Populations in Peril: Decoding Patterns of Forced Displacement in Myanmar May 2024 Document of the World Bank Produced by the Poverty and Equity Global Practice Equitable Growth, Finance and Institutions Vice Presidency East Asia and Pacific Region. 1 © 2024 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Attribution—Please cite the work as follows: Sinha Roy, Sutirtha. 2024. Populations in Peril: Decoding Patterns of Forced Displacement in Myanmar. World Bank. 2 Acknowledgments The report was prepared under the guidance of Mariam J. Sherman (Country Director for Cambodia, Myanmar, and LAO PDR); Lalita M. Moorty (Regional Director for East Asia and the Pacific); Rinku Murgai (Manager, Poverty and Equity Global Practice); and Kim Alan Edwards (Program Leader for East Asia and the Pacific’s Equitable Growth, Finance and Institutions Practice Group). The author is most grateful to Noriko Takagi, Mulugeta Zewdie, Pankaja Bhattarai, Musa Al-Asad, Nii Ako Sowa, and Douglas Bradley Jennings for their guidance and thoughtful feedback throughout the preparation of this report. The author also extends gratitude to UNHCR colleagues and the broader team of UN colleagues for attending a seminar and sharing insightful feedback that led to significant improvements in the quality of the report. The author thanks Micheal Spier, Chris Foulkes, Sanghee Bang, and Heidrun Salzer for sharing critical inputs and helping cross- validate the observed trends in the data. The author also thanks Luc Verna, Virginie Tighe Lafleur, and Judith Johannes for their guidance and for providing valuable inputs during a seminar presentation. The author thanks Roy van der Weide (Senior Economist), and Travis Baseler (Assistant Professor, University of Rochester) for their key contributions to this report and Maria Eugenia Genoni (Senior Economist), Johannes G. Hoogeven (Senior Economist) and Jeffery Tanner (Senior Economist) for their technical support and guidance through the project. Mildred Gonsalvez, Thida Aung, May Oo Mon, and Buntarika Sagarun provided excellent administrative assistance to the project. Finally, the author sincerely thanks a survey agency for granting access to the multi-sectoral needs assessment survey -- the report would not have been possible without their support. 3 Executive Summary The objective of this report is to systematically document the living conditions of the internally displaced population (IDPs) in Myanmar that have been forcefully displaced since the military coup of 2021. IDPs are likely among the poorest and most vulnerable in Myanmar. Most IDPs originate in states and regions that had not experienced forced displacement before February 2021. With official data collection activities suspended, there is a lack of systematic data or information on the displacement experiences of the poorest households in Myanmar and their well-being amid multiple rounds of crises. Most accounts of IDP conditions rely on field-based reports from journalists and other organizations. Moreover, the number of IDPs in Myanmar continues to grow due to ongoing conflict. These circumstances underscore the need to assess the living standards of IDPs in Myanmar. The report uses data from the Multi-Sectoral Needs Assessment (MSNA). This household survey, which prominently includes IDPs in its sample, was conducted in 2023 by a research and data collection agency in collaboration with the United Nations Office for the Coordination of Humanitarian Affairs (OCHA), the United Nations High Commissioner for Refugees (UNHCR), and the Inter-cluster Coordination Group (which includes the Shelter and NFI cluster). This report is the first to put a spotlight on the household well-being levels of IDPs in Myanmar, who have been displaced since the military coup of 2021, and the first to provide a data-driven account of the displacement experience of IDPs in the country. The report highlights differences in demographic, socio-economic, livelihoods, and human capital characteristics, as well as differences in exposure to shocks and coping strategies across IDPs, households that have returned to their pre-displacement locations (“returnees”), and households that have never experienced displacement (“the non-displaced communities”). The report also underscores differences in welfare outcomes across IDPs depending on whether they reside in planned settlement locations, unplanned sites, or non-settlement areas. The report finds cyclic patterns of forced displacement where IDPs are repeatedly forced to seek shelter from violence, conflict and other security reasons. IDPs displaced since February 2021 (“new IDPs” or “newly displaced IDPs”) reported 5-6 displacement events between 2021 and 2023. Displacement was highly localized with most households seeking shelter within the same township as their pre-displacement location. New IDPs are predominantly rural and take shelter in locations that are planned and organized by community-based organizations. IDPs living in such areas typically have a coordination system that can ensure access to public goods and services. The demographic profiles of IDPs shows scars of violence and conflict. IDP families are more likely to be single parent and have more members below 20 years of age, than any other group. These single parent families are disproportionately female-headed and are more likely to report a widowed civil status than other households. The high share of female-headed households could be a result of higher mortality rates experienced by male-heads of IDP families. IDPs in unplanned settlement sites are likely the most vulnerable of all groups. They suffer from poor housing conditions, possess fewer identification documents and report more movement restrictions than any other group of IDPs. The biggest restriction around free internal mobility occurs at designated checkpoint locations. 4 IDPs are more at risk of landmine contamination than any other group. Households in the survey were less likely to report physical harm or injuries due to landmine contamination. Instead, reports of loss of livelihoods due to landmine exposure were higher among IDP families. School dropout rates among IDPs are 3x that of households who have not experienced forced displacements. When IDPs do enroll in school, they are more likely to attend informal institutions. But informal education centers may not be able to cater to the learning needs at more advanced levels. As a result, dropout rates are higher among older children. IDPs also depend heavily on more informal health facilities that are run by community-based organizations. IDPs face more difficulties in paying for health care services than accessing or travelling to health centers due to security challenges. Finally, coordination mechanisms in IDP settlement locations are generally associated with lower school dropout rate and greater access to informal education and health services compared to other displaced groups. Unemployment is the most important challenge for forcefully displaced families. IDPs have unemployment rates that is 3x that of non-displaced households. Male IDPs are more likely to be unemployed than females. Kayah, a state with high conflict intensity, has among the highest unemployment rates in the country. In addition to high unemployment levels, IDPs also have high levels of children participating in labor activities. Unlike non-displaced households, heightened security risks do not affect IDPs’ willingness to find or perform work. However, regulatory barriers and a worsening environment that makes it difficult to conduct business activities causes their unemployment rates to rise. One example of an environment factor that leads to adverse climate for business activities is exposure to IEDs and explosives. Landmine contamination around IDP settlements is robustly associated with higher rates unemployment. However, providing communities with landmine contamination education increases the chance of employment. Self-reported data from the survey shows that these education programs are associated with more employment because they likely may lead to an improvement in surrounding business conditions, for example by allowing greater retail activities to flourish around neighborhoods. This channel could therefore lead to higher employment among IDP households. Earning data from MSNA confirms that IDPs are among the poorest households in Myanmar. Average monthly earnings of non-displaced households are 80 percent higher than IDPs and 50 percent higher than returnee households. Labor outcomes of IDPs continue to lag those who have not been displaced, even if we control for households’ rural/urban status: IDPs living in rural areas report higher levels of unemployment and greater incidence of child labor than non-displaced households living in rural locations. Most IDPs rely on casual wages, followed by agriculture and self-employment opportunities. About one-fifth of all IDPs and one-tenth of returnees reported humanitarian assistance as one of their top 3 income sources. IDPs living in planned locations rely strongly on casual employment and humanitarian assistance. Only a handful of such families are engaged in self-employment or crop- production. In comparison, those in unplanned are more actively engaged in farming. Consumption data shows that even among IDPs, those living in unplanned locations are poorer than others living in planned or non-settlement areas. These households are disproportionately affected by high food and fuel prices and allocate more of their budgets to food and fuel than any other group. As a result, their budgetary allocation to long-term human capital investment in education and health is lower than other groups. 5 IDPs in unplanned locations have larger households and lower food security. The higher levels of farming observed among this group likely reflect subsistence farming rather than commercial, market-oriented crop production. Large household sizes and high food insecurity lead IDPs to adopt emergency-level coping strategies. IDPs in unplanned locations are 4 to 8 times more likely to engage in such coping strategies than other groups. Due to a significantly higher rate of open defecation, IDPs in unplanned locations are at high risk of exposure to fecal bacteria. Exposure to fecal bacteria for children under 5 years carries long- term human capital risk of stunting, wasting, and undernourishment. IDPs are generally well- connected to financial services that can be offered over the phone. Sixty-seven percent of IDPs in planned sites and about 50 percent of those in unplanned sites have access to mobile money platforms. In comparison, about half of all returnee and non-displaced households have accessibility to mobile money channels. Compared to non-displaced households in rural areas, rural IDPs are more likely to have children that are out of school, more likely seek community-based healthcare services, and more likely to suffer more from adverse mental health conditions. In addition, rural IDPs report higher levels of unemployment and greater incidence of child labor than non-displaced households living in rural locations. 