblan Islands in December, and typhoonIn India, the impact of two major flood and storm Goni (local name Ineng) displaced more thanevents were responsible for 81 per cent of the 318,000 in the north of the country in August.displacement, forcing three million people toflee their homes. Heavy rains and flash floods Floods, landslides and the impacts of cycloneassociated with a weak tropical cyclone that Komen displaced more than 1.6 million peopletracked across the Bay of Bengal in November in Myanmar in July and August, resulting in thedisplaced 1.8 million in the states of Tamil fifth highest figure worldwide in absolute termsNadu and southern Andhra Pradesh. Monsoon and the sixth highest in relative terms (see figureflooding associated with cyclone Komen, which 1.8, p.19). Twelve of the country’s 14 states andstruck neighbouring Bangladesh in late July, regions suffered widespread destruction. Thedisplaced 1.2 million, mostly in the northern and government declared the worst-affected statescentral states of West Bengal, Odisha, Manipur, of Chin and Rakhine, and the Magway andRajasthan and Gujarat. Sagaing regions, as disaster zones.Three large-scale typhoons and a flood disaster In Pakistan, two disasters accounted for almosttogether triggered 75 per cent of the displace- all of the displacement that took place duringment in China. Three typhoons, Chan-Hom, the year. A 7.5 magnitude earthquake struck theSoudelor and Dujan, struck four eastern provinces Hindu Kush mountains in October. Its epicentrebetween July and September, destroying homes, was in a remote area of northern Afghanistan,causing landslides and flooding and, between but north-west Pakistan suffered the worst of itsthem, displacing more than 2.2 million people. impacts. Nearly 666,000 people were displacedEarlier in the year, heavy rains and flooding in nine in Khyber Pakhtunkhwa (KP) province and Bajaursouthern and eastern provinces forced another in the Federally Administered Tribal Areas. Snow518,000 people to flee their homes in May. and rainfall over mountainous terrain left many of those affected isolated and acutely vulner- able. In August, flooding in northern and central KP, Punjab, Gilgit Baltistan, Chitral and Kashmir displaced 330,000 people. ON THE GRID: Global internal displacement in 2015 15

Figure 1.7: Countries with most new disaster displacements in 2015 (absolute number) South Asia East Asia and the Pacific Latin America and the Caribbean Sub-Saharan Africa 3,655,000 India 2,623,000 China 2,221,000 3,602,000 Nepal 1,618,000 Philippines 1,047,000 3.5 4 Myanmar 1,002,000 Chile 531,100 Millions Pakistan 486,100 Bangladesh 343,000 Japan Malawi 0.5 1 1.5 2 2.5 3 0 Cyclone Komen was the largest trigger of In Chile, an 8.3 magnitude earthquake struck the displacement in Bangladesh. It struck the south- Coquimbo region in September. This earthquake east of the country at the end of July, displacing and the subsequent tsunami warning led the 331,000 people. authorities to evacuate a million people. Chile is one of only two high-income countries among Asia bore the brunt of new displacements, but the ten with the highest levels of new displace- no region of the world was unaffected. Latin ment in 2015. The other is Japan, where 486,000 America and the Caribbean accounted for eight people were displaced by a range of hazards, per cent of the global figure, with 1.5 million new primarily typhoons Goni, Etau and Nangka, displacements, and sub-Saharan Africa six per which struck in close succession between mid- cent with just over a million. Chile and Malawi July and early September (see figure 1.7). were the only countries from these regions among the ten with the highest number of In southern Africa, seasonal floods in Malawi in people displaced in 2015 (see figure 1.7). January displaced the majority of the 343,000 people who fled their homes in 2015 – the world’s tenth-largest displacement in absolute terms – and caused widespread damage to agri- culture.16 G20R1ID6

Nepal SPOT LIGHTObstacles to protection and recoveryTwo major earthquakes in April and May 2015 hardship. Around 200,000 households were stilland thousands of associated aftershocks took a living in temporary shelters as of November at andevastating toll on the already fragile nation of altitude of over 1,500 metres.36Nepal. They affected almost a third of the country’spopulation of 28.2 million, killed 8,700 people, Since Nepal’s last major earthquake in 1934, onlydamaged or destroyed more than 712,000 houses the last decade has seen significant investmentand displaced more than 2.6 million people.30 They in disaster preparedness and coordination. Muchhit both urban and rural areas hard, and triggered of the focus, however, has been on preparinglandslides and avalanches in high mountain areas, communities in the Kathmandu valley, to therazing entire villages and leaving hundreds of thou- neglect of less populated but highly vulnerablesands of people with acute shelter, livelihood, towns and villages in higher mountain areas.37protection, food, water and education needs. Over the last four years the UK government, oneMost of those displaced stayed at least initially of Nepal’s major bilateral donors, has investedin makeshift or temporary shelters near their more than $30 million in earthquake resiliencedamaged or destroyed homes.31 Those who took programming.38 Despite such initiatives, the 2015shelter in open spaces or public buildings began earthquakes caused large-scale destruction. Theto return to their homes within a week of the delivery of humanitarian assistance was slow andfirst earthquake. Displacement patterns changed, vulnerable groups were not well protected.however, after the second earthquake. Manypeople who feared their homes were structur- The country’s national disaster response frame-ally unsound went back to open areas such as work, which the Ministry of Home Affairsfields. Many others fled the worst-affected rural adopted in 2013, established coordination mech-areas toward displacement sites in the densely anisms that were implemented in the aftermathpopulated urban areas of the Kathmandu valley.32 of the earthquakes.39Among the IDPs who sought shelter in collective Their impact was largely determined by theurban sites were many poor people, including poor quality of buildings unable to withstandDalits and female-headed households, whose the seismic activity, and their exposed locationhomes had been badly damaged or destroyed, on steep mountain slopes.40 The governmentand whose access to jobs and basic services had approved a national building code in 2003, butalso been disrupted. The vast majority did not when the earthquakes struck only 26 of 191own the property they lived in, and the earth- municipalities had begun implementing it.41quakes severely reduced the amount of rental Poverty, rapid urbanisation and weak institu-accommodation available and inflated rents. tional oversight have led to hasty and low-cost construction.42 Enforcement has also been under-Around June, IDPs in just under half the sites mined by corruption, judicial delays and a lack ofassessed by the International Organization for building inspectors.43Migration (IOM) said they had not receivedadequate assistance, and women and children in Political instability and weak institutions havearound half the sites felt unsafe.33 Sixty-eight per also hampered the humanitarian response.44cent of the IDPs surveyed said damage to their National and local government capacity had beenhomes was the main obstacle to their return. significantly weakened by civil war and moreOther issues included personal security, family than 12 years without local elections.45 Afterseparation, damage to infrastructure such as the end of the war in 2006, Nepal struggledroads and a lack of food.34 to adopt a federal constitution and key legisla- tion on disaster risk management was overdue.46In the absence of medium-term plans to settle After the earthquake, the stalled process of final-IDPs, some lived in fear of eviction from their shel- ising the constitution was expedited and it wasters, particularly in Kathmandu.35 The monsoon adopted on 20 September 2015.47and winter seasons that followed brought further ON THE GRID: Global internal displacement in 2015 17

Earthquake survivor Its content, however, sparked protests and a trade public buildings and infrastructure began its worksifts through the rubble blockade on the border with India that lasted with $4.1 billion in aid donations.55of her home in Baluwa, from September 2015 until February 2016.48Gorkha district. Photo: The blockade prevented fuel and other essential The government’s reconstruction policy prom-IOM, May 2015 supplies from entering Nepal, raised commodity ises financial compensation to every homeowner prices and created a rampant black market.49 whose house was damaged beyond repair. Fuel shortages complicated the delivery of much- Vulnerable IDPs, however, including squatters, needed shelter, food and medical supplies.50 undocumented citizens and owners without Delivery was further impeded by lengthy admin- formal title deeds risk exclusion from the policy, istrative procedures and new customs duties along with people whose homes were partially introduced just five weeks after the earthquake.51 damaged and those who lived in rented accom- modation. Despite progress in Nepal’s social legis- Feedback from affected communities on the lation in recent years, women, undocumented fairness of aid distribution and reporting by residents and refugees are also still denied equal international agencies pointed to discrimination inheritance and property rights.56 associated with a failure to recognise people’s specific needs and protection concerns according Rebuilding the country according to building to social caste, ethnicity, gender, disability and and safety standards is of vital importance if the age.52 risk of future disasters and displacement is to be reduced. The value of retrofitting existing build- Human rights advocates also raised concerns ings to make them resilient to earthquakes is well about nepotism and political favouritism, and proven.57 Reconstruction following such a major the exclusion of certain people or groups from disaster can be expected to take many years, and needs assessments.53 The failure to respect the is likely to be delayed because of a shortage of humanitarian principle of needs-based assistance technical and skilled labour, complex land issues made the risk of impoverishment for vulnerable and continuing political instability.58 groups worse.54 Careful land use planning, strengthened tenure Resolving IDPs’ ongoing needs in Nepal will and protection for vulnerable groups, along with require long-term financial and technical engage- support for broader strategies to reduce poverty ment by both federal and local governments and increase community resilience will also be and the international community. In December vital if progress is to made and sustainable solu- 2015, a new authority responsible for leading the tions for IDPs achieved. reconstruction of more than 500,000 homes,18 G20R1ID6

