92605 A Comparative Study of Ex-Combatant Reintegration in the African Great Lakes Region: Trajectories, Processes, AND Paradoxes Randolph Wallace Rhea Centre for Peace Studies, University of Tromsø July 2014 2014 The International Bank for Reconstruction and Development/The World Bank 1818 H Street, NW Washington, DC 20433 www.tdrp.net www.worldbank.org The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the Executive Directors of the International Bank for Reconstruction and Development/The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denomina- tions, 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. Cover Design and Text Layout: Duina Reyes Acknowledgments T he Transitional Demobilization and Reinte- gration Program (TDRP) of the World Bank would like to extend its gratitude to Randolph Rea and Qinyu Cao (Sabrina) for this report. collaborate with numerous Norwegian and interna- tional agencies. Qinyu Cao (Sabrina) synchronized the data collected from different studies conducted in five countries by Randolph Rhea, the author of this study, is a Doctoral TDRP and its predecessor, the Multi-country Demobi- Research fellow at the Centre for Peace Studies at The lization and Reintegration Program (MDRP), creating University of Tromsø and is a member of the Interna- the database for this cross-country analysis. At the time tional Research Group on Reintegration. Mr. Rhea’s of the synchronization of this database, Ms. Cao was research focuses on ex-combatant reintegration pro- a Master’s candidate of International Development cesses in the Great Lakes Region of Africa with special Studies at the Elliott School of International Affairs focuses on social reintegration, methodological and of the George Washington University. Ms. Cao subse- analytical approaches to the study of reintegration, and quently joined TDRP as an Operations Associate in the monitoring and evaluation of reintegration pro- 2013 and is currently working as a Social Development gramming. Specialist at the Post-Conflict and Social Development Practice Group (AFTCS) of the World Bank. The Centre for Peace Studies (CPS) is located at the University of Tromsø, Norway. The CPS was founded The George Washington University’s Elliott School of in 2002 and has grown into a graduate education and International Affairs, located in the heart of Washing- research institution focusing on a broad range of peace- ton DC, offers a course in International Development building and development areas. The Centre for Peace Management Tools and Processes. The Elliot School Studies is home to the International Research Group has a collaborative association with the TDRP, whereby on Reintegration (IRGR), comprised of both faculty every year a group of students work on an operational staff and affiliated scholars and practitioners in the field TDRP project, to design a development sub-project as of reintegration. The CPS & IRGR undertake interdis- per a specific Terms of Reference. During the last five ciplinary and comparative studies of reintegration and years, three of these sub-projects were subsequently implemented by the TDRP. c d Table of Contents List of Acronyms ............................................................................................................... 4 1. Note to the Reader........................................................................................................ 5 2. Executive Summary....................................................................................................... 6 2.1 Key Findings....................................................................................................................................................................... 6 3. Introduction................................................................................................................. 9 3.1 Methodology.................................................................................................................................................................... 10 3.2 Limitations and Challenges.............................................................................................................................................. 11 3.2.1 Individual GLR Country Survey Compatibility..................................................................................................... 11 3.2.2 Individual GLR Country Context Comparability................................................................................................... 12 4.1 Reintegration as a Process................................................................................................................................................ 14 4.1.1 An Individual Process.......................................................................................................................................... 16 4.1.2 A Unique Process?............................................................................................................................................... 19 4.2 Reintegration Trajectories................................................................................................................................................. 20 5. Meta-Analysis of Ex-Combatants’ Reintegration Processes in the GLR. ....................... 21 5.1 From Mobilization to Demobilization in the GLR.............................................................................................................. 21 5.1.1 Mobilization Under the Age of 18....................................................................................................................... 21 5.1.2 The Legacies of Mobilization and Wartime Experiences...................................................................................... 22 5.2 Economic Reintegration in the GLR.................................................................................................................................. 23 5.3 Social Reintegration in the GLR........................................................................................................................................ 25 5.3.1 The Logic of Social Capital in Reintegration........................................................................................................ 26 5.4 Female Ex-Combatants in the GLR.................................................................................................................................... 28 5.4.1 A Broader Societal Transformation of Gender Dynamics..................................................................................... 29 5.5 DRC: Reintegration in the Context of Ongoing Conflict................................................................................................ 30 5.5.1 The Absence of Social Cohesion in DRC: A Splintered Society............................................................................. 31 5.5.2 The Paradox of DRC: Parity versus Process.......................................................................................................... 32 5.6 The Limits of Reintegration Programming.................................................................................................................... 33 6. Conclusions................................................................................................................. 34 6.1 Reintegration: An Imbedded Transition............................................................................................................................ 35 Referenced Works........................................................................................................... 37 Annex I - Great Lakes Region Ex-Combatant Reintegration: Comparative Survey Results and Analysis................. 40 7. Demographics............................................................................................................. 41 7.1 From Mobilization to Demobilization............................................................................................................................... 42 7.2 Marriage and Household.................................................................................................................................................. 45 7.3 Literacy, Education, and Vocational Training..................................................................................................................... 49 7.4 Summary.......................................................................................................................................................................... 51 7.4.1 Vulnerable Subgroups........................................................................................................................................ 52 7.4.2 Unique Country Trends....................................................................................................................................... 53 1 8. Housing, Land, Livestock and Food Security................................................................ 55 8.1 Dwelling, Living Conditions and Land Security................................................................................................................. 55 8.2 Land Access and Food Security.......................................................................................................................................... 57 8.3 Summary.......................................................................................................................................................................... 59 8.3.1 Vulnerable Subgroups........................................................................................................................................ 60 8.3.2 Unique Country Trends....................................................................................................................................... 61 9. Economic Issues ......................................................................................................... 62 9.1 Economic Status and History............................................................................................................................................ 62 9.2 Non-Economically Active Ex-Combatants on Employment Issues..................................................................................... 66 9.3 Female Ex-Combatants on Employment Issues................................................................................................................ 67 9.4 Disabled Ex-Combatants on Employment Issues.............................................................................................................. 68 9.5 Income, Savings and Access to Credit............................................................................................................................... 68 9.6 Economic Associations...................................................................................................................................................... 71 9.7 Summary.......................................................................................................................................................................... 72 9.7.1 Vulnerable Subgroups........................................................................................................................................ 73 9.7.2 Unique Country Trends....................................................................................................................................... 74 10. Social Capital............................................................................................................ 75 10.1 Networks and Sociability................................................................................................................................................ 75 10.2 Trust and Solidarity......................................................................................................................................................... 78 10.3 Social Cohesion and Inclusion ....................................................................................................................................... 79 10.4 Empowerment................................................................................................................................................................ 81 10.5 Social Change................................................................................................................................................................. 84 10.6 Summary........................................................................................................................................................................ 86 10.6.1 Vulnerable Subgroups...................................................................................................................................... 87 10.6.2 Unique Country Trends..................................................................................................................................... 88 11. DDR Experiences....................................................................................................... 90 11.1 Reinsertion..................................................................................................................................................................... 90 11.2 Experiences of Return..................................................................................................................................................... 92 11.3 Summary........................................................................................................................................................................ 92 11.3.1 Vulnerable Subgroups...................................................................................................................................... 93 11.3.2 Unique Country Trends..................................................................................................................................... 93 12. Conclusions............................................................................................................... 94 12.1 Ex-Combatants and Economic Reintegration.................................................................................................................. 95 12.2 Ex-Combatants and Social Reintegration........................................................................................................................ 95 12.3 Female Ex-Combatant Subgroup.................................................................................................................................... 95 Annex II - Great Lakes Region Community Dynamics: Comparative Survey Results and Analysis........................... 97 13. Demographics........................................................................................................... 98 13.1 Marriage and Household................................................................................................................................................ 99 13.2 Literacy, Education, and Vocational Training.................................................................................................................102 13.3 Summary......................................................................................................................................................................104 13.3.1 Vulnerable Subgroups....................................................................................................................................104 13.3.2 Unique Country Trends...................................................................................................................................105 2 14. Housing, Land, Livestock and Food Security............................................................107 14.1 Land Access and Food Security.....................................................................................................................................109 14.2 Summary......................................................................................................................................................................111 14.2.1 Vulnerable Subgroups....................................................................................................................................112 14.2.2 Unique Country Trends...................................................................................................................................112 15. Economic Issues......................................................................................................113 15.1 Economic Status and History........................................................................................................................................113 15.2 Non-Economically Active Community Members on Employment Issues......................................................................117 15.3 Female Community Members on Employment Issues.................................................................................................118 15.4 Income, Savings and Access to Credit...........................................................................................................................118 15.5 Economic Associations..................................................................................................................................................121 15.6 Summary......................................................................................................................................................................123 15.6.1 Vulnerable Subgroups....................................................................................................................................123 15.6.2 Unique Country Trends...................................................................................................................................124 16. Social Capital..........................................................................................................125 16.1 Networks and Sociability..............................................................................................................................................125 16.2 Trust and Solidarity.......................................................................................................................................................129 16.3 Social Cohesion and Inclusion......................................................................................................................................130 16.4 Empowerment..............................................................................................................................................................131 16.5 Social Change...............................................................................................................................................................134 16.6 Summary......................................................................................................................................................................136 16.6.1 Vulnerable Subgroups....................................................................................................................................137 16.6.2 Unique Country Trends...................................................................................................................................137 17. Reintegration Experiences......................................................................................139 17.1 Community Sensitization and Preparedness................................................................................................................139 17.2 Community Perspectives on Ex-Combatant Reintegration and Fear.............................................................................140 17.3 Positive and Negative Perceptions of Ex-Combatants...................................................................................................142 17.4 Summary......................................................................................................................................................................143 18. Conclusions.............................................................................................................144 18.1 The Community and Economic Reintegration...............................................................................................................144 18.2 The Community and Social Reintegration.....................................................................................................................144 18.3 Female Community Member Sub-Group......................................................................................................................145 18.4 DRC – A Splintered Society............................................................................................................................................146 3 List of Acronyms DDR Disarmament, Demobilization and Reintegration DRC Democratic Republic of Congo FCS Fragile and Conflict Affected Situations GLR Great Lakes Region IDDRS Integrated Disarmament, Demobilization and Reintegration Standards IDP Internal Displaced Person IRGR International Research Group on Reintegration MDRP Multi-Country Demobilization and Reintegration Program RoC Republic of Congo TDRP Transitional Demobilization and Reintegration Program 4 1. Note to the Reader T his report is structured in three parts. Part 1 is a summary document, including: (i) an executive summary (ii) an introduction; (iii) a review of core concepts of reintegration that will be referred to in this study; (iv) a meta-analysis of reintegration process- The analysis presented in the study is further compli- mented by the individual reintegration studies per- formed by various partners of the World Bank and DDR commissions in respective countries. For Burundi, see World Bank (2011a) Troisieme Enquete des Beneficia- es in the Great Lakes Region (GLR) vis-à-vis the con- rires du PDRT; for the Democratic Republic of Congo ceptual discussion; and (v) conclusions to the summary (DRC), see World Bank (2011b) Rapport d’Evaluation document. Part 2 (Annex I) comprises an in-depth des Bénéficiaires du PNDDR Réintégration des hommes review and analysis of data on the reintegration process- soldats et ex-combattants démobilisés; for the Republic of es of ex-combatants across the GLR. Part 3 (Annex II) Congo (RoC), see World Bank (2011c) Réintégration is an in-depth analysis of community dynamics across Socio-Economique des ex-Combattants Région du Pool, Ré- the GLR. In brief, Part 1 of the study is a meta-analytical publique du Congo; for Rwanda, see World Bank (2012) and knowledge-focused piece that reflects more broadly Comparative report: Ex-combatant and community study on the detailed analysis of the datasets presented in on the socio-economic reintegration; and for Uganda, see Annexes I and II, therefore, this part can be read as a World Bank (2011d) Reporter Re-integration and Com- freestanding report. However, it’s worth noting that any munity Dynamics Survey Report. reading will benefit significantly from exploring the detailed findings in Annexes I and II. 5 2. Executive Summary T his study explores the reintegration processes that ex-combatants, as well as the communities that receive them, go through in the transition from being soldiers to being civilians across the Great Lakes region (GLR) of Africa (Uganda, Rwanda, DRC, 2.1 Key Findings This study has found that across the GLR, ex-combatants have been largely successful in navigating reintegration processes and have shown a positive trajectory towards RoC, and Burundi). This study uses a cross-country reaching parity with the broader community. Across comparative approach capitalizing on survey data col- the GLR, an overall positive trajectory of reintegra- lected between 2010 and 2012 from nearly 10,000 ex- tion over time is visible despite the fact that most combatants and community members across the GLR. ex-combatants encounter structural barriers and This is the first time that such a large sample of data on serious social and economic challenges in reintegra- ex-combatants from across multiple countries has been tion processes. Regardless, there are also cases where systematically compared and analyzed, thus the study individual ex-combatants fall behind the arc of this gen- represents the cutting edge of empirically driven quan- erally positive reintegration trajectory. In addition, this titative research on the reintegration processes of ex- study has found that communities across the GLR have combatants. played a generally positive role in facilitating the reinte- An important component of the analysis of ex-com- gration of ex-combatants despite certain exceptions. batant reintegration processes revolves around their The communities that ex-combatants return to and the position relative the broader community. As such, this dynamics within these communities are the contexts study compares the reintegration processes of ex-com- into which ex-combatants must reintegrate and, there- batants with those of community members and there- fore, are vital to the understanding of barriers that fore, explores in turn the ways in which these two types ex-combatants face and the processes by which ex- of reintegration processes interact with each other. combatants succeed in reintegrating. It appears that in Notably, the core structure of the analysis presented in the GLR countries, if communities are on a positive the detailed data analysis in Annexes I and II is not only trajectory towards improved stability and social about ex-combatants and the processes through which cohesion, ex-combatants can then root themselves they reintegrate, but also an investigation of commu- in this larger societal transition. However, if this nities themselves, i.e. their willingness and ability to larger transition is less evident, as it is in DRC, there absorb ex-combatants back into society. may be structural barriers to the possibilities of ex-com- This study presents a snapshot of the social and batants’ reintegration. Furthermore, if communities economic dimensions of the overall reintegration are unwilling to receive ex-combatants due to stigma, process of ex-combatants and community members. distrust, economic scarcity or fear, there will be limits However, the conceptual discussion and analysis of to ex-combatants ability to actively engage in the pro- empirical evidence presented consolidates key knowl- cesses of reintegration. In this sense, reintegration is a edge and understanding about the broad trends of ex- two-way process that involves the transformation of ex- combatants’ reintegration processes across the GLR. combatants and communities together. Further, the findings here no doubt carry weight for understanding ex-combatants’ reintegration processes Core to the analysis presented in this study is the idea in contexts beyond the GLR. that the reintegration processes may in part take place independently from reintegration policy and programming - though processes are simultaneously 6 encapsulated within programming. With or without turn building social cohesion more broadly in the assistance from reintegration programming, ex-com- community, are at the core of social reintegration batants in the GLR have returned to communities and processes in the GLR. An essential pathway to expand- navigated the complex set of transitions that reintegra- ing social networks is through marriage and family unit, tion processes entail. Reintegration processes are fun- a domain in which ex-combatants across the GLR face damentally unique in that they are highly contingent on considerable barriers. Again, this does not suggest that both the characteristics of the individual ex-combatants social reintegration is not taking place, as ex-combatants and the contexts into which they reintegrate. However, and community members show key improvements and this study finds that despite the vast range of respec- an overall understanding that social reintegration pro- tive variation in ex-combatants’ reintegration processes cesses take place over a long timeframe. So while social across the GLR, there are also boundaries to such varia- reintegration may show an overall shallow, yet positive, tion, which form the outlines of the broad shape of ex- trajectory, this may be a feature inherent to the nature combatants’ reintegration across the GLR. of the slow processes of reshaping ex-combatants’ self- identity and their identity in the eyes of society, from Indeed, a key finding of this study is that while one of soldier to civilian. Rwanda, Uganda, Burundi, DRC and RoC all display a variety of distinct trends and unique processes The consistent segmentation of ex-combatant and com- in the various dimensions of reintegration, their munity member samples by age, gender, and disability overall reintegration trajectories are remarkably throughout this study has led to the identification of similar. For this reason, the analysis presented in this some distinct subgroups that are at risk. Young ex-com- study focuses primarily on the narrative of the overall batants (age 18-30) lag considerably behind the rest trajectory of reintegration processes in the GLR – only of ex-combatants in terms of social and economic delving into distinct country level processes when they reintegration. This appears to be a result from their diverge significantly from this overall trajectory. The time lost in establishing social and economic footing most notable break in the overall trajectory of reinte- while participating in conflict – as the majority of gration across the GLR at the country level occurs in those aged 18-30 were mobilized as adolescents DRC, where an extremely shallow, and in some regards (under 18). Despite their laggard position in absolute negative, trajectory of social reintegration processes terms, young ex-combatants are found to have an overall stands out. positive trajectory of reintegration across the GLR. While this study finds that ex-combatants across the Likewise, while disabled ex-combatants face unique GLR have had general success in reintegration pro- challenges related to their health, including diminished cesses, they still face many challenges. For example, ex- livelihood potential, they also display a similar positive combatants are worse off than community members in trajectory. terms of overall economic activity, wealth, and income However, both female ex-combatants and female com- security – though this should not indicate that economic munity members break significantly from the overall reintegration is not taking place. In fact, ex-combatants positive trajectory of reintegration in the GLR, and display a steep positive trajectory in economic process- display evidence that suggests a distinct narrative of es. Economic dimensions to reintegration are im- structural disadvantages, especially acute in female ex- portant; however, evidence in the GLR suggests that combatants, that leaves both populations at risk for mar- economic gains of ex-combatants cannot be fully ac- ginalization and social isolation. In this sense, gender tualized without engaging the set of comparatively based disadvantages in the GLR are a broader slowly moving social reintegration processes. issue that is not only exclusive to ex-combatants. In terms of social reintegration processes, ex-combatants Compared to their respective male populations, female face immediate trust and stigma barriers in the commu- ex-combatants and community members consistently nity and are slow to build social networks and establish perform weaker on most social and economic indica- strong social capital in the community. Building social tors. When comparing female ex-combatants and com- capital through expanding social networks, and in munity members , female ex-combatants almost always 7 perform worse. It appears that the stigma associated social capital and social cohesion than are visible else- with ex-combatant status has to an extent an amplifying where in the GLR. The broader societal shift towards effect on already distinct range of gender-based disad- peace and development, which appears to have vantages that female community members face. served as catalyst to ex-combatants’ reintegration in the other GLR countries, is visibly diminished in The core of the structural challenges that female ex- the contexts of continued local violence and inse- combatants face revolve around the stigma barriers curity in Eastern DRC. As a result, although ex-com- to building new familial networks through marriage batants in DRC have quickly caught up to community and in turn the ability to leverage these familial con- members across core reintegration indicators, they have nections towards social and economic outcomes. In had little ground upon which to root during the broader addition female ex-combatants are worse off than female processes of reintegration. community members in terms of education and skills, a gap that will need to be closed if female ex-combatants The case of ex-combatants’ reintegration in DRC hope to gain parity with female community members, let presents a paradox. DRC is the country in the GLR alone males. These dynamics collectively place female that has the greatest level of parity between ex-combat- ex-combatants on a distinctly different overall trajecto- ants and community members across core indicators. ry of reintegration than the rest of ex-combatants that, Though ex-combatants in DRC face an extensive while positive, is so shallow that the disparity between range of disadvantages, the extent of these disadvan- females and males could grow – leaving females at clear tages to the wider community is relatively insignifi- risk for further social and economic marginalization. cant compared to other GLR countries. However, DRC receives special attention in this study as a standout ex-combatants and community members in DRC case of ex-combatants’ reintegration in the GLR. In together are arguably the worst off among those in DRC, both ex-combatants and community members all the other GLR countries. As such, the processes of face social and economic challenges to a degree gener- ex-combatants’ reintegration and broader societal trans- ally beyond what is found in the other GLR countries. formation in DRC appear perhaps the weakest across As such, both ex-combatants and community members the GLR. This study grapples with this seeming contra- perform weaker across almost all indicators of reintegra- diction and its implications for reintegration policy and tion processes explored in this study than in any other programming. It may be that reintegration policy and country in the GLR. The core challenge of reintegration programming have limited abilities to directly shape in DRC is that communities display weaker levels of some elements of reintegration processes. 