6 1. Introduction Myanmar has faced successive challenges since the beginning of the COVID-19 pandemic. The 2021 military coup, coupled with soaring food and fuel inflation, currency depreciation, and repeated waves of conflict, has resulted in a 15 percent contraction in real wages since 2017. Consequently, educated and experienced workers in Myanmar have been compelled to transition from manufacturing to less productive agricultural activities, adopting precarious coping mechanisms to withstand the impact of multiple shocks. New household surveys conducted by the International Food Policy Research Institute (IFPRI) and the World Bank since 2022 have filled important data gaps, as official survey activities in the country have remained suspended. These surveys have utilized extensive telephonic databases to proportionally sample households from states and regions according to their population shares. Survey samples are collected from almost all townships of Myanmar, meaning that household well- being indicators are reliably representative up to state and regional levels and contain sufficient sample sizes to highlight variations across locations. However, it is likely that IDP households are insufficiently represented in these surveys. Despite their scientific design, internally displaced households may not be adequately represented in these surveys for the following reasons. First, representative surveys that do not explicitly target IDP households through appropriate stratification risk excluding IDP families from survey samples because they are likely residing in remote locations to evade detection. Second, anecdotal evidence suggests that IDP households may not have sufficient access to mobile phones or may be in areas without mobile network coverage. This increases the chances of existing telephonic surveys missing out on these populations. Finally, even if IDPs are proportionately represented in representative surveys, they would represent only a small fraction of the overall survey sample size. This means that IDP-centric estimates derived using such surveys will be prone to statistical error and, therefore, be less reliable. IDP households are likely among the poorest and most vulnerable households – meaning that interventions designed for the poorest populations of Myanmar must target this group of households. There are additional reasons to document patterns of forced displacement in Myanmar systematically. According to UNHCR (2023), Myanmar is one of the major contributors to the IDP population around the world. With worsening conflict conditions, its share in the global IDP population is likely to rise further. Before 2021, the majority of Myanmar's IDPs were from Rakhine, Kachin, and Chin. Following the February 2021 coup, households most impacted by forced internal displacement are now found in Kayah, Sagaing, Tanintharyi, Magway, and other states. Many of these locations, now home to "newly" displaced populations, had not experienced forced internal displacement before. Little is known about this population – apart from anecdotal reports -- emphasizing the need to bridge existing data gaps and systematically examine their socio-economic conditions and vulnerabilities. IDPs comprise a remarkably diverse group of households with varied displacement experiences. While some anecdotal reports suggest that IDPs seek protection in remote, hilly, and forested locations 1, others emphasize the lives of IDPs in settlement communities 2. Similarly, some reports 1 https://www.irrawaddy.com/news/burma/exodus-tens-of-thousands-flee-as-myanmar-junta-troops-face-last- stand-in-kokang.html 2 https://asiatimes.com/2023/04/on-the-front-lines-with-the-free-burma-rangers/ 7 indicate that IDPs are engaged in farming activities through community gardens 3. In contrast, others indicate that IDPs who used to be farmers are now taking up casual work activities 4. This report aims to produce a systematic account of the displacement experiences of IDPs in Myanmar and spotlight the unique demographic and socio-economic characteristics of this population. The report relies on the Multi-Sectoral Needs Assessment (MSNA). This household survey counts IDPs as a prominent group in its sample and was conducted in 2023 by a research and data collection agency in collaboration with the United Nations Office for the Coordination of Humanitarian Affairs (OCHA), United Nations High Commissioner for Refugees (UNHCR) and the Inter-cluster Coordination Group (which includes the Shelter and NFI cluster). This 2023 round of MSNA was conducted before the escalation of conflict events in October 2023. As a result, the trends reported in this report do not capture the latest patterns of displacements. The report is organized as follows: chapter 2 describes the MSNA household survey data and adjustments applied to the data to ensure that they offer representative estimates at the subnational and population group levels. Chapter 3 reviews the displacement experiences of IDPs and those who have returned to their pre-displacement locations, i.e., returnees. Chapter 4 provides a demographic sketch of IDPs and returnees and shines a light on their civil registration status and their exposure to movement restrictions and landmine contamination. Chapter 5 explores whether IDPs can access education and health services and if these levels are commensurate with other population groups. Chapter 6 is a deeper dive into livelihoods, employment, earnings, consumption, food insecurity, and coping strategies of IDP families. Chapter 7 reviews access to WASH facilities and whether IDPs can access financial systems through formal, informal, or mobile money channels. Chapter 8 concludes by outlining a policy agenda for improving the lives of IDP households in Myanmar despite the fragile security environment in the country. 3 https://iirr.org/news_and_updates/community-garden-myanmar/ 4 https://www.voanews.com/a/farmers-in-myanmar-s-kayin-state-face-indiscriminate-attacks/6581091.html 8 2. Data Sources and Sampling Weight Adjustments This report uses data from a multisector needs assessment (MSNA) conducted by a survey data collection agency. The objective of the MSNA is to provide a detailed overview of the current humanitarian needs and gaps of the crisis-affected population in Myanmar and to inform humanitarian operations and resource planning for the following calendar year. The MSNA data allows humanitarian agencies to prioritize the needs of vulnerable households, identifying variations among population groups and across geographies. The 2023 round of MSNA consisted of a sample size of 9,230 households from all 18 states, regions, and sub-regions of Myanmar. Data were collected from June 20 to August 2, 2023, using a combination of telephonic (21 percent of samples) and face-to-face (79 percent) modes. The survey had to use dual modes of implementation due to the challenging security conditions in certain township locations. Note that this data does not include the rise in displacement activity due to rising conflict in the second half of 2023. As a result, the trends observed in this report do not speak to the displacement experiences of about 0.5 million individuals who have been affected by violence and displacement since then. The MSNA was stratified at the level of state-regions and population group. The survey considered three groups for stratification – IDP, IDP families who have subsequently returned to their pre- displacement locations, and other crises-affected people (OCAP). The OCAP group includes non- displaced families. The survey initially aimed to enumerate 200 households for each population group and state-region combination. In other words, the MSNA targeted a sample size of 3600 OCAP households across 18 states, regions, and sub-regions; 200 non-displaced stateless population groups in the Rakhine state; 2400 IDP samples from 12 states, regions, and sub- regions where IDPs were known to reside and 2200 returnee households from 11 states, regions and sub-regions where returnees were known to reside. An updated township-level list of the IDP population maintained by UNHCR and a 2014 frame of the OCAP population maintained by UNFPA served as the area frame for the MSNA survey. The dual mode of survey implementation and the stratification at the state/regional and population group levels mean that estimates directly derived using the MSNA are: (1) for all population groups (excluding Rakhine NDSP) are indicative only, since they include a share of telephonic survey conducted using quota sampling; and, (2) not representative at the state and regional levels (instead, they are representative at the state/regional-population group level, only in areas and population group where data collection was exclusively conducted based on face-to-face interviews 5). While the MSNA seeks to inform humanitarian programs at the field level, this report's objective is to derive estimates of the three population groups that are representative of the population at the subnational levels and provide a more strategic overview of household welfare conditions of IDP families. To achieve this objective, the sampling weights in the MSNA had to be adjusted in two steps. First, potential biases introduced by non-representative phone survey samples were corrected using a maxentropy reweighting algorithm 6, applied separately at the state and regional levels. The 5 Additionally, as is the case for most MSNAs globally, they are only representative of HHs residing in those regions/states’ accessible areas. 6 Excluding telephonic interviews from the sample would mean that townships where security risks are high are not incorporated in the analysis. Trading-off the costs and benefits of including non-representative 9 procedure used the following target variables: share of members in the age groups 0-5, 6-10, 11- 15, 16-20, 21-30, 31-50, 51-60, 61-70, 71-80 and over 80; share of females in the household; whether household resides in rural areas; log age of the head of the household; whether the head of the household is married; household size of 1 or 2 and household size of 3 or 4. Representative values of these target variables were obtained using the World Bank’s MSPS surveys conducted in 2022-23 7. The sampling weights in MSNA were then adjusted so that weighted estimates match representative values obtained from MSPS 8. Second, the maxentropy adjusted weights from the first stage were further corrected based on the share of new and old IDP population, IDP returnees and OCAP within a state and region. The updated state-wise population totals for IDPs and returnees were obtained from UNHCR. The state level population for OCAP for each state and region was estimated using the following formula: 2022 = 2019 + 2019−2023 − 2019−2023 − − . The term 2019 represents the total population in the state in 2019, as reported in the Intercensal Censal Survey. 2019−2023 and 2019−2023 represent migrated population entering and existing a state/region since 2019 respectively. These were estimated using data from MSPS- 2022. The migrated population data from MSPS includes those that were forcefully displaced as well as others; the population of IDPs and returnees were therefore subtracted to obtain the final OCAPs population at the subnational level. telephonic samples, the current analysis chose to incorporate these samples into the analysis but adjusted its sampling weights to reflect their representative shares in the population. 7 This step assumes that estimates of target reweighting variables obtained from MSPS 2022-23 are sub nationally representative and represent weighted averages of IDP, returnees and OCAP population of each location. Implicit in this assumption is that the OCAP population reflects all non-displaced populations in Myanmar (that is, those that are neither IDPs nor returnees). See Sinha Roy, 2023 for more details. 8 Raw weights in MSNA were not used as priors in the maxentropy procedure as several observations in MSNA were assigned extreme sampling weights while others even had zero weights. 10 3. Displacement Typologies Key Messages • While Myanmar has had a history of forced internal displacement, the military coup of February 2021 has introduced forced displacement to areas and populations that have never experienced involuntary displacement before. • Areas with high fragility and conflict have more IDPs but the displacement experiences of IDPs vary considerably across the country. • Displacement is a recurring event; newly displaced households reported a median of 5-6 displacement cycles in the 2½ years prior to the survey. • Recurring displacement is most common in Sagaing, with IDPs experiencing a median 8-9 displacement cycles in the previous 2½ years. • Newly displaced IDPs were generally displaced within the same village and lived in planned settlement locations at the time of displacement. These areas often have a coordination mechanism to access public services. • Towns and villages typically had either planned or unplanned IDP sites, with rare instances of both types of settlements coexisting. The military coup of 2021 led to forced internal displacement in areas that had not experienced displacement before. The share of IDPs in the population has increased from 0.6 percent to approximately 4 percent since February 2021. Less than 0.1 percent of all IDPs had returned to their pre-displacement locations ("IDP returned" hereafter). The remaining 94.3 percent of the population has not undergone displacement, encompassing populations vulnerable to conflict and violence. Figure 1 shows the distribution of internally displaced populations in Myanmar. Figure 1: Geography of forced displacement in Myanmar 11 Source: UNCHR, https://data.unhcr.org/en/documents/details/106717 Before the coup, forced displacement was predominantly observed in Rakhine, as well as in Kachin and Chin. Following the military coup in 2021, a significant portion of the population in Kayah, Sagaing, Chin, Kayin, Rakhine, Tanintharyi, Magway, and Kachin experienced forced displacement (Figure 2). Notably, most of these states and regions had limited experience of forced displacement before 2021. In this report, individuals displaced before and after the coup are referred to as "older displaced" (or “older IDPs”) and "newly displaced" (or “new IDPs”); households that have returned to their pre- displacement locations are referenced by the term “returnees.” Finally, households that have never experienced displacement are referred to as “non-displaced” populations. The displacement experiences of IDPs are heterogeneous, with conflict and violence being one of the strongest correlates of forced displacement. Sagaing, which recorded the highest number of conflict incidents in the country, is also home to approximately 820,800 newly displaced persons (Figure 3). The upward trajectory depicted in Figure 2 suggests that, overall, states and regions with more internally displaced persons (IDPs) witnessed more significant conflict since 2021. 