Small countries, big impacts Developing countries hardest hitConsidering new disaster-induced displacementrelative to population size highlights its significant As expected based on patterns in previous years,impact on Pacific small island developing states the vast majority of displacement associated with(SIDSs) including Tuvalu, Vanuatu, Micronesia and disasters in 2015 took place in developing coun-Kiribati (see figure 1.8). SIDSs face disproportion- tries, defined by the World Bank as low and lower-ately high disaster risk because their mostly low- middle income countries. High income countries,lying, coastal populations tend to be exposed accounting for 1.8 million new displacements,to a range of hazards, particularly cyclones, were less affected than low income countries,floods, landslides, earthquakes and tsunamis. which accounted for 3.6 million (see figure 1.9).The devastation wrought on Vanuatu by tropicalcyclone Pam, a category five storm, in March Low and lower-middle income countries have2015 provides a clear case in point. relatively little capacity to meet the protection and assistance needs of IDPs, or to invest inThe disaster forced around a quarter of Vanu- disaster risk reduction measures that would eitheratu’s population to flee their homes and left prevent displacement or mitigate the impacts ofaround 166,000 people on 22 islands in need future shocks and disasters.of emergency assistance, including nearly 65,000IDPs who required emergency shelter. Fifteen Figure 1.9: New displacements associated with disasters bythousand homes were reported as damaged or World Bank-defined income group, 2015destroyed, with almost the entire country severelyaffected. Pam also had impacts on other SIDSs in High income Lower-middlethe region, including a damaging storm surge in economies incomeTuvalu that prompted a state of emergency after55 per cent of its population of around 10,000 1.8m economiespeople were displaced, the highest relative figure (9%) 9.8mworldwide (see figure 1.8). (51%) Upper-middleAs El Niño conditions strengthened towards the incomeend of the year, intensified drought put furtherstress on countries still recovering from the economiesimpacts of cyclones and typhoons, bringing food 4.1minsecurity to Vanuatu, the Solomon Islands and (21%)the Northern Mariana Islands. Low income economies 3.6m (19%)Figure 1.8: Countries with most new displacements associated with disasters in 2015 (relative to population)South Asia East Asia and the Pacific Latin America and the Caribbean Sub-Saharan Africa Tuvalu 54,800 Vanuatu 9,200 20,000 41,700 40,000 50,000 60,000 Nepal 6,500Micronesia 5,800 30,000 3,000 Chile 2,600 Myanmar 2,200 Paraguay 2,200 2,000 KiribatiPhilippines 10,000 Malawi 0 Per 100,000 inhabitants ON THE GRID: Global internal displacement in 2015 19

Figure 1.10: Annual new displacements, 2008 to 2015, by hazard category Millions 45 Weather-related (average 21.5 million per year) Geophysical (average 3.9 million per year) 40 31.1 (15%) 35 Total: 2008-2015 30 172.3 (85%) 25 20 20.8 38.3 15 31.7 20.3 17.4 14.7 10 15.3 4 13.9 0.7 1.8 1.7 4.5 2010 2012 2013 2014 2015 15.8 1.5 1.1 2009 2011 5 0 2008 Exposure to weather extremes respectively, the majority of the latter in West Bengal, Odisha and Manipur states (see figure Disasters triggered by weather-related hazards 1.11). Eight of the ten largest displacements of tend to displace far more people than those the year in relative terms were also caused by brought on by geophysical hazards. They weather-related disasters, with cyclone Pam’s accounted for almost three times as much impacts on Tuvalu and Vanuatu topping the list. displacement in 2015, forcing 14.7 million people to flee their homes compared with 4.5 million for Displacement associated with weather hazards geophysical disasters (see figure 1.10). was lower in 2015 than the annual average level of 21.5 million people for the period since 2008, The same pattern was also seen in the size of despite the onset of the natural global weather displacements in 2015. Disasters triggered by phenomenon known as El Niño (see spotlight). weather-related hazards caused seven of the ten largest events, both in absolute terms and rela- During the short rainy season in equatorial east tive to the populations of the countries in ques- Africa from October to December, local weather tion. The main exception was Nepal, where the systems and sea surface temperature patterns in April and May earthquakes together caused the the Indian Ocean also played an important role in largest displacement of the year associated with 2015. Most of Kenya received higher than average a disaster (see figure 1.11). rains, which caused localised flooding and mud and landslides. More than 240,000 people were Within the category of weather-related hazards, affected, infrastructure was damaged, livestock floods and storms led to the vast majority of lost and cholera spread across at least 21 coun- displacements. Disasters brought on by floods ties. Around 103,500 people were displaced, forced 8.3 million people to flee their homes, and according to the Kenya Red Cross Society. storms 6.3 million. Around 87,000 people were displaced by wildfires (see figure 1.12). That said, the scale of displacement was signifi- cantly lower than expected and the rains bene- There were no mega-scale weather-related fitted arid and semi-arid parts of the country, events, defined here as those which cause more helping crop development, livestock productivity, than three million displacements, but flooding the replenishment of water sources and the in the southern Indian states of Tamil Nadu and recovery of rangeland in pastoral areas. Andhra Pradesh caused 1.8 million displace- ments. Cyclone Komen and monsoon floods Some of the strongest El Niño effects are on triggered disasters in both Myanmar and India, rainfall deficit and agricultural drought, whose causing 1.6 million and 1.2 million displacements indirect impact on displacement is not captured in our global statistics (see part 3 of this report).20 G20R1ID6

Figure 1.11: The ten largest displacement events of 2015, absolute and relative to population size Weather-related GeophysicalAbsolute (total displacements) 2,623,000 Nepal: Gorkha earthquakes 1,801,000 India: Andhra Pradesh and Tamil Nadu floods 1,617,000Myanmar: Monsoon floods and cyclone Komen 1,200,000 1,100,000 India: Monsoon floods and cyclone Komen 1,000,000 China: Typhoon Chan-Hom 938,000 743,000 Chile: Illapel earthquake and tsunami 666,000Philippines: Typhoon Koppu (local name: Lando) 562,000 Philippines: Typhoon Melor (local name: Nona) Pakistan: Hindu Kush earthquake China: Typhoon SoudelorRelative (displacements per 100,000 inhabitants) Tuvalu: Cyclone Pam 55,000 Vanuatu: Cyclone Pam 25,000 Nepal: Gorkha earthquakes 9,200Federated States of Micronesia: Typhoon Maysak 6,500Chile: Illapel earthquake and tsunami 5,600Myanmar: Monsoon floods and cyclone Komen 3,000 Kiribati: Cyclone Pam 2,200 Malawi: Seasonal floods 2,000Paraguay: Flooding in the Southern Cone 2,000 Dominica: Tropical storm Erika 1,000Figure 1.12: New displacements by type of weather hazard, 2015 and 2008 to 2015Wet mass Wildfire Extreme Wet mass Wildfire Extrememovement 87,000 temperature movement 362,000 temperature (0.6%) 704,000 (0.2%) 54,000 2,000 960,000 (0.4%) (0.1%) (0.4%) (0.6%)Storm 2015 Storm6.3m 60.2m(43%) (35%) 2008–2015 Flood 8.3m Flood (56%) 110m (64%) ON THE GRID: Global internal displacement in 2015 21