8 3. Introduction T his study compares the reintegration processes that demobilized ex-combatants across the GLR of Africa navigate as they return to com- munities and transform their identities from soldiers to civilians. The World Bank commissioned this compara- ment in a dialogue between ideas about reintegration processes more broadly, and evidence about reintegra- tion processes in the GLR more specifically. With these overarching goals in mind, this study must tive study to help build on baseline and tracer studies be contextualized outside the traditional paradigm completed in 2010-2012 on reintegration processes as a of most stakeholder-produced literatures on ex-com- batants’ reintegration. Indeed, this study is afforded part of Transitional Demobilization and Reintegration the luxury of engaging in, at times, abstract scholarly Program (TDRP) support for reintegration program- concepts and ideas surrounding reintegration at a level ming across the GLR of Africa - specifically Rwanda, that rarely fits the motive or mandate of DDR reintegra- Uganda, Burundi, DRC and RoC. The overall aim of tion stakeholders. In this sense, this study is designed this study is to help consolidate knowledge and under- as a bridging piece between scholars and practitioners standing of reintegration processes across the broad concerned with the reintegration of ex-combatants. range of contexts in the GLR. The core structure of the analysis presented in Annexes This study utilizes the newly merged TDRP-GLR I and II of this study are not only of ex-combatants database of nearly 10,000 ex-combatants and com- and the processes by which they reintegrate and their munity members across the GLR. This is the first time position relative to communities, but also as investiga- that such a large collection of survey data on reintegra- tion of communities themselves - their willingness and tion processes has been systematically compared and ability to absorb ex-combatants back into society. Part 1 analyzed. As such, this study is on the cutting edge of on this study focuses on high-level analysis of ex-com- empirically driven quantitative research on ex-combat- batants’ reintegration processes across the GLR. This ant reintegration processes. The findings presented in high-level analysis is supported by the more detailed this study are a first pass-over of the enormous amount data analysis in Annexes I and II of this study examin- of data in the TDRP-GLR dataset – a resource that will ing the following areas: no doubt continue to generate insight on ex-combat- ants’ reintegration processes in the future. i. Demographics: Information pertaining to a standard range of This study is not a part of the formal monitoring and demographic factors such as age, gender and evaluation (M&E) of TDRP activities in the GLR, but disability in addition to marital status and levels is rather a knowledge piece focused on building broader of educational achievement and vocational skills. understandings of reintegration as a phenomenon. This includes: (i) engaging and exploring a range of primari- ii. Housing: ly scholarly concepts and ideas around the fundamental Analysis of current living situation including: meaning and nature of what it is to reintegrate (ii) using housing type and tenure. these concepts and ideas to broadly reflect on the social and economic reintegration processes that ex-combat- iii. Land, Livestock and Food Security: ants across the GLR have experienced, including how Analysis of access to arable land and measures of community members across the GLR have perceived general food security. and engaged with these processes (outlined in detail in iv. Economic Issues: Annexes I and II of this study); these two components Analysis of current economic status as well as and the interwoven discussion represent (iii) engage- actual and perceived vulnerability and outlook 9 Table 1: Raw GLR Country Sample Contributions GLR Raw Sample N Burundi DRC RoC Rwanda Uganda GLR Total Ex-Combatants 1256 3625 668 517 410 6476 Community Members 510 722 1456 510 182 3380 Subtotal 1766 4347 2124 1027 592 9856 for the future. In addition, further analysis on of the previous evaluations and studies carried out uti- income, savings, access to credit, and economic lizing the survey data from each of the five GLR coun- associations is undertaken, and a special focus tries. These studies have been further complimented by on the dimensions of gender and disability in a selective review of contemporary academic literatures economic reintegration is explored. on ex-combatants’ reintegration processes as well as informal consultations with experts on ex-combatants’ v. Social Capital: reintegration in the GLR countries. Analysis of the dynamic components of social capital in- cluding: sociability; trust and solidarity; social cohesion The total sample contributions from each of the five and inclusion; empowerment; and social change. GLR countries in this study varied considerably and are visible below in Table 1. The samples from each vi. DDR Experiences: country were drawn with certain purposive sampling biases and all encountered barriers to achieving their Analysis of ex-combatant and community member ex- ideal sample compositions. For further details on these periences of demobilization and reinsertion program- points see the individual ex-combatant and community ming as well as the initial phases of ex-combatants’ member studies from each of the five GLR countries.2 return to the community, including: initial levels of Information concerning more specific details of the de- trust, acceptance and stigma in the community. mographic compositions of each sample, and certain in- 3.1 Methodology stances where data from certain countries is excluded or missing is presented in Annexes I and II of this study. The overall methodology of this study comprises of two main components: To avoid biasing effects of the uneven sample sizes from each of the five GLR countries, data has been i. Document Review weighted for this study. The ex-combatant and com- munity member samples from each GLR country have ii. Quantitative Analysis weighted evenly respectively. The overall trends of re- This study utilizes the newly created TDRP-GLR integration processes across the GLR countries are database – which merges previously collected data from such that this weighting rarely makes a difference to baseline and tracer studies on ex-combatants’ reinte- the overall trends in the region. However, in several key gration in Rwanda, Uganda, DRC, RoC, and Burundi fields it provides an essential component of an accurate carried out by the TDRP from 2010-2012.1 To aid in and nuanced data analysis. the analysis of this vast sample of data of nearly 10,000 ex-combatants and community members, the method- This study of ex-combatants’ reintegration processes ology of this study includes a phase of extensive review across the GLR is a meta-study that synthesizes the findings from previous TDRP baseline and tracer 1   The data collected in each of the five GLR countries was captured in partnership with local DDR commissions and implementing partners. 2   See World Bank (2011a) (2011b) (2011c) (2011d) (2012) 10 studies. As such, this study focuses on combining and Rwanda covers a similar range of topics to the second contrasting the data and results of previous studies in format used in the rest of the GLR countries, but is the GLR in order to draw out a core understanding of often formatted in such a way that the data from Rwanda the processes of reintegration in the GLR countries. is not directly comparable to that from the other GLR This core understanding is contextualized against the countries. For this reason, including Rwanda in the full backdrop of broader scholarly thought on reintegra- range of analysis in this study has proved challenging. tion processes in general. While previous studies and Instances where Rwanda is excluded from detailed data evaluations on ex-combatants’ reintegration in the findings in Annex I and II of this study are footnoted. GLR countries have tended towards a combination of See Section 7 in Annex I for more detail. micro (individual level) and meso (subnational) level analysis, this knowledge-focused study shifts towards The survey data in the TDRP-GLR database for com- a micro / macro approach – analyzing ex-combatants munity members in Burundi has also presented some primarily by demographic segments at a national level issues. In Burundi, community members were surveyed at the cost of much subnational (geographic sub-region by using a shortened version of the complete survey or specific armed group) level analysis. In some regards, format. Furthermore, this format did not include a this approach represents a loss of detail that has been capture of basic demographic details. Due to this produced in earlier studies, but at the same time, this “missing data”, including community members from shift in the grain of analysis allows the opportunity to Burundi in the full range of analysis in this study has speak more broadly about the social-economic process proved challenging. For further detail see Section 13 in of reintegration across the GLR. Nevertheless, the chal- Annex II of this study. lenge depicting the broad generalities of reintegration Because of the two cases discussed above, data format- processes in the GLR countries versus the unique speci- ting differences in Rwanda surveys and missing data ficities of reintegration processes is a source of funda- in Burundi community member surveys, this study mental tension in this study. can at times appear as though it approaches being a 3.2 Limitations and Challenges comparative study of only DRC, RoC, and Uganda. Further, at times there is extended analysis nuancing This study has faced a range of challenges revolving specific points in the survey data that is drawn specifi- around comparing data from different survey formats, cally from Uganda. It should be noted that this does not sample compositions, and contexts for analysis across reflect an overt focus on Uganda, but suggest the fact the five GLR countries. In general, these challenges that the range of survey data collected in Uganda is the create a range of limitations that fall into two catego- most expansive – and in many instances it is the only ries: (i) the compatibility of survey data collected from GLR country with data available to make such further across the GLR countries; and (ii) the comparability of nuances in analysis, as the Uganda survey format is the different reintegration contexts across the GLR coun- most developed iteration used in this study and holds tries. the most comprehensive range of data. 3.2.1 Individual GLR Country Survey Compatibility There are also two issues of sample validity that limit the analysis in this study. First, there is no analysis of ex- The format of the GLR ex-combatant and community combatants under the age of 18 in this study. Across the member surveys has been a process of iterative learning total sample of ex-combatants from across the five GLR and refinement. Rwanda was the first GLR country in countries there were 326 respondents under the age of which the survey format was tested and developed. 18 (300 ex-combatants and 26 community members). The data and learning itinerary that came out of this These 326 have been omitted from the analysis in this original format then influenced the design of a con- study for two main reasons: (i) the systematic capture solidated second format that has seen customization to of information pertaining to the specific dynamics of re- the specific study needs in Burundi, Uganda, DRC and integration facing minors was absent from the surveys RoC; however, its base components have remained con- used across the GLR countries – with the exception of sistent since this consolidated second format. In effect, DRC, where 291 of the total 300 ex-combatants under the data that exists in the TDRP-GLR database for the age were sampled; and (ii) the validity issues that 11 the small sample of ex-combatants under the age of 18 been seen as largely case specific and , therefore, are (again, almost entirely from DRC) makes meaningful excluded from the scope of this study, warranting and valid comparative analysis infeasible. focused inquiry on their own in other studies. In this study, there is no analysis of ex-combatants by armed Data along health and disability demographics also group. At a national level understanding, the unique presents challenges in the total GLR community reintegration challenges faced by the different armed member sample. Health and disability data for com- groups within that nation is essential; at the regional munity members was only collected in Rwanda and level, however, this may lead to comparing apples with Uganda and was absent from Burundi, DRC and RoC. oranges. Different armed groups have different wartime Unfortunately, data from Rwanda and Uganda is limited activities and draw different members to mobilization as only 58 disabled community members were sampled for different reasons. In the GLR, a plethora of armed (49 from Rwanda and 9 from Uganda). Thus, drawing groups of different sizes, with different goals, and with valid comparisons between these two samples of 49 and different modus operandi have gone through reinte- 9 disabled community members is considered as infea- gration processes. All these factors impact the unique sible; furthermore, comparing these 58 disabled com- experiences that specific armed groups have in reinte- munity members to the 454 disabled ex-combatants gration. In the context of this study, analysis by armed in this study will present further issues for validity. For group is considered infeasible. One dimension that these reasons, analysis of community members along future studies could focus on is the range of reintegra- the lines of disability is absent from Annex II of this tion processes that members of regular versus irregular study. armed groups go through – for instance, if they differ and to what extent. For ex-combatants, disability was defined differently for the baseline and tracer studies carried out in each of the In this study there is no direct analysis of the reintegra- GLR countries. Detailed analysis of ex-combatant rein- tion processes that ex-combatants encounter by sub-na- tegration data in Annex I of this study outlines the logic tional geographic region. The social (e.g. ethno-religious by which these different definitions of disability were / cultural) and economic (e.g. urban vs. rural) makeup streamlined for this study. See Section 7 for specific of different regions creates different contexts for rein- details. tegration. These are important dimensions for under- standing varying reintegration processes at a national 3.2.2 Individual GLR Country Context level; however at the regional level of the GLR as a Comparability whole, taking on this frame would increase the scope analysis by an order of magnitude and is thus judged to While the conflicts that have occurred in the GLR be outside the scope of this study. countries are deeply intertwined, the contexts in which they have occurred and the contexts into which ex- In this study, there is no systematic periodization of the combatants reintegrate are fundamentally unique. The reintegration processes taking place across the GLR idiosyncrasies of the GLR country contexts create fun- countries. While the data used for this study was all col- damental challenges for comparing reintegration pro- lected in a roughly similar period of time between 2010 cesses at a fine-grained level. This raises the question of and 2012, some questions in the surveys presented in “how can we systematically compare diverse contexts to Annexes I and II refer to events (such as demobiliza- draw broadly applicable conclusions that also preserve tion) that have occurred in varying time proximity to the the specific details present in each case?” Indeed, the time of sampling. Due to the unique temporal dynamics tension between the general and the specific is perhaps of reintegration processes in each of the GLR countries, the core methodological challenge of this study. This a synchronization of period-specific questions is largely study has tried to balance the detailed data analysis in judged as infeasible for analysis in this study. Annexes I and II with the broad general analysis in the summary document. Lastly, this study does not include a systematic com- parison of DDR programming components in the Beyond this overarching tension between the general GLR countries, as the focus of this study is the pro- and the specific in this study, a number of areas have cesses of ex-combatants’ reintegration occurring in the 12 GLR countries as distinct from the range of policy and ex-combatants and community members across the programming aiming to affect these processes in each GLR represent a leap forward in empirically driven country. Admittedly, programming components are research on ex-combatants’ reintegration processes. By relevant for understanding the reintegration processes systematically comparing the reintegration experiences and they are brought in for contextualization where of ex-combatants and the communities that receive necessary. However, there is no systematic capture them across a broad range of contexts, we can begin as this is outside the fundamental focus of this study. to consolidate understanding about the characteris- Future studies would do well to connect the broad un- tics of ex-combatants’ reintegration processes that go derstanding of reintegration processes here with differ- beyond the specific country context. Taking this explic- ent programming approaches. itly cross-country perspective can essentially help us to consolidate a core understanding of the general shape Despite limitations listed above, the synchronization and nature of ex-combatants’ reintegration processes as and merging of previously collected survey data on they occur in the GLR or even in broader context. 13 4. Core Concepts in Academic Literature on Reintegration Processes The following section of this study outlines several core and identity in the eyes of society, from one of soldier reintegration concepts: to civilian with or without assistance from reintegration programming.3 i. analytical distinction between reintegration as an individual level process that ex-combatants While distinguishing between these two types of reinte- navigate and reintegration as a body of policy gration can seem somewhat esoteric, the deep ontologi- and programming that DDR practitioners cal distinction between reintegration as a programming implement; activity and reintegration as process should not be over- looked.4 At the heart of this distinction is the point that ii. ex-combatant experience of reintegration as a reintegration as a process that individuals and com- process – essentially social and psychological in munities go through is conceptually independent from nature; reintegration programming activities. Ex-combatants, iii. underlying conceptual similarity of ex-combatants’ refugees, IDPs and other groups affected by conflict all reintegration processes to those that other groups reintegrate into society in FCS settings, with or without (e.g. refugees and internal displaced persons) the help reintegration programming.5 To recognize in fragile and conflict affected situations (FCS) this distinction is to acknowledge that reintegration experience more broadly. processes are essentially human processes navigated by individuals, and that their underlying form may not To address the ambiguities of complex reintegration be unique to ex-combatants alone (see Section 4.1.1). processes across a wide range of contexts, the concep- tual tool of reintegration trajectories is outlined. These concepts implicitly inform the discussion in 3   This study is hardly the first to recognize the analytical distinction between reintegration as a program and reintegration as a process – though Section 5 and the more detailed data analysis in Annexes the point is often implicit in many texts - see for example Bowd & Özerdem I and II. (2013). For a particularly good explicit handling see Torjesen (2013). This distinction invites a contradiction. On one hand distinguishing between 4.1 Reintegration as a Process processes and programs creates space for a dialogue between evidence and ideas in this study. However, in practice programming and processes are inherently bound, exerting force upon each other. This this sense this study must consider ceteris paribus when discussing processes in the GLR in a A core idea underlying the analyses presented in this broad fashion. study is the conceptual distinction between two overlap- 4   One way to further assert this ontological distinction is to think ping parts of reintegration. First, there is reintegration epistemologically. Studying the activity primarily involves a focus on policy as part of Disarmament, Demobilization and Reinte- and programming and their interaction with ex-combatants. Inversely, studying reintegration processes is primarily concerned with individual gration (DDR) policy and programming. This type of processes that ex-combatants themselves experience and navigate, and reintegration is the programming activity of aiding com- the interaction of these processes with policy and programming. This ontological distinction underlies much of the conceptual discussion batants’ return to society in their transition from com- discussed in this section of the study. batant to civilian identity. Second, there is reintegration 5   Torjesen (2013) makes the important point that due to the usually as a process. This type of reintegration is the social and voluntary nature of participation in reintegration programming there are many ex-combatants who go through reintegration, in terms of the economic processes by which individual ex-combatants process, even though they may not receive assistance through reintegration return to communities, build economic livelihoods, programming, the activity. This point lends to the distinction between reintegration the activity and reintegration the process. Further, this point and perhaps most importantly, rebuild social connec- identifies the importance of so called “self-reintegraters” as an important tions in the community – reshaping their self-identity area of inquiry themselves. Baas (2012) makes a similar point. 14 Indeed, taking this to its extreme, some scholars have these underlying processes for granted, as reintegration used the distinction between program and process to as an activity and as a process are inherently bound in posit that reintegration processes are fundamentally “… practice. Indeed, it may be that the prospect of under- unrelated to DDR programs, which have little to con- standing local contexts and effectively applying broad tribute towards the return process and in most instances sets of guidelines (such as those in the IDDRS) are the family, and the home community of ex-combatants among the fundamental challenges to implementing bear the onus of responsibility for reintegration success programming in FCS. Practitioners and scholars alike at an individual level.”6 Again, while the two sides of re- often speak about these two types of reintegration inter- integration are analytically and ontologically distinct, changeably and in some cases as one conflated concept. they are nevertheless inseparable in practice and in fact Indeed, when we talk about DDR, distinguishing which dynamically feed into one another. of these two “reintegrations” we refer to, the activity or the process, can be challenging to keep track of but is Literature on reintegration as a programming activity is fundamentally essential. abundant and has been focused on policy and program- ming challenges in the complexities of implementing The point of drawing distinctions between reintegra- reintegration projects. Indeed, much literature on rein- tion as a programming activity and reintegration as tegration finds its origins within organizations serving a process is not to suggest that these two should be as key stakeholders in DDR policy and programming separated in practice. On the contrary, in practice, re- as a part of program evaluations, or from practitioners integration policy and programming activities must be cum scholars - and has thus inevitably adopted analysis grounded in endogenous local reintegration processes. focused on their programmatic and policy priorities.7 However, maintaining an analytical distinction between Consequently, there has been an overwhelming focus the activities and processes helps frame the underlying in reintegration literature on the technical and logisti- question in this study: what is the overarching shape of cal components of program planning and management. those endogenous reintegration processes in the GLR This body of literature often draws from a single case to countries? It is through building an understanding of produce what has often been characterized as “lessons the broad trends in reintegration processes across the learned” or “best practices” literature. Collectively this GLR countries that we can contextualize the extent and body of work has been synthesized together into a broad variation of reintegration processes in more specific base of programming knowledge, culminating perhaps settings – including their interaction with reintegration most notably in the establishment of the Integrated policy and programming. Disarmament, Demobilization and Reintegration Stan- dards (IDDRS).8 However, understanding the relationship between reintegration programs and reintegration processes While these types of studies have a very real value to is no easy task and the answers to many fundamental reintegration practitioners in terms of guidance for re- questions surrounding reintegration processes remain integration policy and programming, they often leave unclear – if even addressed at all. What are the compo- the underlying processes that these policies and pro- nents of reintegration processes? By what mechanisms gramming are meant to affect fundamentally unde- do reintegration processes function? What do reintegra- fined (at least in any explicit sense). It is easy to take tion processes look like empirically? Can we measure reintegration processes? Under what circumstances and to what extent can policy and programming actually 6   Özerdem & Podder (2011) p.5. affect these processes? Are reintegration processes 7   The UNDP and UNDPKO and The World Bank’s TDRP (formerly fundamentally unique to a given context, or is there an MDRP) are examples of such key organizations. The point here is not to underlying structure to them? Do reintegration pro- suggest that a policy and programming focus is inferior to an academic inquiry (which indeed carries its own set of priorities), but rather to cesses at the individual level aggregate to shifts at the merely acknowledge the role these priorities have played in shaping the community, regional, or national levels? The answers practitioner discourse around reintegration. to such questions are of fundamental importance to re- 8   UN (2006) For a good overview of the landscape of academic integration policy and programming. Without a clear literature on DDR see: Muggah (2010) or Schulhofer-Wohl & Sambanis (2010). understanding of reintegration processes, the prospect 15 of creating meaningful benchmarks, or “metrics of son across the GLR countries is methodologically an success”, reintegration programming will remain chal- important step in building knowledge about reintegra- lenging.9 Without clear goalposts, reintegration policy tion processes - as comparison can serve a vital role in and programming can risk becoming decontextualized hypothesis confirmation, modification or falsification. from the process they are meant to affect.10 This is the It is only when we begin to compare outside the first core of the programming / process divide. case that the impact of elements that may have been held constant, and thus invisible, become plain to see. More recent academic studies on reintegration, which Indeed, scholars on reintegration, and DDR as a whole, reflect an implicit awareness of the reintegration pro- have emphasized that “serious comparison across coun- gramming / process divide, have begun to move away tries of relevant aspects of conflicts and the programs from the policy and programming focused case-study, designed to address them will provide invaluable insight which was typical in earlier reintegration scholar- into the complex interaction between DDR programs ship, to a process focused approach, shifting their unit and social processes beyond the insights a single case of analysis to the individual ex-combatants.11 While can provide.”14 these studies may still be fundamentally focused on understanding the effectiveness of reintegration policy A focus on reintegration as a process, distinct from the and programming, such shifts in the scale and unit of programming activity of reintegration embodied in the analysis open the door to the large-scale study of rein- policy and programming which they underlie, is the tegration processes. However, such academic studies frame from which this study departs. While in reality are still not widely applied, as the ability to access and processes and programming are inextricably linked in capture empirical data at the individual level in chal- practice, exploring reintegration processes in isolation lenging FCS environments necessitates the capacity reaffirms the importance of grounding programming of large international organizations – most likely those in a deep understanding of the endogenous local pro- directly involved in the planning and implementation of cesses of reintegration.15 reintegration programming. 4.1.1 An Individual Process From this perspective, data collected by the TDRP across the GLR countries and the studies produced Reintegration is a complex process involving multiple from them are on the cutting edge of empirically driven simultaneous overlapping transitions. Reintegration quantitative research on ex-combatants’ reintegra- processes are multidimensional in that they have social, tion processes.12 While the individual country level political, and economic components.16 Ex-combatants analyses in Rwanda, Uganda, Burundi, DRC, and RoC must reshape their social identity and build social con- have revealed much about the dynamics of reintegra- nections in the community around this new identity. In tion processes in each of these specific contexts, they addition, ex-combatants must reach a level of economic are yet to lead to a broad comparison of individual level stability through building a sustainable livelihood ex-combatants’ reintegration processes across the GLR path and, at the same time, engaging in their commu- as a whole to attempt to distill the core of reintegration processes as they are visible in the context of the GLR and perhaps more broadly as well.13 Indeed, compari- outcome. Hard empirical facts about overarching reintegration processes as a broad phenomenon will likely remain few, however broad understanding of the shape of reintegration processes are the more attainable goal of this piece. The specific contexts and details of reintegration in each setting are so unique that some levels of comparison may prove unfruitful. This speaks to 9   Bowd & Özerdem (2013) the tensions between the general and the specific outlined in Section 2.3.2 of this study. 10  Jennings (2008) 14   Schulhofer-Wohl & Sambanis (2010) p. 42. 11  See for example: Humphreys and Weinstein (2004) (2005) (2006) (2007) (2008), Kaplan and Nussio (2012), or Porto et al (2007). 15   Another way to say this would be that reintegration programming must be rooted in an understanding of the local political economy. 12  The TDRP’s quantitative research benefits from coordination on ongoing qualitative research in the TDRP and partner organizations. 16   The point that reintegration processes consists of social, political, and economic dimensions is ubiquitous in both academic and institutional 13   One may, however, question whether this kind of distillation is literature on DDR. For a particularly thorough handling see Bowd & even theoretically possible. The answer may depend some on the desired Özerdem (2013). 16 nity. These multiple transitions occur simultaneously, For reintegration programming, this meant that if the though not necessarily synchronously. For example, primary reasons that individuals joined armed groups evidence from this study suggests that while ex-combat- were economic, then the best way to help them in the ants in the GLR make very immediate and solid gains in transition to civilian status was through measures like terms of finding employment, returning to their family skills and vocational training, livelihoods packages, and and building trust in the community, the broader pro- micro-credit mechanism designed to help ex-combat- cesses of building an economic foundation for stability ants achieve relative economic security through sustain- and developing social networks in the community take able livelihoods. Admittedly, economic reintegration place on a longer timescale. is an important dimension to reintegration processes, but as evidence across the GLR suggests, it is only one Reintegration processes are overlapping in the sense part of reintegration processes. In fact, it is likely that that the distinctions between social and economic the social dimensions of reintegration processes are the processes are rarely black and white. For example, in most important for realizing the outcomes of economic the GLR countries, ex-combatants’ social networks reintegration. can contribute to economic networks and in turn to economic stability – blurring the lines between the Today most scholars and practitioners alike acknowl- social and economic spheres. Indeed, in settings where edge that understanding individuals’ mobilization there is no broader societal safety net and the ability into armed conflicts in settings like the GLR involves of households to provide a family-based social safety shifting from a singular focus on “rational-actor” frame- net is severely compromised, it is likely that social and works of purely economic perspectives to a broader economic spheres cannot be viewed separately. Lastly, approach incorporating insights from psychology, soci- reintegration processes are transitions, as there is no ology, and history.18 Across the GLR countries, a vast clean break from being soldiers to being civilians, from and rich social history has shaped many contexts in marginalization to inclusion in society, from economic which individuals have mobilized into armed groups – insecurity to relative stability. Rather, the processes of from the most central Hutu / Tutsi divide in Rwanda, to reintegration are a gradual realignment across multiple the incessant local conflicts of Eastern DRC stemming dimensions. from competition for land and economic opportuni- ties. Overall, extreme poverty and few opportunities Scholars and practitioners in the field of reintegration for social, political or economic mobility and empow- have always, at least implicitly, grounded their analysis erment are largely the norms in the GLR. It is not these of the pathways out of conflict on an understanding of contextual factors themselves but rather the individual pathways into conflict. This is logical - to understand experience that drives mobilization into armed groups. how to deconstruct armed groups we should under- Indeed, contemporary scholars are overt in the position stand how they were constructed. In this sense, the that “… in order to understand what it means to be study of reintegration is grounded on the study of the disarmed, demobilized and reintegrated, we must un- causes of war. In earlier periods of reintegration schol- derstand what it means to be armed, mobilized and to arship and programming, this grounding was often become part of a guerilla movement” from the perspec- implicitly signified by an overwhelming focus on the tive of ex-combatants themselves.19 economic dimensions to reintegration, which was tied heavily to “greed-based” explanations for civil war.17 Adopting a perspective based on the individual ex- perience of conflict does not require leaving behind empirical evidence about the economic dimensions of conflict as factors affecting soldiers’ motivations to 17   The so called “greed vs. grievance” discourse was central in the latter half of the 1990’s evolving a large body of discourse. The work of Paul mobilize. What adopting such a perspective entails is Collier is the best example of this discourse. See for example Collier & Hoeffler (2004) or Collier et al (2009). Today, though, purely economic arguments on the causes of civil war are considered somewhat passé due to their weakness in capturing the psycho-social dimensions of mobilization 18   Keen (2008) and conflict. Indeed, many have argued that the economic motives for mobilization that are traditionally understood as a part of “greed” can just as 19   Baas (2012) p. 20. See also McMullin (2013) and Özerdem & well be understood as grievances themselves. Podder (2011) 17 acknowledging that the individual experience of things sense, becoming a soldier can be understood as a form like extreme poverty and political disenfranchisement of personal empowerment.”22 Indeed, Peter Uvin simi- is almost always also social and psychological in nature. larly found that mobilization of young men in Burundi While this point may seem to be a truism, economic was related to addressing emotions of anger and pow- factors, approached through rational actor perspectives, erlessness as well as finding opportunities for income, have often received the most emphasis in the context mobility and empowerment.23 Consequently, the point of academic scholarship on mobilization and demobi- here is that despite vastly different conflict contexts, of lization. Making this point frames the discussion going which economic dimensions may be an important com- forward. Powerlessness, shame, and humiliation are ponent, social-psychological individual experience may powerful human emotions that some scholars believe be at the core of the reason why individuals mobilize may be at the core of understanding the individual ex- into conflict.24 perience that drives mobilization into armed groups.20 Highlighting the individual level factors driving mobi- This discussion brings us back to the programming / lization in Liberia, Mats Utas describes social and psy- process divide, the space between reintegration as an chological challenges that youth faced as they moved activity and reintegration as a process. If reintegration towards mobilization: policy and programming are to resolve the pathways by which civilians are mobilized into armed groups and Possibilities to participate in the wage economy become soldiers through the activity of reintegrating diminished and education ceased having any them into communities, then such policy and program- importance. With this crisis looming, many ming will have to address the individual experience of young men lost even the possibility to estab- anger, shame, insecurity, poverty, hunger, immobility lish themselves as adults, by building a house, and powerlessness. This is of course presuming that or getting married – even though they contin- resolving such pathways is among the strategic goals ued to become fathers, of children for whom of individual reintegration programs.25 Nonetheless, they could not provide.21 core questions remain on whether or not reintegration policy and programming can in fact affect these social Utas’ main point is that while the conditions of extreme and psychological dimensions in a meaningful manner. poverty and political alienation were the backdrop of Evidence from the GLR countries suggest that when mobilization in Liberia, individual experience or mo- reintegration programming is carried out as part of a tivation for joining armed groups was one of moving larger societal transition from war to peace, it can play a away from the periphery of society to its center. Saskia positive role in helping transform the contexts in which Baas’ analysis of the mobilization and demobilization ex-combatants were initially mobilized. However, in of soldiers in South Sudan largely concurs, in that: contexts of continued conflicts and instability, with “… these movements also have something to offer: a Eastern DRC as a chief example in the GLR, it appears meaningful life path as part of a social structure that that there is a limited societal transition, as such, for re- offers real opportunities for upward mobility… In that integration programming to take root in, thus limiting the range of impacts that reintegration activities can have on reintegration processes (see Section 5.6). In 20   Keen (2008) sees the emotional experience powerlessness as core to understanding mobilization and violence. Drawing from the work of psychologist James Gilligan’s (1997) study of violence in criminals and prisons Keen (2008) p. 68 asserts that: “If power corrupts, so too 22   Baas (2012) p. 141. See also Bøås & Dunn (2007) does powerlessness; the two together have proven particularly dangerous. 