12 Figure 2: Population shares by displacement 9 Figure 3: Conflict Incidence and New IDPs Newly IDP Older IDP IDP returned 4 (Log) number of violent conflict 40% shan sagaing 3.5 rakhine yangon events since 2021 kachin magway 3 chin kayin mandalay mon kayah 20% ayeyawad taninthary bago 2.5 y i 2 nay pyi 0% taw 1.5 bago mon shan kayin kayah sagaing magway kachin ayeyawady mandalay yangon chin rakhine tanintharyi nay pyi taw 0 2 4 6 8 (Log) number of New IDPs Notes: Data – MSNA, 2023. Data on conflict events is from ACLED. Estimated using adjusted population weights. Involuntary displacement is not a singular life occurrence but a recurring event for many. Returnee IDPs reported a median of 2 to 3 displacement events over the past 2.5 years, compared to a median of 5 to 6 forced displacement events over the same period amongst new IDPs. The median time passed since the last displacement experience is 13 to 18 months for returning IDPs and 7 months to 1 year for new IDP families. This indicates that IDP households generally wait over a year after their last displacement event, probably to allow conditions to stabilize, before moving back to their pre-displacement locations and transitioning to a 'returnee' status. Unlike older IDPs, newly displaced households often sought refuge in villages within the same township as their pre-displacement residence. Approximately 80 percent of recently displaced IDPs, inhabited rural areas, contrasting with older IDPs, 60 percent of whom lived in urban settings. Additionally, 80 percent of newly displaced individuals, compared to 60 percent of older IDPs, stayed within the same township 10 as their pre-displacement residence. Notably, only a small number of IDPs, whether new or old, crossed administrative state and regional boundaries during their displacement. Table 1: Displacement and residential patterns among IDP groups Location Currently living in Displacement No. of Displaced Displacement Non- times Planned Unplanned Group Rural across S/R within the settlement displaced site site borders township location in past 2.5 years Newly IDP 79% 15% 81% 59% 34% 7% 5.7 Older IDP 38% 16% 60% 63% 35% 2% - - IDP returned 94% 11% 90% 3% 1% 97% 2.1 4 9 The selection of sample clusters was conducted based on information provided by data collection partners on places hosting IDPs - as such, some biases may remain but cannot yet be quantified. 10 Townships are the third administrative division in Myanmar. The comprise of village tracts and urban wards. Overall, there are 330 such townships in Myanmar. 13 Notes: Displaced across the S/R border indicates a change in state and region of residence due to forced displacement. Planned sites include settlements that are organized by an organization, agency, or group. The unplanned site indicates no such arrangement and includes boarding or collective sites – areas that were not originally intended to be residential locations but have now been repurposed to house IDP households. Non-settlement locations could include IDPs living in houses owned by family members, relatives, friends, etc. Data – MSNA, 2023. Estimated using adjusted population weights. Types of Settlement locations IDP households reside in three main types of settlement locations: • planned sites – are locations managed by communities, organizations, agencies, or religious groups. • unplanned settlements – are areas where IDPs have settled collectively without seeking authorizations. • non-settlement locations -- include IDP households residing in individual locations that are not settlement units. These households have most likely sought shelter with relatives, family, or friends. Approximately two-thirds of all IDPs resided in planned sites. Around 7 percent of recent IDPs resided in non-settlement areas, highlighting that the majority sought shelter in planned or unplanned communal sites during forced displacement (Table 1). Seventy percent of IDPs residing in these locations have organized themselves under a coordination system to manage provisions of public services. In comparison, only 35 percent of the IDP population in unplanned settlements developed a coordination system to organize public services. Upon returning, almost all returnee households chose to live as individual family units rather than in clustered settlement locations. The displacement experiences of new IDPs vary significantly across states and regions. In Bago and Rakhine, new IDPs were more likely to reside in urban settlements, while in Kachin, they are predominantly rural and residing in planned locations, possibly due to the presence of UNHCR camps that predated the February 2021 coup (Figure 4a). No new official camps have been built since February 2021 – including in Sagaing, which is home to the highest number of new IDPs. Thus, 60 percent of new IDPs in Sagaing reside in unplanned rural settlements (Figure 4b). New IDPs in Sagaing have also reported a median of 9 to 10 displacement events in the past 2.5 years, compared to 2 to 3 in other areas, reflecting the highest level of insecurity across all other states and regions. 14 Figure 4a: Locations of New IDPs Figure 4b: Current living situation of New IDPs 100% 12 100% 80% 10 80% 8 60% 6 60% 40% 4 40% 20% 2 0% 0 20% 0% Rural Displaced across S/R boundaries No. of times displaced in past 2.5 years (RHS) Planned Site Unplanned site Non-settlement site Notes: Displaced across the S/R border indicates a change in state and region of residence due to forced displacement. Planned sites include settlements organized by an organization, agency, or group. The unplanned site indicates no such arrangement and includes boarding or collective sites – areas that were not originally intended to be residential locations but have now been repurposed to house IDP households. Non-settlement locations could include IDPs living in houses owned by family members, relatives, friends, etc. Data – MSNA, 2023. Estimated using adjusted population weights. 15 4. A Profile of Internally Displaced Populations Key Messages • IDP families are more likely to be single-parent families and have children below 20 years of age than any other group. This demographic structure likely reflects a disproportionate mortality risk carried by male headed IDP families. • IDPs living in unplanned locations suffer from poor housing conditions, fewer identification documents and report more movement restrictions than IDPs living in any other group. The biggest restriction around free internal mobility occurs at designated checkpoint locations. • IDPs face the highest risk of landmine contamination. Interestingly, this contamination does not result in physical harm or injuries but rather leads to the loss of livelihoods among IDP families. 4.1 Age, gender and civil status One-third of all IDPs are under the age of twenty, 6 percentage points more than non-displaced communities (Figure 5a). Although the proportion of females and female-headed households is similar across IDP and non-displaced populations, female-headed IDP households are 12 percentage points more likely to be widowed than non-displaced communities (Figure 5b). By comparison, marriage or widowed status of male-headed households are comparable across both groups. This suggests that female-headed IDP families are more likely to be single-parent family units than non-displaced communities – which means that they have fewer earning members, fewer caretakers to provide support and education to children, and are likely more prone to shocks. Figure 5a: Age Distribution and Gender Ratios Figure 5b: Share of population living in female- headed households by their civil status 40% 3.90 IDP IDP-returnee Non-displaced 52.9% 35% 3.40 30% 46.2% 2.90 25% 43.8% 42.9% Population shares Gender Ratio 41.3% 20% 2.40 15% 1.90 35.9% 10% 1.40 5% 0% 0.90 IDP IDP_returnee Non-displaced 0-19 20-39 40-59 60-79 80+ Age Cohort female HoH, married female HoH, widowed Notes: Data – MSNA, 2023. Estimated using adjusted population weights. Instances of children below 18 years living away from their families are high among returnees and IDPs. Seven percent of IDPs and 11 percent of returnees indicated that they have a child below 18 16 living apart from the household (Table 2) – compared to only 4 percent in the non-displaced group. It is more likely for male children to be living away from the family than female children across all groups. Table 2: Share of households reporting having a child (<18 years) living away from the family. % households with children % households with at least 1 % households with at least 1 (<18 years) away from boy child away from family girl child away from family family (Conditional on the child living away (Conditional on the child living away from family) from family) IDP 7% 63% 59% IDP Returnee 11% 69% 61% Non-displaced 4% 71% 55% Notes: Data – MSNA, 2023. Estimated using adjusted household-level weights. The last two columns of the table show results for households that have at least one child living away from the family. The reasons why children stay away from their families vary considerably by gender and population groups (Table 3). For instance, over 80 percent of returnee households and more than 50 percent of non-displaced had children living away from the family due to educational reasons. IDPs, on the other hand, were more likely to report children living away from the family due to riskier reasons such as child labor: 55 percent of IDPs reported having children living away from home due to employment. A third of all IDP households reported having their girl child living in another country for reasons unrelated to education, employment, marriage, conscription, abduction, detainment, or reportedly missing. Table 3: Reasons for children (below 18 years) living away from family. Marriage Employment Studying Move Joined involuntary Abducted Missing Detained abroad army conscription BOYS IDP 3% 55% 35% 13% 2% 0% 0% 0% 0% Returnee 0% 10% 89% 2% 11% 0% 0% 0% 0% Non- 4% 32% 53% 5% 5% 0% 0% 0% 0% displaced GIRLS IDP 3% 31% 38% 34% 4% 0% 0% 0% 0% Returnee 8% 5% 83% 2% 12% 0% 0% 0% 0% Non- 9% 30% 57% 3% 5% 0% 0% 0% 0% displaced Notes: Data – MSNA, 2023. Estimated using adjusted household-level weights. The MSNA instrument allowed respondents to select multiple options for reasons why child/children are living away from the household. As a result, the sum of cells can add up to more than 100% across rows. The shares are conditional on families having at least 1 boy or girl child living away from family. 4.2 Shelter Conditions Forty-five percent of all IDPs live in makeshift shelters 11 with low-quality roof and wall materials. In contrast, non-displaced and returnee households are 25 percentage points more likely than IDPs to reside in structures fortified with improved quality roofs and wall materials (Figure 6a). Among IDPs, those residing in unplanned locations face a higher risk of poor-quality roofs, as only 36 percent of such households have access to a fortified roof, which is less than half compared to all other groups (Figure 6b). Unplanned IDP locations predominantly consist of makeshift shelters, 11 The main shelter categories for IDPs include collective shelters (16%), long houses (9%), makeshift shelters (13%), mason houses (7%), permanent solid structures (19.35%), semi-permanent structures (26%) and tents (6%). 