SPOT El NiñoLIGHT Aggravated risk associated with extreme weather From March 2015 to the first half of 2016, a the number of hurricanes in the Atlantic. Both new phase of the naturally occurring climatic were true in 2015, with an unprecedented 21 cycle known as the El Niño Southern Oscillation category four and five storms in the north Pacific, (ENSO) contributed to the highest average global breaking the previous record of 17 set in 1997.61 temperatures on record, about 1°C above those of the pre-industrial era, and disrupted weather El Niño episodes are associated with above patterns worldwide.59 average sea surface temperatures in the central and east-central equatorial Pacific, while the El Niño’s effects depend on the season and opposite phase of the cycle, known as La Niña, vary from one cycle to the next, but evidence is associated with cooler than average waters. El from past events suggests that the most likely Niño episodes typically occur every two to seven impacts tend to be lower than average rainfall years. They last for anything from nine months over Indonesia and northern South America, and to two years, reaching their maximum strength the opposite in south-eastern South America, between October and January and then contin- the southern US and eastern equatorial Africa.60 uing for some months before decaying. A strong El Niño also tends to increase the The latest El Niño peaked in November and number of cyclones in the Pacific and reduce December 2015, but its impacts on agricultureEl Niño and RainfallEl Niño conditions in the tropical Pacific are known to shift rainfall patterns in many different parts of the world. Although they vary some-what from one El Niño to the next, the strongest shifts remain fairly consistent in the regions and seasons shown on the map below. Wet Dry Dry Nov. to following March Wet Wet Dry Jan. to April July to Dec. Dec. to March Jan. to May July to following April Jan. to April Nov. to following April Dry Dry June to Sept. June to following MarchJuly to Sept. Jan. to June to following Jan. April April to Dry June Wet Wet June to following April Wet DryOct. to following Jan. Oct. to Dec. July to following March Sept. to following March Dry WetNov. to following March June to Sept. July to Nov. Wet July to following Jan. Sept. to following Jan.Original map: International Research Institute for Climate and Society, Earth Institute, Columbia UniversitySources:1. Ropelewski, C. F., and M. S. Halpert, 1987: Global and regional scale precipitation patterns associated with the El Nino Southern Oscillation. Mon. Wea. Rev., 115, 1606-1626;2. Mason and Goddard, 2001. Probabilistic precipitation anomalies associated with ENSO. Bull. Am. Meteorol. Soc. 82, 619-63822 G20R1ID6

and deteriorating food and livelihood security will future interactions between El Niño, La Niña and People dependentcontinue to be felt throughout 2016 and longer long-term climate change will play out.65 Mean- on agriculture inin some parts of the world. As of February 2016, while, as the World Meteorological Organiza- the Philippinesthe phenomenon was expected to transition to tion’s secretary general has put it: “El Niño is who were initiallya neutral ENSO state during the second quarter turning up the heat even further.”66 displaced byof the year.62 Evidence then points to a possible conflict in 2015 alsoLa Niña setting in later in the year, though at For the people most exposed and vulnerable to faced effects fromthe time of writing it was too early for experts rainfall extremes and higher temperatures asso- El Niño and laterto be certain. ciated with the phenomenon, its effects have flooding, leading been devastating and have led to displacement to crop productionClimate patterns, however, are more complex in many parts of the world. losses estimated atthan El Niño and La Niña alone can account for. more than 24,000Other local or regional tropical weather systems metric tonnes.also affect rainfall patterns. The Indian Ocean Photo: © M,dipole and the Tropical Atlantic sea surface Navales/FAO,temperature, for example, may affect the climate July 2015on adjacent land masses, and winter conditionsin the northern hemisphere are influenced by theso-called Arctic and North Atlantic oscillations.63Across decades, changes in the global climatebrought on by human activity also play a part.Some studies suggest that El Niño episodes arebecoming more intense as a result of changesin the global climate,64 but there is no scientificconsensus on the extent to which this may bethe case. It is simply not known how past and ON THE GRID: Global internal displacement in 2015 23

Large-scale geophysical Earthquakes have caused 97 per cent of the hazards: few in number, many displacement associated with geophysical displaced hazards since 2008, and for 2015 the figure was 99 per cent. Volcanic activity and erup- The average number of people displaced during tions caused around 32,000 displacements in disasters triggered by geophysical hazards was 2015, and dry mass movements and landslides 3.9 million a year between 2008 and 2015 (see a residual number. figure 1.10, p.20). The 2015 total of 4.5 million was higher than average, driven up by the Nepal The ten largest displacements associated disas- earthquakes, which alone caused the displace- ters in 2015 also include earthquakes in Chile and ment of around 2.6 million people. The scale Pakistan (see figure 1.11, p.21). A magnitude 7.5 of displacement they caused was also very high earthquake struck the Hindu Kush mountains in relative to the size of the country’s population, October. Its epicentre was in a remote area of of which 9.2 per cent were displaced (see figure northern Afghanistan, but north-west Pakistan 1.11, p.21 and Nepal spotlight). suffered the worst of its impacts, including the displacement of nearly 666,000 people. Khyber24 G20R1ID6

Pakhtunkhwa (KP) province, where human devel- meant that early warning systems were effec- An internallyopment is low compared with the rest of the tive and response planning was good. Around displaced mancountry, was hardest hit, particularly the districts a million people were evacuated pre-emptively, waits for a winterof Chitral, Dir Lower, Dir Upper, Swat, Shangla, saving many lives. The enforcement of better shelter pack inMalakand and Buner. construction standards and codes also meant the Malakand that recently constructed buildings were better District of KhyberChile appears on the list for the second year in able to withstand the seismic activity. Tradition- Pakhtunkhwaa row. The 8.3 magnitude Illapel earthquake ally built adobe homes, however, tend to be very province as part ofand five-metre tsunami it triggered affected the vulnerable and further efforts to strengthen the the Pakistan RedCoquimbo region of the country in September resilience of poor rural communities would help Crescent Society’s2015. Most people were evacuated in good to reduce the disproportionate affects of such emergencytime, but several coastal cities and ports suffered events on them. response to theextensive damage. 2015 earthquake. Photo: SajidThe country’s disaster preparedness framework, Qayyum/ IFRC, Apriladopted after a catastrophic earthquake in 2010, 2015 ON THE GRID: Global internal displacement in 2015 25

26 G20R1ID6 People internally displaced as a result of conflict and violence as of 31 December 2015 (Total: 40.8 million) Azerbaijan Afghanistan Russia Nepal Armenia 564,000 1.2m 27,000 50,000 Kosovo 17,000 8,400 Bosnia and Georgia Herzegovina 239,000 98,000 Macedonia 200 Turkey Ukraine Pakistan 954,000 1.7m 1.5m Cyprus 272,000 Syria 6.6m Lebanon 12,000 Iraq 3.3m Palestine Mexico 221,000 Sudan Yemen Egypt 287,000 Libya 3.2m 2.5m 78,000 500,000 Guatemala Honduras Mali South Abyei Philippines 251,000 174,000 50,000 Sudan 82,000 62,000 El Salvador Colombia 1.7m Indonesia 289,000 6.3m Senegal Nigeria Bangladesh 6,100 24,000 Niger 2.1m DRC 426,000 1.5m 153,000 Ethiopia 450,000 Côte d’Ivoire Togo Somalia Sri Lanka Papua 303,000 3,000 1.2m 44,000 New Guinea Cameroon India Myanmar 6,300 124,000 612,000 644,000 Chad Kenya 107,000 309,000 Peru Thailand 60,000 35,000 Uganda 30,000 Congo 7,800 CAR Burundi 452,000 99,000

Total headcountPeople living in displacement as of end 2015There were 40.8 million people internally of the ten largest internally displaced populationsdisplaced worldwide as a result of conflict and every year since 2003.67 Their persistent inclu-violence as of the end of 2015 (see map, opposite sion points to a depressing reality of intractablepage). This represents an increase of 2.8 million conflict in which many people may have beenon 2014 estimates and the highest figure ever forced to flee more than once. These five coun-reported since IDMC began monitoring internal tries alone accounted for almost 40 per cent ofdisplacement in 1998. It is also twice the number the world’s IDPs, or nearly 16 million people, asof refugees in the world. of December 2015.Protracted crises swell Figure 1.14: Number of people internally displaced by 6.6unprecedented numbers conflict and violence as of the end of 2015 (millions)The total number of IDPs has doubled over the Syriapast 15 years. From below 20 million in the 1990sit rose to 27.5 million by 2010, the result of new Colombia 6.3and protracted displacement caused by long-running internal conflicts (see figure 1.13). Five Iraq 3.3years later, it has reached more than 40 million, 3.2in large part the result of conflict and violence in Sudan 2.5the Middle East following the Arab spring upris- 2.1ings that began in late 2010. Yemen 1.7 1.7Of the total, around three quarters, or 30 million Nigeria 1.5people, were located in just ten countries (see South 1.5figure 1.14). Half of them – Colombia, DRC, Iraq, SudanSudan and South Sudan – have featured in the list Ukraine DRC Pakistan 01234567 MillionsFigure 1.13: Refugees and IDPs displaced by conflict and violence, 1990 to 2015 45MILLIONS 40 IDPs Refugees 35 30 25 20 15 10 5 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015Sources: UNHCR, UNRWA for refugee figures; IDMC and USCR for internal displacement figuresNote: Refugee data for 2015 was not available at the time of writing ON THE GRID: Global internal displacement in 2015 27