23   Uvin (2007) A sense of powerlessness can feed into the assertion of power through violence; frequently the perpetrator feels powerless in relation to one set 24   Indeed, this point is not new and is not unique to contemporary of people but exerts power over a separate and vulnerable group (usually conflicts in Africa. Human experiences may be at the core of understanding unarmed). This helps explain why the perpetrator frequently feels like a voluntary individual mobilization throughout history. victim. In the midst of carnage, morality does not always or completely collapse; instead, it is typically twisted, distorted, perverted and even 25   However, whether or not resolving the pathways by which ex- inverted, with attack perhaps redefined as self-defense and restraint as combatants are mobilized into armed groups is an explicit strategic goal, it weakness.” is almost always part of the underlying logic by which such programs are legitimated (with the notable exception of the reintegration of national 21   Utas (2005) p. 150. armed forces). 18 these contexts, reintegration can actually end up leading reintegration that are related to their individual history to undesired remarginalization.26 of mobilization into armed groups and participation in violent conflicts. Indeed, as argued above, these factors The processes that ex-combatants navigate as they return compose the core frame through which ex-combatants to communities and transition from being soldiers to navigate the process of reintegration. However, if the being civilians are also rooted in the boarder context process of reintegration involves reshaping identity and of the post-conflict setting. Communities that ex-com- building senses of empowerment and social cohesion batants return to and the dynamics within the commu- in the community and society at large, then there may nities are the contexts into which ex-combatants must reintegrate and, in this sense, are vital to understanding be other groups, such as returning IDPs and refugees, the barriers they face and the processes by which they including the community as a whole, that face similar succeed. If communities are unwilling to receive ex- processes. combatants due to stigma, distrust, economic scarcity Indeed, the label of ex-combatant is a social con- or fear, there may be limits to the agency that ex-com- struct that we use to distinguish ex-combatants for batants can assert in the process of reintegration. In the security threat that they are thought to pose in the contrast, it appears that in the GLR countries, if com- post-conflict environment. In this light, it is interesting munities are on a positive trajectory towards improved to note that though the analytical distinction between stability and social cohesion, ex-combatants can root “ex-combatants” in the global south and “veterans” in themselves in this larger societal shift. In this sense, re- the global north is often thin, both are soldiers return- integration is a two way process that involves reintegra- ing from combats who face the challenge of reshaping tion of ex-combatants and communities together.27 It their identity and reentering society; however, the dis- is for this reason that the core structure of the analysis presented in Annexes I and II are not only of ex-com- courses surrounding the security threats they pose, their batants and the processes by which they reintegrate and role in development, and the processes by which they their position relative to communities, but also as inves- reintegrate into society are separated by vast chasms in tigation of communities themselves - their willingness the global south and the global north.28 However, the and ability to absorb ex-combatants back into society. point is that while there are very real contextual dif- ferences between “ex-combatant reintegration” into 4.1.2 A Unique Process? settings of extreme poverty and social marginalization, and “veteran reintegration” into settings of high levels While reintegration processes are fundamentally in- of social and economic development, the underly- dividual in nature, it may be that the range of possible ing human process that they go through can be quite individual processes is bounded to the limited array of human social and psychological processes. If reintegra- similar. tion is an individual human process, then we must ask ourselves if this is a process that ex-combatants alone go 28   McMullin (2013) gives an extensive account of the role of veterans through. Or is there an underlying shape to the process- in state development in USA and England – contrasting this against the es of reintegration that exists across different contexts discourse on contemporary ex-combatant reintegration. This reading is complimented by the likes of Campbell (2003) and Hannoum (2013). As of ex-combatant reintegration? Further, are these pro- an interesting side note, a group that faces remarkably similar context of cesses similar to transitions that other groups, or the reintegration (reintegration into poverty, insecurity, social marginalization, and stigma) as well as a similar discourse of securitization, accompanied community at large, face? This is not to suggest that by a massive project to aid their reintegration into society, is that of ex- ex-combatants do not experience specific barriers to prisoners. Ex-prisoner reintegration in the field of criminology has long addressed the processes by which individuals carry the legacies of their crimes and time in prison as they attempt to return to family structures, find economic independence, build ties in the community and contribute 26   Many scholars have made the point that reintegration can often to society. Indeed, in reading contemporary literature on ex-prisoner mean reintegration into the contexts of poverty and marginalization that reintegration one could be forgiven for mistakenly thinking they were served as the context for mobilization in the first place. McMullin (2004) is reading a work about ex-combatant reintegration (See for example Visher often cited for the idea of “reintegration back into basic poverty.” However & Travis (2003)). Discourse on ex-prisoner reintegration has much to the specific phrasing of remariginalization comes from Utas (2005). offer the study of ex-combatant reintegration – which would stand to benefit from utilizing a long history of established sociological concepts in 27   Finn (2012) understanding reintegration 19 If we understand reintegration as a human process, it cesses in the GLR may be a way forward.30 appears that although physical contexts of reintegra- tion vary vastly, this process may not be exclusive to One advantage to the heuristic value of using reinte- ex-combatants. Learning from a long tradition of estab- gration trajectories is that it can be applied at various lished concepts in sociology revolving around identity, scales. We can talk about the specific areas of reinte- communities, and belonging can offer valuable insights gration and then aggregate these to larger processes into the process of reintegration that ex-combatants in through the same conceptual tool. For example, we the GLR face. In fact, evidence in this study suggests can talk about ex-combatants in the GLR’s trajectory that though ex-combatants across Rwanda, Uganda, in terms of economic status moving positively from Burundi, DRC, and RoC face unique environmental unemployment to self-employment in agriculture or contexts of reintegration, the overall shape of the pro- non-agricultural small business.31 This is only one piece cesses that they encounter is remarkably similar. of the puzzle when considering ex-combatants in the GLR’s overall moderately positive economic reintegra- 4.2 Reintegration Trajectories tion trajectory – despite numerous disadvantages and barriers to economic reintegration that ex-combatants A heuristic concept used in both academic scholarship across the GLR face. Further, we can continue to use and stakeholder literature on reintegration, though the concept of reintegration trajectories when we speak seldom with any explicit definition, is that of reintegra- about reintegration even more broadly, such as in the tion “trajectories”.29 The concept of reintegration trajec- dynamic interaction of the social and economic spheres tories provides an analogy that allows us to speak about of reintegration. the overall arc or narrative of ex-combatants’ experience of, and engagement in, reintegration processes. Positive While the analysis presented in this study looks across trajectories indicate progress towards overall reintegra- a broad range of country contexts with distinct groups tion into the community along social and economic imbedded in them, the concept of reintegration tra- lines, while negative trajectories indicate a discord in jectories will play an organizing role by which portray reintegration processes and the presence of barriers to the overall narratives of reintegration processes in the ex-combatants’ reintegration. Steep trajectories indicate GLR – especially in Annexes I and II. Indeed, a key robust and decisive processes, and shallow trajectories finding in this study is that while Rwanda, Uganda, indicate more subtle or slow moving processes. Burundi, DRC and RoC all display a variety of distinct trends and unique processes in the various dimensions Adopting an explicit focus on reintegration as a process of reintegration, their overall reintegration trajectories, and the consequent acknowledgement of individual and the large-scale processual narrative that they rep- psychosocial experiences of this process across a wide resent, are remarkably similar. The most notable break range of contexts creates considerable complexity in in the overall trajectory of reintegration across the GLR comparing ex-combatant reintegration across the GLR. at country level occurs in DRC, where an extremely Indeed, one could question whether the particular shallow, and in some regards negative, trajectory of contexts of conflict, mobilization, and return at the social reintegration processes for ex-combatants and group and individual levels in each of the GLR coun- community members stands out. tries are fundamentally comparable at all. However, it may be that adjusting the grain of our analysis to build an overall understanding of the arc of reintegration pro- 30   Torjesen (2013) argues that the very complex nature of reintegration processes may preclude the possibility of building formal models of reintegration processes and that instead focusing on building 29   In institutional literature see for example Finn (2012), for explicit overall understanding is perhaps the most useful approach to studying treatment in scholarship see Torjesen (2013) who builds on workshop reintegration processes at this time. based discussions with the International Research Group on Reintegration (IRGR). 31   See Section 9.7 in Annex I for more detail. 20 5. Meta-Analysis of Ex-Combatants’ Reintegration Processes in the GLR. T his section of the study provides a comparative meta-level analysis of ex-combatants’ reinte- gration processes in the GRL and proceeds in six parts: (i) processes of mobilization and demobiliza- and seven years on average with armed groups, meaning the majority of them (78.4%) return to communities between the ages of 31-40. These trends are remark- ably durable across the GLR countries. Though these tion; (ii) a review of processes of economic reintegra- ex-combatants return to communities to face a range of tion; (iii) an in-depth discussion of social reintegration social, political and economic disadvantages compared processes in the GLR; (vi) female ex-combatants as a to their peers in the community, and tend to be behind vulnerable subgroup, followed by a special discussion of community members in terms of core demographic in- (v) DRC as an outlying case and how it contributes to dicators such as marriage, literacy, and education, their an overall understanding of (vi) the limits of reintegra- overall reintegration trajectory is positive. At the same tion programming. time, it is worth noting that the ex-combatants who face 5.1 From Mobilization to Demobilization even greater challenges from the outset of reintegration processes are those who were mobilized under the age in the GLR of 18. Individuals mobilize into armed groups through 5.1.1 Mobilization Under the Age of 18 numerous pathways. There are those who join for ideo- logical reasons, there are some that join for protection Indeed, one third (33.4% on average and just over 40% in the context of intense insecurity, there are others that for females) of all ex-combatants across the GLR were join out of anger or in seeking revenge, and still, there mobilized as children under the age of 18.33 While there are individuals who join armed groups to escape the is much that can be discussed about when adulthood misery of extreme poverty, in addition, there are also begins across the different cultural contexts in the GLR, people that are forcibly mobilized into armed groups. figures show that of those ex-combatants mobilized The pathways by which individuals mobilize, and the into armed groups under the age of 18, the average age individual experiences of those pathways, create lasting of mobilization was only 13 – clearly in adolescence. legacies that individuals must face as they return to com- These findings are robust across the GLR countries and munities as ex-combatants. Unfortunately, while we drive home the point that the wars in the GLR have, to know some about when ex-combatants joined armed a significant extent, been fought with children - even groups and how long they spent with them in the GLR though many were demobilized as adults.34 countries, there has been no systematic capture of their mobilization pathways and wartime experiences in the Those ex-combatants in the GLR who were mobilized TDRP-GLR database utilized for this study. In the GLR countries, the majority of ex-combatants treated as a broad indication of the age at mobilization in the GLR countries rather than the hard facts. were mobilized as adults between the ages of 18-30 (46.1%).32 These ex-combatants spend between three 33   This number would in fact be higher if child soldiers, those under 18 at the time of demobilization, were included in this study. 34   A full discussion of child mobilization in the GLR countries, however, would necessitate an examination of those ex-combatants 32   As outlined in Section 7.1 of Annex I, there are considerable mobilized under the age of 18 who also demobilized under the age of 18 challenges to be taken into account when trying to systematically capture – a category that this study does not examine due to data restrictions. See the age at mobilization. The exact numbers presented here should be Section 7.1 in Annex I for more detail. 21 under the age of 18 spend on average just under seven about the return of ex-combatants, such fears dissipated years with armed groups, meaning that the majority of quickly upon their return.37 them leave armed groups between the ages of 18-30 (64.1%). However, evidence suggests that though 5.1.2 The Legacies of Mobilization and Wartime these individuals return chronologically as adults, their Experiences starting point for reintegration processes along social, The evidence outlined above gives supports to the economic, and political lines is somewhat worse than idea that individual mobilization and wartime experi- that of ex-combatants mobilized as adults. Evidence in ences hold significant legacies for ex-combatants in this study indicates that mobilization at such a young the process of reintegration after conflict. However, we age carries considerable legacies – as through their have revealed little regarding the actual effects of mo- absence while in armed groups, these children ex-com- bilization and wartime experiences on reintegration. If batants have missed out in the processes of education, we are to take seriously the point that “… reintegration maintaining and building familial connections through does not happen in a vacuum and is not isolated from marriage, building social networks, engaging in com- munity structures, and building a basic economic track previous experiences of recruitment and involvement record. Indeed, in most indicators of these processes, in armed groups”, then we must admit that the findings ex-combatants who were mobilized under the age of here are only scratching the surface.38 There remains 18 perform considerably worse than their community much to be learned about processes that youths and member peers in the same age categories – as well as adults go through as they are socialized in the use of compared to older ex-combatants in general.35 violence and the individual legacies that these socializa- tion processes hold for the prospect of reshaping iden- Therefore, in this sense, younger ex-combatants in the tities – especially for youths who may have missed the GLR, most of whom were mobilized under the age of opportunities for learning “regular” societal norms.39 18, start the process of reintegration with a handicap, Further, we have little knowledge about how these indi- having missed opportunities for personal development vidual legacies affect the families of ex-combatants and while in combat during their formative years. These ex- in turn the communities at large.40 combatants face a double transition of relearning, or often learning, new and societal norms of adulthood while simultaneously reshaping their identity from 37   See Section 17.2 in Annex II for more detail. While there is no soldier to civilian.36 What is remarkable, however, is that analysis in this study by armed group, one could speculate that there could be considerable differences in the level of acceptance ex-combatants receive while younger ex-combatants may start with a range of from community members if disaggregated by irregular versus regular disadvantages compared to other ex-combatants, they armed forces. Even more so, the level of abusive violence that armed groups have carried out towards communities may be paramount ex-combatants are quick to build momentum in an overall positive being welcomed in communities – see Humphreys and Weinstein (2004). trajectory of reintegration towards parity with commu- 38   Özerdem & Podder (2011) p. 313. While the point that nity members – displaying an overall trajectory of rein- reintegration processes cannot be separated from the context of mobilization and wartime experiences may be an inherent truism that tegration similar to their older ex-combatant peers. In DDR programmers and policy makers understand, the implementation fact, upon returning to communities, the vast majority of this point in programming contexts has an arguably poor track record – indeed this is perhaps among the core challenges of reintegration of ex-combatants in the GLR report being welcomed programming. However, the point here is that a scientific understanding of home by family members. The broader community cor- the relationships between varieties mobilization / wartime experiences and roborates this as they report that though they had fears reintegration experiences is only burgeoning. 39   Vermeij (2011) provides a particularly good account to the dynamics of socialization of those mobilized as youths in Northern Uganda and the challenges that this creates for reintegration as these ex-combatants 35   See Section 7 in Annex I for more detail may not be returning to a previously learned set of norms and customs, but entering a largely foreign context. However, it should be stressed that these 36   This supports Özerdem & Podder’s (2011) position that: “… in are largely individual level legacies. post conflict settings youth face a dual and complex transition, while life –stages preceding adulthood are characterized by complex and challenging 40   Forthcoming research from LOGiCA will contribute to filling transitions, conflict exacerbates the transition to adulthood by breaking knowledge gaps on family dynamics in ex-combatant reintegration. down social norms and cultural practices, disrupting education systems This may be an especially important area for continued research. As an and employment opportunities and for many youth, promoting a sense of additional side note, criminology, for example, has long emphasized the identity based on the exertion of power through violence.” p. 9. importance of the family for successful ex-prisoner reintegration. 22 Appreciating the role that mobilization and wartime ex- these issues are well acknowledged, the challenge is periences have on reintegration processes goes beyond getting knowledge and practice of reintegration pro- acknowledging opportunities missed. For example, in cesses to align (praxis). a 2005 study of ex-combatants in Sierra Leone, Hum- phreys and Weinstein found that individuals’ expe- 5.2 Economic Reintegration in the GLR riences of war, particularly the extent to which they Generally speaking, while ex-combatants across the engaged in violence towards communities, was the GLR countries remain economically disadvantaged greatest determinant of reintegration success.41 Indeed, to community members in terms of core indicators individuals who return to communities from overtly like education and employment levels, they show a abusive armed groups can face considerable barriers moderately steep trajectory towards gaining economic to building trust in the community. However, perhaps parity in the future.44 Indeed, ex-combatants generally the biggest blind spot in terms of the individual legacies have a positive outlook on their future. Like commu- of mobilization and wartime experiences is that of psy- nity members, the primary economic pathway of ex- chological illness – as epidemiological research has consistently shown that mental disorders are common combatants in the GLR countries is towards small-scale in war affected populations.42 This is an issue that affects agriculture, though there is also a minority who move families, community members, and ex-combatants and into non-agricultural services or retail. Although dif- which remains all but absent in reintegration program- ferent segments of the ex-combatant population (e.g. ming. Harold Hinkel paints the situation starkly: disabled and females) across the GLR countries experi- ence unique barriers to attaining economic parity with What all psychiatric illnesses have in common community members, ex-combatants and community is the fact that they impair the sufferer in every members alike understand that the largest barrier to day functioning: reducing the capacity to reaching economic stability is a lack of opportunity – a sustain intimate relationships and friendships, distinct characteristic of the severe development chal- hindering successful participation in work, lenges present in the GLR countries. lowering scholastic achievements, limiting the ability to participate in communal life and It appears that social capital (see Section 5.4.1) plays an impairing the ability to plan and follow-up on important role in the barriers that ex-combatants across realistic goals for the future. In this way, the the GLR encounter in moving towards economic rein- mental consequences of war, terror and or- tegration.45 While ex-combatants report broad accep- ganized violence on the individual are long- tance and support from their immediate family, they are term and psychiatric illness is often chronic. still disadvantaged, as they are slow to build extended If mental health is not addressed in ex-com- kinship networks due to comparatively low marriage batant rehabilitation, the effort of improving rates, and broader social and economic networks can in social capacities and reducing poverty is clearly turn be essential to accessing and actualizing economic weakened.43 opportunities.46 With this overall weaker footing in the community, ex-combatants in the GLR are significantly The challenge of incorporating knowledge about psy- more willing to migrate for labor than their peer com- chosocial legacies and processes into reintegration munity members, who predominately cite family re- programming more broadly may revolve around how sponsibilities for an overall unwillingness to migrate. 47 to implement knowledge building strategies about the individual legacies of mobilization and wartime experi- For ex-combatants across the GLR countries, closing ence that feed back into grounded programming. While 44   See Section 9 in Annex I and 15 in Annex II for more detail. 41   Humphreys and Weinstein (2004) 45   See Section 10 in Annex I and 16 in Annex II for more detail. 42   Schulhofer-Wohl & Sambanis (2010) 46   Porto et al (2007) 43   Hinkel (2003) p.5. 47   See Section 9.1 in Annex I and 15.1 and Annex II for more detail. 23 education and skills gaps is an important part of moving marginalization and risk in the context of the GLR. In towards economic parity with community members.48 addition, when ex-combatants are sole breadwinners, However, there exist limits to closing these gaps. they are less likely to meet their monthly expenses, are Although the overall education level of all ex-combat- short on meeting their monthly expenses by a greater ants is skewed lower than that of community members, margin, and are more likely to resort to borrowing to only younger ex-combatants are significantly active meet their monthly expenses. Ex-combatants have half in closing this gap by pursuing continued education. the rate of application to micro-credit, but a similar However, benefits that young ex-combatants may enjoy rate of successful application to community members. from education exist outside the time bounds of this All these factors contribute to an overall higher level of study, and the role that education plays in the reintegra- income insecurity for ex-combatants across the GLR tion of ex-combatants in the GLR countries is somehow countries than that of community members.51 dilemmatic. On one hand, even low level of literacy and educational achievement may be especially helpful for Regardless of ex-combatants’ overall positive economic ex-combatants in pursuing entrepreneurship endeavors trajectory towards parity with community members and outside formal employment. On the other hand, in the their corresponding positive outlook on their economic context of the GLR countries, the importance of educa- prospects in the future, ex-combatants across the GLR tional achievement for broader employment prospects countries still perceive themselves as an economically may only truly come into effect at the completion of disadvantaged group in society.52 It is likely that this secondary school – when the doors to civil sector em- perception carries considerable weight for the overall ployment open.49 In regards to skills, while ex-combat- process of reintegration that ex-combatants in the GLR ants receive vocational trainings more frequently than must navigate. Essentially, the reality of economic rein- community members, they are less likely to use the vo- tegration in the GLR may reflect the oft-cited quip of cational skills. Ex-combatants primarily claim that this “reintegration back into basic poverty.”53 Indeed, the is due to lack of tools or workplace, but it is also likely overall economic setting that ex-combatants and com- this is related to their lack of economic networks in the munity members alike face in the GLR countries is community through which they could leverage their severely challenging. No reintegration program alone skillsets, or there may be a lack of demand for such skill- will be able to solve large-scale economic problems; sets in the community in the first place.50 rather, it must dovetail with broader stabilization pro- gramming.54 One way to think about this is that reinte- Ex-combatants across the GLR countries have weak gration programming can affect the supply side of labor social capital relative to community members (see through training, micro economic activities, and micro- Section 5.3). This weak social capital results in numerous credit, but these measures do little to affect the over- economic consequences, including a heavy reliance on arching demand side of labor in the GLR economies.55 the immediate family (when the family is willing) for The inherent danger of this prospect is that reintegrat- economic support – although this is also the case to a ing back into the same economic setting that may have slightly lesser extent for community members. With contributed to their mobilization into armed groups in this heavy reliance on family, ex-combatants are to some the first place (even if now with a new set of skills and extent less likely than community members to be sole knowledge) can potentially shape feelings of helpless- breadwinners for their households, an economic status that is largely synonymous with exposure to economic 51   See Sections 9.5 in Annex I and 15.4 in Annex II for more detail. 52   See Section 9.1 in Annex I for more detail. 48   See Section 7.3 in Annex I and 13.2 in Annex II for more detail. 53   McMullin (2004)(2013) 49   Unvin (2007). This is of course presuming that there is a civil sector present. Many of the areas across the GLR are devoid of meaningful 54   Shibuya (2012) connection to civil sector employment. Thus, this point may be most relevant in more urban areas of the GLR. 55   There are considerable efforts being made in the GLR counties, and elsewhere, to fine tune the supply and demand of labor in reintegration 50   Anecdotal evidence suggests that when ex-combatants receive programming through economic opportunities mapping that are matched vocational training that includes tool packages they often sell these tool to ex-combatant profiles in interfaces like the TDRP’s ICRS (Internal packages for immediate economic support. Counseling and Referral System) and UNDP’s DREAM database. 24 ness and marginalization among ex-combatants, which, ing more slowly. Ex-combatants have overall smaller in turn, possibly feeds into remobilization or even radi- social networks than community members and tend calization.56 to be more reliant on their family, if they have one, for social support compared to community members.58 On average ex-combatants in the GLR are quick to Due to their smaller social support networks, when move in a positive trajectory towards economic parity ex-combatants are not accepted into families, they fair with community members; though they experience considerably worse than community members who are some unique challenges and limitations related to social outside of family structures – especially when it comes capital. In the context of reintegration programming’s to income security. Although a comparatively high limited ability to shape the larger economic situation ratio of ex-combatants report being accepted by their in FCS and development setting, a core goal of the family networks upon returning to communities, they economic support that ex-combatants (and commu- still have less familial contact overall and marry less fre- nity members) receive should revolve around shaping quently compared to community members. clear expectations about the economic outcomes of re- integration programming.57 Shaping expectations can Marriage is a core component to the process of social play an important role on ex-combatant’s perception of reintegration in the GLR. Indeed, marriage is a central their economic situation. In this context it may be that pathway through which ex-combatants extend their ex-combatants and community members’ perceptions, familial networks, signal a shift in identity to the rather than the absolute economic outcomes alone, community, and hopefully build broader social and are important parts to consider in order to understand economic networks, which will in turn open the possi- economic reintegration processes in the GLR. bility of leveraging the value of these networks towards economic and social outcomes. For these reasons, the 5.3 Social Reintegration in the GLR fact that across the GLR countries ex-combatants marry significantly less frequently than community members Social reintegration is the processes by which ex-com- is troubling. Indeed, ex-combatants and community batants reshape their identity from soldier to civilian members alike communicate their hesitancy around and weave themselves into the social fabric of society, the idea of marrying an ex-combatant with stigma as both in their own eyes and the eyes of the community. a central issue.59 Female ex-combatants are the only This is no simple task, and involves multiple processes segment of ex-combatants or community members by which ex-combatants reconnect to, or build, family across the GLR countries that marry ex-combatants, and social networks. These steps in turn contribute despite the fact that they are dramatically less likely to to building trust in the community and cultivating a be married in the first place (see Section 5.4).60 broader sense of inclusion that shapes personal empow- erment and an overall ability to contribute to society. The dynamics of ex-combatant marriage as a com- This is a complex and slow moving set of processes that ponent of social reintegration presents a quandary. the experiences of ex-combatants in the GLR countries Although marriage is considered as a central pathway for speak to clearly. ex-combatants to reshape their identities, erode stigma, and connect into society, pathways into marriage for ex- As discussed, according to economic and other core combatants appear to be blocked until they can begin to indicators, ex-combatants across the GLR see a steep trajectory of improvement over time, and they are moving quickly towards parity with community members in general. In the realm of social reintegra- 58   The family can take vastly different forms across the GLR. Complex tion processes, however, ex-combatants are progress- family structures and marriage dynamics (e.g. polygamy) mean that the specific social dynamics that ex-combatants face in reintegration will vary considerably. Never the less, the family is a key stepping stone in the process of social reintegration through the expansion of social networks. 