17 including tents and semi-permanent structures, while planned and non-settlement locations are more likely to have permanent structures (Figure 7). Within rural areas, the share of households with fortified roofs among IDPs and displaced populations is 65 and 90 percent, respectively, and those with fortified walls are 39 percent and 66 percent. Figure 6a: Share of households using fortified Figure 6b: Share of households using fortified materials for roof and wall – by population materials for roof and wall – by IDP group group 89% 90% 89% 80% 65% 66% 46% 39% 39% 38% 42% 36% Within IDPs / Returnee roof walls roof walls IDP IDP returnee Non-displaced planned unplanned non-settlement site Notes: Data – MSNA, 2023. Estimated using sampling weights. Fortified materials for roofs include cement, ceramic tiles, calamine cement fibers, roofing shingles, wood, wooden planks, palm and bamboo, CGI, and metal tins. Materials used in roofs that are not fortified include bulrush leaves, rustic mat, sod, tarpaulin, thatch palm leaf, and other roofing materials or households with no roof. Similarly, fortified walls include bricks, cane palm trunks, cement, cement blocks, CGI, covered and uncovered adobes, stone with lime cement and mud, and wood plank shingles. Finally, walls with non- fortified walls include materials such as bamboo with mud, cardboard, dirt, reused wood, tarpaulin, and other materials, or houses with no walls. Figure 7: Type of shelter by IDP group 45% 40% 35% Share of IDP population 30% 25% 20% 15% 10% 5% 0% house collective shelter make shift permanent semi permanent unfinished, other shelter or tent shelter shelter or no shelter hosted or planned unplanned non-settlement site Notes: Data – MSNA, 2023. Estimated using sampling weights. 18 4.3 Possession of Key Identification Documentation IDPs lack essential identification documents necessary for accessing public services and, at times, for domestic travel, where screenings occur at designated security checkpoints. Seventy-three percent of IDPs, in contrast to 87 percent of the non-displaced community, possessed a valid Citizenship Scrutiny Card (CSC), Associate Citizenship, or a Naturalized citizen card. Similarly, 41 percent of IDP families reported that every member held a valid birth certificate, compared to 65 percent among the non-displaced community. Ownership rates of CSC and other citizenship cards are lowest in Tanintharyi (62 percent), Sagaing (64 percent), and Magway (65 percent). Identification document possession is lower among IDPs in unplanned sites. Only 60 percent and 33 percent of such IDP households reported that every member possessed a CSC, Associate Citizenship, or Naturalized citizen card and a Birth Certificate, respectively. In comparison, 80 percent and 45 percent of IDP families living in planned sites reported similar figures. Possession of identification documents can improve IDPs’ access to public services and job opportunities but can also expose families to security risks by revealing sensitive information about their ethnic or religious profiles. Consequently, IDPs who choose to live in unplanned settlements are likely to mitigate security risks through the anonymous and informal nature of such sites and are also less likely to report owning identification documents. Approximately 60 percent of all IDPs in unplanned settlements reported possessing identification documents, compared to 80 percent among IDPs living in planned locations. 4.4 Restrictions on Movement IDPs and returnees encounter limitations on free movement. Fifteen percent of all households in the survey reported safety or security restrictions on their ability to move freely in the three months preceding the survey. Mobility restrictions were much higher among returnees and IDP households – a third of all returnees and over a quarter of all IDPs reported mobility issues (Figure 8). In comparison, only 4 percent of non-displaced households reported mobility restrictions in the survey. IDPs residing in unplanned locations were most at risk, with half of all such households reporting issues with free movement. Figure 8: Share of households reporting safety or security restrictions in their ability to move freely in the past 3 months. 50% 35% 28% 22% 17% Within IDPs 4% IDP returnee Non-displaced IDP hosted or unplanned non-settlement planned site Notes: Data – MSNA, 2023. Estimated using sampling weights. The three bars on the right show the fraction of IDP households that reported movement restrictions based on their current site location. The figure excludes individuals who preferred not to answer this question – about 106 households (1.15 percent of the sample) belong to this group. 19 The primary factor impeding mobility is stoppages at checkpoints. Among those encountering mobility restrictions across Myanmar, 65 percent identified checkpoints as one of the main causes (Table 4). This factor surpasses all other reasons for movement restrictions. Furthermore, in Sagaing, Chin, Tanintharyi, Mon, and Kayin, more than one-fifth – and up to half of the entire population – experienced movement restrictions. In these states, mobility was hindered both by impeded mobility at checkpoints and the direct effects of conflict events. Table 4: Share of population experiencing movement restrictions and its causes. Causes of movement restrictions Movement Blocked state Covid Conflict Violent Checkp Valid Mine In-voluntary Other restricted Bribery by armed rules events event oints IDs blast conscription reasons forces sagaing 46 0 56 6 71 0 0 28 16 1 0 chin 46 3 39 1 64 0 1 34 17 0 0 tanintharyi 30 0 44 9 88 1 19 28 18 2 0 mon 27 0 48 0 24 0 0 48 3 0 1 kayin 21 4 70 7 20 0 0 10 17 0 1 bago 13 2 58 6 81 0 0 47 44 2 0 kayah 12 1 40 22 81 6 3 1 22 0 3 kachin 12 0 65 37 72 0 2 9 26 2 0 magway 8 0 6 4 100 21 0 15 23 4 0 shan 7 1 22 1 65 0 1 5 23 0 6 nay pyi taw 2 0 0 20 60 0 0 20 0 0 0 yangon 1 0 0 0 67 0 0 0 0 0 33 ayeyawady 1 0 0 0 100 10 10 0 0 0 0 mandalay 1 0 0 0 100 0 0 0 0 0 0 rakhine 1 0 0 22 56 0 0 0 33 0 0 Myanmar 15 1 46 6 65 1 2 27 19 1 1 Notes: Data – MSNA, 2023. Estimated using sampling weights. All figures are in percentages. The table excludes individuals who preferred not to answer the question on mobility restrictions – about 106 households (1.15 percent of the sample) belonged to this group. If households indicated that their movement was restricted, they could choose multiple options from the list of causes. As a result, percentages across rows of causes do not add up to 100. 4.5 Threats from Explosives and IEDs IEDs and explosives affect the majority of all IDPs, especially those living in unplanned locations. IED exposure likely depresses local demand for goods and services and adversely affects livelihoods. Half of all returnees and a quarter of all IDPs have been exposed to landmines, bombs, missiles, IEDs, bullets, or other explosive weapons in the 12 months preceding the survey. In contrast, only 4 percent of the non-displaced population encountered the effects of IEDs (referring here to a whole range of explosive ordinances) during this time. The majority of IDP and returnee households indicated that these IEDs impacted their livelihoods and movements more than causing health damages, injuries, or mortality. This could suggest prior knowledge of IEDs amongst IDP and returnee households, which is used to mitigate direct mortality and injury-related risks. Indeed, reported levels of awareness are higher among IDPs, with 24 percent of such households and 20 percent of returnees reportedly receiving mine risk education within their communities. In comparison, non-displaced households were less aware of such services, with only 4 percent reporting such education in their town or village. IDPs living in unplanned locations and non-settlement sites are 5 and 8 percentage points more vulnerable to such threats than those in planned locations. About a quarter of IDPs in unplanned locations reported effects on livelihoods, while one-fifth of IDPs in non-settlement sites experienced movement restrictions due to the presence of explosives in their surroundings. Reported rates of health damages or mortality induced by explosives were low amongst all three IDP groups. 20 5. Access to Education, Health and Psychosocial Services Key Messages • Although school enrollment rates have improved between academic year 2022-23 and 2023-24, dropout rates among IDPs are 2x that of non-displaced communities. • When IDPs do enroll in school, they are more likely to attend informal institutions. But informal education centers may not be able to cater to the learning needs at more advanced levels. This may explain the higher dropout rates among older children. • Gaps in enrollment rates between IDP and non-displaced populations are particularly high in rural areas. One-third of rural IDP children between 6 to 17 years are reportedly out of school – about twice as many as non-displaced children in the same age cohort. • IDPs depend heavily on community health facilities, facing affordability challenges more than accessibility or travel insecurity issues. • Urban IDPs seek more healthcare services – both in public as well community health centers – than non-displaced households living in such areas. • IDPs in areas with established coordination mechanisms are less prone to school dropouts and more likely to access health services compared to other groups. • About one-third of all IDPs and about 40 percent of rural IDP households are suffering from bad mental health conditions. Mental health needs of rural IDPs are significantly more than their urban counterparts as well as non-displaced communities living in rural areas. • Despite the major need for psychosocial assistance, all groups of households – IDPs, returnees and non-displaced – reported negligible levels of mental health support services available in their communities. 5.1 Education School enrollment rates have recovered between academic years 2022-23 and 2023-24 across all population groups. The Net Enrollment Rate (NER) for 6- to 17-year-olds has risen from 68 percent to 83 percent between the two academic years 12. The rise can be attributed to a 9 percent increase in formal enrollment and a 5.5 percent increase in informal education. The former includes re- enrollment into institutions accredited by the de-facto authorities, while the latter includes centers operated by the community, non-profit organizations, and self-tutoring. High-school and senior secondary school-aged children were most likely to re-enroll in a formal school, while informal school enrollment has risen the most for 6 and 7-year-olds. 12 Enrollment data for the two academic years were collected in MSPS and MSNA surveys respectively. The two surveys have slightly different questions related to enrollment. The large increase observed between the two surveys is unlikely to be due to differences in instrument alone. Upcoming data from the 2023-24 round of MSPS is expected to shed further light on these trends. 21 School dropout rates are about twice as high amongst IDP families. IDP families are three times more likely to attend informal institutions compared to non-displaced communities. Only half of all school-aged children of IDP families have enrolled in formal education -- roughly 30 percentage points lower than that of non-displaced communities (Table 5). The non-enrollment rate among IDPs is particularly high around age 14, signaling a trend of children exiting the educational system just before starting high school (Figure 9). Enrollment in informal education is likely more effective at younger ages for children of IDP families. Still, it is unlikely informal centers provide the level of education that is needed by higher age groups. This could explain the bigger gaps in enrollment rates between IDP and non-displaced households among older cohorts of students. Table 2: Net Enrollment Rates during the 2023-24 academic year Enrolled in formal Enrolled in informal Group Not Enrolled institutions institutions IDP 30% 46% 24% Returnees 21% 73% 6% Non-displaced 16% 77% 7% Notes: Data – MSNA, 2023. Estimated using adjusted population weights. Figure 9: Formal, Informal, and Non-enrollment rates for IDP, Returnees and non-displaced communities Notes: Data – MSNA, 2023. Estimated using sampling weights. The graph shows enrollment rates for the 2023-24 academic year for children aged between 6 and 17. The enrollment gap between IDP and non-displaced children is significantly higher in rural areas. Seventeen percent of non-displaced rural children are reportedly out of school, but the same cohort amongst the IDP population has dropout rates that are twice as high. By comparison, the gap in enrollment rates between the two groups is considerably narrower in urban areas. The urban cohort of children in IDP and non-displaced populations are 5 percentage points apart. 