SPOT DRCLIGHT Multiple and chronic displacementDisplaced women Conflict in DRC is intractable, and the internal turn means IDPs have varying needs and chal-living with host displacement crisis it has caused is multifaceted. lenges. Both national and international responsesfamilies take part in Despite apparent similarities among the waves of need to take this complexity into account if alla focus group dis- displacement, their causes, dynamics and perpe- IDPs are to achieve durable solutions.cussion in Swima, trators vary from one situation to the next. OurSouth Kivu. Photo: estimates show that over the past 15 years, there There were an estimated 1.5 million IDPs in theIDMC, March 2015 have consistently been more than a million IDPs country as of December 2015, the ninth highest in the country, and in peak years such as 2003 figure worldwide. Most have fled violence and more than three million. human rights abuses committed by armed groups and the military, but inter-communal tensions The implication is that DRC is faced with a range and disputes over land and the control of natural of both protracted and new displacement situ- resources have also caused displacement, as ations, sometimes in the same places, which in have natural hazards. Conflict and violence are concentrated in eastern DRC, as are the country’s IDPs. More than half live in the provinces of North and South Kivu, and the remainder in Orientale, Katanga, Maniema and Equateur. Multiple and chronic displacements are common- place, particularly in the Kivus but also in other provinces. Many IDPs in North Kivu have been displaced two, three or even more times in the last 18 months alone, and surveys suggest that a significant majority in North Kivu, South Kivu and Ituri have been displaced more than once since 1993.68 Each time people flee, they lose almost every- thing and are forced to start rebuilding their lives from scratch. They lose their homes, mate- rial assets and often their jobs and livelihoods. Their children’s education is interrupted, and they become separated from their communities and support networks. Their sense of cultural iden- tity may suffer, and the trauma and upheaval of fleeing conflict and violence leave many with psychological and physical scars. Whether it is the first or the fifth time that a person is displaced in DRC, they face a long struggle to try to recover their assets, access basic services and re-establish their social and psychological wellbeing. The longer and more often they are displaced, the more difficult these challenges become, and the more their ability to cope is eroded.28 G20R1ID6

Of the other five countries on the list, displace- Invisible IDPs: protractedment in Nigeria, Syria, Ukraine and Yemen is displacement followinglargely the result of armed conflicts that erupted disastersor escalated within the last five years. Pakistan isthe exception. If history is any indication, there The estimates for displacement associated withis a risk these relatively new conflicts will also disasters presented in this report are aggregatedbecome intractable, and that the people they incidents of new displacements over the coursehave forced to flee, who make up roughly a third of 2015. There is very little information on peopleof the overall global estimate, face a future of still displaced in December 2015 following disas-protracted and possibly multiple displacement. ters during the year or in previous years, which means it is not possible to give cumulative figuresSub-Saharan Africa: Chronic for the total number of people displaced by disas-displacement keeps figures ters as of the end of the year, as there are forstubbornly high IDPs displaced by conflict and violence.A regional analysis provides a different perspec- Data that monitors IDPs’ situations over time,tive on the same point (see figure 1.15). As of the whatever the causes of their displacement, tends toend of 2015, the Middle East and north Africa become more scarce the longer they are displaced.accounted for around a third of the world’s IDPs, In the aftermath of disasters, the assumption thator 13.2 million people, largely the result of the displacement is generally temporary makes suchsurge in new displacements over the last few information scarcer still. The assumption is madeyears in Iraq, Syria and Yemen. based on observations of displacement over short distances and the fact that a significant proportionFigure 1.15: Number of people internally displaced by of those who flee return quickly to their homes. Itconflict and violence by World Bank-defined region is also influenced by the fact that people displaced by disasters tend to flee to diverse locations and East Asia undertake complex movements that make them and Pacific difficult to identify and track. 754,000 These factors increase the risk that some of the most vulnerable IDPs will be left behind in long- South Asia term recovery, disaster risk reduction and develop- 3.8m ment processes.69 Among a sample of 34 ongoing cases of displacement following disasters docu-Europe and mented in 2015, there were hundreds of thou-central Asia sands of people identified as living in protracted displacement for periods ranging between one 3.9m and 26 years.70 This points to the likelihood of hundreds of thousands more yet to be recorded. Middle East and People displaced for long periods of time areLatin America north Africa particularly difficult to identify and access for a and the variety of practical and political reasons. They 13.2m may also be relatively fewer in number compared with the overall or peak number of IDPs, but suchCaribbean “residual caseloads” following major disasters, and people displaced by recurring local or smaller 7.3m Sub-Saharan disasters, are some of the most vulnerable. Africa The capacity of governments, civil society and affected communities to achieve solutions varies 11.9m widely. The cases of Japan and Haiti highlight some of the human impacts of protractedMeanwhile, at close to 12 million, the number of displacement, its disproportionate effect on someIDPs in sub-Saharan Africa has more or less flat- of the most vulnerable people and common chal-lined over the last decade, underlining the chronic lenges in resolving it (see Japan/Haiti spotlight).nature of displacement in the region. Failure toaddress the causes of protracted displacementis one of the main factors behind the ever-increasing number of IDPs worldwide, and thestubbornly high figures for Africa. Finding theright solutions requires a better understandingof the phenomenon, from clarifying conceptsto recognising its complexity and diversity (seeDRC spotlight). ON THE GRID: Global internal displacement in 2015 29

SPOT For richer or poorerLIGHT Protracted displacement in Japan and Haiti Most of the protracted displacement associated Those able to afford it, most often younger fami- with disasters is in low and lower-middle income lies, have rebuilt their lives elsewhere, but others developing countries, but there have also been have been unable to afford reconstruction and significant cases in some of the world’s richest have moved into rental accommodation provided and poorest countries which reveal some striking by the government.71 similarities and differences. In Fukushima prefecture, 99,000 of the 160,000- Japan’s Fukushima IDPs plus people evacuated from contaminated areas around the crippled nuclear plant are still living In March 2011, a devastating magnitude 9.0 in displacement.72 They face a difficult deci- earthquake and tsunami struck Japan’s Tohoku sion in whether to return home or not to areas region, triggering the meltdown of three reactors where government evacuation orders have at the Fukushima Daiichi nuclear power plant been lifted, but where radiation risk remains a and major radiation leaks. The disaster displaced concern, particularly for younger generations. around 470,000 people. Those affected also worry about the lack of basic infrastructure such as schools and hospitals in Five years later, tens of thousands of people their former home areas, and about becoming whose homes and villages were destroyed across isolated given that few of their family members, three prefectures have yet to re-establish their former neighbours and friends plan to return.73 lives in new or former homes and communities. Plans to rebuild destroyed homes and relocate A lack of trust in official information and poor displaced communities to higher ground or other consultation with affected communities have also areas have been reduced by 30 per cent over the delayed solutions for IDPs unable or unwilling to past three years and in some cases abandoned, return, and social tensions with host communi- the result mainly of soaring construction costs ties have left social and psychological scars. A and the time local authorities have taken to over- 2015 survey of evacuees revealed that many were come land issues and prepare plots on new sites. suffering from anxiety, loneliness and depres- sion.74 In Fukushima, the number of people who have killed themselves or succumbed to health problems related to the disaster exceeds the death toll from its direct impacts.Temporary housingcomplexes such as thisone are still a commonsight one in Iwateprefecture.Photo: IFRC, March2016https://flic.kr/p/E9K941(edited contrast andbrightness)30 G20R1ID6