59   See Section 7.2 in Annex I and 13.1 in Annex II for more details. 56   On radicalization see Özerdem & Podder’s (2011b). Ex-combatants and community members in Uganda overall are the exception to this this trend in that they 57   This is a point consistently articulated in both scholarly and stakeholder literature on reintegration programming 60   For more detail see Section 7.2 in Annex I and 13.2 in Annex II. 25 do the very things that marriage is the central pathway of severe development challenges in the GLR countries. to achieving (reshaping identities, eroding stigma, and However, among ex-combatants, these slightly stronger connecting into society). Addressing the barriers that senses of empowerment do not manage to translate into ex-combatants face in accessing marriage is a slow higher levels of happiness or an improved perception of process of confrontation and atonement that may occur their standing in communities. Indeed, ex-combatants largely outside the direct influence of reintegration pro- are more likely to report that they are unhappy and un- gramming - but that can be supported by it.61 satisfied with their lives in general, and that they have a negative impact on the community.65 Moreover, ex- Trust does not appear to be a long-term problem for combatants perceive themselves as considerably worse social reintegration of ex-combatants in the GLR coun- off than community members in wide range of catego- tries.62 Even though community members consistently ries, and community members also perceive ex-com- report high levels of fear regarding the return of ex-com- batants as disadvantaged. Despite these disadvantages, batants before their arrival, few report having such ex-combatants, together with community members, fears after ex-combatants’ return and more often than have a positive outlook on their future with a clear un- not community members describe the positive con- derstanding that social change occurs over a long period tributions that ex-combatants make to communities. of time – on the scale of years. This is important because, Indeed, both community members and ex-combatants like economic reintegration, perceptions are key. report high levels of overall trust in the community (and improvement in trust over time) that contrib- Generally speaking, communities in the GLR countries utes to an overall sense of togetherness. These are very have provided a positive social context into which ex- positive findings indeed. In addition, ex-combatants combatants can begin to reintegrate (with the exception across the GLR do not appear to be forming exclusive of DRC, see Section 5.6) – though there are significant social groups. Only a minority of ex-combatants identi- barriers to their social reintegration in the long run, pri- fies most of their friends as other ex-combatants, being marily those related to marriage and building broader twice as likely to identify their friends as being at the social and economic networks in the community same age (approximate) or gender.63 Interestingly, com- that can in turn lead to positive social and economic munity members across the GLR predominantly claim outcomes. Indeed, evidence from the GLR countries that they have no ex-combatant friends.64 These two gives credence to the idea that while ex-combatants can sides of social reintegration appear somewhat at odds make quick gains in education, skills, access to land, etc., with each other and would benefit from further inquiry these gains cannot be leveraged to actualize improve- in future studies. ments in income and food security without building social capital, which is one of the primary components Ex-combatants across the GLR countries report similar, of social reintegration.66 if not higher, senses of empowerment to affect change in their lives than community members do. This likely 5.3.1 The Logic of Social Capital in Reintegration links back to the point that ex-combatants’ wartime ex- periences can carry considerable legacies for reintegra- Social reintegration is perhaps the pivotal step in tion – in that participation in armed groups can offer linking individual level reintegration processes to the a sense of personal control and empowerment absent larger prospect of building peace and reconciliation, from the lives of many individuals living in the context while also leveraging these towards societal reconstruc- tion and development. Many scholars and practitioners alike acknowledge the centrality of social reintegration 61   Speaking the point of part of social reintegration being a processes, yet the logic of social reintegration is rarely confrontation, Porto et al (2007) draw from Lederach (1997) to assert: “the notion of reconciliation as an encounter where space for acknowledging the past and envision the future are the necessary ingredients for the reframing of the present.” p. 152. 65   Though, it is likely that ex-combatants’ overall weaker economic 62   See Section 10.2 in Annex I and 16.2 in Annex II for more detail. situation may contribute to these indicators. 63   See Section 10.1 in Annex I for more detail. 66   Put in other words, it appears that social reintegration is an intervening variable to successful economic and political reintegration 64   See Section 16.1 in Annex II for more detail. processes. 26 handled explicitly.67 Indeed, understanding social re- transcend specific kinship and ethno-religious networks integration is deeply challenging because it involves and connect individuals to the broader community. Ex- delving into an amalgamation of complex social ideas combatants in the GLR countries vary in the extent to deeply rooted in social science, with the concepts of which they are able to build bridging social capital. In social capital and social cohesions serving as the nuclei Rwanda and Burundi, where regular communal labor is among them. Exploring the logic of social capital in institutionalized and serves as a platform for address- reintegration will shed much light on the social reinte- ing community issues, ex-combatants and community gration processes and the interconnectedness between members alike appear to have higher levels of bridging social and economic reintegration processes that ex- social capital. In contrast, ex-combatants and commu- combatants in the GLR countries navigate. Indeed, it nity members in DRC, RoC and Uganda appear to have is important to note that while social reintegration can considerably less bridging capital.69 Bridging social be key to accessing economic opportunities and pro- capital can have very concrete impacts on ex-combat- gressing towards economic reintegration, economic ant’s access to economic opportunities – as economic prosperity itself can grant a degree of social capital. The opportunities across the GLR are largely dependent on social and economic spheres overlap and dynamically access to informal networks, including those beyond feed into each other. immediate family networks.70 In this means, social reintegration can have very real meaning as far as ex- The core of social reintegration revolves around two combatants access to labor markets. Moreover, bridging concepts: social capital and social cohesion. Social social capital is the core of building a sense of overall capital is the idea that social networks have both tangible togetherness and cohesion in the community. and intangible values. Social cohesion is the idea that individual social networks come together to weave the If bonding social capital is about connections between cohesive social fabric of society. Behind the idea that people, and bridging social capital is about connections social networks contribute to social capital and in turn, between communities, then linking social capital is social cohesion is the distinction between three types about the connections between communities and higher of social capital: bonding capital, bridging capital, and level social units – most notably the state (though there linking capital.68 are many strata in between). This is the upward logic of social cohesion that connects individual level networks Bonding social capital is that between familial networks to those between communities and the country as a as well as religious or ethnic groups. Even though they whole.71 There has been very little data collected in the face limits in expanding this capital through marriage, GLR countries in this study that contribute to an un- it seems that ex-combatants in the GLR countries are derstanding of linking social capital. One can speculate quick to build bonding social capital, as implied by that the large scale institutionalization of communal the quick acceptance of ex-combatants from existing labor and meeting forums in Rwanda and Burundi con- familial networks and ex-combatants’ heavy reliance on tribute to the overall functioning of the state – though family. supporting this speculation empirically is outside the Bridging social capital is about building crosscutting ties bounds of this study.72 Indeed, building linking social though networks that are not defined by inherent char- acteristics of ex-combatants and community members 69  In Rwanda this institutionalized national practice of communal such as ethnicity or religion. Bridging social capital can labor is called Umuganda. In Burundi a similar national practice is called be built, for example, through membership in economic Travaux Communataires. associations, participation in community projects, or 70   Lamb (2011), Porto et al (2007). civic engagement in community issues. These arenas 71   See Woolcock (2000) or Colletta and Cullen (2002) for especially good handlings of social capital and social cohesion in the contexts of post- conflict development and peacebuilding. 67   Jennings (2008), Bowd & Özerdem (2013) 72   Barnhart (2011) argues that Umuganda (institutionalized communal labor and community meetings) in Rwanda has been ineffective 68   Bowd & Özerdem (2013). The idea of macro, meso, and micro in the project of nationbuilding, the project consolidating and overarching indicators of social reintegration processes is largely analogous to the national identity, though she implicitly discusses its contributions to state distinction between bonding, bridging, and linking social capital. functions. 27 capital is perhaps the most long-term component of the likely demographic group, from ex-combatants and process of social reintegration, and is a project that is by community members alike, to be married, and the most no means exclusive to ex-combatants – occurring well likely to be divorced, separated, or widowed. Indeed, beyond the mandates of reintegration policy and pro- while male ex-combatants in the GLR see a positive gramming. trajectory of improved marriage rates over time, female ex-combatants’ marriage rates are near stagnant. Male 5.4 Female Ex-Combatants in the GLR ex-combatants and community members alike commu- nicate their general unwillingness to consider marriage This study finds that across the GLR countries, female with female ex-combatants due primarily to stigma- ex-combatants, almost without exception, stand out as related issues – lending explanation to female ex-com- the most extensively and consistently disadvantaged batants’ marginal improvements in accessing marriage segment (including all age and disability segments, male over time.75 This of course leaves unproblematized the or female) of ex-combatants according to core indica- presumption that female ex-combatants are interested tors that reflect economic standing and social position in attaining marriage and the likely return to traditional in the community, and are at clear risk for economic gender roles connected to it – a point that would benefit and social isolation and marginalization. from further analysis in future studies. Across the GLR countries, female ex-combatants are This social disadvantage in terms of marriage for female more likely to be mobilized under the age of 18, though ex-combatants dynamically interacts with economic they spend notably less time with armed groups in factors, especially access to arable land. Movement general compared to their male peers (4.95 years on into small-scale agriculture is the dominant economic average versus 7.37 years on average for males). While trajectory for all ex-combatants across the GLR (see the legacies of mobilization and wartime experiences are Section 5.3). As such, access to arable land is an im- important for understanding the pathways to reintegra- portant indicator of economic stability, and growth in tion among male ex-combatants, the challenges these access to arable land is in turn one indicator of a positive legacies present are acute for female ex-combatants economic trajectory. For male ex-combatants, marriage in the GLR. The widespread experience of socializa- and inheritance are the primary pathways to expand- tion in violence and hyper-exposure to sexual violence ing land access. However, female ex-combatants, as that female ex-combatants may experience as a part of the demographic group with the lowest marriage rate conflict can contribute to stigma-based barriers to rein- across the GLR countries, do not have this pathway to tegration upon their return. Indeed, in some contexts increased land access, and generally do not inherent in the GLR, this stigma is so strong that some female land due to traditional gender structures and in some ex-combatants avoid self-identifying as ex-combatants cases stigma. These factors contribute to an overall and therefore, forfeit their access to targeted assistance shallow reintegration trajectory for female ex-combat- for ex-combatants.73 ants in the GLR. This dynamic interaction of gender, Perhaps the most measurable disadvantage that female stigma, marriage and land access in the GLR countries ex-combatants face in the GLR is their lower level of is among the core structural barriers to reintegration access to marriage.74 While female ex-combatants in processes that female ex-combatants must face, which the GLR countries generally report acceptance into can also run the risk of becoming intergenerational dis- existing kinship networks upon return to communities, advantages for the children of female ex-combatants.76 they face distinct barriers in expanding these networks Further, female ex-combatants across the GLR have through marriage. Female ex-combatants are the least 75   Though, it is likely that economic considerations for attitudes 73   While avoiding self-identification is an approach that many ex- towards marrying female ex-combatants play a role here too – as evidence combatants, regardless of gender, take – the point here is to emphasize that in this study shows that female ex-combatants are disconnected from this may be a particularly prevalent approach for females ex-combatants in inheritance structures and thus carry little economic value through some contexts across the GLR. marriage. 74   See Section 7.4.1 in Annex I for more detail. 76   See Section 12.3 in Annex I for more detail. 28 lower literacy and education levels than male ex-com- this is only half of the story. To truly appreciate the batants as well as male and female community members context of the gender dimensions of reintegration in – a factor which they identify as a barrier to gaining the GLR, we have to include a comparison of female employment. Not only are female ex-combatants more ex-combatants to female community members in our likely to be unemployed, they are also much less likely analysis – rooting it in the wider context of gender to be a sole household breadwinner, and thus, are less dynamics in the GLR countries. exposed to the economic risk that this status represents in the GLR countries (see Section 5.3). It seems that 5.4.1 A Broader Societal Transformation of Gender female ex-combatants are less commonly sole bread- Dynamics winners because of their heavy reliance on immediate Across the GLR countries, female community members family networks. Indeed, while female ex-combatants have a very similar range of disadvantages compared to report general acceptance from their immediate family male community members as female ex-combatants do networks, they have few other social or economic to male ex-combatants. In addition, the space between networks and their overall social capital in the commu- female ex-combatants and female community members nity is weak.77 across a broad range of indicators of social and economic Female ex-combatants report lower levels of trust in the processes is often little – though female ex-combatants community and lower perceptions of improvement in are almost always more disadvantaged compared to trust over time than male ex-combatants. With a lack female community members. Perhaps the core advan- of social capital and a lack of trust in the community, tage that female community members hold is their female ex-combatants report weaker overall feelings of higher rates of marriage, better prospects for marriage empowerment and control in their lives compared to in the future, and an overall better integration into male ex-combatants. Further, these lower levels of trust extended social networks that contributes to stronger can have economic consequences, for example low trust social capital in the community.78 In other dimensions, can make it hard for female ex-combatants to attract such as gender discrimination in the workplace, female customers when opening small businesses. Female ex- community members and ex-combatants have similar combatants across the GLR countries understand the experiences of gender-based disadvantages – though range of disadvantages that they face, as implied by the it appears that ex-combatant status and the stigma that fact that they consistently report perceptions of their it carried can have an amplifying effect on the deeply overall weaker situation compared to others in society. ingrained gender inequalities that exist across the GLR countries. Despite the expansive range of disadvantages that female ex-combatants face across nearly all indicators of social There are of course subtleties to this overall picture and economic reintegration processes, it is interesting to of female disadvantage in the GLR countries. For find that they in fact consistently report a stronger sense example, female ex-combatants express senses of em- of overall happiness, a stronger sense of overall life satis- powerment to control their lives and everyday activities faction, and a better outlook on the future compared to at levels skewed slightly higher than female commu- male ex-combatants. In future studies, building an un- nity members. This finding supports other research derstanding of female ex-combatants’ coping strategies that suggests that in some circumstances, conflict can for arriving at these stronger perceptions of themselves actually create opportunities for female empowerment, could prove illuminating for the prospect of under- while at the same time as it shapes other negative gender- standing ex-combatants’ perceptions of themselves and their prospects going forward as a whole. In almost every aspect, female ex-combatants are disad- 78   Again, this leaves the assumption that marriage is in fact something that female ex-combatants in the GLR are interested in attaining. Also, vantaged compared to male ex-combatants. However, it is possible that female ex-combatants may have other disadvantages to female community members in that because many of them were mobilized at a young age (see Section 5.1.1) they may have missed the opportunity to learn the cultural customs of their community – for example, in the different expectations for men and women in terms of community respect 77   See Section 10.6.1 in Annex I for more detail. and etc. 29 and stigma- based challenges.79 Another subtlety is that of those very local gender dynamics for achieving the female ex-combatants and community members in the transformation of gender norms more broadly. GLR countries are more likely to be involved in micro- economic activities than males. This also aligns with 5.5 DRC: Reintegration in the Context of previous research in the region that suggests that the Ongoing Conflict large-scale mobilization of men into conflict, and the overall displacement dynamics in some instances in Since the end of the Second Congo War in 2003, DRC the GLR, have created an opportunity vacuum which has been on a path of peacebuilding and slow recovery females have been able to exploit through the attain- from the social and economic legacies of colonialism ment of microeconomic activity programming, for and the widespread corruption of the 1970’s and 1980’s. which they might have otherwise had to compete with Indeed, in 2006 and 2011, democratic elections were males to a greater extent.80 held peacefully in DRC, and security in Kinshasa has improved greatly. However, in Eastern DRC, insecurity The fact remains that in the GLR countries, female ex- and violence have persisted – shaped by a range of local combatants and community members share a broad and international actors.82 Since 2004, the Government range of gender-based disadvantages.81 These disadvan- of DRC has worked to demobilize and reintegrate ex- tages are structurally ingrained and culturally produced. combatants involved in the local conflicts in the eastern In this sense the landscape that female ex-combatants in parts of the country, especially the Kivu provinces, and the GLR countries face as they navigate is fraught with has made considerable programming gains in this effort. the challenges of overarching gender disadvantages. If However, due to the context of continued and quickly female ex-combatants are to gain parity with female shifting violence in Eastern DRC, many of the ex-com- community members, the amplifying effects of stigma batants that go through reintegration programmes begin will no doubt serve as a barrier to entering the com- the process of reintegration in the context of ongoing munity. However, if female ex-combatants are rather to insecurity and severe development challenges at a level reach parity with male community members, a much generally beyond most other areas in the GLR. deeper set of social-structural barriers stand in their way – barriers that they and their female community Analysis in this study reveals that while ex-combatants member counterparts face simultaneously. in DRC do well, perhaps even the best across the GLR countries, in moving towards parity with community This evidence suggests that while female ex-combatants members in terms of core indicators, the process of in the GLR do experience unique dynamics of reinte- reintegration as it is visible across the rest of the GLR gration and special challenges in accessing marriage, countries is not occurring. In almost all dimensions of addressing gender issues in reintegration programming reintegration, ex-combatants and community members should be contextualized as an opportunity for contrib- in DRC are significantly worse off, despite the smaller uting to a community-wide transformation of gender space between them, than those in the rest of the GLR dynamics in the GLR countries. This is no easy task, as countries.83 Essentially, it appears that because com- reintegration must simultaneously take on a complex munities in Eastern DRC are considerably worse off balance of recognizing traditional gender norms and re- than most others across the rest of the GLR, the bar is flecting them in programming, at the same time it must set low as ex-combatants move towards parity. In this use reintegration programming as an opportunity in the context, parity between ex-combatants and community transitional phase to address the negative implications members may not be indicative of the range of social and economic reintegration processes associated with it elsewhere in the GLR. 79   See Specht (2013). 80   See International Alert (2010). A possible challenge for this 82   Indeed, the complex constellation of violent actors in Eastern dynamic is that when male ex-combatants return to communities newly DRC is a challenge to keep track of, let alone understand. Autesserre empowered female community members may lose this advantage – (2010) provides a particularly compelling narrative the local and possibly creating tension on the community. international dimensions to violent conflict in DRC. 81   See Section 18.3 in Annex II for more detail. 83   See Section 16.6.2 in Annex II for more detail. 30 Ex-combatants in DRC have on average spent the them. Ex-combatants in DRC are the most susceptible longest amount of time with armed groups before de- group among the GLR countries to 1) have no one to mobilization and are more likely to have been mobi- turn to for economic help, 2) have the weakest feeling of lized under the age of 18 compared to those in the other togetherness with the community, 3) feel they have the GLR countries. Despite this, ex-combatants in DRC do least amount of power to make important decisions in not appear to have significant disadvantages to commu- their life, 4) perceive the weakest ability to control their nity members in terms of education, literacy, or access everyday activities; in addition, they are 5) the least to land. When it comes to economic reintegration, likely to perceive that they make a positive impact on ex-combatants in DRC are among the worst off in the the community, 6) the least likely to gather to express GLR in terms of employment, though their overall tra- political concerns, 7) the least likely to feel their voice jectory towards small-scale agriculture fits with that of being taken into account by leaders, 8) the most likely the rest of the GLR countries. However, ex-combatant to think their overall situation will deteriorate in the and community members in DRC are the most likely to future and therefore, they 9) have the lowest level of life be a sole breadwinner for their household and are more satisfaction across the GLR countries. What’s worse, in exposed to the income insecurity associated with this terms of social change, DRC is the only GLR country status in the GLR (see Section 5.3). where ex-combatants see drops in their perception of their overall situation relative to the rest of society in Indeed, in the context of continuing insecurity in the the year prior to sampling in the categories of food, Eastern DRC, ex-combatants and community members clothing, and finance. alike are exposed to a level of economic hardship that is considerably worse than that of the rest of the GLR 5.5.1 The Absence of Social Cohesion in DRC: A countries.84 These are the challenges ex-combatants and Splintered Society community members face in DRC. However, the crux of understanding the weight of both ex-combatants’ and In the context of ongoing conflict and insecurity in community members’ disadvantages in the context of Eastern DRC, ex-combatants and community members ongoing conflict is social reintegration. Ex-combatants alike face a baseline in social and economic processes in DRC have the weakest familial networks in the GLR far below their neighbors in the rest of the GLR coun- in terms of level and frequency of familial contact. This tries. While it appears that ex-combatants in the rest of is likely an effect of both the challenging social geogra- the GLR countries are making quick gains in core indi- phy of eastern DRC – severely dilapidated transport in- cators and moving towards parity with the community, frastructure and mountainous terrain make travel near they are still disadvantaged across most dimensions impossible when combined with heavy seasonal rains – in general. What ex-combatants in the GLR countries and the large-scale nature of displacement in the context often lack is the social capital relative to communities of continuing violence and insecurity. In this sense, ex- and the networks and connections in society to trans- combatants in DRC are missing the basic social footing late their gains into social and economic outcomes. that many ex-combatants in the other GLR countries In DRC, however, it appears that not only do ex-com- are afforded – a family network to return to (at least at batants but also community members lack the social the level it is visible in the other GLR countries). capital necessary to leverage outcomes. In effect, this study finds that in DRC there is little in the way of social Though ex-combatants in DRC build new family capital and social cohesion for ex-combatants or com- networks through marriage on a similar level to com- munity members to leverage compared to other GLR munity members, the absence of pre-existing family countries. In a sense, DRC is a splintered society, where networks appears to have serious consequences for there is weaker social fabric for ex-combatants to rein- tegrate into. 84   It is worth nothing the persistent conflict and insecurity in Eastern It appears that the upward logic of bonding, bridging, DRC does have spill-over effects in border regions neighboring countries. In Uganda, for example, continued conflict in Eastern DRC has contributed and linking social capital (see Section 5.4.1) is not the mobilization of ADF forces. visible in the context of continued insecurity preva- 31 lent in Eastern DRC. Ex-combatants and community to be addressed by innovative new approaches. members remain extremely marginalized, living on the fringes of society. Indeed, numerous scholars have char- 5.5.2 The Paradox of DRC: Parity versus Process acterized the social, political, and economic dimensions DRC reveals a paradox of understanding the successes of Eastern DRC as formed by local dynamics that are and failures in reintegration policy and programming. essentially disconnected from DRC as a state.85 Large- This paradox lies in the potential incongruity of parity scale displacement and the social geography of Eastern between ex-combatants and community members as DRC keep kinship networks dispersed, and they thus an indication of reintegration processes. Reintegration do not appear to be serving the role of bonding social programming aims to aid ex-combatants in return- capital. As such, there is a limited platform on which ing to communities and moving towards parity with ex-combatants and community members together can community members in terms of social and economic expand out from their family networks. With the lack characteristics, essentially to remove ex-combatants as of a broad societal platform in DRC, the prospects of a vulnerable group in society so that larger peacebuild- building bridging social capital between cohesive social units and linking social capital to larger societal units ing and development initiatives can take over. Parity seem dire. between ex-combatants and community members is the overarching mandate and among the key dimen- In the context of ongoing conflict in Eastern DRC, it sions along which impact evaluation in the GLR has appears that the extremely weak social fabric and splin- been made. In this regard, DRC is the country in the tered characteristics of society are a serious impedi- GLR with the greatest level of parity between ex-com- ment to the process of ex-combatants’ reintegration. batants and community members. Though ex-combat- These findings suggest that ex-combatants’ minimum ants face a range of disadvantages, the extent of these level of social capital and cohesion in the community disadvantages to the wider community is relatively little may be a necessary condition for ex-combatants’ rein- compared to other GLR countries. tegration processes to occur. Continued insecurity and violence in Eastern DRC make the establishment of However, as outlined above, ex-combatants and com- this basic necessary condition extremely problematic. munity members in DRC together are the worse off In Eastern DRC where both ex-combatants and com- than those in all the other GLR countries. It is to some munity members face a similar challenge of rebuilding, extent by comparing DRC to the wider context of the rather than merely connecting back into, the founda- GLR with a distinct focus on processes that we are tional fabric of society, community-based ex-combat- able to see that though the highest level of parity has ants’ reintegration approaches that focus on benefiting been achieved in DRC, this parity is a limited indica- a broader segment of society as well may be the most tor for the broader processes of reintegration occurring. beneficial.86 However, it is possible that such large- Relative to communities in the other GLR countries, scale social transformations of post-conflict, or in this communities in Eastern DRC are considerably weaker case ongoing conflict, societies may be fundamentally in terms of economic security and social support. Thus, outside the bounds within which traditional reintegra- the scope improvements that ex-combatants must make tion programming can affect and would perhaps need to gain parity with community members are relatively little compared to those that ex-combatants in the other GLR countries must make. Ex-combatants in DRC do make quick gains, but evidence in this study suggests 85   Autesserre (2009)(2010), Marriage (2007), Reno (1999) that this relative parity to community members is not 86   This evidence supports Shibuya’s (2012) assertion that: “As the distinction between “combatant” and “civilian” becomes less meaningful indicative of the broader and slow moving set of social within the community, financial and other assistance should be disbursed and economic reintegration processes observable in towards the larger community rather than to individual combatants. In post-conflict situations where this distance is not great in the first place, the other GLR countries. It should be noted that from reintegration assistance that targets individual combatants can cause a programming impact perspective (based on parity), resentment, and even in situations where special incentives for combatants DRC has the most successful country program in the is a good idea, over time doing so will perpetuate the distinction that the assistance is meant to eliminate.” p. 134-135. GLR; while in terms of affecting reintegration process- 32 es as appear in the other GLR countries, DRC appears this navigation through support grounded in local rein- to be the least successful case in the GLR.87 tegration processes.89 5.6 The Limits of Reintegration Learning about the shape and trajectories of reinte- Programming gration processes can contribute much to the effort of building metrics for success in reintegration program- The process of reintegration may in part take place inde- ming – though clear successes and failures may always pendently from reintegration policy and programming remain elusive.90 Indeed, what establishing metrics may - though it is simultaneously encapsulated within them. mean in this context is building an understanding of With or without assistance from reintegration programs, the limits of reintegration policy and programming, ex-combatants in the GLR have returned to communi- defining its scope of potential impact. No reintegration ties and navigated the complex set of transitions that program will ever steer 100% of ex-combatants into a reintegration processes entail. Economic dimensions positive reintegration process outcome, but how much to reintegration are important; however, it may be that breathing space do reintegration practitioners have in the economic gains that ex-combatants make cannot be affecting reintegration processes in a given context?91 fully actualized without engaging in the set of slower In brief, what DRC shows us is that without a necessary moving social reintegration processes. In this sense, re- minimum level of social cohesion in communities and integration as a process is primarily an individual social society more broadly, the likelihood of ex-combatants process of reshaping identity and reconnecting into the meaningfully navigating social reintegration processes social fabric of society. appears small, and reintegration programming may have With an independent focus on reintegration processes a limited scope of potential impact on the social pro- that is distinct from specific reintegration programming cesses. Even the best designed reintegration program in mind, it may be that reintegration policy and pro- in DRC will not alleviate the poor security situation gramming have a limited ability to directly shape some or lack of opportunities to make a living – though it can elements of reintegration processes – most notably contribute to these ends.92 Reintegration programming the individual legacies and social dynamics of return. focused on combining both ex-combatants and com- Ex-combatants, their families and communities are munities may prove the most fruitful - in a sense helping the primary agents by which reintegration processes foster a larger societal transformation that appears to fa- occur, and they have the most direct influence on the cilitate ex-combatants’ reintegration processes. trajectories of ex-combatants’ reintegration in the given context. Considering these factors, it may be that better reintegration programming requires learning how to 89   Özerdem & Podder (2011) best facilitate the endogenous processes of reintegra- 90   Özerdem & Podder (2011) have gone as far as to posit that “… tion, in other words, to capitalize on the groundswell of reintegration remains essentially a processes outside of numeric outputs.” p. 313. Indeed, there are many scholars that might agree. In a well-known already occurring processes.88 Reintegration processes study Humphreys and Weinstein (2007) found that there was no evidence take place in contexts of constant flux – flux that ex- that exposure to DDR programming had any measurable effect at the individual level in Sierra Leone. However, many scholars (e.g. Stankovic combatants, their families and communities navigate and Torjesen 2010) have pointed out that in the post-conflict peacebuilding together, and reintegration programming help can guide setting numerous projects and aid occur simultaneously. From this view the prospect of isolating the effects of reintegration programming on ex-combatants and communities suffers from a fundamental attribution problem. 