22 Informal education networks have supported school enrollment amongst IDPs in many areas, but gaps remain in other states and regions. IDP enrollment rates in Mon, Magway, Kayah, Chin, and Sagaing lag significantly behind non-displaced communities (Figure 10). In Kayin, Shan, Bago, and Tanintharyi, non-enrollment rates for IDPs are nearly equal to non-displaced communities, thanks to informal education channels. In Kayah and Sagaing, IDP enrollment in informal education exceeds that of the non-displaced community, but overall enrollment rates for IDPs remain below those of non- displaced communities. Notably, in Kachin and Rakhine, areas with a history of forced displacement prior to 2021, IDPs have reported lower dropout rates than non-displaced communities and, notably, have higher formal enrollment rates. This means that in areas with pre-existing displacement experience, households are continuing to invest in human capital accumulation. This evidence is also supported by the fact that returnee households – concentrated in Kayin, Rakhine, and Chin – are most likely to have children living away from the family to pursue education. Figure 10: Excess dropout rate and incidence of informal education among IDPs and non-displaced communities. Notes: Data – MSNA, 2023. Estimated using adjusted population weights. The figure is restricted to states and regions with reported IDP populations. IDPs residing in areas that have a mechanism to coordinate public services are observed to have a lower school dropout rate of 15 percent, compared to 38 percent in locations where no mechanisms exist (Figure 11). In the latter areas, a third of all children are enrolled in informal education channels. Such channels are less effective at keeping children in school at higher age groups (Figure 9) – meaning that children at higher ages are more likely to drop out in such areas. Figure 11: IDP educational status with coordination mechanism 23 Notes: Data – MSNA, 2023. Estimated using adjusted population weights. The figure is restricted to age groups 6 to 17. 5.2 Health Services Most households in Myanmar seek healthcare services at private facilities, but for IDPs, access to community health centers was more important than any other group. Households in Myanmar utilize a combination of different types of facilities when seeking healthcare. Respondents in the MSNA survey were allowed multiple choices of response for access to healthcare facilities such as public, private, community, and faith-based centers. Consultations for acute illnesses – such as fever, diarrhea, etc. (69 percent of all individuals) and chronic illnesses such as diabetes and hypertension (17 percent) – were the main requirements across all respondents. The proportion of individuals reporting the need for such consultations was approximately similar across IDPs, returnees, and non-displaced communities. Approximately half of all households reported using private health facilities, including pharmacies, hospitals, general practitioners, and clinics. Over a third of all returnee families relied on public healthcare facilities, more than any other group (Figure 12a). For IDP families, however, community and NGO-based healthcare systems were crucial as they were 17 percentage points more likely to be served as such centers than other groups. Thanks to access to such facilities, only 21 percent of IDP households reported not being able to access healthcare facilities, compared to 28 percent of non-displaced communities returning a similar figure. Urban IDPs seek more healthcare services than non-displaced households living in urban areas. Twenty-six percent of rural IDPs accessed community health centers, compared to 8 percent of rural non-displaced households. The latter group was 14 percentage points more likely to visit public health centers than the IDP group. Similar comparisons in urban areas reveal opposite trends. Twenty-eight percent of urban IDPs seek public health care services compared to only 15 percent in the non-displaced group. In addition, 16 percent of urban IDPs seek community health services, but almost none in the non-displaced group reported using such services. This means that IDPs living in urban areas are much more likely to seek healthcare facilities than non-displaced households. MSNA is unable to answer why the healthcare demands of urban IDPs are so much more than those of non-displaced households. IDP households living in planned sites are able to access public and private facilities more than other groups (Figure 12b). In contrast, one-third of IDPs living in non-planned locations did not seek any form of healthcare, including community-based health services. 24 Figure 12a: Families seeking healthcare Figure 12b: IDPs seeking healthcare 53% 56% 50% 52% 51% 41% 36% 34% 32% 29% 28% 25% 25% 26% 23% 23% 21% 20% 19% 18% 15% 7% 6% 7% 8% 8% 4%3% 4% 2% public private community faith did not seek public private community faith did not seek care care IDP IDP returnee Non-displaced hosted or planned unplanned non-settlement site Notes: Data – MSNA, 2023. Estimated using adjusted population weights. The affordability of health services could be a bigger challenge than accessibility for IDPs. Forty- eight percent of IDPs encountered difficulties in accessing healthcare services, in contrast to 17 percent in the non-displaced communities. Among the challenges reported by IDP households, issues pertaining to affordability (cited by 30.2 percent of all IDPs) and supply constraints at healthcare centers (19.8 percent) were the most prominent. These constraints included factors such as the unavailability of specific medicines, long waiting times, incorrect medications, the number and training levels of staff at health facilities, and language barriers. Interestingly, the proportion of IDPs reporting issues with accessibility or facing insecurity during travel was small and comparable to the shares reported by non-displaced communities. In general, those living in unplanned locations were more likely to encounter barriers at medical facilities and less likely to report challenges related to affordability than other IDP groups. 5.3 Psychosocial services One-third of all IDPs reported experiencing fear, anger, fatigue, disinterest, hopelessness, and other negative emotions in the past 30 days, adversely impacting their ability to undertake regular activities such as taking care of children, sleeping, participating in community activities or self-care. Those living in rural areas are facing greater psychological stresses – about 38 percent of IDP households in these areas are suffering from poor mental health conditions. In comparison, only 14 percent of non-displaced households nationally and 13 percent in rural areas reported such a figure. Mental health conditions are significantly better amongst urban IDPs, with 20 percent of households reporting negative emotions. This is only 2 percentage points more than the share of urban non-displaced households affected by similar conditions. The share of households that have received support for their mental health conditions across IDP, returnee, and non-displaced groups is almost negligible and close to zero. This underscores a significant gap between the need and availability of such services. 25 6. Livelihoods and Living Standards Key Messages • Unemployment rate among IDPs is three times as much as that of non-displaced households. Male IDPs are more likely to be unemployed than females. Kayah, a state with high conflict intensity, has among the highest unemployment rates in the country. Unfortunately, incidences of child labor remain persistently high across household types. • Unlike non-displaced households, IDPs’ willingness to work is not influenced by increased security risks. However, regulatory barriers and a worsening environment for conducting business activities raises their unemployment rates. • Communities that provide landmine contamination education to IDP households are likely to have more chance of employment. Self-reported data shows that this is not because of lower risk of mortality or injury but likely due to an improvement in the business environment. • Earnings data confirms that IDPs are among the poorest households in Myanmar. Average monthly earnings of non-displaced households are 80 percent higher than IDPs and 50 percent higher than returnee households. • Most IDPs rely on casual wages, followed by agriculture and self-employment earnings. About one-fifth of all IDPs and one-tenth of returnees reported humanitarian assistance as one of their top 3 income sources. • IDPs living in planned locations rely strongly on casual employment and humanitarian assistance. Only a handful of such families are engaged in self-employment or crop- production. In comparison, those in unplanned are more actively engaged in farming. • Consumption data shows that even among IDPs, those living in unplanned locations, are poorer than others living in planned or non-settlement areas. These households are disproportionately affected by high food and fuel prices and allocate more of their budgets to food and fuel than any other group. As a result, their budgetary allocation to long term human capital investment in education and health is lower than other groups. • Consumption data also shows that despite being able to purchase food from markets, IDPs in unplanned locations have larger household sizes and face food insecurity. However, because of higher levels of farming observed among this group, their food insecurity levels are lower than those in planned locations. • Higher vulnerabilities lead to risker forms of coping strategies. IDPs in unplanned locations are 4 to 8 times more likely to engage in emergency-level coping strategies than other groups. 26 6.1 Patterns of Employment Unemployment is approximately 3 times higher amongst IDPs than other groups in the country. Specifically, 46 percent of males and 40 percent of females IDPs aged 18 and above are currently unemployed and actively seeking opportunities (Figure 13a). The average unemployment rate is close to 25 percent for IDPs who have returned to their pre-displacement locations. Among non- displaced communities, unemployment rates are considerably lower at about 15 percent 13. IDPs living in unplanned sites are least likely to be employed with approximately one-third of all males and 22 percent of all females looking for jobs (Figure 13b). Employment patterns for IDPs living in non-settlement and planned locations are somewhat better than the former group but are found to be significantly lagging when compared to non-displaced communities. Gaps in unemployment rates persist but are slightly smaller when comparing non-displaced rural individuals to IDPs. Among males, the unemployment rate of rural (urban) IDPs is 24.6 (41.5) percentage point higher than those that have not experienced forced displacement. Similarly, among rural (urban) females, the unemployment rate is 25.2 (31.3) percentage points more than similar non-displaced groups. Figure 13a: The unemployment rate among Figure 13b: Unemployment rate by settlement working-age adult IDPs, returnees, and non- sites of IDPs and returnees displaced communities 46% 30% 40% 26% 25% 23% 22% 19% 20% 23% 22% 14% 16% 15% 11% Male Female All Male Female IDP IDP returnee Non-displaced planned unplanned non-settlement site Notes: Data – MSNA, 2023. Estimated using adjusted population weights. The sample in Figure Xb is restricted to IDP and returnee households. In Kayah – the state with the most IDPs as a share of its population – a third of all displaced are unemployed and looking for work. In absolute terms, Sagaing has approximately 820,000 IDPs, the highest among all other states and regions, and a quarter of these IDPs in the state are unemployed. Unemployment rates among IDPs living in states and regions that have historically experienced displacement prior to 2021 – i.e., Rakhine and Kayah – are also amongst the highest in the country. The fact that IDPs in regions with prolonged displacement experiences continue to find it difficult to find work suggests that the labor market frictions that predated the military coup of 2021 have not attenuated over time. Male unemployment rates are higher than females; the regulatory environment and difficult conditions for conducting business activities impact employment activities. In general, unemployment rates of female IDPs are lower than males across most states and regions (Table 3), and male IDPs living in rural areas are 10 percentage points less likely to be unemployed than those in urban areas. Strikingly, there is almost no difference in the unemployment rate of female 13 See Sinha Roy et al (2023) for more aggregate unemployment rates across populations. 