Haiti’s earthquake IDPs Obstacles to urban integration and other solu- One of the victims tions for camp residents include the poor condi- of the devastatingIn January 2010, a magnitude 7.0 earthquake tions in the neighbourhoods where they lived 2010 earthquake instruck Haiti, killing more than 160,000 people before the earthquake and high unemployment Haiti, waiting to beand displacing nearly 1.5 million. Six years later, levels. Unemployment among IDPs is estimated relocated. Photo: ©there were still nearly 62,600 people living in at 83 per cent, more than double the rate for the Frédéric Loward,deteriorating conditions in 36 displacement urban work force.77 Some, meantime, are gradu- April 2015camps in and around Port-au-Prince. IDPs in 16 of ally turning their temporary shelters into morethese camps are highly vulnerable to the impact permanent, unplanned informal settlements.of future natural hazard events.75 Ultimately, real solutions for Haiti’s remaining IDPsLiving conditions in the camps were always poor, and as many as 3.5 million others living in urbanbut have become worse still as basic services poverty with similar vulnerabilities will depend onare wound down. There is ever less humani- reducing socio-economic deprivation, insecuritytarian funding available and fewer organisations and disaster risk. Improving national capacity toproviding assistance. plan and manage land issues to complement the implementation of the 2013 National HousingPeople living in the camps are exposed to criminal and Habitat Policy would enable access to safe,gang violence, abuse, exploitation and forced affordable housing and tenure security for Haiti’seviction that displaces them again. Women poorest people, including IDPs.78 Political insta-and girls have been particularly susceptible to bility and declining development assistance,increasing insecurity and health hazards. IDPs’ however, have wide-reaching implications forlack of civil documentation also continues to the responses required, without which the risk ofhamper their access to basic services, their chil- another major disaster is high. As the head of thedren’s enrolment for school exams, and their right UN Office for the Coordination of Humanitarianto own land, vote and open bank accounts.76 Affairs in the country has said: “Haiti cannot afford to become a forgotten crisis.”79 ON THE GRID: Global internal displacement in 2015 31

Forty-five displacedfamilies live in a formerschool in the al-Waerneighbourhood of Homs,Syria. Photo: EmmanuelBargues/OCHA,December 2015

Part 2Inside the GRID Caveats for the estimatesBehind all of the figures in this report are people most countries have yet to designate a specificwhose lives have been disrupted, in many cases government agency to systematically collect andseverely, by traumatic events. Given the propen- share comprehensive data.80sity of displacement to become protracted, theupheaval and its consequences can be long- As a result, we and our partners are left with anlasting if not permanent. Becoming displaced not imperfect set of tools that are inconsistently used.only means losing one’s home and other material We compile our figures based on the best, mostassets. Many IDPs also lose their jobs, livelihoods, credible data we are able to obtain, but they aresocial support networks and documentation that ultimately only estimates. We round our figuresthey are likely to need to start rebuilding their to help emphasise that fact, and the numbers welives elsewhere. Their children’s education is publish are deliberately conservative.often interrupted, families are broken up, theirhealth suffers, and the trauma and upheaval of Reporting accurately means treading a tightropeflight leave many with psychological and physical between under- and over-estimating the scale ofscars. displacement, with significant human implica- tions for those affected. Under-estimates meanWith this human toll in mind, we take our respon- IDPs go unseen and unaccounted for when itsibility to monitor internal displacement seriously comes to providing assistance. Over-estimatesand strive to report on it in a comprehensive risk misdirecting scarce resources away fromand accurate way. That said, displacement is a those most in need.complex, fluid and politically sensitive phenom-enon and as such it is difficult to measure. Our In this part we outline our efforts to improve theestimates are our best attempt to do so – to coverage and transparency of the global evidencecount vulnerable people who are on the move base on internal displacement. By providing theand who have no official status, with the ultimate breakdown of the age of our figures for theaim of their being provided with the protection first time in this report, we are appealing to theand assistance they need. governments concerned and to our partners in the field to contribute to this ongoing effort.Monitoring and reporting on IDPs is very differentfrom doing so for refugees. IDPs are seldomregistered and they are often difficult to identify.Some may not even want to put their head abovethe parapet by being counted. Some govern-ments too resist efforts to monitor and reporton displacement. A 2015 UN General Assemblyresolution encourages states to “ensure theprovision of reliable data on internal displace-ment”, including by collaborating with IDMC, but INSIDE THE GRID: Caveats for the estimates 33

Key findings and Messages ||Our ability to obtain data on the number ||We were able to obtain updated informa- of IDPs and the processes responsible for tion in 2015 for nearly 31.7 million of the increases or decreases in the size of the 40.8 million people who we estimated were IDP population is limited. living in displacement as of the end of the year as a result of conflict and violence. ||The Guiding Principles and several UN General Assembly Resolutions have recognised that ||The age of the most recent data for the sovereign states bear the primary respon- remaining 9.1 million IDPs varies widely and in sibility for collecting and sharing data on some cases is significantly out of date. The data internal displacement. This should, of course, on nearly a million IDPs in Turkey dates back at include regular updates on the number of people least to 2006, and some for Guatemala goes who have become newly displaced or achieved back as far as 1997. durable solutions, as well as data disaggregated by sex and age. At present, displacement data ||The issue of outdated or decaying data in several countries is already outdated, is of particular concern with Colombia, a and it is at risk of becoming outdated in others, country that has been among the five countries including countries with large IDP populations with highest number of people displaced by such as Afghanistan. In order to avoid this, more conflict every year since we began monitoring resources and capacities are needed at country internal displacement in 1998. level to collect displacement data and keep it up to date. ||Outdated or decaying data is a problem in 12 of the 53 conflict- or violence-affected ||We have difficulty in obtaining data on the countries in this report, accounting for processes that lead to the end of displacement approximately 20 per cent of IDPs worldwide. and the number of IDPs who have fled across The countries concerned are Armenia, Bangla- international borders. There is also little infor- desh, Congo, Cyprus, Guatemala, Macedonia, mation available about the number of children Nepal, Papua New Guinea, Thailand, Togo, born to IDPs and the number of people who Turkey and Uganda. die in displacement. ||The lack of updated data, particularly on ||Our estimates for the number of people displacement that has become protracted, is internally displaced by conflict and violence one of the main gaps we face in both conflict are deliberately conservative. When we and disaster contexts. We have also found that receive information that IDPs have returned, inte- people displaced by intractable conflicts grated locally or settled elsewhere, we subtract around the world tend to fall off the radar. them from our totals regardless of whether they We have been unable to obtain return figures are known to have achieved a durable solution. for a number of countries, including Bangla- We do this because reporting on the end of desh, Burundi, Guatemala and Turkey. displacement and the processes that lead to it are open to different interpretations. ||As a global monitor we want to call attention to such situations, and their inclusion also consti- ||To generate global estimates, we have histori- tutes an explicit plea for updated data and cally attempted to account for new displace- information. ments associated with disasters without indi- cating the length of people’s displacement. This means our figures are the sum of all displacements triggered by a particular disaster or event, and do not account for any outflows such as returns or onward move- ments.34 G20R1ID6

IDMC’s data modelCapturing the human toll of displacementTo paint a comprehensive global picture of partial (see Syria spotlight). For each of the situa-internal displacement associated with both tions we reported on last year, we were able toconflict and disasters, we obtain data from our estimate the number of IDPs as of 31 Decembersources and relate it to the generic displacement 2015 – this “stock” of people is represented bymodel below (see figure 2.1). Obtaining data on the orange box in figure 2.1 – and the incidentseach of the relevant processes or “flows”, which of new displacement, based on direct reportingdetermine displacement patterns, is a crucial part from the field or by inference from increases inof accurate reporting. Not doing so would mean the size of displaced populations.we lose sight of what is happening to tens ofmillions of people around the world every year. We have much more difficulty in obtaining data on the processes that lead to the end of displace-Conservative estimates for ment and the number of IDPs who have fleddisplacement associated with across international borders. These flows areconflict represented by the dark blue arrows in figure 2.1. There is also little information availableFor displacement caused by conflict and violence, about the number of children born to IDPs andwe try to obtain data on the number of IDPs the number of people who die in displacement.and the processes responsible for increases or Explicitly disaggregated information was onlydecreases. Our ability to do this, however, is only available for relatively few of the 52 countries and one region (Abyei) for which we provide estimates for 2015 (see table 2.1, p.36).Figure 2.1: IDMC’s displacement data model IDPs settled elsewhere Failed settlement Settlement elsewhere elsewhereInternal displacement Returns ReturneesChildren born IDPs (includes secondary Failed returns / Locally to IDPs and tertiary displacement) returnee integrated displacement IDPs Local integrationDeaths Cross-border Failed local flight integration Cross-border return to displacement People displaced across borders (e.g., refugees) INSIDE THE GRID: Caveats for the estimates 35