87   On way to think about this difference may be to distinguish 91   The idea of breathing space is also presented in some literature as between quantitative parity versus qualitative parity. “acceptable levels of failure.” See, for example, Shibuya (2012). 88   Another way to say this is that specific reintegration programming 92   This finding supports broader arguments of Vries & Wiegink needs to be deeply contextualized to fit local political economies. (2011). 33 6. Conclusions T his study has drawn from a survey of nearly 10,000 ex-combatants and community members across Rwanda, Uganda, Burundi, DRC, and RoC to formulate a broad understanding of And ex-combatants display a decreasing level of un- employment over time as they move, primarily, into small-scale agricultural activities. However, it should also be noted that ex-combatants across the GLR coun- the range of reintegration processes that occur in the tries have weaker economic networks than community GLR. The focus of this study has been the individual- members, and tend to be more singularly dependent scaled processes of social and economic reintegration on their immediate kinship networks for economic in this region, rather than the range of reintegration pro- support. The limited pool of support that ex-combat- gramming and policy carried out to aid this process in ants can draw from leaves them at a generally higher each GLR country. Indeed, a core finding of this study level of exposure to income insecurity than community is that despite the disparate post-conflict contexts across members. However, ex-combatants and community the GLR countries, ex-combatants display a broadly members alike understand that it is the context of over- analogous range of reintegration processes that collec- arching development challenges, instead of ex-combat- tively form a similar overarching trajectory, or narrative, ant status itself, that creates the primary restriction to of reintegration. their ability to achieve economic stability. Despite this understanding, ex-combatants still perceive themselves In the context of the GLR countries, it is evident that as an economically disadvantaged group in society. This mobilization and wartime experiences shape the perception is key, as in the end, ex-combatants’ percep- baseline at which ex-combatants begin the reintegra- tions of their economic standing in society that may tion processes. A direct example of this is that of young matter as much, if not more, as their actual trajectory. ex-combatants, of whom the majority were mobilized in early adolescence. The opportunities missed during While the economic dimensions of reintegration pro- adolescent and young adult years and, maybe more cesses are important, this study finds that a slower set of importantly, the experience of socialization in the social reintegration processes may be at the core of what context of extreme violence carry lasting legacies for it means to reintegrate in the GLR. In the GLR coun- ex-combatants as they demobilize and begin their pro- tries, accessing marriage, expanding kinship networks, cesses of reintegration. While young ex-combatants are and in turn strengthening ties with the broader commu- disadvantaged compared to older ex-combatants and nity are the core processes of building social capital and community members across a broad range of core in- social cohesion for individuals as well as for society as a dicators, the psychosocial challenges that they face may whole. Stigma weighs heavily, as ex-combatants return be acute. This study is essentially blind to the scope of to communities and face barriers to accessing marriage. psychological trauma that ex-combatants in the GLR Without sufficient access to marriage, many ex-com- have endured and the legacies from wartime experi- batants across the GLR lack the entry into the larger ences as they interact with their families and commu- process of expanding kinship and communal networks nities. This trauma may be among the cruxes of many as part of building social capital in an overall contribu- ex-combatants’ reintegration into communities. tion to social cohesion in society. From this evolves a paradox: ex-combatants have to build social capital In absolute terms, although ex-combatants across and erode stigma through marriage, yet they must the GLR are economically disadvantaged to commu- have social capital and erode stigma to access marriage. nity members along economic lines, they do display Across the GLR countries, social reintegration is a slow a positive trajectory towards gaining economic parity. moving process of atonement for the past, readjustment 34 to community norms, and reconciliation for the future. all social and economic indicators in this study, it is clear that ex-combatants and community members in DRC Across the GLR countries, female ex-combatants are are the worst off compared to those in the other GLR unequivocally the single most disadvantaged group and countries. In the context of continued violence and in- face a set of deeply ingrained structural barriers to re- security in Eastern DRC, it appears that a basic level of integration upon their return to communities. Female social cohesion that appears to facilitate ex-combatants’ ex-combatants face both the stigma associated with slow process of reintegration into communities is all but ex-combatant status and that of having stepped outside absent. As such, reintegration processes, as they take societal gender norms as combatants. In the GLR coun- place in the rest of the GLR countries, do not appear to tries, this double stigma appears to be a potent mix, as be occurring at a meaningful scale in DRC – at least to it amplifies the already inherent gender specific disad- the extent that they are empirically visible in this study. vantages of being a female in the GLR country settings. Conceptually speaking, what is occurring in DRC is Indeed, while female community members across better characterized as remarginalization, and in terms the GLR countries display a range of disadvantages of reintegration programming, what is occurring in compared to male community members, such disad- DRC is better defined as reinsertion. The paradox that vantages are almost always further exaggerated in the this reveals is that in terms of parity, ex-combatants and case for female ex-combatants relative to male ex-com- community members in DRC are the closest across batants. Female ex-combatants are the least marrying most indicators as compared to the ex-combatants and group among ex-combatants and community members communities in the other GLR countries. However, the and, unlike male ex-combatants who see some improve- overall trajectories of reintegration processes in DRC ments in access to marriage over time, they see near are extremely weak relative to the other GLR countries. stagnant marriage rates over time. Additionally, female This weak trajectory may be inherent to the context of ex-combatants have extremely weak social capital that continued violence and insecurity in Eastern DRC, and may be indicative of the broader limits of reintegration contributes to their continued marginalization in the programming. GLR countries. The effects of this social capital deficit for female 6.1 Reintegration: An Imbedded Transition ex-combatants can spill over from the social to the The process of reintegration that ex-combatants across economic realm, as marriage is a key pathway to land the GLR face entails ongoing transformations along access, employment opportunities and customer bases social and economic lines. These ex-combatant trans- in the GLR countries. Limited land access mobility and formations are imbedded in the larger transformations education and skills gaps render female ex-combatants of society in post-conflict and development settings. economically vulnerable. These social and economic To understand reintegration processes requires un- dimensions interact to contribute to female ex-com- derstanding the context in which they occur. Building batants’ overall perception of disadvantage in the com- sustainable livelihoods and income security must be munity. Female ex-combatants are well aware of these understood as a piece in the process of larger economic disadvantages and may avoid identifying themselves development. Building social capital and social as ex-combatants at all, forfeiting consequently their cohesion must be understood as a piece of reweaving access to differentiated reintegration support. Interest- the social fabric of society. Closing gender gaps must ingly, despite the deep set of disadvantages that female be understood as a piece of a larger transformation of ex-combatants face in the GLR countries, they con- societal gender norms. All these dimensions are things sistently report higher levels of overall happiness and that ex-combatants, communities, and broader society a better outlook for the future, which is a point that across the GLR countries face together. While ex-com- warrants further inquiry in future studies. batants face a range of social and economic disadvan- tages to community members in FCS settings, what Focusing on the country level, DRC is the clear outlier is perhaps most important for ex-combatants is the to the broader trajectory of reintegration processes in perception that they are included in this larger societal the GLR countries. Taking into consideration of almost transformation. 35 To acknowledge the imbedded nature of ex-combatant As the case of DRC demonstrates, thinking about rein- reintegration processes is to contextualize them in an tegration in terms of parity alone may be insufficient to overall societal trajectory. Indeed, understanding the understand the scope of potential impact for reintegra- state of broader societal transformations in the GLR tion programming in a given context. Rather, parity is countries is a starting point to understanding the range complimented by an understanding of the overall pro- of reintegration processes that can occur there, the cesses that ex-combatants navigate, and the extent to space that exists for reintegration programming to aid which they are connected with broader societal transfor- in these endogenous processes, and in return the variety mations. In this light, it is possible that ex-combatants’ of possible reintegration programming approaches that perceptions of their inclusion and positive trajectory in might be successful in this range and space. This is what society are more important than their absolute parity the scope of potential impact refers to in a given context. with community members. 36 Referenced Works Autesserre, S. (2009). Hobbes and the Congo: frames, local violence, and international intervention. International Organization, 63(2), 249-280. Autesserre, S. (2010). The Trouble with the Congo: Local Violence and the Failure of International Peacebuilding. Cam- bridge University Press. Bleie, T. (2013) New Insights from Comparative Reintegration Research: Getting “the Community” right, Paper pre- sented at the Inter-University Seminar on Armed Forces & Society, 3O Sept 2013, Chicago Baas, S. (2012). From Civilians to Soldiers and from Soldiers to Civilians: Mobilization and Demobilization in Sudan. Amsterdam University Press. Barnhart, J. (2011). Umuganda: The Ultimate Nation-Building Project?. Pursuit-The Journal of Undergraduate Research at the University of Tennessee, 2(1), 3. Bowd, R., & Özerdem, A. (2013). How to Assess Social Reintegration of Ex-Combatants. Journal of Intervention and Statebuilding, 7(4), 453-475. Buxton, J. (2008). Reintegration and Long-Term Development: Linkages and Challenges. Bradford: Centre for Inter- national Cooperation and Security Thematic Working Paper, 5. Bøås, M., & Dunn, K. C. (Eds.). (2007). African guerrillas: raging against the machine. Boulder, CO: Lynne Rienner. Campbell, A. (2003). Where Do All the Soldiers Go? Veterans and the Politics of Demobilization, Chapter in: Davis & Pereira (eds.) Irregular Armed Forces and Their Role in Politics and State Formation, Cambridge University Press. Colletta, N. J., & Cullen, M. L. (2002). Social capital and social cohesion: case studies from Cambodia and Rwanda. Chapter in: Grootaert & Bastelaer (eds.) The Role of Social Capital in Development: An Empirical Assessment, Cambridge University Press. Collier, P., & Hoeffler, A. (2004). Greed and grievance in civil war. Oxford economic papers, 56(4), 563-595. Collier, P., Hoeffler, A., & Rohner, D. (2009). Beyond greed and grievance: feasibility and civil war. Oxford Economic Papers, 61(1), 1-27. Finn, A. (2012). The Drivers of Reporter Reintegration in Northern Uganda. World Bank, Gilligan, J. (1997). Violence: Reflections on a national epidemic. New York: Vintage Books. Hannoum, A. (2013) DDR before DDR: North African veterans from colonial mobilization to national reintegration. World Bank Humphreys, M., & Weinstein, J. (2004). What the fighters say: a survey of ex-combatants in Sierra Leone, June-August 2003. Survey by Center on Globalisation and Sustainable Development, Columbia University. 37 Humphreys, M. & Weinstein, J. (2005). Disentangling the determinants of successful demobilization and reintegration, No 69, Working Papers, Center for Global Development . Humphreys, M., & Weinstein, J. M. (2006). Handling and manhandling civilians in civil war. American Political Science Review, 100(03), 429-447. Humphreys, M., & Weinstein, J. M. (2007). Demobilization and reintegration. Journal of conflict resolution, 51(4), 531-567. Humphreys, M., & Weinstein, J. M. (2008). Who fights? The determinants of participation in civil war. American Journal of Political Science, 52(2), 436-455. International Alert (2010) Changing fortunes: women’s economic opportunities in post-war northern Uganda. Investing in Peace, 3 Jennings, K. M. (2008). Unclear ends, unclear means: Reintegration in postwar societies-The case of Liberia. Global Governance: A Review of Multilateralism and International Organizations, 14(3), 327-345. Kaplan, O., & Nussio, E. (2012). Community Counts: The Social Reintegration of Ex-Combatants in Colombia. Avail- able at http://ssrn. com/abstract, 2138188. Kaplan, O., & Nussio, E. (2013). Explaining Recidivism of Ex-Combatants in Colombia. Available at SSRN 2350318. Keen, D. (2008). Complex emergencies. Polity Press. Lamb, G. (2011). Assessing the Reintegration of Ex-Combatants in the Context of Instability and Informal Economies: The Case of the Central African Republic, the Democratic Republic of Congo and South Sudan (Washington, DC: World Bank, 2012), 60. Lederach, J. P. (1997). Building peace: Sustainable reconciliation in divided societies (pp. 23-35). Washington, DC: United States Institute of Peace Press. Marriage, Z. (2007). Flip-flop rebel, dollar soldier: demobilisation in the Democratic Republic of Congo. Conflict, Security & Development, 7(2). McMullin, J. (2004). Reintegration of combatants: were the right lessons learned in Mozambique?. International Peace- keeping, 11(4), 625-643. McMullin, J. (2013). Ex-combatants and the Post-conflict State: Challenges of Reintegration. Palgrave Macmillan. Muggah, R. (2010). Innovations in disarmament, demobilization and reintegration policy and research. NUPI working paper 774. Özerdem, A. (2012). A re-conceptualisation of ex-combatant reintegration: social reintegration’approach. Conflict, Security & Development,12(1), 51-73. Özerdem, A., & Podder, S. (Eds.). (2011). Child soldiers: From recruitment to reintegration. Palgrave Macmillan. Özerdem, A., & Podder, S. (2011b). Disarming Youth Combatants: Mitigating Youth Radicalization and Violent Ex- tremism. Journal of Strategic Security, 4(4), 63-80. Porto, J. G., Alden, C., & Parsons, I. (2007). From soldiers to citizens: Demilitarization of conflict and society. Ashgate Publishing. 38 Reno, W. (1999). Warlord politics and African states. Lynne Rienner Publishers. Schulhofer-Wohl, J., & Sambanis, N. (2010). Disarmament, Demobilization, and Reintegration Programs: An Assess- ment. Folke Bernadotte Academy Research Report. Shibuya, E. Y. (2012). Demobilizing Irregular Forces. Polity Press. Specht, I. (2013). Gender, Disarmament, Demobilization and Reintegration and Violent Masculinities. Cadernos, 61. Stankovic, T., & Torjesen, S. (2010). Fresh Insights on Disarmament, Demobilization and Reintegration: A Survey for Practitioners. Oslo: NUPI. Torjesen, S. (2013). Towards a theory of ex-combatant reintegration. Stability: International Journal of Security & De- velopment, 2(3). UN (2006) Integrated Disarmament, Demobilization and Reintegration Standards Utas, M. (2005). Building a future? The reintegration and re-marginalisation of youth in Liberia. Chapter in: Richards (ed.) No Peace, no war: an anthropology of contemporary armed conflicts. Uvin, P. (2007). Ex-Combatants in Burundi: Why the joined, how they left, why they fared, MDRP Report (No. 3). Vermeij, L. (2011) Socialization and reintegration challenges: a case study of the Lord’s Resistance Army. Chapter in: Richards (ed.) Child soldiers: From recruitment to reintegration. Palgrave Macmillan. Visher, C. A., & Travis, J. (2003). Transitions from prison to community: Understanding individual pathways. Annual review of sociology, 89-113. De Vries, H., & Wiegink, N. (2011). Breaking up and going home? Contesting two assumptions in the demobilization and reintegration of former combatants. International Peacekeeping, 18(1), 38-51. Woolcock, M., & Narayan, D. (2000). Social capital: Implications for development theory, research, and policy. The world bank research observer,15(2), 225-249. World Bank (2011a) Troisieme Enquete des Beneficiarires du PDRT, Unpublished Internal Document World Bank (2011b) Rapport d’Evaluation des Bénéficiaires du PNDDR Réintégration des hommes soldats et ex-combat- tants demobilize, Unpublished Internal Document World Bank (2011c) Réintégration Socio-Economique des ex-Combattants Région du Pool, République du Congo, Un- published Internal Document World Bank (2011d) Comparative report: Ex-combatant and community study on the socio-economic reintegration, Un- published Internal Document World Bank (2012) Reporter Re-integration and Community Dynamics Survey Report Reporter Re-integration and Community Dynamics Survey Report, Unpublished Internal Document 39 Annex I - Great Lakes Region Ex-Combatant Reintegration: Comparative Survey Results and Analysis 40 7. Demographics T he following is a capture of the demographics of the ex-combatant sample for this comparative study. The demographics reflected in the sample here are not those of the overall ex-combatant populations in each of the five GLR countries at the time of study, but rather reflect certain purposive sampling biases. For more information about the specific sampling methods and decisions in each of the GLR countries please see the individ- ual survey studies for ex-combatants in each of the five GLR countries.93 For a brief introduction to the reintegration programming context in each of the GLR countries see section 5.1. The unweighted ex-combatant sample contributions from the five GLR countries for the total sample of 6,475 ex- combatants in this study is as follows: Burundi comprises 19.4% (n=1,256) of the total raw sample, DRC 56% (n=3,625), Republic of Congo 10.3% (N=667), Rwanda 8% (N=517) and Uganda 6.3% (N=410). However, in an effort to create valid cross-country analysis of ex-combatants across the GLR, and especially for comparison to the community member sample, which contains proportionally different sample contributions from the five GLR countries, the raw sample contributions from each country have been weighted evenly. Integrating the full range of data from Rwanda has proved challenging in this study. The evolving format for the individual GLR country surveys has been a continual process of learning and iterative refinement. The Rwanda survey format is the starting point from which surveys evolved in RoC, Burundi, Uganda, and DRC. So, while data content in the Rwanda surveys is very much in line with the rest of the GLR countries, much of the specific question formatting is often different enough that a direct comparison of data is not feasible. Such instances are explained in footnotes. Collectively the data restrictions present in this study of ex-combatants across the GLR countries mean that the task of this study is to present a mosaic of findings. Up close, the pieces of the picture are not always complete and data is not always congruent. Nonetheless, there are clear data trends that represent a distinct narrative of ex-combatant reintegration across the GLR countries. Of the total sample of ex-combatants 88.1% were male and 11.9% were female. While across all individual coun- tries the disparity of representation of males and females was high, this is most true in Rwanda where only 2.5% of the country sample was female. The sample contribution from Uganda was comprised of the largest proportion of women (25.4%). The remaining countries fell closer to the overall sample composition.94 Table 1 above gives a cross-tabulated breakdown of the age, sex and disability of the ex-combatant sample from each of the five GLR countries.95 93   For Burundi see (2011) Troisieme Enquete des Beneficiarires du PDRT; for DRC see (2011) Rapport d’Evaluation des Bénéficiaires du PNDDR Réintégration des hommes soldats et ex-combattants démobilisés; for RoC see (2011) Réintégration Socio-Economique des ex-Combattants Région du Pool, République du Congo; for Rwanda see (2012) Comparative report: Ex-combatant and community study on the socio-economic reintegration; and for Uganda see (2011) Reporter Re-integration and Community Dynamics Survey Report Reporter Re-integration and Community Dynamics Survey Report. 94   It is important to note that a portion of ex-combatants never participate in formal reintegration programming. This is especially true of female ex-combatants who are hyperaware of the heavy stigmatization that can accopmony self-identification as an ex-combatant. As such, the actual proportion of female ex-combatants in the GLR is likely higher than the figures above suggest. 95   For the purposes of this report as a comparative study the demographic breakdown of the Ex-combatant sample by armed group will not be included for systematic analysis. The contexts of the different armed groups within the five GLR countries are seen as unique to each country context thus, while important units of analysis within each GLR country, not systematically comparable across the GLR countries. Further details on the Ex-combatant sample by armed group within each country are available in some of the five GLR country survey reports. 41 Table 1: Ex-Combatants - GLR Country Demographics Country Burundi DRC RoC Rwanda Uganda GLR Total Male 91.6% 86.1% 91.0% 97.5% 74.6% 88.1% Female 8.4% 13.9% 9.0% 2.5% 25.4% 11.9% Subtotal 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Age 18-30 65.4% 39.4% 31.3% 22.6% 38.6% 39.5% Age 31-40 26.9% 29.0% 40.6% 53.4% 24.6% 35.0% Age Over 40 7.7% 31.6% 28.1% 24.0% 36.9% 25.6% Subtotal 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Disabled 7.8% 4.6% 4.7% 24.1% 17.1% 10.9% Not Disabled 92.2% 95.4% 95.3% 75.9% 82.9% 89.1% Subtotal 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% In the total sample of ex-combatants 10.9% were categorized as disabled, the remaining 89.1% were categorized as not disabled. Most GLR countries were composed of a similar proportion of disabled ex-combatants, though Rwanda and Uganda had higher representations of disabled ex-combatants – 24.1% and 17.1% respectively. However, these higher compositions of disabled ex-combatants may be an artifact of the process by which disability categorizations were combined across the total sample.96 Of the total sample of ex-combatants 39.5% are between the ages of 18 and 30, 35% are between 31 and 40 years of age, and 25.6% are over the age of 40.97 Most of the GLR countries’ age compositions follow the trend of the total sample split, with two notable exceptions. Burundi’s age composition is heavily skewed towards those aged 18-30 (65.4%), and Rwanda is particularly heavy in the 31-40 years of age category (53.4%). In the case of Rwanda this age composition is likely a result of ex-combatants prolonged time spent participating in conflict – 50.9% of ex- combatants in Rwanda having spent between 10 and 20 years participating in conflict. The aspects of the lives of ex-combatants discussed in the following sections are key indicators of the process that ex-combatants experience in accessing pathways to reintegration across the GLR countries. Within the following sections of this report age, gender and disability dimensions to these processes to reintegration are explored to extract key trends across the GLR countries. The family and community, education and training, and addressing health needs are all seen as key pathways to reintegration of ex-combatants that will temper this discussion. 7.1 From Mobilization to Demobilization The following is a brief snapshot of the ex-combatants’ time with armed groups and the ways in which pertinent demographic details play in their experiences. Indeed, understanding the dynamics of the pathways into mobiliza- tion can add considerable nuance to our understanding of the specific challenges that ex-combatants can face at the time of demobilization. 96   Criteria for disability varied slightly from country to country across the GLR. To create a consistent categorization of disabled ex-combatants across the GLR countries, disability status was computed using the disability criteria from the Ugandan Amnesty Commission which included: (i) amputees; (ii) blind and partially blind; (iii) paralysis and partial paralysis; and (iv) body and head injury. 97   Across the total sample of ex-combatants from across the five GLR countries there were 300 under the age of 18. These 300 have been omitted from the analysis in this study for two main reasons: (i) the systematic capture of information pertaining to the specific dynamics of reintegration facing minors was absent from the surveys used across the GLR countries – with the exception of DRC, where 291 of the total 300 ex-combatants under the age were sampled; and (ii) the validity issues that the small sample of ex-combatants under the age of 18 (again, almost entirely from DRC) make meaningful comparative analysis infeasible. 42 Table 2: Ex-Combatant Age at Mobilization Age at Mobilization Mobilized Under Age 18 Mobilized Age 18-30 Mobilized Age 31-40 Mobilized Over Age 40 Proportion Average Age Proportion Average Age Proportion Average Age Proportion Average Age of Sample of Sample Of Sample of Sample Male 32.30% 13.83 48.40% 22.64 10.50% 34.71 8.70% 51.71 Female 40.10% 13.58 30.90% 22.74 11.80% 35.07 17.30% 50.11 Age 18-30 64.10% 13.75 35.90% 20.86 XXX XXX XXX XXX Age 31-40 8.80% 15.09 78.40% 23.41 12.90% 33.41 XXX XXX Age Over 40 3.90% 15.52 30.30% 24.01 27.00% 35.51 38.80% 51.34 Disabled 30.10% 13.39 41.50% 22.67 8.30% 34.02 20.10% 51.18 Not Disabled 33.60% 13.82 46.60% 22.65 10.80% 34.79 9.00% 51.39 Burundi 35.20% 14.53 59.30% 21.85 4.30% 33.97 1.20% 45.01 DRC 43.00% 13.96 48.10% 22.21 7.00% 34.53 1.90% 48.69 RoC 32.40% 13.3 46.00% 23.36 13.30% 34.8 8.30% 56.07 Uganda 22.00% 12.87 29.80% 24.08 18.80% 35.02 29.50% 50.43 GLR Average 33.40% 13.79 46.10% 22.66 10.60% 34.76 9.90% 51.34 In this table the use of XXX indicates a logically impossible field. While obtaining reliable information about ex-combatants’ age at mobilization, especially younger ex-combatants who may have been only adolescents, is a challenging endeavor we can pull out general trends for comparison across the GLR countries. The following data should be treated with caution and be regarded as a rough picture rather than concrete truth of age of mobilization in the GLR countries.98 The average age at mobilization was 23.8 across the GLR countries, however this figure masks considerable nuance in the age at mobilization.99 There are two steps to understanding age at mobilization more deeply. First is to understand the proportion of ex-combatants mobilized in different age brackets and then to understand the average age at mobilization within each bracket – this data is displayed in Table 2. The largest proportion of ex-combatants across the GLR countries (46.1%) was between the ages of 18 and 30 at the time of mobilization (on average aged 22.6). Indeed, with all GLR countries the 18-30 group is the largest. However, it is important to note the sizable number of ex-combatants who were mobilized under the age of 18 (38.9%) and who were very young at the time (on average at age 13.79) – a factor that can have a profound impact on their psy- chosocial wellbeing and in turn prospects for reintegration. The under 18 category was second largest within all GLR countries and in some almost even with the 18-30 category – for example in DRC where 43% of ex-combatants were aged under 18 at the time of mobilization (on average 13.79) and 48.1% were aged 18-30 (on average 22.21).100 Mobilization into violent conflict at an adolescent age can have a profound impact on the social and psychologi- cal development of individuals as they mature and, in turn, carry considerable weight for their ability to interact 98  For example, in Uganda 41.95% of ex-combatants were unsure of their age at the time of mobilization. However, working backwards from the current age of ex-combatants we can subtract away the time since demobilization and the time spent with armed groups to calculate an approximate age at mobilization for all ex-combatants. 99  Rwanda is excluded from all findings on age at mobilization due to lack of directly comparable data.  100  Disaggregation of the average age at mobilization in Uganda by armed group provides some necessary nuance here. The average age at mobilization for members of the LRA in Uganda, known for their strategy of youth abduction for mobilization, was 18.38 years. This stands in contrast to the ADF who had an average age of mobilization at 31.02 years, West Nile Bank Front with 41.34 years, and UNRF with 42.18 years. Further examination reveals that 51.3% of LRA ex-combatants were mobilized under the age of 18, with an average age of 12.78. 43 with communities upon their return. While data on abduction versus voluntary mobilization was not available for comparison across the GLR countries, this data may have added considerable nuance in exploring further gendered dynamics of mobilization. For example in Uganda, the only GLR country in which such data is available, where abduction is a well-known tactic for recruitment and mobilization 92.9% of females sampled between the age of 18 and 30 reported being abducted. Though as a counterpoint, there is reason to be cautious to such data. There are enormous social pressures at work and ex-combatants may fear stigma, retribution, or denial of amnesty as a result identifying themselves as willing participants in conflict – possibly inflating the proportion of ex-combatants that report abduction. As illustrated in Table 2, it appears that there are certain gendered dynamics to the age of mobilization. Across the GLR sample, female ex-combatants were more frequently mobilized under the age of 18 (40.1%, average age 13.58) when compared to male ex-combatants (32.3%, average age 13.83), and with decreasing frequency as the age at mo- bilization increases – 64.1% between the ages of 18 and 30, 8.8% between the ages of 31 and 40, and 3.9% over the age of 40. This trend holds true in all of the GLR countries except for the Republic of Congo in which the pattern of age at mobilization follows more closely to the male ex-combatants’ trend in which the majority of ex-combatants are mobilized between the ages of 18 and 30. Thought he exact reason for this gendered dimension to the age of mobilization is unclear, it is likely related to the benefits of mobilizing child soldier from an armed group’s perspec- tive – though child soldiers may be less effective soldiers in the traditional sense they are also easier to intimidate, indoctrinate, and misinform than adults.101 Not surprisingly current age showed a positive correlation to age at mobilization – meaning that on average the older an ex-combatant was at the age of mobilization the older they were at the time of sampling. Disability did not show any relationship to age at mobilization. Across the GLR countries the average number of years ex-combatants had spent with armed groups varied. At a cross-country level ex-combatants in the GLR countries spent an average of 7.08 years with armed groups. DRC and Rwanda stand out on the high end of this cross-country average with 11.16 years and 9.09 years spent with armed groups on average (respectively). Ex-combatants in Uganda spent the least amount of time on average with armed groups (4.38 years). A full table of the average time spent with armed groups is presented in Table 3. Drawing from DRC and Rwanda we can observe that those ex-combatants that were members of national armed forces (FAC in DRC and RPA in Rwanda) spent longer on average participating in conflict than those in other irregular armed groups (mean 18.99 years vs. 5.27 years in DRC and mean 12.57 vs. 8.15 years in Rwanda).102 The number of years spent with armed groups displayed a gendered trend across the GLR countries. Female ex- combatants spent a lower average number of years (mean = 4.95 years) with armed groups compared to their male ex-combatant counterparts (mean = 7.37 years). This trend holds across the GLR countries with the exception of Uganda – in which female ex-combatants spent on average slightly more years (mean = 4.84 years) than their male ex-combatant counterparts (mean = 4.22 years). A more detailed breakdown of years spent with armed groups across cross-cutting demographic lines can be found below in Table 3. Again, as with age at time of mobilization, current age showed a positive correlation to years spent with armed groups at a cross-country level. Though RoC and Uganda stood apart from this trend – in Uganda there was even a negative relationship between current age and average years spent with armed groups. Those ex-combatants who were categorized as disabled spent slightly longer on average (mean = 9.40 years) compared to their non-disabled counterparts (mean = 7.27 years). 101   See Beber and Blattman (2013) The Logic of Child Soldiering and Coercion, International Organization, 67, pp 65-104. 