27 IDPs living in rural or urban locations. Thus, male IDPs living in urban areas, looking for work in the non-agricultural sector, are most at risk of unemployment. Male unemployment rates are also influenced by environmental factors such as exposure to explosives and whether the current township is the same as the IDPs’ pre-displacement location. Female employment, in comparison, is less sensitive to such factors. Additionally, regulatory conditions related to employment, such as possession of verification documents like a valid CSC, Associate Citizenship, or Naturalized Citizen card, are associated with a greater likelihood of employment among both genders. IDPs living in settlements with an existing coordination mechanism among residents are more likely to be unemployed, suggesting that these mechanisms are currently geared to ensure access to public services for IDP households rather than effectively addressing their livelihood challenges. Table 3: Unemployment rates among Male and Female IDP Conditions intermediating employment status Male IDP unemployment rate Female IDP unemployment rate Urban 34% 21% Rural 24% 20% No mine/IED exposure 25% 20% Exposed to mines/IEDs in the community 34% 22% Displacement to another township 34% 21% Displacement within the same township 25% 20% There is no coordination mechanism at the 24% 19% settlement site A coordination mechanism exists at the location 33% 22% Movement not restricted 27% 19% Movement restricted 27% 21% Without identification documentation 35% 25% Possessing identification documentation 24% 18% Notes: Data – MSNA, 2023. Estimated using adjusted population weights. The sample is restricted to IDP households. Movement restrictions are challenges in mobility experienced by the household in the past 3 months. Identification documentation includes possession of a valid CSC, Associate Citizenship, or Naturalized citizen card. Regulatory barriers and a worsening environment for doing business raise IDPs’ unemployment rates. Table 4 attempts to untangle some key determinants of the labor market by observing relationships between household unemployment rates and (i) the regulatory environment – comprising variables related to identification documents; (ii) the ease of doing business – which includes factors such as exposure to movement restrictions, mines and IEDs, whether the household’s current township was the same as it was before displacement and whether there was a coordination mechanism at the site; and; (iii) individual-specific variables related to supply of labor – such as self-reported concerns about security and protection 14. Columns (1) and (3) of Table 4 restrict the sample respectively to male and female members of IDP households, while columns (2) and (4) relate to the genders of non-displaced families. Concerns about security lead non-displaced households to reduce employment, but IDPs’ willingness to work is unaffected by security considerations. In samples restricted to IDP households, the regression coefficients for various labor supply variables related to violence, fatality, harassment, and prosecution consistently yield significant negative values. This implies that despite facing substantial security risks (as seen in section 4.5), IDP households continue seeking 14 The MSNA is not a survey designed to study labor market outcomes. An employment specific survey of IDP, returnee and other households will be needed to cross-examine the labor markets trends reported in this section and highlight more traditional sources of labor market frictions. 28 employment opportunities. In contrast, many of the coefficients related to security perceptions have a positive coefficient for non-displaced communities – indicating that security concerns deter non- displaced households, raising their unemployment rate. Regulatory barriers raise unemployment rates among IDPs more than the non-displaced population. Table 4 shows that households that possess relevant identification documentation are more likely to be employed. Moreover, male unemployment shows no significant correlation with movement restrictions. This suggests that unemployment is associated with identification documents as they are likely prerequisites for jobs rather than being needed by households for unrestricted movement. Movement restrictions are significantly related to female unemployment rates, likely reflecting social norms. An adverse environment for doing business – such as landmine contamination in the vicinity of IDP households – raises unemployment rates. Table 4 also indicates that areas exposed to mines and IEDs create a challenging business environment, resulting in higher unemployment rates among IDPs. In section 4.5, we highlighted that a quarter of all IDPs and only 4 percent of non-displaced households faced exposure to IEDs and explosives. Consequently, table 4 shows landmine contamination affecting unemployment among IDP households more significantly than non- displaced communities. The adverse consequences of landmine exposure on economic activity are not unique to Myanmar. Past studies in Mozambique have found landmine clearance programs significantly improve market access for firms in densely populated areas and lead to substantial enhancements in overall economic activities (Chiovelli, Papaioannou, and Michalopoulos, 2018). Overall, the evidence presented in table 4 shows that regulatory barriers and difficulties in the business environment such as landmine contamination have a strong influence on IDPs’ unemployment patterns. In contrast, labor supply factors such as perceptions of security risks are insignificant to their employment decisions. Table 4: Selected correlates of male and female unemployment share in the household IDP: (1) Non-displaced: (2) IDP: (3) Non-displaced: (4) male male female female unemployment unemployment unemployment unemployment Faced movement 0.0182 0.0266 0.0805** 0.0629*** restrictions (0.60) (1.74) (3.08) (4.82) -0.120*** -0.0705*** -0.0199 -0.0483*** ID documentation (-4.77) (-6.96) (-0.93) (-5.55) -0.0591 0.00526 -0.0740** -0.00881 Living in a village (-1.89) (0.70) (-2.88) (-1.36) 0.134*** 0.0157 0.0306 -0.0160 Exposed to mines/IEDs (4.30) (0.90) (1.07) (-1.07) Same township as pre- -0.119*** -0.0815** displacement (-4.17) (-3.28) 0.0912** 0.0162 Coordination mechanism (3.11) (0.67) Security concerns related -0.176*** -0.0129 -0.0278 -0.0146* to violence (-5.20) (-1.70) (-1.23) (-2.30) 29 Security concerns: 0.0224 0.0126 -0.0181 0.0103 harassment, prosecution, (0.89) (1.46) (-0.88) (1.54) and discrimination Security concerns: - -0.167*** 0.0417*** 0.0711*** mortality, injury, or 0.0835*** (-6.09) (4.37) (7.08) explosives (-3.63) Security concerns: substance abuse, curtailed -0.109*** 0.0106 0.0328 0.0346*** opportunities, social (-4.17) (1.54) (1.43) (5.50) isolation 0.553*** 0.129*** 0.316*** 0.101*** Constant (12.93) (10.69) (8.59) (9.71) Observations 980 5096 1089 5435 Notes: Data – MSNA, 2023. Estimated using adjusted population weights. t statistics in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001. The dependent variable is male/female unemployment share at the household level, calculated as the sum of male/female individuals above 18 years, currently unemployed but looking for work opportunities – divided by the total members in the family above 18 years. Community services to make the environment more conducive for business activities – such as mine risk education – can reduce IDP unemployment rates. The MSNA shows that many IDPs are offered community services, presumably by community-based organizations. Among these, access to birth registration documents and mining risk education were the most common, and about 23 percent of all IDPs reported receiving such services. Additionally, about 10-15 percent of IDPs received general awareness-building and community-specific information, and 2.5 percent of IDPs received legal aid services. Approximately 40 percent did not receive any of these services. Table 5 shows that among all these services, the effect of mine risk education on unemployment was most significant. IDP households exposed to mine risk education community services had 6.7 percentage points fewer unemployed adults in their households than other displaced households within the same township that did not receive such services. Table 5: Community services received by IDP households and their association with unemployment rates. Unemployment shares at the household level Community service: Assistance with birth registration -0.0324 (-1.43) Community service: Mine Risk Education -0.0673** (-2.65) Community service: General awareness-raising 0.0325 programs (1.33) Community service: Community-specific knowledge and 0.0321 information (1.08) Community service: Legal aid services 0.0757 (1.82) Community service: None -0.0379 (-1.46) 30 IDP household living: Unplanned sites -0.0238 (-1.16) IDP household living: non-settlement location -0.0802** (-3.05) Constant 0.273*** (14.01) Observations 1906 Notes: Data – MSNA, 2023. Estimated using adjusted population weights. t statistics in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001. The dependent variable is unemployment share at the household level, calculated as the sum of male/female individuals above 18 years, currently unemployed but looking for work opportunities – divided by the total members in the family above 18 years. 6.2 Child labor A quarter of all IDPs reported incidences of child labor in their vicinity. Although no child below the age of 14 is legally permitted to work in Myanmar, there was one child worker for every 11 children in the country in 2015 (ILO, 2014). More recent data from the MSNA shows that the incidence of child labor may have risen further, with 23 percent of IDPs and about 16 percent of returnees and non-displaced families reporting children between the ages of 12 to 17 years in their town or village that were engaged in employment activities (Table 6). Only 15 percent of IDPs and 10 percent of returnees and non-displaced communities were aware of the negative consequences of child labor on the physical and mental well-being of children. This evidence is complementary to results in Table 3, which shows that 55 percent of boys and 33 percent of girls below 18 years old and belonging to IDP families were living away from the household for employment-related reasons. Overall, child labor activities are, on average, higher among IDPs living in rural areas and among households with more unemployed adult members. All this evidence suggests that trends associated with child labor activities have not shifted in the right direction over the years and may have worsened in recent years. Table 6: Incidences of child labor observed by respondents in their vicinity Child labor Negative consequences of child labor (12 to 17 years) (12 to 17 years) IDP 23% 14% Returnee 16% 10% Non-displaced 17% 10% Notes: Data – MSNA, 2023. Estimated using adjusted population weights. The first two columns show the share of households that agreed or strongly agreed with the statement: “In the past 7 days, it was common in your village for children aged 5-11 and12-17 years to work and contribute to household income”. The last two columns show the share of households that agreed or strongly agreed with the statement: “In the majority of the cases, this work might have a negative impact on their physical/mental health, well-being, or development.” 