Table 2.1: Summary of data on 2015 flows that influence the total number of displacements associated with conflict or violence Conflict- or violence- related flows Number of countries or territories for which data was obtained out of 52 countries and New incidents of displacement one region Returns 28 Local integration 20 Settlement elsewhere 1 Cross-border flight 2 Children born to IDPs 0 Deaths in displacement 2 1 Not every flow is relevant to every situation we Data gaps for displacement report on. The absence of data on new displace- associated with disasters ment may simply mean that no new displacement has taken place. Births and deaths may have been We use a different methodology to monitor included in some of the data we obtained, but displacement associated with disasters, one of not in a way that allowed us to disaggregate it the implications of which is that our coverage of from other flows. That said, a cursory glance at the data model is more limited. Our figures are table 2.1 reveals significant data gaps. This is the sum of all of the people newly displaced by a particularly true for information on the processes particular disaster – all of the people in the orange that lead to the end of displacement, though box in figure 2.1, p.35 – without accounting for their impact on the overall number of IDPs may any of the outflows from that stock. be relatively small compared with the ever- increasing number of people newly displaced This means we are unable to report on the by conflict and violence. duration of displacement at the global level, or provide a cumulative figure for the number of Our estimates are deliberately conservative. people displaced as of 31 December 2015. We When we receive information that IDPs have have, however, gathered evidence from dozens returned, integrated locally or settled elsewhere, of case studies that shows there are hundreds of we subtract them from our totals regardless of thousands of people still living in displacement whether they are known to have achieved a following disasters in previous years and decades. durable solution. We do this because reporting on the end of displacement and the processes Our estimates do not reflect where people shelter that lead to it are open to different interpreta- or live while they are displaced, or where and tions. when they eventually settle again. The figures may include people who fled disasters to other This has sometimes led in the past to the applica- countries, but we found no such cases in the tion of different criteria for subtracting people process of our 2015 data collection. from the displaced population. A profiling exercise might find that a percentage of those We are currently expanding our data collection displaced have returned but still not achieved in an effort to capture all of the outflows in our a durable solution. Continuing to count these model for displacement associated with disas- people as IDPs creates a different – and higher – ters. This includes IDPs who return to their home benchmark for assessing returns compared with areas, integrate locally, settle elsewhere in the another situation in which an authority simply country or continue their flight across an interna- reports that IDPs have “returned” or are “no tional border. Doing so will enable us to paint a longer displaced”. more comprehensive picture of situations as they evolve and enable comparisons between them.36 G20R1ID6

Syria SPOT LIGHTTrapped in the country, and out of the pictureBy any measure, the humanitarian situation in mean that the humanitarian response, which isSyria worsened significantly in 2015. The coun- already overwhelmed, is unlikely to be using thetry’s civil war is now in its sixth year, with four resources available efficiently.of the five permanent members of the UN Secu-rity Council actively engaged in the hostilities, The limitations of current data collection effortsand it has caused one of worst displacement worldwide, as outlined in our confidence assess-crises since World War Two.81, 82 As of December, ment tool (see methodological annex), can beintense fighting and violence had forced more broadly grouped into three categories in Syriathan 10.9 million people, or over half of the coun- – security and access restrictions, the politicaltry’s pre-war population, to flee their homes. Put environment and methodological challenges.another way, an average of 50 families have beendisplaced every hour of every day since 2011.83 Of More than 4.5 million people were living inthe total, at least 6.6 million people have been areas of the country that the UN considered tointernally displaced.84 be either difficult or near impossible to reach in 2015, including besieged cities, and humani-The bulk of international attention has focused tarians had less access than in 2014.89, 90 Theon the millions of people who have risked their number of people living in areas OCHA classifiedlives and those of their children to seek safety as besieged more than doubled from 2014 toelsewhere in the region or in Europe, with almost 500,000 people in early 2016, of whomdiminishing hope of finding safety, acceptance less than one per cent received food aid.91and opportunity. Having initially admitted largenumbers of refugees, however, neighbouring Monitoring internal displacement was furthercountries have increasingly restricted the flow hampered by the intensity of the conflict andof people out of Syria, or sealed their borders the volatility of its frontlines. The presence of thealtogether.85 Islamic State (also known as ISIL and ISIS) made the north-eastern governorates of Ar-Raqqa andAs a result, hundreds of thousands of people are Deir Ez-Zor particularly difficult to access, and thetrapped inside the country, abandoned in camps lack of data collection in such areas is likely toor staying with host communities near border have led to significant under-reporting.points with no legal escape route and often livingin subhuman conditions.86 The unpredictable complexity of Syria’s polit- ical environment also impedes the collection ofThe main causes of casualties and displacement reliable data. OCHA’s displacement estimates,in Syria are well known. They include indiscrimi- for example, which are only aggregated at thenate attacks in populated areas, the deliberate country level once a year, are based on informa-targeting of civilians and civilian infrastructure tion gathered from various government entities,such as schools and healthcare facilities, and UN agencies and the Syrian Red Crescent Society.sieges during which people are deliberately In areas under opposition control, it has also haddeprived of aid and basic services such as food, to rely on NGOs active there and local authori-water and medical care.87 Such acts were relent- ties. As such, data collection and reporting areless in 2015, and as of October, at least 1.3 million subject to the influence of parties to the conflict,people had been newly displaced, many for the including some that have played a central role insecond or third time.88 causing displacement in the first place.92Despite a broad awareness of these drivers, there Methodological challenges meantime may resultis relatively little understanding of their specific in under-reporting or double counting, and aconsequences: who the IDPs are, where they flee distorted understanding of the needs of peopleto and in what number, and what their needs fleeing within and beyond Syria. The estimatedare. This incomplete picture and the failure to 6.6 million IDPs in the country as of the end ofconduct an accurate assessment of the situation 2015 is fewer than the 7.6 million at the end INSIDE THE GRID: Caveats for the estimates 37

TURKEY ]] ] ] ] ] ] ]] ] Al-Hasakeh ] ] P IRAQ ] Aleppo Ar-Raqqa 301,600 ]] P P Deir-ez-Zor ] Idleb 1.25 million 201,400 P P ] ] ]]] 318,700 704,500 Lattakia ] P 378,000 Hama 500,800 258,500 P P Tartous HomP s 526,500 6.6 million ]] ] IDPs as of 31 December 2015Mediterranean Sea ] SYRIAN ARAB REPUBLIC LEBANON ] P Rural Damascus Damascus 1.27 million ] 436,170 Source: OCHA Humanitarian Snapshot Quneitra (as of 31 December 2015) P 38,700 The boundaries and names shown and the Dar'a P As-Sweida designations used on this map do not imply 320P,800 61,700 official endorsement or acceptance by IDMC. ] LEGEND ] ]] IDPs by governorate JORDAN Country capital Governorate capital International boundary Hard to reach areas Border crossing open Border crossing closed Border crossing sporadically open or restricted of 2014,93 but the figure relative to the popula- count these IDPs effectively are hampered both tion as a whole has most likely increased, given by the methodological challenge of identifying the number of people who have fled abroad. them among the besieged population as a whole, The current reporting systems for refugees and and by the lack of access to areas under siege. asylum seekers also make it very difficult to know how many were formerly IDPs. As such, when Compared with the attention given to Syrian refu- figures for IDPs and refugees are combined, gees, the country’s IDPs have been neglected, many people are counted twice. with significant implications for humanitarian funding and assistance, not to mention the The fact that many, if not most IDPs have been lives of those affected.94 The pace of displace- forced to flee more than once presents another ment remains relentless, and people are likely to methodological challenge. Multiple displace- continue to uproot their families at a similar rate ments are difficult to track in any context, and unless the fighting is brought to an end.95 Despite particularly so in Syria. One the one hand, such needs increasing throughout 2015, it was harder people may not be counted at all because they than ever to get aid to the most desperate.96 live in host communities where they are largely invisible, but on the other multiple displacement Data gathering is a vital part of saving lives. may mean that people are counted more than Timely and reliable information on the trajecto- once – each time they are displaced. ries of families fleeing violence within Syria and the tipping point to cross the country’s border Agencies that estimate the number of IDPs in contribute to a better understanding of their situ- different parts of the country use different meth- ation. This in turn improves the quality of advo- odologies, and those trapped in besieged cities cacy and programming on their behalf, and ulti- will have been displaced relatively short distances mately the likelihood that efforts to protect and given their inability to leave the area. Efforts to assist them will be better resourced and targeted.38 G20R1ID6