102   The caseload of ex-combatants in Uganda and RoC consisted almost wholly of ex-combatants from irregular armed groups (such as the LRA and ADF in Uganda and the Ninjas in RoC), thus a valid comparison of their average time spent with armed groups compared to national armed forces is not feasible here. 44 Table 3: Ex-Combatant Average Years Spent with Armed Group Average Years with Armed Group Burundi DRC RoC Rwanda Uganda GLR Total Male 6.4 12.12 4.55 9.11 4.22 7.37 Female 4.67 5.19 4.27 8.08 4.84 4.95 Age 18-30 5.01 4.3 4.62 8.51 5.88 5.4 Age 31-40 7.14 8.48 4.77 8.56 3.79 6.77 Age Over 40 13.85 24.74 4.17 10.81 3.27 10.58 Disabled 8.43 12.57 4.5 13.24 6.41 9.4 Not Disabled 6.07 11.09 4.54 12.48 3.96 7.27 Country Average 6.25 11.16 4.55 9.09 4.38 7.08 On average across the GLR countries it had been 4.05 years since ex-combatants were formally demobilized at the time of sampling. In Uganda the time since demobilization was about half the cross-country average (1.87 years) while in RoC it was roughly twice the cross-country average (8.07 years).103 It is important to remember that some ex-combatants may spontaneously self-demobilize during conflict, leaving behind their armed groups. In addition, after the cessation of violence, ex-combatants may leave armed groups and return to their home community, or another place, on their own initiative. A considerable amount of time may pass between these ‘informal’ demobili- zations and the point at which ex-combatants take part in a formal demobilization process. Being able to measure this gap may prove an important indicator in assessing dynamics of return within the GLR countries. Unfortunately, while there is data on formal demobilization across the GLR countries, there is only data in Uganda on both informal and formal demobilization collected – 42.1% having informally demobilized as much as a decade or more before participating in formal demobilization processes. These findings however are most likely relevant to the specific dynamics of return and reintegration in Uganda where ex-combatants escape from armed groups (primarily the LRA) and return directly to their communities and then retroactively applying for amnesty, reinsertion assistance and possibly attain further referral to reintegration programming – often with a lengthy time-lapse. In contrast, other reintegration programs in the GLR leave few opportunities for accessing reintegration benefits without participating in a formal demobilization process and a fairly linear supply of reinsertion and reintegration assistance upon their return.104 7.2 Marriage and Household Marriage dynamics are an important indicator of ex-combatants’ basic social standing. Indeed, marriage dynamics can tell us much about ex-combatants’ ability to leverage familial, economic, and social networks towards the attain- ment of marriage and in turn their ability redouble their engagement in these social structures through marriage – all indicators of a strong footing in the community. Across the GLR countries there is a clear trend of increasing marriage and cohabitation rates among ex-combatants at three time points: before demobilization, at demobilization, and at sampling. As is visible in Table 4, the propor- tion of ex-combatants that were married across the GLR countries increased from 33.9% prior to demobilization, 103   Though examination of trends by armed group is not included in the analysis here, it is worth noting that in Uganda membership to different armed groups appeared to play an important role in the time since demobilization and could serve as a valuable line of inquiry for more focused analysis within each of the individual GLR countries. 104   For a thorough examination of trends of formal and informal demobilization in Uganda see: (2011) Reporter Re-integration and Community Dynamics Survey Report Reporter Re-integration and Community Dynamics Survey Report. 45 to 36% at demobilization, and 46.8% at the time of sampling.105 These increases in marriage (and cohabitation) rates among ex-combatants are matched by an even clearer decline in the proportions that were single and or never married. There are two noteworthy trends in regards to the trajectory of ex-combatant marital status with the specific GLR countries. First, in RoC the proportion of married ex-combatants at all three time points was much lower than the GLR cross-country average (6.7% prior to demobilization, 5.6% at demobilization, and 5.6% at the time of sampling), instead the decrease in the proportion of ex-combatants that were single / never married were absorbed into the category ‘living together’ (47.1% prior demobilization, 60.1% at demobilization, and 75.3% at the time of sampling). Second, Rwanda is the only GLR country in which ex-combatants are more frequently married than community members at the time of sampling (77.4% vs. 46.9%) – a point that will receive attention in the summary of the ex-combatant portion of this study. There are certain demographic trends that can be extracted regarding marriage. Concerning gender, female ex-com- batants are less likely to be married at all time points than male ex-combatants. As is visible in Table 4, this disparity between female and male ex-combatants grows from 7.7% prior to demobilization, to 13.2% at the time of demo- bilization, and 24.2% at the time of sampling. This growing disparity between male and female ex-combatants can be explained in part by looking at the proportion of female-combatants who were divorced, separated, or widowed compared to male ex-combatants – female ex-combatants are the most likely to be divorced, separated, or widowed at any time period. Essentially it appears that while male ex-combatants’ marital trajectory across the three time points is primarily one of moving from single / never married to married or living together, female ex-combatants by contrast see only very marginal increases in marriage and cohabitation – instead their decreases in the single / never married category are absorbed into the divorced or separated, or widowed categories. These differing trajectories flag female ex-combat- ants across the GLR countries as facing clear barriers to accessing marriage and in turn the primary social unit for reintegration, the family, leaving them at increased risk for social isolation and marginalization. It is likely that stigma plays a core role in female ex-combatants very shallow trajectory towards marriage compared male ex-combatants.106 While male and female ex-combatants alike carry the burden of stigma and distrust as perpetrators of violence, female ex-combatants can face the additional cultural stigma of having stepped out of traditional gender roles. Turning now to age demographics, at all time points age shows a positive relationship to the likelihood of being married and accordingly a negative relationship to the likelihood of being single / never married (as is visible in Table 4). While those aged 18-30 are the least likely age demographic to be married at all time points they have the most positive trajectory towards marriage across age demographics – there is a 25.6% increase in the rate of marriage between prior to demobilization and the time of sampling among those aged 18-30 versus a 22.1% increase in those 31-40, and only 1.9% increase in those over the age of 40. So while it appears that younger ex-combatants face considerable challenges in accessing reintegration pathways through marriage compared to other age demograph- ics, their rate of change towards the near stagnant levels of marriage among those over 40 is the greatest – giving credence to the idea that one dimension to younger ex-combatants’ lag behind their elder peers, struggling to make up for time lost in conflict. As seen above divorce was low across the GLR countries, however of those who were divorced 26% of ex-combatants across the GLR countries said that their divorce was related to their ex-combatant status.107 When asked to explain 105   Rwanda is excluded from findings on marital status before demobilization and at demobilization due to lack of comparable data. In addition, Burundi is excluded from findings on marital status at demobilization due to lack of directly comparable data. 106   For a good introduction to the range of gender specific challenges that female ex-combatants can face see Coulter, Perssan and Utas (2008) Young Female Fighters in African Wars: Conflict and Its Consequences, Nordiska Afrikainstitutet, Uppsala. 107   Burundi is excluded from findings on ex-combatant divorce due to lack of directly comparable data. 46 Table 4: Ex-Combatant Marital Status at Three Time Points Marital Status Before Demobilization Married Living together Divorced or Separated Widowed Single/Never married Male 35.00% 15.10% 1.00% 0.30% 48.60% Female 27.30% 14.30% 4.50% 9.10% 44.80% Age 18-30 14.40% 8.10% 0.60% 0.40% 76.50% Age 31-40 38.80% 23.20% 2.10% 0.70% 35.30% Age Over 40 65.00% 18.60% 2.30% 4.80% 9.20% Disabled 36.10% 10.10% 1.30% 3.10% 49.40% Not Disabled 33.80% 15.30% 1.50% 1.40% 47.90% Burundi 29.90% 4.50% 0.20% 0.10% 65.30% DRC 45.50% 7.40% 2.00% 1.50% 43.60% Republic of Congo 6.70% 47.10% 3.40% 1.30% 41.50% Uganda 51.20% 3.70% 0.70% 3.40% 40.90% GLR Average 33.90% 15.00% 1.60% 1.60% 48.00% Marital Status at Demobilization Male 38.20% 24.10% 3.80% 1.30% 32.70% Female 25.00% 15.00% 11.30% 19.10% 29.60% Age 18-30 15.70% 17.70% 4.70% 1.70% 60.30% Age 31-40 38.50% 32.90% 5.60% 2.00% 21.00% Age Over 40 60.70% 19.70% 5.10% 9.60% 4.90% Disabled 30.60% 13.10% 7.60% 12.20% 36.50% Not Disabled 36.70% 23.40% 4.80% 3.30% 31.80% DRC 51.20% 8.60% 3.90% 1.90% 34.40% Republic of Congo 5.60% 60.10% 5.00% 3.10% 26.20% Uganda 48.40% 2.70% 6.40% 7.40% 35.10% GLR Average 36.00% 22.70% 5.10% 4.20% 32.10% Marital Status at Sampling Male 55.90% 23.30% 2.90% 1.20% 16.60% Female 31.70% 17.60% 13.00% 16.30% 21.30% Age 18-30 40.00% 21.50% 4.00% 1.10% 33.40% Age 31-40 60.90% 27.30% 4.10% 1.60% 6.10% Age Over 40 66.90% 18.40% 4.70% 8.10% 2.00% Disabled 57.70% 16.50% 6.10% 7.70% 12.10% Not Disabled 51.00% 23.90% 4.00% 2.60% 18.50% Burundi 60.00% 11.50% 0.80% 0.60% 27.00% DRC 59.90% 9.30% 4.10% 2.70% 24.00% Republic of Congo 5.60% 75.30% 7.40% 2.80% 9.00% Rwanda 77.40% 13.00% 1.70% 0.60% 7.40% Uganda 60.70% 5.90% 6.80% 8.30% 18.30% GLR Average 46.80% 25.10% 4.80% 3.60% 19.60% Rwanda is not calculated into the cross-country statistics for marital status at the time of sampling. Essentially, even with the sample weighting, including Rwanda in cross-country figures on marital status at the time of sampling can make it appear as though across the entire GLR ex-combatants marry more often than com- munity members – even though Rwanda is the only country in which this is actually true. 47 more specifically the most common responses were: (i) Stigma or the influence of the spouses family (19%); (ii) the emotional abuse and fear that spouses married with ex-combatants faces; (iii) lack of tools or money. Female ex-combatants most notably cited that they were in the bush with their spouse, but escaped leaving them behind (29.9%). On average across the GLR countries, 13.5% of ex-combatants who were married had a spouse who either was then, or had at one point been a combatant.108 The GLR countries deviating notably from this trend were RoC, in which a slightly higher proportion of ex-combatants had a spouse who was or had at one point been a combatant (24%), and Rwanda where rates of marriage with other ex-combatants were considerably lower (3%). The proportion of ex- combatants with a current or past spouse who is or was a combatant was fairly even across all demographics except for sex. Female ex-combatants were vastly more likely to currently have, or at one point have had, a spouse who was a combatant (53%) compared to male ex-combatants (9.1%). There are two dynamics which likely play some role in these findings: (i) in conflict where the proportion of females to males is relatively low it may be that female combat- ants marry at a higher rate than male combatants and (ii) upon return to the community female ex-combatants may face higher barriers (e.g. stigma) to marriage with non ex-combatants (see below). Across the GLR countries, ex-combatants’ attitudes towards marrying another ex-combatant varied considerably – in DRC as low as 25% would consider marrying another ex-combatant, and as high as 54.2% in Uganda. However, on average 32.2% of ex-combatants across the GLR countries would consider marrying another ex-combatant.109 Con- current to the rate at which ex-combatants marry other ex-combatants outlined above, attitudes towards marrying ex-combatants in the future display a distinctly gendered dynamic – 50.2% of female ex-combatants across the GLR countries would consider marrying an ex-combatant versus 29.8% of males. When asked to explain negative attitudes towards marrying another ex-combatant respondents most commonly cited the misbehavior of ex-combatants (22.7%) or stigma related the perceived criminality of ex-combatants (16.7%).110 Uganda was the only GLR country that departed from this trend – as stigma was only cited by 3.1% of ex-combatants and instead risk associated with living with ex-combatants (12.5%) was cited most commonly. Of female ex-combatants that would not consider marrying an ex-combatant, 28.4% cited stigma due to their perceived criminality as an explanation, compared to 15.6% of male ex-combatants. Across the GLR countries ex-combatants most commonly saw themselves as the household head (52.8%) – re- sponsible for household food and finances – followed by those who saw themselves and their spouse as respon- sible (19%) and those who saw only their spouse as responsible (6.7%).111 Across demographic lines there are clear trends: (I) female ex-combatants are less than half as likely as male ex-combatants to cite themselves as the household head (25.4% vs. 57%); (ii) female ex-combatants are vastly more likely to cite solely their spouse as the household head than male ex-combatants (21.9% vs. 4.4%); and (iii) both disabled ex-combatants and those aged 18-30 are far more likely to cite their parents and grandparents than non-disabled and other age demographic ex- combatants (14% of disabled vs. 5.7% of non-disabled, and 15.3% of those aged 18-30 vs. 1.3% of those 31-40 and 0.5% of those over 40). 108   Burundi is excluded from findings on combatant status of spouse due to lack of directly comparable data. 109   Burundi is excluded from findings on attitudes towards marrying an ex-combatant due to lack of directly comparable data. 110   Burundi is excluded from findings on explanations of attitudes towards marrying an ex-combatant due to lack of directly comparable data. 111   In Burundi, DRC, and RoC inquiry about household head as constituted by who was primarily responsible for household finance and food was asked in one question, whereas in Uganda these were two separate questions (finance and food respectively). However, due to the high correlation between the two answers in Uganda (over 80%) they were recoded as one question for direct comparability with the other GLR countries. Rwanda is excluded from findings on household food and finance responsibility due to lack of directly comparable data. 48 7.3 Literacy, Education, and Vocational Training Levels of literacy, educational achievement, and vocational training are important indi- Table 5: Ex-Combatant Literacy cators of ex-combatants’ basic life chances Literacy and their ability to engage with educational Neither Read nor Read Only Read and Write and vocational structures, to the extent Write they exist, in the different GLR country Male 17.30% 8.60% 74.10% contexts and to leverage the dividends of Female 36.90% 6.30% 56.80% this engagement towards further economic Age 18-30 22.60% 7.80% 69.60% and social opportunities – in the end solidi- Age 31-40 15.60% 8.20% 76.10% fying their footing in the community. Age Over 40 20.60% 8.40% 70.90% While literacy was generally high across the Disabled 27.40% 10.20% 62.40% ex-combatant samples in the GLR coun- Not Disabled 19.40% 8.10% 72.60% tries (71.6 % could read and write), female Burundi 11.90% 4.20% 83.90% ex-combatants had the lowest literacy rate DRC 13.10% 8.50% 78.50% (56.8%) and were most likely to be com- Republic of Congo 18.70% 13.10% 68.30% pletely illiterate (36.9%) compared to any Uganda 37.00% 7.40% 55.60% other crosscutting demographic (17.3% of GLR Average 20.10% 8.30% 71.60% males for example).112 Disabled ex-com- batants and ex-combatants age 18-30 also scored notably lower on literacy. Across the GLR countries these three categories (female, disabled and age 18-30) of ex-combatants were consistently poor performers on literacy, though they closely switch places for worst performer within the individual GLR countries. These trends are displayed in Table 5. Regarding educational achievement there was very little change across all demographic groups between level of educational achievement at demobilization and at the time of sampling. The largest portion of all ex-combatants had some primary level of education (33.6%) at the time of demobilization and at the time of sampling (34.2%), followed by some secondary education (26.3% at demobilization and 23.4% at sampling).113 However, as is visible in Table 6, there is considerable variation in the individual GLR countries as far as the levels of ex-combatant educa- tional achievement. Generally speaking, ex-combatants in Burundi and Uganda had educational achievement levels skewed more towards partial or complete primary education, while those in DRC and RoC were more skewed towards partial or complete secondary education. Again, while there was generally very little change in ex-combatants’ educational achievement levels in the time between demobilization and sampling (which as discussed above was on average 4.05 years) DRC and RoC stand out in that ex-combatants across all levels of educational achievement at the time of demobilization were absorbed substantially into professional level achievement at the time of sampling (visible in Table 6), though this was especial- ly true for male ex-combatants in DRC.114 Across the GLR countries female ex-combatants educational achievement levels were skewed lower than their male counterparts at demobilization and the time of sampling. Ex-combatants 112   Rwanda is excluded from findings on literacy due to lack of directly comparable data. 113   Rwanda is excluded from findings on educational achievement at demobilization and the time of sampling due to lack of directly comparable data. 114   Movement into professional level educational achievement is likely related to vocational training provided as a part of reintegration programming. Vocational training is a component of most reintegration programs in the GLR countries, however they were an especially large component in DRC specifically – where vocational training was given to ex-combatants and community members together in combination with the formation of related economic associations. 49 Table 6: Ex-Combatant Educational Achievement Levels Education Level at Demobilization higher education Some secondary Some primary Some higher Professional Completed Completed Completed secondary education Islamic or religious primary None Male 7.7% 1.5% 32.2% 20.3% 27.8% 7.9% 1.1% .6% .9% Female 15.6% 4.9% 41.2% 12.2% 18.6% 6.1% .8% .2% .5% Age 18-30 9.4% 1.0% 39.4% 20.6% 25.2% 3.2% .9% .0% .2% Age 31-40 6.0% 1.7% 28.2% 19.7% 30.7% 11.5% 1.2% .6% .4% Age Over 40 10.4% 4.3% 28.2% 15.2% 25.0% 11.7% 1.4% 1.5% 2.3% Disabled 12.5% 2.4% 42.2% 21.6% 16.7% 2.5% 1.6% .3% .3% Not Disabled 8.6% 2.0% 32.8% 18.8% 27.3% 8.1% 1.0% .6% .8% Burundi 8.4% 2.5% 27.2% 34.8% 24.0% 2.0% .9% .1% .2% DRC 8.5% .2% 21.4% 14.0% 38.2% 12.9% 1.6% 1.7% 1.5% RoC 3.9% 0.0% 12.8% 20.7% 26.3% 31.8% 1.1% 0.0% 3.4% Uganda 11.3% 3.9% 58.0% 7.9% 16.7% 1.5% .7% 0.0% 0.0% GLR Average 8.9% 2.0% 33.6% 19.0% 26.3% 7.6% 1.1% .5% .8% Education Level at Sampling Male 7.7% 2.0% 32.8% 19.5% 24.8% 5.7% 1.3% .9% 5.4% Female 17.4% 6.0% 42.1% 9.6% 16.2% 2.7% .9% .6% 4.6% Age 18-30 7.9% 1.1% 35.3% 19.2% 26.5% 5.3% 1.2% .4% 3.3% Age 31-40 6.2% 2.6% 34.4% 21.1% 19.9% 5.4% 1.9% 1.1% 7.3% Age Over 40 15.7% 6.6% 33.3% 12.8% 18.4% 4.3% .8% 1.7% 6.4% Disabled 13.9% 2.8% 40.2% 19.1% 16.0% 2.1% 1.1% 1.0% 3.8% Not Disabled 8.6% 2.6% 33.5% 17.8% 24.4% 5.6% 1.3% .8% 5.4% Burundi 8.4% 2.5% 26.8% 33.8% 24.5% 2.8% .9% .2% .2% DRC 3.0% .4% 4.7% 2.8% 32.4% 17.6% 2.6% 2.3% 34.3% RoC 5.1% 0.0% 6.8% 5.1% 23.7% 33.9% 3.4% 1.7% 20.3% Uganda 12.4% 3.7% 53.5% 7.7% 19.6% 1.2% 1.0% 1.0% 0.0% GLR Average 9.2% 2.6% 34.2% 17.9% 23.4% 5.2% 1.2% .8% 5.3% aged 18-30 were also more clearly represented in slightly lower levels of educational achievement than their older counterparts as were disabled in relation to non-disabled. In line with the very low levels of mobility in ex-combatants’ levels of educational achievement between demobi- lization and sampling, only 15.2% of ex-combatants across the GLR countries reported that they were continuing education since demobilization.115 DRC, where 30.2% of ex-combatants were continuing education, was the only GLR country that stood out significantly from this trend. Ex-combatants aged 18-30 were the most likely (18.8%) demographic group across the GLR countries to be continuing education since demobilization. Of those who were continuing education since demobilization across the GLR countries, the most notable pathways were: (i) pursuing professional qualifications (34.8%), or (ii) pursuing normal academic qualifications (34.1%). 115   Uganda is excluded from findings on rates and varieties of continuing education due to lack of directly comparable data. 50 Across the GLR countries the majority of ex- Table 7: Ex-Combatants’ Participation in Skills combatants (58.1%) did not partake any form & Vocational Training of vocational training as part of the reintegration process. This is not to suggest that vocational Participated in Skills or Vocational Training training was not available as all reintegration Yes No programs in the GLR offer, or in some cases Male 41.20% 58.80% serve as a referral to, some form of vocational Female 47.30% 52.70% training.116 However, as visible in Table 7, there is considerable variation between the respec- Age 18-30 39.90% 60.10% tive GLR countries. Notably in DRC, where Age 31-40 42.40% 57.60% vocational training was a large component of Age Over 40 44.30% 55.70% reintegration programming, ex-combatants had Disabled 47.10% 52.90% indeed received vocational training at a higher Not Disabled 41.40% 58.60% rate. Rwanda also displayed higher rates of vo- DRC 66.10% 33.90% cational training – though it is unclear whether Republic of Congo 20.20% 79.80% this is due to reintegration programming.117 By Rwanda 76.40% 23.60% contrast in Uganda, where reintegration services Uganda 32.60% 67.40% merely served as a referral to existing vocational GLR Average 41.90% 58.10% programs for the general population, there was considerably lower reported participation. While at a cross-country level it appears as though there is a slight positive relationship between age and the likeli- hood of receiving vocational training, this is only truly evident in DRC. Female ex-combatants, however, are slightly more likely to receive vocational training compared to male ex-combatants (47.3% vs. 41.2%), with the exception of RoC where they pair only slightly lower (18% vs. 20.6%). Of those ex-combatants who had received skills or vocational training, the majority (62.7%) were utilizing these skills and training.118 Female ex-combatants were using skills and training slightly less than male ex-combatants (61.3% vs. 63%). In addition ex-combatants were using their skills and training progressively more across age demo- graphics (53.9% for ages 18-30, 56.8% for ages 31-40, and 76.3% for those over 40). Disabled ex-combatants utilized their skills and training at a lesser frequency (54.5%) than their non-disabled counterparts (63.8%).119 Of those ex- combatants who were not using their skills and vocational training the most common explanations were: (i) 29.4% lost necessary tools and have no money for new ones, (ii) 21.6% lack of capital, and (iii) 10.9% lack of facilities for carrying out the vocation and skills they were trained in. 7.4 Summary Conflict represents an immense social disruption that often results in the disintegration of families, communities, and the broader fabric of society. The process of DDR is aimed at reconnecting the fractured pieces of these social entities so that collective norms and processes can be re-solidified. For ex-combatants facing this transition from conflict to peace by returning to families, gaining economic independence and participating in their communities represent the core challenges of reintegration. While there are few that do not feel the effects of conflict across the 116   Burundi is excluded from findings on vocational training received due to lack of directly comparable data. 117   In Rwanda entrepreneurship training, with the end result of a business plan and small grant were a core part of reintegration programming. It is unclear however if this programmatic component is higher for Rwanda’s higher vocational training rates 118   Burundi is excluded from findings on use of vocational training due to lack of directly comparable data. 119   Though unclear, it is possible that this trend is an indication of inadequate targeting of disabled ex-combatants for skiling and training specific to their unique needs and abilities. 51 GLR countries, it is noteworthy that ex-combatants face a range of distinct challenges in the process of reintegra- tion. A large proportion of ex-combatants across the GLR countries were mobilized into conflict under the age of 18 (who were on average only early adolescents around 13 years old) and spent a number of their formative years as adults socialized in a context of violence. One way to view this is that a significant proportion of ex-combatants have missed the opportunity of the socialization of adult norms and behavior during normal peacetime, setting them with a steep learning curve upon return for socializing to these norms and values that they may have never learned in the first place – due to their absence from traditional family and community structures during their formative adult years in conflict. Further study into the specific modes of mobilization, for example abduction, may add consider- able explanatory power to the specific challenges the ex-combatants who were mobilized at a young age face. While the evidence presented on the age at mobilization is not conclusive, this line of inquiry deserves further attention in future studies. Ex-combatants across the GLR countries, with the exception of Rwanda, are married less frequently than commu- nity members. However, ex-combatants generally show a positive trajectory towards marriage and cohabitation over time. It appears that across the GLR, ex-combatants’ largest obstacle in reintegrating into, or in most cases building the familial unit is making up for the time lost by participation in conflict, especially for younger ex-combatants. Ex-combatants’ levels of educational achievement and literacy are skewed slightly lower than community members’ – with lower levels of partial secondary, secondary, partial tertiary, or complete tertiary level achievement. While educational mobility was very low in general, it appears as though time lost while in conflict is a significant barrier to educational achievement – especially for younger ex-combatants who are a step behind their older peers, but are more aggressive about closing this gap through continuing education since demobilization. 7.4.1 Vulnerable Subgroups As discussed above, while younger ex-combatants (aged 18-30) tend to be a step behind their older peers as far as access to marriage and educational achievement, this appears to be a product of the time lost while in conflict. In addition, though younger ex-combatants tend to be a step behind, they share the same general positive trajectory as their older peers. Female ex-combatants, however, show a widening gap in relation to male ex-combatants – who themselves generally have a positive trajectory in terms of education and marriage. In assessing trends in marriage across the GLR countries we can summarize several key points regarding female ex-combatants: (i) female ex-combatants are less likely than male ex-combatants to be married or cohabitate; (ii) the gap in marriage and cohabitation rates between male and female ex-combatants has grown over time from prior to demobilization to the time of sampling; and (iii) female ex-combatants are the most likely group to be divorced, widowed, or separated. The weight of female ex-combatants’ disadvantage in these regards is exaggerated further when compared to female community members marriage rates – who themselves rank lower in marriage rates than male community members, though still notably higher than female ex-combatants (31.7% of female ex-combatants are married versus 38.1% of female community members). Essentially while male ex-combatants are making steady progress towards parity with community members in terms of marriage, female ex-combatants’ progress is extreme- ly shallow. Female ex-combatants also have the lowest prospects for marriage in the future, as attitudinal indicators reveal that male ex-combatants are much less likely to be willing to marry another ex-combatant than female ex-combatant (29.8% vs. 50.2%). When analyzed against the back drop of community members’ ranking on the same attitudinal indicator (25.7% of male community members and 25% of female community members would consider marrying an ex-combatant) we can see that female ex-combatants have a considerably smaller pool of individuals who are at- titudinally open to marrying them compared to male ex-combatants. 52 Collectively these findings cement the fact that in the GLR countries female ex-combatants are not only the least likely group across all demographics (ex-combatants and community members alike) to be married and have a family, but also the group that faces the largest barriers to accessing marriage in the future – placing them outside of the primary unit of reintegration and at substantial risk for marginalization and social isolation. As an additional note, female ex-combatants lag behind male ex-combatants with lower levels of literacy, and educa- tional achievement – also lagging behind the female community members on both measures as well. As with those aged 18-30, female ex-combatants are slightly more aggressive than their male counterparts (16.5% vs. 15.1%) in pursuing further education to close this gap. 7.4.2 Unique Country Trends Rwanda stands out from the rest of the GLR countries as the only country where ex-combatants appear to be more frequently married than community members – and to a considerable extent (77.4% versus 46.9%). However, there are reasons to be skeptical to these figures. In the Rwandan sample female ex-combatants (a group that consistently displayed the lowest marriage rates across the other GLR countries) were severely under represented (only 2.5% or n= 13 of the total 517 Rwandan ex-combatants) compared to female community members (31.2% or n=159 or the total community member sample). In addition, those ex-combatants aged 18-30 (who across all other GLR coun- tries were the least likely age demographic to be married) were more than twice as represented in the community member sample (57% or n=290) as in the ex-combatant sample (22.6% or n=132). In effect, these facets of the demographic representations in the ex-combatant and community member samples may have inflated the rate at which it appears that ex-combatants marry, and deflated the rate at which community members appear to marry. However, there are further contextual details to consider in terms of marriage in the case of Rwanda. In Rwanda males are required to have access to adequate housing in order to get married. However, the formal regulations for what qualifies as adequate housing in Rwanda are somewhat narrowly defined under the policy of imidugudu – a large scale body of housing policy aimed at consolidating dispersed land and housing in an overall effort toward villagization. The result has been inflation in adequate housing prices and in turn a severe crisis in the availability of adequate housing overall that in effect is locking many Rwandans out of official marriage – though they may co- habitate without formalized marital status.120 This dynamic may further deflate the rate of marriage in Rwanda for community members. In contrast to community members, most ex-combatants are returning to Rwanda from Eastern DRC, where they have been away for an average of nine years. In this time some ex-combatants have married and when returning to Rwanda bring their spouse with them. The legal status of these marriages in Rwanda is unclear, however it is possible that some ex-combatants unwittingly navigate past the formal barriers to marriage that community members face – in turn accounting for their slightly higher marriage rates. While it is likely that the interaction of housing policy, marriage, and dynamics of return are key in understanding why ex-combatants marry more than community members in Rwanda, this exact narrative must be treated as conjecture. In the future a more focused inquiry into the dynamics of ex-combatant and community member marriage in Rwanda could prove prudent. For if indeed ex- combatants have been more successful than community members in accessing pathways to marriage in Rwanda the details of this finding could hold considerable explanatory value in analyzing other cases and, not least, in develop- ing reintegration programming in the future. RoC, where marriage rates for both ex-combatants and community members alike were drastically lower than in other GLR countries (5.6% of ex-combatants and 18.5% of community members at the time of sampling were 120   This narrative of the interrelated nature of housing policy and marriage in Rwanda is well documented in Sommers (2012) Stuck: Rwandan Young and the Struggle for Adulthood, Ga: University of Georgia Press. 53 married versus 60.2% and 64.3% on average of the remaining respective ex-combatants and community members), stood out as well. What is notable in RoC is that while marriage rates are much lower than average across the GLR countries, a much higher proportion of both ex-combatants and community members are cohabitating, but are not married, than on average across the GLR countries (75.3% of ex-combatants and 53.5% of community members in RoC were cohabitation versus 8.9% and 4.9% on average of the remaining respective ex-combatants and community members across the GLR). It appears that while community members do access marriage at a higher rate than ex-combatants in RoC, commu- nity members also face considerable barriers to accessing marriage themselves. Instead the most significant marital status for both community members and ex-combatants alike is cohabitation. As ex-combatants’ levels of cohabi- tation in RoC increase from 47.1% prior to demobilization, to 60.1% at demobilization, and 75.3% at sampling, it would appear as though this is the primary pathway to accessing the familial unit. Further study to explore the dynamics of formal and informal marriage in RoC would prove illuminating – especially if formal marriage, largely understood as the primary pathway to accessing the familial unit, is not necessary for reintegration in the RoC context. It should also be noted that without a clear explanation or triangulation for RoC’s departure in terms of marriage rates the data should also be treated carefully. It is possible that there are unbeknownst errors in data capture or coding that have produced these findings. Lastly, in Uganda ex-combatants are considerably more likely than average across the GLR countries to report will- ingness to marrying an ex-combatant in the future. Though there is no direct explanation it is possible that the specific dynamics of combatant mobilization in Uganda may play a role in this trend. In Uganda abduction was a well-known tactic of mobilization, especially by the LRA. Though abductees may have committed violent acts against their communities, often forcibly, there is evidence that ex-combatants are simultaneously understood as victims and perpetrators by community members – a factor that has reportedly contributed to a general willingness to accept returning ex-combatants back into communities.121 This dynamic, combined with the extensive use of traditional reconciliation ceremonies (not necessarily part of reintegration programming), may contribute to com- munity members in Uganda’s openness to marriage with ex-combatants. Futures studies could flag this conjecture for further analysis. 