6.3 Patterns of Earnings Monthly household incomes of IDPs are significantly lower than those of returnees or non- displaced families. Non-displaced households, on average, have incomes that are 80 percent higher than IDPs and 50 percent higher than returnee households. Incomes of IDPs in Kayah and Shan are among the lowest in the country; the median IDP income in these two states is approximately 1/8th the median income of all households across Myanmar. Conversely, incomes for IDPs in Rakhine, Sagaing, and Tanintharyi are higher. But even in the states, the median IDP income reaches only up to ¾th of the median national income. 31 Most IDPs rely on incomes from casual employment, with only a small percentage reporting on agricultural or self-employment-related earnings. Casual employment ranks as one of the top 3 income sources for over two-thirds of all IDP households (Figure 14). Agriculture is the second most significant source, with 30 percent of IDPs and around 33 percent of returnees and non- displaced households considering it a top-3 income source. Self-employment is the third most important source of income among IDPs, with fewer than 20 percent listing it among their top income sources. In contrast, approximately 45 percent of non-displaced households cited self- employment as a top 3 income source. Ten percent of all three groups reported remittances as a top income source. Finally, humanitarian assistance reached more IDPs and returnee households than non-displaced ones. This suggests that the aid is well-targeted and can reach the intended audience more than others. However, only 22 percent of IDPs and 10 percent of returnees cited humanitarian assistance as one of their top 3 income sources, suggesting that either many IDP families do not receive humanitarian aid (potentially indicating exclusion errors) or the value of the assistance received is small compared to other income streams. Figure 14: Top three income sources of IDP, Returnee, and Non-displaced Households Notes: Data – MSNA, 2023. Estimated using adjusted household level weights. The sample excludes households that reported a missing monthly household income value. This affected 547 observations – less than 6 percent of the full sample. Most IDPs in planned locations are engaged in casual employment. In contrast, those living in unplanned sites and non-settlement locations are respectively more likely to work in agriculture and self-employment activities than any other group. Seventy percent of IDPs in planned locations rely on casual work as a top income source, while 40 percent of IDPs in unplanned sites reported agriculture earnings as one of the top sources (Figure 15). IDPs not living in settlement clusters are more likely to be self-employed than any other group, and only a handful of them rely on humanitarian assistance as a top income source. In comparison, more than 20 percent of IDPs living in planned and unplanned settlements relied on humanitarian assistance – which is not surprising given the high levels of unemployment among these groups (Figure 13b). Figure 15: Top three income sources of IDP, Returnee, and Non-displaced Households 32 Notes: Data – MSNA, 2023. Estimated using adjusted household level weights. The sample excludes households that reported a missing monthly household income value. This affected 547 observations – less than 6 percent of the full sample. 6.4 Consumption, market integration, and food security IDPs are among the poorest in the country, especially those living in unplanned settlements. Two- fifths of all IDPs have average consumption per capita below the 20th percentile of the national consumption distribution (Figure 16a). In comparison, only one-fifth of non-displaced households reported consumption below the 20th percentile. Poverty levels are particularly high amongst IDPs living in unplanned locations (Figure 16b). Fifty percent of IDPs living in unplanned areas have average per capita monthly consumption below the 20th percentile of the national distribution. In comparison, 40 percent of IDPs in planned sites and less than 10 percent of those living in non- settlement locations average consumptions below this minimum threshold. These trends indicate that IDPs, especially those forced to reside in unplanned settlement locations, rank among the poorest in the country. The concentration of poorest households among returnees is lower than among IDPs, indicating improvements in living standards upon returning from forced displacement events. Conversely, wealthier households (those at the top of the national consumption distribution) are more likely to be among non-displaced groups. Figure 16a: Composition of IDP, returnee, and Figure 16b: Composition of IDP households non-displaced households based on their rank in based on their settlement type and rank in the national consumption distribution the national consumption distribution 33 Notes: Data – MSNA, 2023. The legend shows households that fall below the 5 quintiles of per capita monthly consumption expenditures (MPCE) in the national consumption distribution. The consumption values are captured in MSNA on a mixed recall basis using a combination of 30-day, and 6-month recalls across 15 food and non-food items. Missing consumption values are imputed to zero before estimating a household consumption value. Households are assigned a national quintile rank based on nominal consumption values without adjusting for temporal or spatial price differences. Adjusted population level weights are used to assign quintile ranks based on monthly consumption per capita values. The most economically disadvantaged IDPs are those residing in unplanned locations. These families bear the brunt of food and fuel price inflation. IDPs in unplanned locations allocate 66 percent of their monthly per capita consumption on food and fuel expenses (Table 6). In contrast, non-displaced households, comprising more affluent households, spend 5 percentage points less on such necessities. Moreover, IDPs in unplanned locations devote 3 percentage points more of their budget to transportation and communications compared to their counterparts in planned sites. Consequently, food and fuel price inflation put a heavier load on the household budget of IDPs residing in unplanned locations than any other group. Note that despite spending a greater share of their expenditures on transportation, 50 percent of IDPs in unplanned locations have experienced safety or security restrictions in moving freely over the past three months (Figure 7). The high share of transportation expenditures could, therefore, comprise hidden compensations that may be required to mitigate the risks associated with travel. Given their high expenditures on food, fuel, travel, and communication, IDPs in unplanned locations can allocate only 5 percent of their monthly consumption to productive human capital investments like health and education. This is 4 percentage points lower than the human capital expenditures made by IDPs in planned sites and returnees. Table 6: Share of household expenditure on key consumption items by population group Food and Fuel Transportation and Health and Education communication IDP: Planned location 64% 8% 9% IDP: Unplanned location 66% 11% 5% IDP: Non-settlement areas 63% 11% 5% Returnee 60% 10% 9% Non-displaced 61% 10% 6% Notes: Data – MSNA, 2023. The consumption values are captured in MSNA on a mixed recall basis using a combination of 30-day and 6-month recalls across 15 food and non-food items. Missing consumption values are imputed to zero before estimating a household consumption value. Households are assigned a quintile rank based on nominal consumption values without adjusting for temporal or spatial price differences. Adjusted population level weights are used to assign quintile ranks based on monthly consumption per capita values. Most IDPs can access markets and purchase food with cash; those residing in unplanned locations are most likely to cultivate their own food. About 90 percent of IDPs in planned locations used cash to purchase food; a quarter of such households received in-kind transfers to supplement their food requirements further. IDPs in unplanned locations also use cash to purchase food but are more likely to cultivate their own crops than other groups. This explains why IDPs in unplanned locations are more engaged in agriculture than any other group of IDPs (Figure 14). About a quarter of these households receive additional in-kind food assistance. Most IDPs living in non- settlement sites do not report food assistance, consisting of minimal income transfers reported by households in Figure 14. Why do so many IDPs in unplanned sites resort to subsistence farming despite a large proportion of households purchasing food from markets and receiving in-kind food transfers? One explanation 34 is that these IDP families have larger household sizes and need more food than other groups. This is plausible since the average household size for this group is approximately 4.8 members per household, whereas IDPs in planned and non-settlement locations, returnees, and non-displaced households have an average of 4.5, 3.6, and 4.2 members, respectively. Table 7: Top-3 sources of food across population groups Purchased Self- Gifted/exchanged Aid production IDP: planned sites 89% 54% 59% 26% IDP: unplanned sites 95% 67% 60% 19% IDP: non-settlement areas 97% 49% 44% 2% Returnees 93% 61% 31% 7% Non-displaced 95% 58% 31% 2% Notes: Data – MSNA, 2023. Estimated using adjusted household level weights. Another complementary explanation is that IDPs in unplanned locations suffer from low food security. IDPs living in unplanned locations are amongst the poorest in the country (Figure 15b) and are most affected by high food inflation (Table 6). This could adversely impact their ability to purchase sufficient food from markets, forcing families living in unplanned sites into subsistence farming. Food consumption scores (FCS) reported in MSNA suggest this is the case 15. Table 8 shows that 30 percent of IDPs living in planned and unplanned sites have FCS scores that are “below adequate” based on Myanmar-specific thresholds found in past literature. For 12 percent of IDPs in unplanned sites, the FCS score is in the “poor” category, driven mainly by underconsumption of dairy, meat, fish eggs, and sugar relative to other household groups. The food insecurity levels in Table 8 are approximately similar for IDPs in planned and unplanned sites. However, note that in Table 7, IDPs living in planned sites are 7 percentage points more likely to receive food aid than non-displaced ones. Lower levels of food aid and comparable levels of food insecurity lead 67 percent of IDPs in unplanned locations to engage in agricultural self- production (Table 7). Table 8: Food consumption scores across population groups Poor FCS score Borderline FCS Adequate FCS IDP: planned sites 10% 22% 68% IDP: unplanned sites 12% 18% 69% IDP: non-settlement areas 3% 10% 87% Returnees 5% 16% 79% Non-displaced 3% 11% 87% Notes: Data – MSNA, 2023. Estimated using adjusted household level weights. Despite receiving humanitarian aid and in-kind transfers, hunger levels among IDPs in planned locations are high – likely because, unlike IDPs in unplanned locations, they are unable to supplement food through subsistence farming. Seven percent of these households reported instances in the past 30 days when there was no food in the house or a member had to sleep hungry. Instances of hunger can often be repeated. In 20 percent of cases, households reported no 15 The FCS is a measure of dietary diversity and food frequency, calculated as the weighted sum of the frequency of food groups eaten over the seven days prior to survey. The nutritional values associated with each food item serves as the weight in the aggregate index. Households are then assigned to poor, borderline and adequate categories based on their FCS and threshold values as reported in IFPRI (2023). 35 food between 3 to 10 times in the past 30 days, while in 33 percent of cases, a member went to sleep hungry 3 to 10 times over the same period. Instances of members staying hungry through the day and night were fortunately limited to only 2 percent of IDPs living in planned locations. Table 9: Household Hunger Score (HHS) and 30-day recall questions There was no food of any kind Household members Household members went full day in the house went to sleep hungry & night without food IDP: planned sites 7% 7% 2% IDP: unplanned sites 2% 1% 0% IDP: non-settlement areas 4% 3% 0% Returnees 4% 5% 1% Non-displaced 3% 2% 1% Notes: Data – MSNA, 2023. Estimated using adjusted household level weights. 