Working withdecaying dataThe lack of updated data, particularly on displace- recent data for the remaining 9.1 million IDPsment that has become protracted, is one of the varies widely and in some cases is significantlymain gaps we face. We have tried to address it out of date. The data on nearly a million IDPs inas consistently and transparently as possible by Turkey dates back at least to 2006, and some forpresenting stratified bar graphs of the number Guatemala goes back as far as 1996.of IDPs based on the age of the data for eachsituation. We have chosen to continue reporting Despite these and other sources being out ofon situations for which we have not received any date, we continued accounting for the IDPsnew information, but we call attention to the fact concerned for two reasons. As a global monitorthat the data may be out of date. we want to call attention to such situations, and their inclusion also constitutes an explicit plea forWe were able to obtain updated information in updated data and information. We hope that by2015 for nearly 31.7 million of the 40.8 million presenting our data in this way, our readers willpeople who were living in displacement as of the be able to draw their own conclusions about theend of the year as a result of conflict and violence displacement situations covered, and decide how(see figure 2.2). For an annual report targeting much emphasis to put on evidence that may beglobal policy processes, this information can years out of date.be considered up to date. The age of the mostFigure 2.2: People internally displaced as a result of conflict and violence as of 31 December 2015, by year of latest data update Data dated 2015 Recent data Outdated data 23.1m (56%) Data Data dated dated 2014 19968.6m (21%) 1.2m (3%)0 5,000,000 10,000,000 15,000,000 20,000,000 25,000,000 30,000,000 35,000,000 40,000,0002015 (displaced in 2015) 2015 Year of latest data update 2011 2010 2009 2008 2007 2006 2005 2004 2014 2013 2012 2000 1999 1998 1997 1996 2003 2002 2001 INSIDE THE GRID: Caveats for the estimates 39

Figure 2.3: Countries with the highest number of people internally displaced by conflict, stratified Syria 2015 Colombia 2014 2013 2012 2011 2010 2009 2008 2007 2006 Iraq Last update of IDP status between 2006 and 1996 2015 Sudan 2015 Yemen 2014 Nigeria 2015 S. Sudan 2015 Ukraine 2015 DRC 2015 2015 Pakistan 2010 0 2014 1,000,000 2,000,000 3,000,000 4,000,000 5,000,000 6,000,000 IDPs Year of latest data update 2015 2014 2013 2012 2011 2010 2009 2008 2007 2015 (displaced in 2015) 2004 2003 2002 2001 2000 1999 1998 1997 1996 2006 2005 A more nuanced illustration of the ten countries Each year, IDMC reaches out to UN Member with most people internally displaced by conflict States inviting them to share their displacement (see figure 2.3) points to the fact that some of the data, and each year only a handful reply. This year, stock data is relatively old and possibly decaying. only five governments responded with their data This was the case for 12 of the 52 countries and – Azerbaijan, Bosnia and Herzegovina, Georgia, one region (Abyei) in this report, accounting for Ireland and Mexico. Governments in several other less than 20 per cent of IDPs worldwide. The countries – Afghanistan, CAR, Colombia, Cyprus, countries concerned were Armenia, Bangladesh, the DRC, Honduras, Macedonia, Mali, Nigeria, Congo, Cyprus, Guatemala, Macedonia, Nepal, Peru, the Philippines, Russia Sri Lanka, Togo Papua New Guinea, Thailand, Togo, Turkey and and Ukraine – designated national authorities Uganda. to collect and publish this data or to collaborate with others to do so. Eighty-one per cent of the data used to compile our country estimates has been updated within Particularly in protracted crises, displacement the past two years, but in some cases part or all data often becomes outdated when govern- of the latest available information is more than ment authorities and international actors lose two years old. This is the case for Colombia (see the capacity needed to collect it. This can be box) and the 15 other countries shown in figure due to attention and resources being allocated 2.4. The upshot is that the estimates we gener- to more visible or pressing crises. When UNHCR ated for some countries are more reliable than shared its 2015 IDP data for Afghanistan, it noti- for others. fied us that IDP profiling and data collection had ceased, partly due to lack of funding. We have also found that people displaced by protracted conflicts around the world tend to The solution to this is more frequent collection of fall off the radar. Colombia is clearly not the displacement data that accounts for the number only country to have outdated or decaying data of IDPs as well as the flows leading into and out for its stock of IDPs. We have been unable to displacement. By providing the breakdown of the obtain return figures for a number of countries, age of our figures for the first time in this report, including Bangladesh, Burundi, Guatemala, Thai- we are appealing to the governments concerned land and Turkey. and to our partners in the field to contribute to this ongoing effort. Donor governments should The figures for these countries highlight the need ensure that designated authorities have the for improved and updated data on displacement. resources and capacity to collect displacement As clearly stated in the Guiding Principles on data and keep it up to date. Data-gathering Internal Displacement and reaffirmed in succes- agencies should, in turn, give warning several sive United Nations General Assembly Resolu- months prior to halting their data collection to tions, sovereign states are primarily responsible give time to address this impending gap. for maintaining up-to-date statistical information on their displaced populations.40 G20R1ID6

Figure 2.4: Countries with fewer than a million IDPs displaced by conflict, for which some data or all is older than 2014Turkey 2006 2014India 2015 2013Bangladesh 2012 2009 Cyprus 2014 2009Guatemala 2014 1997 Georgia 2015 2008Palestine 2015 2014 2011 2007Burundi 2011Nepal 2013Thailand 2007Uganda 2012 Russia 20210513Armenia 2005Congo 2006Togo 2008 0 100,000 200,000 300,000 400,000 500,000 600,000 700,000 800,000 900,000 IDPsYear of latest data update 2014 2013 2012 2011 2009 2008 2007 2006 1997 2015 (displaced in 2015) 2015Decaying displacement data inColombiaThe issue of decaying data is of particular concern with Colombia (see figure 2.3), a country that hasbeen among the five countries with the highest number of people internally displaced by conflictevery year since we began monitoring internal displacement in 1998.To its credit, the government has maintained a sophisticated and detailed account of the country’sdisplaced population. The data in the latest iteration of its registry for IDPs, part of the nationalvictims’ registry administered by the country’s victims unit, is disaggregated by age, gender, prov-enance and resettlement location, and paints a highly detailed picture.The registry, however, is primarily intended as a tool to facilitate the government’s provision of victims’reparations, in accordance with law 1448 of 2011. As such, it does not take into account peoplewho are no longer displaced, whether because they have achieved a durable solution, or becausethey have died. This means that the number of IDPs in the country never decreases.Anecdotal evidence suggests that many people displaced by Colombia’s conflict, now in its sixthdecade, have resettled in the country’s cities, but it is impossible to gauge with any certainty howmany of the 6.3 million or so people who fled their homes between 1996 and 2015 still live indisplacement. The estimate for 2015 is likely to be significantly inflated and should be interpretedwith caution. INSIDE THE GRID: Caveats for the estimates 41

A local resident, who wasevicted from Vila Autódromoin Rio de Janeiro, Brazil,addresses the hundredsof residents and policegathered in February 2006for the demolition of theneighbourhood association’sbuilding. Decrying thegovernment’s actions, shesays: “I am embarrassed bythis country,” and “my housefell, but I will never stopstruggling.” Photo: MeganHealy/CatComm/Rio on Watch,February 2016.

PART 3Off The Grid The world’s overlooked IDPsIDMC has to date monitored displacement associ- chronic crises. Such drivers are also likely to haveated with conflict, generalised violence and disas- contributed to some of the conflict and violenceters brought on by rapid-onset natural hazards. that has forced people to flee their homes, andAs such, alarming though the scale and trends set the general upward trend in global displacement.out in part one of this report are, the global snap-shot is far from complete. In addition to the data The notion of an IDP is based on two corelimitations explained above, the global figures do components: that their movement is forced,not capture many other contexts in which people to distinguish them from economic and otherare forced to flee their homes. voluntary migrants; and that they remain within internationally recognised state borders, to distin-In recent years, we have been building up guish them from refugees and other people whoevidence on displacement associated with move across them.criminal violence, development projects andslow-onset crises related to drought and envi- Differentiating forced from voluntary movementronmental change. In this part we explain why is not always straightforward, and displacementpeople displaced in such contexts should be associated with slow-onset, frequently recurrent,recognised as IDPs, and we explore some of and cyclical crises are just some of the situationsthe challenges inherent in making their protec- in which it can be particularly difficult. As wetion and assistance needs more visible. We also highlight below, migration and displacement arediscuss some of the consequences of failing to better understood as sitting on a predominantlydo so – for those displaced, the governments voluntary to predominantly forced continuum.responsible for them and others working in thehumanitarian and development fields. Raising awareness and understanding of people displaced in these contexts among policy-makers,This constitutes a step toward a more compre- practitioners, donors and the internationalhensive picture of internal displacement, with the community is important for three reasons.aim of ensuring that all IDPs in need of protectionand assistance, and those vulnerable to displace- First, it helps guide data collection as the basis ofment, are not excluded from efforts to prevent evidence for both policy and operational decision-and respond to the phenomenon. making at all levels. Second, a broader picture means limited resources can be better prioritisedDisplacement associated with criminal violence, and allocated, including for further data collectiondrought and development projects has not been and research to address knowledge gaps. Third,systematically quantified and monitored, in part insights into displacement as a multi-dimensionalbecause of constraints on our resources, but also and cross-cutting issue help policy-makers iden-because of limited access to data, conceptual ambi- tify links between agendas and objectives in areasguities about what constitutes displacement in including humanitarian action, sustainable devel-some contexts, and methodological issues related opment, peace-building, disaster risk reductionto the various drivers of slow-onset disasters and and climate change adaptation. OFF THE GRID: The world’s overlooked IDPs 43