121   See for example: Finnegan, A. C. (2010), Forging Forgiveness: Collective Efforts Amidst War in Northern Uganda. Sociological Inquiry, 80: 424–447. 54 8. Housing, Land, Livestock and Food Security The following is an examination of the core dimensions of: (i) the types of dwellings that ex-combatants live in and related issues such as ownership and tenure; (ii) access to land for agricultural production and (iii) its connection to food security. 8.1 Dwelling, Living Conditions and Land Security In examining who ex-combatants live with across the GLR countries the three most common categories are: (i) with the same family as before conflict (29.2%), (ii) with a family but different to that from before conflict (24.3%), and (iii) with a partner (19.9%). Two countries across the GLR stood out from this general trend. First in RoC the majority ex-combatants reported living with a partner (43.6%) at a proportion more than double the cross-country average. Second, in Rwanda the majority of ex-combatants reported living with a family that was different from the one before conflict (57.7%) at a proportion more than double the cross-country average. As is visible in Table 8 above, female ex-combatants were more likely to be living with a family either the same or dif- ferent to the one before conflict than male ex-combatants – though less likely to be living with a partner or a family that consisted of a partner and children. It is also noteworthy that disabled ex-combatants were the least likely demo- graphic group to be living with a partner – at a proportion less than half the cross-country average. Regarding housing types, there were diverse compositions across the GLR countries; however at a cross-country level ex-combatants were most commonly living in: (i) a house (48.2%); or (ii) in a hut or tent (30.8%). Only Table 8: Ex-combatant Household Membership Who Do You Live With? With same With family but Family (unidenti- With With Alone family as before different to that from fied) / Partner and Other partner Friends the conflict before the conflict children Male 5.00% 20.00% 28.40% 23.80% 3.50% 17.70% 1.50% Female 3.30% 18.90% 35.20% 27.60% 2.20% 12.00% 0.80% Age 18-30 5.70% 20.20% 24.70% 17.20% 4.00% 26.60% 1.50% Age 31-40 4.80% 20.10% 25.40% 31.10% 2.00% 15.10% 1.60% Age Over 40 3.40% 20.20% 41.10% 27.10% 1.30% 5.70% 1.10% Disabled 5.80% 8.40% 38.30% 32.60% 1.80% 12.40% 0.70% Not Disabled 4.70% 21.20% 28.10% 23.40% 3.50% 17.70% 1.50% Burundi 4.50% 11.50% 0.00% 0.00% 3.20% 78.90% 1.80% DRC 5.60% 22.40% 41.70% 21.50% 8.20% 0.00% 0.60% RoC 8.20% 43.60% 16.60% 25.70% 1.70% 0.90% 3.30% Rwanda 4.40% XXX% 32.80% 57.70% 3.60% 0.00% 1.60% Uganda 1.20% 16.20% 56.00% 26.50% 0.00% 0.00% 0.00% GLR Average 4.80% 19.90% 29.20% 24.30% 3.30% 17.00% 1.50% The use of XXX indicates that respondents in Rwanda were not given the option to respond that they live with a partner or spouse – the responses that would have been in the feild are likely absorbed into the categories of those who live with a family either the same or different from the one before conflict. 55 Table 9: Ex-Combatant Housing Ownership Who Owns the Housing You Currently Live In? Family I live Spouse, Joint Non-family My relatives, I own it with such as son, My friend Other partner owners member family mother, parents Male 43.70% 1.70% 4.20% 6.80% 0.00% 2.10% 18.10% 23.40% Female 22.80% 17.60% 3.00% 13.70% 0.00% 0.60% 16.30% 25.90% Age 18-30 31.10% 4.30% 3.20% 11.40% 0.00% 1.90% 26.90% 21.10% Age 31-40 44.20% 3.10% 4.70% 4.40% 0.00% 2.60% 13.00% 27.90% Age Over 40 55.10% 3.80% 4.70% 4.90% 0.00% 1.10% 8.30% 22.10% Disabled 40.40% 5.30% 3.10% 15.00% 0.00% 3.40% 14.20% 18.60% Not Disabled 41.20% 3.60% 4.20% 6.80% 0.00% 1.80% 18.30% 24.20% Burundi 51.00% 2.00% 0.40% 0.00% 0.00% 1.00% 21.70% 23.90% DRC 28.60% 3.80% 4.40% 10.20% 0.00% 2.40% 17.00% 33.70% RoC 24.70% 3.50% 11.30% 1.80% 0.00% 2.00% 29.70% 27.00% Rwanda 39.80% 1.40% 0.00% 12.70% 0.00% 5.40% 13.80% 26.80% Uganda 59.80% 7.40% 3.20% 15.20% 0.00% 0.00% 5.90% 8.60% GLR Average 41.00% 3.70% 4.10% 7.70% 0.00% 2.00% 17.80% 23.70% Uganda stood out significantly from this trend, with the majority of ex-combatants (77.6%) living in a hut or tent. Across demographic categories there was fairly even membership to types of housing categories. Though, female ex- combatants were notably less likely to live in a house compared to male ex-combatants (41.8% vs. 49.1%) and more likely to live in a hut or tent (43.7% vs. 29%). There were varying rates of housing ownership across the GLR countries. Generally speaking self-ownership was the most common across the GLR countries (41%), followed by family ownership (17.8%). However, in RoC family member ownership was most common (29.7%) followed by self-ownership (24.7%). As is visible in Table 9, housing ownership rates were consistently lower for female ex-combatants than male ex-com- batants across the GLR countries; 22.8% of female ex-combatants owned their land versus 43.7% of male ex-com- batants – though Rwanda is an exception from this trend where 63.9% of female ex-combatants owned their land versus 39.1% of male ex-combatants. In contrast female ex-combatants were more likely to cite that their housing was owned by their spouse (17.6% vs. 1.7%) or by family that they live with (13.7% vs. 6.8%) when compared to male ex-combatants. Concerning age dynamics of housing ownership, there was a positive correlation visible between age and rate of housing ownership. Inversely, as age increased ex-combatants were less likely to rely on their relatives or family. When housing ownership was cross-tabulated against marital status a clear trend emerged. Of those ex-combatants who reported self-ownership of their housing, 68.4% were married and 16.9% were cohabitating – only 5.8% of those who reported self-ownership were single / never married. Inversely, when we look at the marital status of those who reported family ownership of their housing, 52.2% of those who reported their housing as owned by family they live with and 39.8% of those who reported family ownership were single / never married. Marriage rates show a clear correlation to housing ownership. This evidence supports the idea that marriage is a key pathway to housing, land access and security. Ex-combatants had a standard distribution of perceptions of their current living situation relative to perceptions at the time of demobilization across all the GLR countries – 21.8% of ex-combatants saw their current living situation 56 as better than at the time of demobilization, 49% saw it as the same, 26.2% saw it as worse, and only 2.9% pointed out that they did not have housing at the time of demobilization.122 Across demographic categories these perceptions were remarkably even as well. When examining ex-combatants’ perceptions of their own living situation compared to their neighbors, the majority saw themselves as well off, or worse off.123 Only 10.9% of ex-combatants saw their neighbors as having a better living situation, 47.8% saw it as the same, and 40.8% saw it as worse. There was some notable variation in ex-combatants’ perceptions of their living situation relative to their neighbors within specific GLR countries. In Uganda and Rwanda ex-combatants were more likely to have seen their living situation as worse than their neighbors (53.7% and 52.6% respectively) than in DRC and RoC (31.4% and 29.1% respectively). Looking at specific demographic differences it is apparent that disabled ex-combatants more commonly saw their situation as worse than their neighbors (59.6%), than non-disabled ex-combatants (38.3%). 8.2 Land Access and Food Security Gaining access to land for agricultural production is seen as a key pathway to both economic mobility and food security for ex-combatants. However, comparing land ownership across, and even within, the GLR countries can prove challenging. In many areas land ownership structures vary considerably and thus across the context of findings. For example, in many areas land ownership is organized around clans and infrequently owned on a private basis. However, land tenure can be very secure because of the clan structure despite the absence of deeds or titles. Though there is no systematic capture of the types of ownership structures across the GLR countries, these must be kept in mind when viewing the findings in this section. Land access for cultivation purposes was universally high across GLR countries and within crosscutting demograph- ic categories, with 92.6% of ex-combatants having access to land for cultivation purposes. In Uganda a more in-depth questioning of the tenure status of the land ex-combatants used for cultivation showed ex-combatants aged 18-31 were more likely to have a title for the land they cultivated (58.6%) compared to those 31-40 (29%) and those over 40 (30%). However these older age demographic groups were more likely than their younger counterparts to use communally owned land: 40% of those aged over 40, 28% of those aged 31-40, and 16.6% of those 18-31. Disabled ex-combatants also more frequently accessed communally owned land than non-disabled counterparts (35.7% vs. 26.5%). Though there is no comparable data for the other GLR countries these findings from Uganda may lend some nuance to the land ownership dynamics across demographic lines. Of the ex-combatants who did not have any access to land for cultivation at all in the DRC and RoC, lack of interest (29.9%) and lack of capital (27.2%) were the most common explanations.124 Other notable trends were that female ex-combatants more commonly cited fear of conflict (35.7%) than male ex-combatants (13.5%). Also, disabled ex- combatants more commonly cited distance / living in the city (28.6%) as an explanation for their lack of access to land for cultivation than their non-disabled counterparts (8.1%). In examining changes in ex-combatants’ access to arable land over a two-year period it was found that a significant proportion of ex-combatants (38%) had experienced an increase in their access to arable land over the last two years.125 Despite this general trend, the GLR countries vary considerably on this point – see Table 10. On the one hand, in DRC and RoC the majority ex-combatants had seen an increase in their access to land and, on the other 122   Rwanda is excluded from findings on perceptions of current living situation compared to at the time of demobilization due to lack of data directly comparable data. 123   Burundi is excluded from findings on perceptions of current living situation compared to neighbors due to lack of data. 124   DRC and RoC are the only GLR countries with data available on reasons for lack of access to arable land. 125   Rwanda is excluded from findings on changes in access to arable land due to lack of directly comparable data. 57 hand, in Burundi and Uganda the majority had Table 10: Ex-Combatant Change in Access to Arable Land not seen an increase in their access to land.126 Change in Access to Arable Land On average female ex-combatants had less often More Access Same or Less Access experienced an increase in their access to land (26.5%) than male ex-combatants (39.8%) – Male 39.80% 60.20% though Uganda was the only country where this Female 26.50% 73.50% trend was not displayed (12.4% of female ex- Age 18-30 34.70% 65.30% combatants having an increase vs. 9.7% of males). Age 31-40 43.50% 56.50% Similarly, disabled ex-combatants also tended Age Over 40 35.50% 64.50% to less often have experienced increases in their Disabled 22.30% 77.70% access to arable land (22.3%) when compared to Not Disabled 39.40% 60.60% non-disabled ex-combatants (39.4%). Burundi 30.40% 69.60% There is a diverse range of findings across the DRC 63.50% 36.50% GLR countries when looking at ex-combatants’ Republic of Congo 72.20% 27.80% explanations for gains in access to land for culti- Uganda 10.40% 89.60% vation.127 In general it appears as though capital GLR Average 38.00% 62.00% gained through strong agricultural yields has served as ex-combatants’ primary pathway to in- creased access to land for cultivation across the GLR countries. Looking to DRC, RoC, and Uganda we can observe that 42.2% of ex-combatants explained their increased access to land as a result of a combination of factors: (i) capital accrued from bountiful agricultural yields and (ii) the desire to produce more agriculturally for both subsis- tence and commercial purposes. Likewise, when explaining unchanged or decreased access to land for agricultural production in DRC and RoC 48.3% of ex-combatants cited lack of capital or resources as their primary barrier to land access mobility. While capital, especially that acquired through strong agricultural production, appears to be an important explana- tion for ex-combatants’ upward land access mobility across the GLR countries, two other explanations also deserve attention: (i) inheritance dynamics and (ii) marriage. These two pathways to land mobility appear especially relevant to female ex-combatants and young ex-combatants (age 18-30). In DRC and RoC 28.7% of ex-combatants (40.3% in DRC alone) cited inheritance as their pathway to increased land access. This was especially true for younger ex-combatants (aged 18-30), of which 32.8% cited inheritance. Further, while inheritance was only cited by 19.3% of ex-combatants in Uganda as their explanation for upward land access mobility, 53.5% cited regulated division of their land, such as inheritance, sharing and dividing, as the reason for their decreased access to land for cultivation. Female ex-combatants were significantly less likely than male ex- combatants (19.7% vs. 29.6%) to cite inheritance as a pathway to increased land access – which could suggest a lack of access to inheritance structures. Ex-combatants aged 18-30 were the most likely age demographic to cite inheri- tance (32.3%) as their pathway to increased land access. In terms of marriage, while in Uganda only 12.9% of ex-combatants cited marriage as a pathway to increased land access an examination of demographic subgroups reveals that only 3.4% of male ex-combatants cited marriage as their pathway to increased land access compared to 36.4% of female ex-combatants. In addition, 38.5% of those aged 18-30 cited marriage as their pathway to increased land access compared to 0.0% of those aged 31-40 or over 40. 126   With specific reference to Uganda, a more finely grained scale reveals that 10.4% had more land access, 63.4% had the same level of access, and 26.3% had less access. 127   Burundi and Rwanda are excluded from findings regarding explanations for access to more or less arable land due to lack of directly comparable data. 58 Though findings are scattered across Table 11: Ex-combatant Household Hunger the GLR countries, collectively they form a mosaic that suggests that How Frequently Do People in Your Household Go Hungry? capital is a primary enabler of ex- Always Often Seldom Never combatant land access mobility. For Male 13.50% 37.20% 27.80% 21.60% young ex-combatants and female- Female 10.00% 38.10% 33.90% 18.00% combatants, the two demographic Age 18-30 12.50% 40.80% 26.40% 20.30% subgroups least likely to see in- 12.60% 37.80% 27.90% 21.70% Age 31-40 creases in land access, inheritance Age Over 40 14.50% 32.40% 32.30% 20.80% and marriage also appear to play a distinct role. Disabled 9.80% 38.10% 33.60% 18.50% Not Disabled 13.20% 37.20% 28.30% 21.30% Livestock ownership excluding 11.60% 49.10% 20.70% 18.60% Burundi poultry was at 35.7% across the GLR DRC 21.00% 35.50% 24.50% 19.00% countries, though generally higher in Burundi (40.2%) and Uganda Republic of Congo 16.70% 43.20% 24.00% 16.10% (52.7%).128 Age showed a positive Uganda 2.70% 21.60% 45.30% 30.40% relationship to the likelihood of GLR Average 13.00% 37.30% 28.70% 21.00% owning livestock across the GLR countries – 33.5% of those aged 18-30, 35.1% of those aged 31-40 and 42.2% of those over 40. Increases in livestock in the last two years were cited by 54.2% of ex-combatants across the GLR with a similar distribution across demo- graphic lines. Of those ex-combatants who had no livestock, poverty and lack of resources was the most common explanation (56.7%) followed by insecurity due to conflict (11.8%). Beyond access to land for cultivation and the ownership of livestock, another important indicator of food security is the level of household hunger and nutrition – presented in Table 11. Across the GLR countries 13% of ex-com- batants explained that people in their household always go hungry, 37.3% they often went hungry, 28.7% that they seldom went hungry and 16.1% that they never went hungry.129 The exception to this distribution is Uganda, where the majority seldom went hungry (45.3%). In general rates of household hunger were very even across demographic lines. In regards to household nutrition 24.6% of ex-combatants said that in the last two years nutrition had improved, 43.8% that nutrition was unchanged and the remaining 31.6% that nutrition had worsened.130 Again, the only excep- tion is Uganda, in which the proportion of ex-combatants with improvements in household nutrition was greater (36.9%). Of those ex-combatants for which household nutrition had gotten worse in the last two years disabled ex- combatants (39.2%) were more commonly represented compared to non-disabled ex-combatants (31%). 8.3 Summary Ex-combatants display a very high level of access to housing and land for cultivation across the GLR countries. The majorities of ex-combatants across the GLR countries are living in permanent housing, with a family or spouse and see their living situation as equal to their neighbors – and in this sense have reached considerable parity with community members. Assessing the security of their housing tenure, however, is more challenging. The variety of housing ownership structures that exist across, and within, the GLR countries create unique contexts to land tenure 128   Rwanda is excluded from findings regarding livestock ownership due to lack of directly comparable data. 129   Rwanda is excluded from findings regarding household hunger due to lack of directly comparable data. 130   Rwanda is excluded from findings regarding household nutrition due to lack of directly comparable data. 59 security. Owing to a lack of systematic capture of land ownership structures a direct comparison is not possible here. However, what we can note is that lack of housing and land title does not necessarily indicate a lack of tenure security – there are other structures such as clans that can insure land tenure. While at a cross-country level ex-combatants have exhibited a significant level of upward mobility in terms of their access to land for cultivation there remains a divergence between DRC and RoC, on the one hand, which showed very high rates of increased access to arable land and Burundi and Uganda, on the other, which displayed much lower rates of increased access to arable land. Accounting for this divergence is puzzling. While the absolute availability of land is important dimension of increased land access (for example in Rwanda land scarcity is a well identified issue, while in DRC there are large tracts of uninhabited land) it is likely that the local dynamics of negotiating access to land through various pathways is equally if not more important component (for example in DRC ex-combatants and community members alike must navigate between both customary and statutory land access regimes that can stand in direct contradiction to each other).131 As gaining access to land is a key pathway to ex-combatants’ economic stability, food security, and contribution to the community then further investigation of this divergence could prove important for future programming. While access to arable land and livestock ownership are generally considered important indicators of the food security of ex-combatants, it appears as though there is little correlation between these indicators and ex-combatants’ levels of household hunger and nutritional improvement. While ex-combatants’ access to arable land was very high across the GLR countries, nearly on par with community members, they were significantly more likely to face hunger and nutrition problems. Future inquiry into the sources and nature of household hunger and nutrition problems to nuance these findings could prove insightful. 8.3.1 Vulnerable Subgroups Female ex-combatants face a unique set of challenges in regards to access to arable land. Female ex-combatants across the GLR are less likely than male ex-combatants to see increases in their access to land for cultivation (26.5% vs. 39.8%) – the low level of female ex-combatants’ land access mobility is even more stark when they are compared to female community members (45.4%), who are themselves on par with male community members. Scattered evidence suggests that, as with male ex-combatants, female ex-combatants view capital as their primary pathway to increased land access. However, as female ex-combatants are the least likely group to see increases in their access to land this could suggest that they also face considerable barriers in access to capital. There is evidence to suggest that female ex-combatants experience additional barriers to land access mobility, especially in terms of (I) inheritance dynamics and (ii) marriage. Female ex-combatants less frequently cite inheritance as a pathway to land access mobility than male ex-combat- ants and female community members – who are on par with male community members. This could suggest that female ex-combatants face challenges in accessing land inheritance structures that are open to not only community members, but male ex-combatants as well. Lack of access to capital and inheritance structures for female ex-combat- ants is accentuated further when contextualized against marriage dynamics. Female ex-combatants in Uganda are more than ten times as likely to cite marriage as their pathway to increased land access compared to males – however (as discussed in section 7.2) female ex-combatants remain the least marrying demographic group with the weakest prospects for marriage in the future. Collectively this evidence suggests that female ex-combatants face a diverse range of barriers to land access. Future study to confirm and nuance these findings could prove beneficial for devel- oping gender focused reintegration programming. 131   Some scholars have posited that local – national contradictions in land access and ownership structures have played a role in shaping new power structures in effect shaping and sustaining insecurity in some parts of the GLR. See for example: Huggins and Clover (2005) From the Ground Up: Land Rights, Conflict and Peace in Sub-Saharan Africa, ISS Africa 60 Young ex-combatants (aged 18-30) tend to be a step behind their elder peers in terms of many housing, land access, and food security indicators. Young ex-combatants are less likely to own their housing, less likely to have livestock, and less likely to see increases in their access to land. However it appears that these disadvantages, as with marriage and education, may be a product of their years lost in conflict – as they now struggle to make parity with elder ex- combatants and show a clear trajectory of improvement – most notably in terms of accessing marriage, and in turn the familial unit, and access to land for agricultural production, which is tied to the primary economic pathway for ex-combatants across the GLR: small scale agriculture. Supporting the findings that disabled ex-combatants are slightly more likely to be married, so too are they slightly more likely to be living in household with a family. It appears that the majority of disabled ex-combatants fall in line with line with their non-disabled peers in terms of housing, access to land, and livestock. However, few those who do fall behind do so at varying levels – likely commensurate to their particular level of disability. Overall, disabled ex-combatants saw similar levels of access to land for cultivation, but fewer increases in their land access in the years prior to sampling. 8.3.2 Unique Country Trends While across the GLR countries ex-combatants were most likely to be living with the same family as prior to conflict, Rwanda and RoC stand out from this trend. In Rwanda ex-combatants were more likely to be living with a family, but one different from prior to conflict. Have ex-combatants in Rwanda faced challenges in reintegrating into the same familial unit as prior to conflict? While there is no clear evidence in this study, it is possible that this may in part be a product of the 1994 genocide in Rwanda, and the prolonged period of time that ex-combatants have been away from communities since the first and second Congo Wars. By contrast, in RoC ex-combatants were more likely to be living with a spouse (though unmarried, as detailed in section 7.2). Have ex-combatants in RoC been more success- ful in accessing the familial unit, even if it is not officiated in marriage? As accessing the familial unit is understood as a key pathway to reintegration, further investigation into these diverging trends could prove instrumental. 61 9. Economic Issues A ttaining a level of economic stability through employment, access to credit and participation in economic as- sociations are seen as key elements to the economic prospects of ex-combatants and essential for peace and development. As such, the analysis here is presented in five main parts: (i) an examination of ex-combatants’ employment statuses and general outlooks on employment; (ii) an examination of the barriers that non-econom- ically active ex-combatants face to gaining a stable economic status; (iii) an examination of female ex-combatants’ specific economic issues; (iv) an examination of ex-combatants’ levels of income, savings, and access to credit as in- dicators of their general economic stability and ability to leverage economic opportunities; and (v) an examination of ex-combatants’ level of engagement with economic associations as an extended support / opportunity network. In the context of the severe development challenges that characterize the GLR countries, attaining economic rein- tegration (parity with community members) and economic stability may not necessarily be the same thing. Thus, to truly identify the economic challenges that are specific to ex-combatants, we must understand ex-combatants economic prospects in relation to the wider community. As such, this section should be read in conjunction with section 15 on economic issues in the Community Dynamics Comparative Survey and Analysis in Annex II of this report. 9.1 Economic Status and History Concerning employment status, at a cross-country level prior to conflict ex-combatants were most commonly studying or training (37.6%), self-employed in agriculture (26.3%), or unemployed (12.9%).132 At the time of de- mobilization the number of ex-combatants studying or training had dropped to 2.2%. Those who had previously been studying or training prior to conflict were effectively absorbed into the categories of self-employed in agricul- ture (which grew to 33.3%), unemployment (which grew to 31.1%), and employed working in the public sector (which grew from 3.1% prior to the conflict to 11% at demobilization. At the time of sampling unemployment had shrunk to 21.3% and the number of ex-combatants working in the public sector had shrunk to 1.7%. These changes in ex-combatant employment status continued to be absorbed into the categories of self-employed in agriculture which grew to 36.7% and other self-employed in non-agricultural services categories which had grown to 10.4% from the time of demobilization (see Table 12 below) – RoC is an exception to this trend towards self-employment in agriculture with retail instead being the primary pathway. Though levels of employment varied from country to country, with each GLR country ex-combatants followed the same arc in their employment trajectory – a spike of unemployment at the time of demobilization, to a drop in unemployment at the time of sampling that was slightly worse than pre-conflict levels. This unemployment trend coupled with a continual growth in self-employment in agriculture, services, and retail. In examining the demographic trends in employment status across the GLR countries at these three time points we can observe some trends. Female ex-combatants are slightly more frequently unemployed than male ex-combatants prior to conflict and at demobilization, though slightly less so at the time of sampling. Though Rwanda, where 54.5% of female versus 38.2% of male ex-combatants were unemployed at the time of sampling, stood apart in this regard. Younger ex-combatants (age 18-31) are most frequently studying compared to other age groups at all time points. 132   Rwanda is excluded from findings regarding employment status prior to demobilization and at demobilization and Burundi is excluded from findings regarding employment at demobilization due to lack of directly comparable data. 62 Table 12: Ex-Combatant Economic Status at Three Time Points GLR Ex-Combatant Employment at Three Time Points informal economy i.e. agriculture –manufac- Working in the home for employer. private income through sub- agriculture –services Studying or Training for employer. public economically active economically active Employed working Employed working Employed working Self-employed non Self-employed non Self-employed non in or reliant on the for employer. agri- Hustle or involved agriculture –retail in informal sector sistence activities Retired. but not Supplementing Self-employed Housewife or Unemployed agriculture culture turing sector sector Employment Status Prior Conflict Male 12.80% 3.90% 5.60% 3.40% 26.20% 6.10% 3.20% 0.60% 0.10% 0.00% 37.20% 0.70% 0.00% Female 13.00% 2.30% 2.50% 1.30% 26.90% 4.00% 2.90% 0.50% 0.50% 0.50% 40.20% 5.30% 0.00% Age 18-30 12.20% 3.30% 3.90% 0.80% 16.90% 3.80% 2.10% 0.30% 0.00% 0.00% 55.00% 1.80% 0.00% Age 31-40 14.10% 4.90% 6.50% 3.20% 28.80% 7.10% 4.50% 0.70% 0.30% 0.00% 29.20% 0.80% 0.00% Age Over 40 10.70% 3.20% 6.60% 7.40% 43.00% 8.80% 4.20% 1.10% 0.30% 0.30% 13.10% 1.40% 0.00% Disabled 7.80% 3.60% 3.30% 1.90% 32.00% 4.10% 2.40% 0.20% 0.80% 0.80% 40.30% 2.90% 0.00% Not Disabled 13.30% 3.70% 5.30% 3.20% 25.80% 6.00% 3.30% 0.60% 0.10% 0.00% 37.50% 1.30% 0.00% Burundi 7.10% 2.50% 5.90% 1.80% 26.60% 7.60% 4.00% 0.60% 0.00% 0.00% 42.80% 1.20% 0.00% DRC 26.50% 2.30% 5.30% 2.90% 9.60% 7.00% 5.00% 1.40% 0.00% 0.00% 37.70% 2.20% 0.00% RoC 8.60% 25.40% 15.10% 9.20% 41.60% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% Uganda 6.10% 0.20% 1.50% 2.90% 38.60% 4.40% 1.50% 0.00% 0.50% 0.20% 42.80% 1.20% 0.00% GLR Average 12.90% 3.70% 5.20% 3.10% 26.30% 5.80% 3.20% 0.60% 0.10% 0.10% 37.60% 1.40% 0.00% Employment Status at Demobilization Male 30.30% 1.60% 5.00% 12.30% 32.70% 6.40% 4.00% 0.50% 0.20% 0.00% 5.40% 0.90% 0.60% Female 34.10% 0.40% 2.10% 5.90% 35.70% 5.70% 2.60% 0.60% 1.30% 0.00% 4.10% 7.60% 0.00% Age 18-30 43.00% 0.60% 4.70% 8.20% 21.80% 4.80% 2.80% 0.40% 0.00% 0.00% 9.40% 4.10% 0.20% Age 31-40 29.30% 1.20% 3.40% 10.60% 34.20% 8.20% 6.00% 0.90% 0.00% 0.00% 4.20% 1.30% 0.60% Age Over 40 16.40% 2.40% 3.10% 13.60% 49.20% 7.20% 3.40% 0.50% 1.10% 0.00% 1.40% 1.10% 0.60% Disabled 42.30% 0.50% 1.30% 5.20% 35.80% 2.60% 0.50% 0.30% 2.30% 0.00% 3.80% 2.60% 2.80% Not Disabled 29.60% 1.40% 4.80% 11.70% 33.10% 6.70% 4.10% 0.60% 0.10% 0.00% 5.30% 2.20% 0.20% DRC 32.20% 2.20% 7.70% 21.30% 12.20% 8.40% 6.90% 1.10% 0.00% 0.00% 4.80% 2.80% 0.40% RoC 28.60% 0.00% 0.00% 0.00% 14.30% 42.90% 0.00% 0.00% 0.00% 0.00% 14.30% 0.00% 0.00% Uganda 30.00% 0.50% 1.20% 0.70% 54.80% 3.70% 0.50% 0.00% 0.70% 0.00% 5.50% 1.70% 0.50% GLR Average 31.10% 1.40% 4.50% 11.00% 33.30% 6.30% 3.70% 0.60% 0.40% 0.00% 5.20% 2.20% 0.50% Employment Status at Current Male 21.70% 4.50% 6.90% 1.70% 35.80% 10.60% 7.00% 1.40% 0.50% 1.70% 6.70% 0.80% 0.60% Female 18.20% 2.10% 1.20% 1.80% 42.50% 9.10% 7.10% 1.50% 1.00% 0.50% 4.50% 10.10% 0.60% Age 18-30 19.40% 3.10% 7.50% 1.10% 33.60% 10.60% 7.90% 1.00% 0.60% 1.20% 10.40% 3.50% 0.00% Age 31-40 23.30% 5.70% 5.50% 2.00% 35.80% 10.50% 8.60% 1.70% 0.20% 2.30% 3.10% 1.10% 0.30% Age Over 40 20.00% 4.50% 4.80% 2.60% 45.90% 10.50% 3.80% 1.90% 1.10% 1.30% 0.90% 0.80% 1.90% Disabled 29.20% 5.10% 3.70% 1.00% 38.00% 5.40% 5.20% 1.90% 1.10% 2.50% 3.30% 2.20% 1.50% Not Disabled 20.10% 4.10% 6.50% 1.80% 36.60% 11.10% 7.30% 1.40% 0.50% 1.40% 6.90% 2.00% 0.40% Burundi 13.70% 4.10% 11.00% 2.30% 34.50% 12.20% 13.60% 1.00% 0.00% 0.00% 5.70% 1.60% 0.20% DRC 28.40% 3.60% 5.90% 1.70% 20.40% 16.40% 7.80% 2.50% 0.00% 0.00% 9.10% 3.50% 0.60% RoC 12.50% 0.00% 12.50% 0.00% 25.00% 50.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% Rwanda 38.70% 10.40% 4.10% 1.10% 22.30% 3.60% 2.50% 2.20% 1.60% 6.30% 5.80% 1.10% 0.30% Uganda 9.20% 0.50% 2.70% 1.70% 66.20% 6.70% 2.70% 0.20% 1.00% 1.20% 5.00% 1.70% 1.00% GLR Average 21.30% 4.20% 6.10% 1.70% 36.70% 10.40% 7.00% 1.40% 0.60% 1.50% 6.40% 2.10% 0.60% The use of the phrase “at three time points” indicates that respondents were surveyed at one time point with questions regarding three different time points. 63 Older ex-combatants (over the age of 40) are most frequently in the self-employed agriculture group at all time points. Disabled ex-combatants were more frequently unemployed at the time of demobilization and the time of sampling compared to non-disabled ex-combatants. These trends are visible in Table 12. Of those ex-combatants who were unemployed at the time of sampling the explanations most commonly given were lack of work opportunities (61.2%) followed by financial problems (12.2%).133 Uganda departed from this cross- country trend and instead health and disability (46.9%) was most commonly cited as the reason for not working, followed by financial problems (21.9%). Within demographic categories female ex-combatants were slightly less likely to perceive their unemployment as a result of a lack of opportunity (44.1%) and slightly more likely to view it as a result of financial problems (21.3%) or lack of skills (14.5%) than male ex-combatants (63.8%, 10.9% and 7.6% respectively). Disabled ex-combatants were much more likely to perceive health and disability constraints (58.4%) as their primary reason for unemployment compared to non-disabled ex-combatants (3.2%). On average across the GLR countries 31.7% of ex-combatants relied on more than one income earning activity.134 RoC, where 93.6% of ex-combatants relied on more than one income generating activity, departed dramatically from this cross-country trend. Female ex-combatants were slightly less likely to rely on multiple income generating activities (27.1%) compared to male ex-combatants (32.5%). Again, RoC is the exception to the gendered trend for multiple income sources as 100% of female ex-combatants relied on multiple income sources as compared to 92.8% of males. Disabled ex-combatants were slightly less likely to rely on multiple income generating activities (25.6%) compared to non-disabled ex-combatants (32.1%). On average across the GLR countries, 40.6% of ex-combatants returned to their pre-conflict employment / type of work. On average younger ex-combatants were less likely to return to their previous field of work or employment: 28.9% of those aged 18-31 versus 40.1% of those 31-40 and 56.4% of those over 40. Viewed in the context of age at mobilization into conflict, though approximate, it is perhaps understandable that younger ex-combatants do not return to the same employment type – indeed, as 64% of ex-combatants aged 18-30 were under the age of 18 at the time of mobilization they may not yet have had an established employment type. Further, those ex-combatants aged 18-30 were also slightly more likely to be studying or training at the time of sampling, likely returning to study inter- rupted by mobilization (10.4% vs. 3.1% of those aged 31-40 and 0.9% of those aged over 40). When asked to explain why they had chosen to return to their previous line of work after conflict in DRC and RoC ex-combatants most commonly cited three key explanations: (i) lack of other opportunities (51.8%); followed by (ii) that it was a reliable job (18.6%); and (iii) out of economic necessity to take care of the family (16.4%).135 Across the GLR countries ex-combatants communicated that they on average would be very willing to move to another part of their own country for a better job (75.4%) – though significantly lower in Uganda (40.7%).136 Though there is no cross-country data for comparison on explanations for ex-combatants’ attitudes towards migra- tion, looking at Uganda alone may provide some initial insights. The most common explanation for willingness to migrate in Uganda was that ex-combatants were willing to move for financial reasons and the prospect of improving their standard of living (58.4%). A smaller proportion of ex-combatants (15.5%) was bored of their environment 133   Rwanda is excluded from findings regarding reasons for unemployment due to lack of directly comparable data. 134   Rwanda is excluded from findings regarding multiple incomes sources due to lack of directly comparable data. 135   Questions regarding ex-combatants reasons for returning to the same job as prior to conflict were only asked in DRC and RoC. 136   This lower willingness to migrate among ex-combatants in Uganda may be related to their overall higher levels of social capital compared to other GLR countries – discussed more in depth in section 10. Rwanda is excluded from findings regarding willingness to migrate for work due to lack of directly comparable data. 64 and wanted a life change. In Uganda of those ex- Table 13: Ex-Combatant Perception of Relative combatants who were not willing to move for a Difficulty of Finding Employment better job the most common explanation (29.6%) Do you think ex-combatants find it harder than was that they had a lack of education or qualifica- others to get a job? tions followed by having family responsibilities Yes No that prevent them from moving (26.6%). Across Male 65.20% 34.80% the GLR countries female ex-combatants were Female 61.20% 38.80% considerably less likely to be willing to migrate for a job than male ex-combatants (57.6% vs. Age 18-30 71.50% 28.50% 78.4%) – though there is no clear explanation for Age 31-40 63.00% 37.00% why.137 Age Over 40 56.50% 43.50% Disabled 72.90% 27.10% Across the GLR countries, 64.6% of ex-com- Not Disabled 64.00% 36.00% batants perceived that they have a harder time Burundi 83.60% 16.40% finding a job than community members.138 It DRC 50.30% 49.70% appears as though there is a division between Republic of Congo 42.70% 57.30% Burundi and Uganda on the one hand, where ex- Uganda 78.70% 21.30% combatants frequently perceived that they have a harder time than community members, and DRC GLR Average 64.60% 35.40% and RoC on the other, where this frequency was still significant but considerably lower than in their neighbors to the east. These findings here are presented in Table 13 above. Age showed a negative relationship to the likelihood of thinking that ex-combatants have a harder time finding employment. Of those ex-combatants who thought that ex-combatants find it more difficult to find a job than community members there were diverse explanations across the GLR countries however, the common thread through all countries was stigma or distrust towards ex-combatants at varying levels – though in Uganda the most common explanation was ex-combatants’ low education levels (59.7%).139 A table of the proportion of ex-combatants from each GLR country Table 14: Ex-combatant Stigma / Distrust as a Barrier to Gaining Employment Stigma / Distrust is the Reason Ex-Combatants Find it Difficult to Gain Employment Burundi DRC RoC Uganda GLR Total Male 59.1% 43.9% 88.5% 18.5% 52.5% Female 53.9% 40.0% 100.0% 14.3% 52.5% Age 18-30 60.8% 39.3% 91.9% 9.5% 50.4% Age 31-40 54.3% 46.5% 88.0% 26.3% 53.8% Age Over 40 53.5% 45.2% 86.0% 20.4% 51.3% Disabled 41.8% 42.3% 90.0% 14.8% 48.1% Not Disabled 60.0% 43.4% 88.7% 18.1% 52.6% Country Average 58.7% 43.4% 88.9% 17.6% 52.2% 137   In Uganda both male and female ex-combatants both identify lack of education / qualifications and familiy responsibities as the primary reasons for unwillingness to migrate at almost identical levels. Uganda is the only GLR country in which this question was asked. 