6.5 Coping Strategies IDPs living in planned sites were most likely to undertake riskier coping strategies. Higher levels of hunger and their inability to carry out subsistence farming meant that IDPs living in planned sites were most likely to sell assets, spend savings, purchase food on credit, or borrow food. Almost two-thirds of these households reported spending savings to access food – 3 times as many households in the non-displaced group (table 10). Among returnee households, 46 percent reported spending savings for food – lower than other IDP households – but almost twice more than non-displaced households. This means that food security concerns can linger even after IDPs return to their pre-displacement locations. Table 10: Coping Strategies adopted to address lack of food or money to buy food COPE CRISIS EMERGENCY Purchase food on Borrow Sell Reduce Sell High Sold HH Spend Child credit or money productive Health house risk Migrate assets savings labor borrow for food assets Exp. or land activities food IDP: planned 30% 62% 41% 48% 16% 24% 14% 4% 8% 7% sites IDP: unplanned 11% 52% 32% 36% 5% 6% 12% 1% 1% 1% sites IDP: non- settlement 10% 48% 27% 16% 2% 5% 5% 0% 3% 3% areas Returnees 20% 46% 27% 29% 6% 11% 8% 1% 5% 2% Non- 6% 21% 19% 18% 2% 8% 4% 1% 2% 0% displaced Notes: Data – MSNA, 2023. Estimated using adjusted household level weights. 36 7. Access to Water, Sanitation, and Financial Access Key Messages • Most households, including IDPs, are able to access improved sources of water. However, those in unplanned locations are at high risk of exposure to fecal bacteria due to a significantly higher rate of open defecation. Exposure to fecal bacteria for children under 5 years carries long term human capital risk of stunting, wasting and undernourishment. • IDP households are generally well connected to financial services over the phone. Mobile money connectivity rates are lowest among unplanned IDPs and returnees, where about half of all household can access such services. 7.1 WASH Services Most households, including IDPs can access improved water sources in monsoon and summer months; instances of drinking water shortages are rare – but returnee households might be exposed to inferior quality of drinking water. Households in the non-displaced community are more likely to purchase water than rely on improved water sources, such as boreholes, tube wells, piped connected to the dwelling unit or the neighbor’s house, protected spring, protected well, public standpipes or rainwater collection. As a result, these households are more likely to treat their water before consumption. The group that may be most at risk of poor water quality are returnees: 66 percent of these households relied on improved sources or purchases of water, the least among all other groups. The remaining households in this group were reliant on water sources of inferior quality, such as, surface water, unprotected spring, unprotected well or other. Only 17 percent of returnees reported treating water before consumption – raising the risk of borne of diseases. Table 11: Sufficiency, safety, supply source and treatment of drinking water Drinking Improved Purchased Treated UN or Public Private water water source water water Community source Source sufficiency before Organization drinking IDP: planned 95% 80% 2% 9% 17% 9% 38% sites IDP: unplanned 97% 72% 3% 2% 14% 2% 44% sites IDP: non- settlement 98% 74% 16% 25% 2% 0% 22% areas Returnees 96% 60% 6% 17% 6% 4% 40% Non-displaced 94% 67% 16% 17% 3% 2% 35% Notes: Data – MSNA, 2023. Estimated using adjusted household level weights. See section 7.1 for descriptions of improved water and purchased water sources. Most households reported access to improved toilet facilities except those in unplanned locations. About 16 percent of households in this category reported open defecation or other toilets. This group is therefore more exposed to bacterial infections, which can lead to stunted growth and malnutrition among children. Open defecation rates are also higher among returnee households compared to non-displaced and IDPs living in non-settlement locations. Moreover, over 12 percent 37 of returnee households do not have a proper access to handwashing facility which can further expose them to harmful infections and diseases. Table 12: Toilet and handwashing facilities Open defecation or other Improved toilet facility No handwashing facility types of toilets IDP: planned sites 98% 2% 20% IDP: unplanned sites 84% 16% 10% IDP: non-settlement areas 96% 4% 7% Returnees 93% 7% 12% Non-displaced 97% 3% 7% Notes: Data – MSNA, 2023. Estimated using adjusted household level weights. Households living in areas with coordination mechanism are 5 percentage points more likely to access improved water and toilet facilities. The coordination mechanism is especially effective at organizing access to water sources from UN agencies and other community-based organizations (Figure 17). However, the coordination system lags in providing improved toilet facilities to households as these households are 7 percentage points less likely to get access to an improved toilet facility. Figure 17: Coordination committee provide households with greater access to improved water and toilet facilities. 90% 97% 72% 67% 30% 32% 35% 3% Safe water source Improved toilet Water from a Water from a UN/community private source organization Coordination mechanism No coordination mechanism Notes: Data – MSNA, 2023. Estimated using adjusted household level weights. 7.2 Financial Services More than half of all households can access financial services through their mobiles. Mobile money access is higher among IDPs in planned and non-settlement locations, likely because these groups rely more on humanitarian assistance and remittances, respectively. In contrast, IDPs in unplanned locations are least connected to mobile money networks and more reliant on informal financial networks – such as hundi, member run savings groups, services provided by community members and others – than any other group. Figure 17: Access to financial services. Formal financial facility Informal financial facility Mobile Money IDP: planned sites 55% 34% 67% 38 IDP: unplanned sites 29% 53% 49% IDP: non-settlement areas 70% 18% 79% Returnees 47% 43% 52% Non-displaced 53% 32% 53% Notes: Data – MSNA, 2023. Estimated using adjusted household level weights. Formal financial facilities include banks, formal money transfer agents, microfinance institutions and credit unions/savings and credit cooperative organizations. See section 7.2 for examples of informal financial facilities. There is large heterogeneity in IDP households’ access to mobile money networks. In Kayah, only 5 percent of all IDP families could access mobile money systems (table 13); almost 60 percent of these households relied on informal financial networks. Mobile money access is higher in Kayin, Chin and Sagaing than Kayah but still, less than 50 percent of IDPs could access such systems. Aid and financial assistance packages that rely exclusively on mobile transfers could miss a large share of IDPs in these regions. Table 13: IDP access to financial services by states and regions State/region Formal financial facility Informal financial facility Mobile Money Bago 83% 3% 94% Chin 33% 27% 44% Kachin 84% 34% 66% Kayah 7% 61% 5% Kayin 13% 55% 42% Magway 79% 19% 100% Mon 51% 66% 59% Rakhine 75% 22% 74% Sagaing 21% 55% 49% Shan 43% 33% 51% Tanintharyi 47% 52% 72% Notes: Data – MSNA, 2023. Estimated using adjusted household level weights. Formal financial facilities include banks, formal money transfer agents, microfinance institutions and credit unions/savings and credit cooperative organizations. See section 7.2 for examples of informal financial facilities. 8. Conclusion There is scope for livelihood and informational interventions to address the economic challenges IDPs face. IDPs experience acute levels of unemployment and rely on subsistence activities, especially when living in unplanned settlements. Many IDPs remain unable to return to their homes for long periods of time, with a median time since the last displacement of over one year. Humanitarian aid is scarce, with only 22 percent of IDPs reporting aid as one of their top 3 sources of income. It is, therefore, critical that IDPs establish secure livelihoods that can support their basic needs during displacement. Previous research, including in settings with recent conflict experiences, indicates that livelihood and landmine mitigation programs can assist vulnerable households in establishing productive livelihoods. Livelihood interventions can create substantial long-run economic benefits, including in settings where there is cessation of violence and multiple displacements or cases where IDPs have returned or resettled in other locations. Considering the successes observed in graduation-style programs, there is a compelling rationale to introduce a tailored initiative in Myanmar for IDPs facing extreme 39 poverty. Previously tested programs--- including features such as grants, group savings promotion, coaching, entrepreneurship training, and targeted psychosocial interventions---have been shown to substantially increase economic activity and income among ultra-poor households and have demonstrated cost-effectiveness. In a study in postwar Uganda, a graduation-style program produced significant positive changes after 16 months. Participants experienced a doubling of earnings, a rise in nonfarm businesses from 39 percent to 80 percent, and a conservative estimate of almost a third increase in household consumption (Blattman et al., 2016). Another livelihoods program in postwar Uganda, which included grants for vocational training or business creation, found similarly promising results (Blattman et al., 2014). A recent study in Niger found that a graduation-style program combining psychosocial interventions with cash grants generated substantial improvements in economic and psychosocial well-being (Bossuroy et al., 2022). Proximity to landmines constrains economic activity, and landmine education programs offer a promising avenue to alleviate these constraints. Half of the returnees and a quarter of IDPs reported exposure to landmines or other explosive weapons in the past year, with most exposed individuals reporting that these incidents impacted their livelihoods and movements more than causing injuries. These findings are consistent with the impacts of a landmine clearance program in post-war Mozambique, which found large economic benefits from mine clearance, particularly when transportation networks were affected (Chiovelli et al., 2018). In Myanmar, proximity to landmines is associated with higher unemployment, but this impact is mitigated among communities that have received landmine education programs. Scaling up these programs could allow IDPs living near landmines to safely establish new livelihoods. 40 9. References Blattman, Christopher, Eric P. Green, Julian Jamison, M. Christian Lehmann, and Jeannie Annan. 2016. "The Returns to Microenterprise Support among the Ultrapoor: A Field Experiment in Postwar Uganda." American Economic Journal: Applied Economics, 8 (2): 35-64. Blattman, Christopher, Nathan Fiala, Sebastian Martinez, Generating Skilled Self-Employment in Developing Countries: Experimental Evidence from Uganda , The Quarterly Journal of Economics, Volume 129, Issue 2, May 2014, Pages 697–752, https://doi.org/10.1093/qje/qjt057. Bossuroy, T., Goldstein, M., Karimou, B. et al. Tackling psychosocial and capital constraints to alleviate poverty. Nature 605, 291–297 (2022). https://doi.org/10.1038/s41586-022-04647-8 Chiovelli, Giorgio, Stelios Michalopoulos, and Elias Papaioannou. Landmines and spatial development. No. w24758. National Bureau of Economic Research, 2018. ILO (2024). Child labour in Myanmar. Accessed on 22nd January 2024. https://www.ilo.org/yangon/areas/childlabour/lang--en/index.htm Raleigh, C., Kishi, R. & Linke, A. Political instability patterns are obscured by conflict dataset scope conditions, sources, and coding choices. Humanit Soc Sci Commun 10, 74 (2023). Sinha Roy, S. (2023). Myanmar Subnational Phone Surveys (MSPS) of the World Bank: Coverage, Reliability, and Representativeness. Sinha Roy, S., Demarchi, G., & Rhoads, E (2023). A Growing Crisis: Work, Workers, and Wellbeing in Myanmar. World Bank. UNHCR. (2023). UNHCR Global Website. UN Refugee Agency. Retrieved 13 July from https://www.unhcr.org/ 41