Key findings and messages ||Global displacement figures do not capture ||Recognising people as internally displaced as many other contexts in which people are opposed to voluntary migrants helps to identify forced to flee their homes. More compre- them as people in need of particular attention hensive monitoring of displacement is needed from governments, humanitarians and develop- to ensure that all IDPs, and people vulnerable to ment organisations, and who should be priori- displacement, are included in efforts to respond tised for protection and assistance. to their needs and address longer-term develop- ment objectives. Displaced by development projects Displaced by criminal violence ||Rather than being priority beneficiaries, people displaced by development projects usually ||Criminal violence associated with drug traf- end up worse off, undermining development ficking and gang activity had displaced at least gains. They suffer a range of human rights viola- a million people in El Salvador, Guatemala, tions, and solutions are as elusive as for displace- Honduras and Mexico as of December 2015. ment associated with conflict and disasters. ||Studies have established a direct link between ||The most frequently cited global estimate criminal violence and migration, but such for people displaced by development proj- displacement in the region tends to remain ects is 15 million people a year since the unquantified and unaddressed for reasons mid-2000s. ranging from political to methodological. ||Accumulated figures for people displaced by ||Conceptual and information gaps result in a development projects appear only to be avail- lack of protection that means people fleeing able for China, where the total is 80 million criminal violence fall through the cracks, between 1950 and 2015, and in India, where leaving them with little choice other than to the total is 65 million between 1947 and 2010. embark on dangerous migrations, risking traf- ficking and murder, to neighbouring countries ||Indigenous people and the urban poor are or the US. particularly affected by displacement associ- ated with development projects because they Displaced by drought-related tend to live on land that is valuable in terms of disasters natural resources or as real estate. ||Drought is not a direct “cause” of displacement ||The planning that goes into development in and of itself, but has impacts on food and projects provides an opportunity to mitigate livelihood insecurity, including increasing poten- displacement and prepare for durable solutions tial for conflict over scarce resources. from the outset. If projects are undertaken with political commitment, adequate skills, sufficient ||Displacement might be identified as a tipping financial and institutional resources, a participa- point where abnormal movement patterns tory approach and respect for human rights, indicate the breakdown of normal coping the displacement they cause can result in strategies under severely stressed conditions. beneficial and sustainable development.44 G20R1ID6

Displaced bycriminal violencePeople flee criminal violence in a number of because of a general erosion of their day-to-forms, from gang violence and drug traffickers’ day quality of life and livelihood opportunities.103turf wars in Mexico and central America to clan Many flee after refusing to sell their land to drugfeuds in the Philippines and armed banditry in traffickers and receiving death threats as a result,CAR, but their migration is not systematically or to keep their children safe from gang recruit-monitored worldwide.97 This “unseen” flight has ment and violence.104 Some move in anticipa-widespread repercussions for individuals and tion of violence in the neighbourhoods wheresocieties, and raises the question of where the they live or work, some as a result of its impacts.phenomenon should fall within the displacement Others only flee when friends or family membersframework. have been attacked or killed.105Data shows that there is far less information Several studies have shown a direct empiricalon people who flee criminal violence than on link between criminal violence and migration inthose displaced by conflict, and an even weaker the region. A 2012 survey across 12 Mexicanresponse to their plight. There are probably many states established a clear association betweenmore people affected globally than the current violence – defined for the purpose of the surveydata reflects. as homicides, threats, extortion and a general atmosphere of violence – and a net migrationMexico and Central America: rate. It established that, once the effect of socio-a million displaced by economic conditions normally associated withorganised crime internal migration in Mexico was controlled for, the proportion of people moving from the mostOrganised criminal violence associated with violent municipalities was 4.5 times higher thandrug trafficking and gang activity has reached in those with similar conditions but lower levelsepidemic proportions in Mexico, El Salvador, of violence.106 This study provided one of theGuatemala and Honduras in recent years.98 As first evidence-based indications that significanta result, there were at least a million IDPs in the population losses in some areas of the countryregion as of the end of 2015, up from 848,000 were directly linked to violence perpetrated byat the end of 2014, many of them driven from organised crime groups.cities suffering the highest homicide rates in theworld and levels of violence comparable with a Unseen and in displacementwar zone.99, 100 limboThe generalised nature of the violence is well- Despite this evidence, displacement associatedestablished. Numerous articles and reports with generalised criminal violence in the regiondescribe the phenomenon, including the scale tends to remain hidden and unquantified. Peopleand diversity of criminalised zones, corridors and flee unseen and their subsequent protection andmicro-territories.101 The perilous trafficking areas assistance needs go unaddressed for a numberfor migrants in Mexican states such as Oaxaca of reasons.and Tabasco, extortion rackets in marginalisedareas of El Salvador, criminal turf wars in urban In some countries, there is a general lack of recog-Honduras and drug cartel feuds on Guatemala’s nition that criminal violence causes displacement.borders are some of the worst sources of large- Mexican authorities acknowledge the phenom-scale criminal violence.102 enon at a regional level, but not within their own borders, and Guatemala is similarly reluctant.107The intense nature of this generalised criminalviolence has driven population movements in a Honduras is currently the only Central Americanvariety of ways. Some people move in response country to officially recognise the phenomenon,to direct coercion and physical threats, others and in 2013 it set up a cross-institutional commis- OFF THE GRID: The world’s overlooked IDPs 45

After entering Mexico sion to develop policies to prevent and respond Such movements are not as explicitly forced asillegally near Ciudad to it.108 El Salvador has also been more respon- those triggered directly by an attack, but peopleHidalgo on their way sive, as evidenced by its decision to welcome who move in search of income and who wouldto the US border, many permanent delegations from the International not have done so were it not for the impactCentral and South Committee of the Red Cross (ICRC) and various of insecurity and violence on their livelihoodsAmerican migrants NGOs. Both countries, however, continue to warrant protection as IDPs. This is on the groundscontinue their journey pursue state policies that prioritise strong security that they were or felt obliged to flee, rather thanon the freight train measures over redress for victims.109 exercising a free choice to move solely to improveknown as La Bestia, the their economic circumstances.114Beast. The train begins The nature and scale of the displacementits journey in the town involved is also a factor. That triggered by conflict Piecemeal dataof Arriaga in Chiapas and disasters tends to result in large-scale, rela-State, where migrants tively visible population movements, but people Vague concepts and the perception of displace-climb on top of the fleeing criminal violence often do so in small ment as politically inconvenient in some countrieswagons, exposing numbers and keep a deliberately low profile. In combine to mean that quantitative evidence ofthemselves to the Honduras, individuals and families tend to leave people fleeing criminal violence in the region iselements and extortion their communities discreetly to avoid alerting the generally insufficient and inaccurate. The plightby criminal gangs lying dangerous groups from whom they are fleeing.110 of many has likely not been documented, andin wait along their Many are also reluctant to report the violence the scale of the phenomenon and the protectionroute. they have suffered for fear their persecutors may needs of those affected underestimated.115Photo: IOM/Keith track them down and exact retribution.Dannemiller, April 2014 Such cases are seldom reported in the media or The figures that do exist point to an alarming situ- elsewhere.111 Without official records it is diffi- ation. Research to quantify the scale of displace- cult to provide evidence of the true scope of the ment in Mexico indicates that around two per problem, but interviews and what little data is cent of the country’s population, or 1.7 million available suggest that displacement is a wide- people, were forced to migrate between 2006 spread and in some cases a daily occurrence.112 and 2011 because of the threat or risk of violence – an average of 330,000 people a year.116 Humanitarians’ focus on forced or coerced move- ment also often fails to capture the complex The fragmented quantitative data available for circumstances in which people flee generalised El Salvador is equally telling. A relatively robust criminal violence. An attack or atrocity may lead national survey in 2012 revealed that 2.1 per cent directly to displacement, but economic costs – of the country’s population, or around 130,000 when sales drop because customers are afraid to people, changed their place of residence in that go out in the street, or when criminals demand year alone as a result of threats. Almost a third a cut of profits – and lack of hope for the future of those affected had been displaced more than may also influence people’s decision to leave.113 once.117 There were more than 289,000 IDPs in El Salvador, a country described as the world’s most deadly outside a war zone, as of the end of 2015.11846 G20R1ID6