138   Rwanda is excluded from findings regarding perceptions of relative challenges of finding a job due to lack of directly comparable data. 139   Rwanda is excluded from findings regarding explanations for why ex-combatants find it more difficult to find a job due to lack of directly comparable data. 65 that cited stigma or distrust of ex-combatants as the reason why they find it more difficult than non-ex-combatants to find a job is presented in Table 14. On average across the GLR countries, 52.2% of ex-combatants saw stigma / distrust as a barrier to employment – though this was higher in RoC (88.9%) and lower in Uganda (17.6%).140 Concerning outlook for economic prospects in the future, across the GLR countries 73.7% of ex-combatants per- ceived that their economic situation would improve in the near future. Across demographic lines the perceptions of ex-combatants about their economic prospects in the future were remarkably even. However, across and within the GLR countries, disabled ex-combatants perceived slightly weaker economic outlooks (66.5% versus 74.3% at a total sample level). When asked to explain the main reasons for if they perceived their economic situation improving in the future ex- combatants across the GLR countries gave a wide range of responses.141 Very generally speaking, we can say that in Uganda ex-combatants with both positive and negative outlooks for the future saw this as tied to their ability to participate and produce in agriculture. In contrast, in Burundi, DRC and RoC ex-combatants more commonly expressed a range of explanations for positive and negative outlooks more closely tied to their attainment of employ- ment and capital. Across all countries disabled ex-combatants saw health as a key barrier to their economic future. Looking specifically across Burundi, DRC and RoC we can observe that on average ex-combatants work 9.34 months of the year in paid employment – a proportion roughly reflected across all three countries.142 However in contrast, when looking at the number of months that ex-combatants spend participating in unpaid labor, for example subsistence farming or labor in trade for food or housing, there is a division that emerges. In DRC and RoC the majority of ex-combatants spend on average 3.97 months in unpaid labor through the year. However in Burundi the ex-combatants work 10.93 months a year in unpaid labor; in fact the vast majority (78.4%) spend 12 months of the year working in unpaid labor. By cross tabulating months of the year spent in paid versus unpaid labor we find that 83.8% of those who spend 12 months of the year in unpaid labor (heavily represented in Burundi) do this in addition to working 12 months of the year for paid labor. Of those who work for paid labor for 9 months of the year (heavily represented in DRC and RoC), 96.2% do so in addition to working 9 months of the year for unpaid labor. These trends are likely indicative of regional and seasonal farming and employment practices. 9.2 Non-Economically Active Ex-Combatants on Employment Issues When non-economically active ex-combatants across the GLR countries are asked how they get by when they are not working the most common responses are: (i) 29.3% reply that they rely on their family cash contributions; (ii) 18.6% have to borrow money; and (iii) 12.8% say they just find a way to cope.143 Looking within gender demograph- ics we can see that female ex-combatants more commonly rely on family cash contributions (49.9%) compared to male ex-combatants (26.4%). Younger ex-combatants (age 18-30) are also more likely to rely on family cash contri- butions (37.1% vs. 22.1% of those 31-40 and 26.8% of those over 40). At a cross-country level of those ex-combatants that are not economically active, 33.2% of them feel that being an ex- 140   The analysis throughout annex I & II suggest that stigma and distrust are considerably less prominent in Uganda – it is possible that this is related to the dynamics of ex-combatant return in Uganda – where ex-combatants who were abducted return to communities and are seen as both victims and perpetrators. In some communities this dynamic can play a role in greater overall acceptance of ex-combatants in Uganda. In addition, the extensive use of traditional reconciliation ceremonies in Uganda, though not a part of reintegration programming, may play a role in explaining this stark contrast against RoC. 141   Rwanda is excluded from findings regarding reasons for outlook on future economic situation due to lack of directly comparable data. 142   Rwanda and Uganda are excluded from findings regarding annual time spent working for pay and without pay due to lack of directly comparable data. 143   These are only the three most common explanations across the GLR countries. Rwanda is excluded from findings on how non-economically active ex-combatants get by due to lack of directly comparable data. 66 combatant contributes to them not working. Table 15: Ex-combatant Status Contributes to Unemployment However, a closer examination of these percep- tions within individual GLR countries show Do you feel that being an ex-combatant contributes to you not working? a sharp split between Burundi and Uganda in which 70.9% and 66.7%, respectively, felt that Yes No their ex-combatant status contributed to their Male 34.50% 65.50% unemployment, versus DRC and RoC where Female 25.00% 75.00% only 22% and 21.1% respectively. Future in- Age 18-30 38.90% 61.10% vestigation into the reason for this divergence Age 31-40 32.10% 67.90% in ex-combatants’ perception of ex-combatant Age Over 40 27.60% 72.40% status playing a role in unemployment would Disabled 43.50% 56.50% add considerable explanatory value in future Not Disabled 31.90% 68.10% studies.144 Burundi 70.90% 29.10% Across and within the GLR countries male DRC 22.00% 78.00% ex-combatants and younger ex-combatants Republic of Congo 21.10% 78.90% more commonly see their ex-combatant status Uganda 66.70% 33.30% as contributing to their unemployment. The GLR Average 33.20% 66.80% extent of these trends can be seen in Table 15. Attempting to account for the sharp split between Burundi and Uganda, on the one hand, and DRC and RoC, on the other, in the perception of ex-combatant status playing a contributing role in the unemployment of non-econom- ically active x-combatants is challenging. In Burundi the most common explanation for why ex-combatant status contributes to unemployment is unspecified political problems (44.6%) followed by stigma and distrust in ex-com- batants (34.7%). In Uganda, lack of skills and education are the most common explanations (43.5%), followed by poor health (26.1%) and stigma/ distrust in ex-combatants (21.7%). By contrast, in DRC and RoC stigma accounts for the vast majority of explanations for why ex-combatant status contributes to unemployment (80.2% in DRC and 79.6% in RoC). In summary, in DRC and RoC where the likelihood of perceiving ex-combatant status as con- tributing to unemployment was dramatically lower than in other GLR countries – the perception that stigma and distrust in ex-combatants was the reason why ex-combatant status contributed unemployment was dramatically higher. While the relationship between perceptions of stigma and the perception of ex-combatant status playing a contributing role to being non-economically active is unclear here, stigma should at the very least be flagged as an important dimension. At a cross-country level there is a very even split in the perceptions of non-economically active ex-combatants on their future prospects of gaining employment – 50.4% saying that they had a good chance of finding a job in the future, 2.4% saying they had a neither good or bad chance, and 48.6% saying that they had a poor chance. Disabled ex-combatants consistently expressed a less positive outlook towards future employment – 25.4% of disabled ex- combatants across the GLR countries having a positive outlook versus 54.3% of non-disabled ex-combatants. While there is no data to directly compare across GLR countries, it is notable that in Uganda 100% of disabled ex-combat- ants explained their positive outlook on gaining employment in the future on improved health and/or healing. 9.3 Female Ex-Combatants on Employment Issues Of female ex-combatants who were not economically active 36.5% feel that they are discriminated against as a female 144   To further complicate these findings on the role that ex-combatant status plays in gaining employment for those who are non-economically active, they stand in contrast to the similar findings on the role that stigma / distrust plays in gaining employment presented in Table 14. 67 – though in Burundi this number was significantly higher (66%) and in Uganda significantly lower (16.7%).145 Simi- larly, of female ex-combatants who are economically active 24.5% feel they are discriminated against as a female in the workplace. In both instances female ex-combatants between the ages of 18 and 30 are the most likely age demographic to perceive discrimination (44.7% of those unemployed and 23.2% of those employed). In addition, in both of these instances 34.8% perceive their status as not just a female, but a female ex-combatant is related to the discrimination they encounter. While there is no data for direct comparison across GLR countries as to who female ex-combatants see as the main people discriminating against them, Uganda can offer some leads for further investigation. In Uganda 50% of un- employed female ex-combatants see female employers or bosses as the main group discriminating against them, the other 50% see everyone as discriminating against them. Of those female ex-combatants who were employed 30% saw female co-workers as the main group discriminating against them, followed by 15.4% who saw all employers at the main group discriminating against them, 15.4% who saw male co-workers discriminating against them, and 15.4% who saw everybody discriminating against them. What is notable is that, at least in the case of Uganda, in both instances of female ex-combatants who are employed and those who are unemployed, the group most commonly perceived as discriminating against them is other females – be they employers or co-workers. This point could be related to female community members’ overall higher levels of fear, and perhaps in turn discrimination, surrounding the return of ex-combatants discussed in section 17.2. 9.4 Disabled Ex-Combatants on Employment Issues Of disabled ex-combatants who are not economically active 62.4% feel they are discriminated against as a disabled person and 37.6% feel they are not.146 Of those disabled ex-combatants who are economically active the propor- tions of those who feel they are discriminated against is almost perfectly inverse, with 34.2% saying that they per- ceived being discriminated against and 62.5% saying they did not. When asked if discrimination was related to their ex-combatant status 51.1% of disabled ex-combatants perceive that this discrimination has to do specifically with them being a disabled ex-combatant rather than merely disabled. Female disabled ex-combatants were less likely to perceive discrimination linked to their ex-combatant status, 37.5% versus 58.7%. Again, as with the case of female ex-combatants on employment issues, there is no data to directly compare across the GLR countries as to who disabled ex-combatants see as the main groups discriminating against them. However, again looking at data from Uganda can offer some initial insights. In Uganda 80% of non-economically active disabled ex-combatants who perceived discrimination see all employers or bosses as discriminating against them, with the remaining 20% seeing everyone as discriminating against them. Of those economically active ex-combatants who perceived discrimination 40% saw the discrimination as coming primarily from all employers or bosses, 20% from male co-workers, 20% from all co-workers, and 20% from everybody. Confirming these trends across the GLR coun- tries would require further triangulation. 9.5 Income, Savings and Access to Credit In the context of the severe development challenges that characterize most of the Great Lakes Region, ex-combat- ants’ economic statuses are a good starting point for understanding basic individual and household economic stabil- ity. However, a deeper examination of ex-combatants’ income, savings and access to credit can begin to reveal some about their ability, or inability, to move beyond mere subsistence by leveraging economic opportunities. 145   Rwanda is excluded from findings regarding perceptions of discrimination among both economically active and non-active disabled ex-combatants due to lack of directly comparable data. 146   Rwanda is excluded from findings regarding perceptions of discrimination among both economically active and non-active female ex-combatants due to lack of directly comparable data. 68 Across the GLR countries 49.6% of ex-com- Table 16: Ex-Combatant Sole Breadwinner Status batants identified as the sole breadwinner of their household with the remaining 50.4% Are you the sole, or only, breadwinner or do others in your household also earn an income? saying that their household relied on multiple incomes. As is visible in Table 16, in Rwanda Sole Others Assist and DRC ex-combatants are notably more Male 53.00% 47.00% likely to identify themselves as the sole bread- Female 29.10% 70.90% winner (79.6% and 71.6% respectively) – Age 18-30 41.10% 58.90% generally an indicator of household income Age 31-40 56.10% 43.90% instability. Across the GLR countries female Age Over 40 56.70% 43.30% ex-combatants were dramatically less likely to Disabled 42.40% 57.60% identify as the sole breadwinner (29.1%) when Not Disabled 50.80% 49.20% compared to male ex-combatants (53%). In Burundi 42.90% 57.10% some GLR countries this disparity between DRC 71.60% 28.40% male and female ex-combatants was even more Republic of Congo 63.80% 36.20% accentuated – for example in Rwanda 100% of female ex-combatants said their household Rwanda 79.60% 20.40% relied on multiple incomes compared to 19.6% Uganda 40.40% 59.60% of male ex-combatants. GLR Average 49.60% 50.40% Of those ex-combatants who identify themselves as the sole breadwinner in their household across the GLR coun- tries 39.3% say that they usually have to borrow money to meet their monthly household expenses, 22.4% say that they usually break even, 20.4% rely on family money transfers, 13.5% usually have to use past savings, and only 4.4% have money left over.147 As displayed in Table 17, these trends were remarkably durable within each of the GLR countries with the exception of RoC in which a similar proportion of ex-combatants (41.2%) had to borrow to meet monthly expenses, but in contrast to the cross-country trend 41.2% of ex-combatants usually had money left over.148 Across the GLR countries, female ex-combatants were more likely to rely on family money transfers (27.2%) when compared to male ex-combatants (19.1%), and less likely to use past savings to meet monthly expenses (7.8%) when compared to male ex-combatants (14.6%). Younger ex-combatants were also more likely to rely on family money transfers than their elder peers (27.8% vs. 12.4% of those 31-40 and 15.4% of those over 40). Of those ex-combatants who were sole breadwinners and did not earn enough to meet monthly household expenses across the GLR countries, they were on average short by 41% of their income.149 DRC and, to a larger extent, RoC sat below this cross-country average, with average sole breadwinner income shortages of 23% and 7% respectively. Across demographic lines these disabled ex-combatants had notably higher income shortages on average (52%) compared to non-disabled ex-combatants (40%). Of those ex-combatants who were sole breadwinners and did meet monthly household expenses there was a clear trend in which ex-combatants had a surplus on average of 22%.150 However, as is visible in Table 18, in RoC income surpluses were on average only 5%. Of those 49.6% of ex-combatants across the GLR countries who say that their household relies on multiple incomes, there was an average contribution of 46% of their total household income. As is visible in Table 19, Rwanda and DRC stand out with smaller average non-sole breadwinner household income contributions (35% and 37% respectively). 147   Rwanda is excluded from findings regarding sole household breadwinners due to lack of directly comparable data. 148   Rwanda is excluded from findings regarding sole household breadwinners due to lack of directly comparable data. 149   Rwanda is excluded from findings regarding monthly income deficits due to lack of directly comparable data. 150   Rwanda is excluded from findings regarding monthly income surpluses due to lack of directly comparable data. 69 Table 17: Ex-Combatant Sole Breadwinner Meeting Monthly Expenses At the end of each month, do you meet your household expenses? Usually have money Usually break even Usually have to use Rely on family Usually have to left over past savings money transfers borrow Male 4.80% 21.50% 14.60% 19.10% 40.10% Female 2.80% 27.20% 7.80% 27.20% 35.10% Age 18-30 3.60% 17.60% 14.10% 27.80% 36.90% Age 31-40 5.20% 22.30% 17.30% 12.40% 42.80% Age Over 40 5.10% 30.80% 8.70% 15.40% 40.00% Disabled 5.10% 28.90% 9.30% 22.60% 34.10% Not Disabled 4.30% 21.50% 14.10% 20.10% 40.00% Burundi 1.40% 13.90% 20.50% 18.90% 45.30% DRC 6.80% 22.60% 20.40% 14.60% 35.70% Republic of Congo 41.20% 0.00% 11.80% 5.90% 41.20% Uganda 2.50% 32.70% 6.50% 24.20% 34.10% GLR Average 4.40% 22.40% 13.50% 20.40% 39.30% Generally speaking female ex-combatants and younger ex-combatants contributed less on Table 18: Ex-Combatant Sole Breadwinner Average Monthly Income Surpluses and Deficits average of total household income (39% and 42% respectively). Average Monthly Income Average Monthly Income Shortage Surplus Since reintegration programming 31.7% of ex- Male 40% 22% combatants have had to borrow money to help Female 45% 21% meet their daily needs.151 Across age, gender Age 18-30 42% 22% and disability demographics all groups lay very Age 31-40 39% 22% closely to the total sample trend. In Burundi, Age Over 40 41% 22% DRC and RoC, of those who did borrow Disabled 52% 23% 34.6% borrowed from a friend, 28% borrowed Not Disabled 40% 22% from family, 11.7% borrowed from community leaders – only 4% borrowed from some form of Burundi 40% 26% formal credit institution. Ex-combatants aged DRC 23% 24% 18-30 were most likely to borrow from family Republic of Congo 7% 5% (34.8%) compared to those 31-40 (22.7%) and Uganda 50% 31% those over 40 (19.7%). GLR Average 41% 22% In terms of the use of funds borrowed since re- integration programming packages there were three key uses: (i) subsistence; (ii) business investment; and (iii) familial support. 34.6% of ex-combatants identified their first use of borrowed funds as mere subsistence, 22.5% as business investments, and 18.9% as assistance for their family.152 Similarly, 27.8% of ex-combatants identified their second use of borrowed funds as to assist their family, 24.7% as subsistence, and 11.3% as a business investment. As a third use of borrowed funds 17.1% used funds as subsistence, 15.1% as assistance to their family, and 10.9% 151   Due to the varying contexts of reintegration programming across the GLR countries this question should be treated as a broad indicator of the rate at which ex-combatants need to borrow money after the bulk of immediate reintegration assistance (including reinsertion) has passed. Rwanda is excluded from findings regarding money borrowing due to lack of directly comparable data. 152   Rwanda and Uganda are excluded from findings on the most common uses of reinsertion payments due to lack of directly comparable data. 70 as business investments. The drops in these Table 19: Ex-Combatant Average Non-Sole Breadwinner categories were absorbed into, among others, Household Income Contributions education for children (10.1%) and medical expenses (7.6%) – especially among disabled Average Non-Sole Breadwinner Household Income Contribution Percentage ex-combatants. These spending patterns for borrowed money overlap strongly with the Male 47% spending patterns of reinsertion payments in Female 39% section 11.1 – indicating a key set of immedi- Age 18-30 42% ate costs those ex-combatants across the GLR Age 31-40 51% countries face. Age Over 40 48% Disabled 41% Only 6.7% of ex-combatants across the GLR Not Disabled 47% countries have ever applied for micro-credit from a financial institution. This figure is re- Burundi 50% flected in all GLR countries except for Uganda DRC 37% and Rwanda where higher proportions of ex- Republic of Congo 48% combatants (18.4% and 13% respectively) had Rwanda 35% at some point applied for micro-credit. Ex-com- Uganda 43% batants over the age of 40 were the most likely GLR Average 46% age demographic to have applied for micro- credit (11%) when compared to those aged 31-40 (6.5%) and those aged 18-40 (4.6%). This trend is further ac- centuated in the cases of Uganda (27.9% vs. 13.6% and 9.8%) and Rwanda (18.6% vs. 11.5% and 10.3%) where the overall proportion of ex-combatants who had at some point applied for micro-credit was significantly higher. Of those ex-combatants who had applied for micro-credit at some point, 76.6% had had a successful application. The only GLR country that did not reflect this average was RoC in which only 28.6% of micro-credit applications were successful – the explanation for this is unclear. Female ex-combatants more commonly had successful micro credit applications (86.9%) than male ex-combatants (78.5%). There were no other consistent demographic trends across the GLR countries. 9.6 Economic Associations Across the GLR countries 37.3% of ex-combatants were currently involved in some form of micro-economic activity. Burundi departed most significantly from this trend with 78.6% of ex-combatants being involved in a micro-eco- nomic activity.153 In Uganda and Burundi female ex-combatants (39.4% and 100% respectively) were more likely to be currently participating in some form of micro-economic activity than male ex-combatants (32.5% and 76.9% respectively). In contrast, in DRC and RoC female ex-combatants (22.2% and 0.0% respectively) were less likely to be involved in some form of micro-economic activity (46.2% and 80% respectively). Since receiving reinsertion packages, 72.2% of ex-combatants across the GLR countries have never been a member of an economic association, 21.9% are currently members, and 5.9% have previously been members but are not currently.154 Across all GLR countries ex-combatants aged 18-30 were least frequently currently a member of an economic association (15.6%) compared to those aged 31-40 (23%) and those over 40 (29.1%). This could be an indication of older ex-combatants’ generally longer economic track record with economic associations and access to credit. Looking to Rwanda and Uganda we can see that local savings and credit associations and farmers associations 153   Rwanda is excluded from findings regarding current participation in micro-economic activities due to lack of directly comparable data. 154   Burundi is excluded from findings regarding membership in economic associations due to lack of directly comparable data. 71 are the varieties of economic associations Table 20: Ex-Combatant Economic Association Membership that ex-combatants are members of. Since reinsertion, have you ever been a member of Across the GLR countries the most an Economic Association? common benefits that ex-combatants Yes, have been a No, have never been Yes, am currently identified receiving as a member of their member previously, a member of an a member now economic associations were economic but not now economic association networking (34.2%) followed by social Male 6.10% 21.60% 72.40% networking (21.1%) and financial Female 4.60% 24.40% 71.00% support (17.9%).155 Ex-combatants in Age 18-30 4.20% 15.60% 80.30% Uganda were the least likely to cite social Age 31-40 6.50% 23.00% 70.50% and economic networking as the primary Age Over 40 7.20% 29.10% 63.80% benefit of economic associations (3.1% Disabled 7.10% 24.60% 68.40% and 4.6% respectively) and more likely Not Disabled 5.80% 21.60% 72.60% to cite financial support as the primary Burundi 0.00% 0.00% 0.00% benefit (42.4%). Female ex-combatants DRC 2.30% 19.70% 78.00% across the GLR countries were more Republic of Congo 7.90% 12.10% 80.00% likely to identify economic networking Rwanda 13.00% 36.60% 50.40% (42.5%) and less likely to identify social Uganda 2.70% 21.80% 75.50% networking (12.1%) than male ex-com- GLR Average 5.90% 21.90% 72.20% batants (33% and 22.4% respectively). Of those ex-combatants across the GLR countries who were members of an economic association, there were varying compositions of ex-combatant versus non-ex-combatant membership in the given economic association.156 As is visible in Table 21 below, at a cross-country level there was a typical bell curve distribution between categories of economic association membership composition. However, closer inspection within countries shows that there are diverging trends. What is noteworthy is that the level at which ex-combatants move into economic associa- tions with only other ex-combatants is low – suggesting that the majority of those ex-combatants who do join an economic association have the benefit of social interaction with community members, building social and economic networks, in addition to the economic benefits of associations. 9.7 Summary Across the GLR countries the general economic trajectory of ex-combatants is positive. The number of ex-combat- ants who are unemployed is shrinking – these ex-combatants are most commonly being absorbed into self-employ- ment in agriculture, followed by self-employment in non-agricultural business. In line with this positive trajectory, ex-combatants generally have a positive outlook on their economic situation in the future; 76.3% saying that they expected their situation to improve in the future. When ex-combatants explain this positive outlook they generally cite improved agricultural production and improved access to capital and credit – two explanations that can be tied to ex-combatants’ main paths of economic reintegration: self-employment in agriculture and small business. This signals that ex-combatants’ perceptions of their future economic situation and the pathways to attaining it are rooted in their collective trajectory towards self-employment in agriculture or small business. This generally positive economic trajectory has seen ex-combatants reach near parity, but slightly weaker across 155   Burundi is excluded from findings regarding the benefits of economic associations due to lack of directly comparable data. 156   Burundi is excluded from findings regarding the membership composition of economic associations due to lack of directly comparable data. 72 all indicators, in levels of economic stability compared to community members. Ex-combatants are more likely to be unemployed, less likely to meet their household expenses, and more likely to borrow from family than formal economic institutions to close this income gap than community members. Ex-combatants are considerably less likely to participate in micro-economic activities or belong to economic associations than community members – an indication of their considerably shortened economic track record and the time lost while in conflict for establish- ing themselves in formal economic institutions. Regarding ex-combatants perceptions of the barriers they face to gaining productive economic status, the majority of ex-combatants cite lack of opportunity, signaling that they generally identify the barriers to economic improve- ment as contextualized in larger development challenges that affect the entire community. Simultaneously, ex-com- batants perceive themselves as a disadvantaged group that have a harder time finding a job and are subject to stigma and distrust in the community. By contrast community members are much more likely to cite lack of access to credit and lack of skills as key barriers to their economic stability. It appears that ex-combatants understand the dual dimensions of the barriers they face to gaining a productive economic status – the larger context of severe development challenges that characterize the GLR countries, and the context of being an ex-combatant in this development setting – facing challenges with stigma and distrust in the community. Indeed, ex-combatants are far more likely to identify the social networking value of economic interac- tions, bringing about the slow set of social interactions that erode stigma and facilitate social reintegration, than community members. However, in this it appears that with weaker economic track records ex-combatants also fail to recognize capital and credit barriers to economic prosperity to the same extent as community members. In summary, ex-combatants’ economic trajectories are generally positive though in absolute terms they are disad- vantaged to community members. The barriers to reaching true parity with community members revolve around: (i) closing literacy, education and skill gaps with community members; (ii) establishing an economic track record; (iii) accessing credit and other financial institutions; and (iv) eroding stigma and distrust through the slow process of confrontation that social reintegration entails. 9.7.1 Vulnerable Subgroups In this analysis of the economic dimensions of reintegration, female ex-combatants continue to stand out as a key vulnerable group. Female ex-combatants are more likely to be unemployed than male ex-combatants and more likely to see lack of skills as among their core barriers to reaching economic stability – this aligns with earlier analysis re- vealing that female ex-combatants have significantly lower literacy and educational achievement levels compared to male ex-combatants. Female ex-combatants and community members alike report significant levels discrimination on the basis of gender as a barrier to gaining employment, though female ex-combatants report this at twice the rate of community members. This may suggest that stigma associated with ex-combatant status may have an amplifying effect on already entrenched gender inequalities present in the community. Interestingly while female community members that experienced gender-based discrimination identified it as coming primarily from males, female ex- combatants identified female community members as the main sources of discrimination. This point may be related to female community members’ overall higher levels of fear surrounding the return of ex-combatants to their com- munity discussed in section 17.2 of the community dynamics annex of this study. Collectively, these issues represent a clear set of challenges for female ex-combatants in achieving economic reintegration. Health is a key barrier to economic reintegration for disabled ex-combatants who are the most likely demographic group, of ex-combatants and community members alike, to be unemployed at the time of sampling across the GLR countries. Accordingly disabled ex-combatants are the least likely demographic group to have a positive outlook on their economic future. In addition disabled ex-combatants who are unemployed report high levels of discrimination in seeking employment on the basis of their disability (twice the proportion of females that perceive discrimina- tion). However – those who are employed perceive discrimination on the basis of their disability at half the rate (on 73 par with females). This may suggest that while there is clearly discrimination in terms of gender and disability, there is an amplifying effect that overall levels of stigma and distrust in the community have on these dynamics. 9.7.2 Unique Country Trends In terms of economic reintegration ex-combatants in the Republic of Congo display a number of unique trends – though they do not necessarily depart from the dominant narrative of economic reintegration for ex-combatants across the GLR. Most notably perhaps is that while self-employment in agriculture is still an important economic pathway for ex-combatants in RoC, self-employment in non-agricultural services is the dominant economic path. It is possible that migration to the urban capital of Brazzaville among ex-combatants has removed agriculture as a viable economic activity – lending some explanation to this trend of economic status tending away from the self- employment in agriculture. 74 10. Social Capital E xamining the social dynamics of ex-combatant reintegration requires the exploration of a range of concepts including: (i) social networks, (ii) trust, (iii) social cohesion, (iv) social inclusion, and (v) empowerment. Collectively these various concepts come together to represent social capital, essentially the idea that social networks have value, both tangible and intangible, for individuals and communities and are a key indicator of the overall social health of ex-combatants – and in turn their ability to leverage this social capital towards social and economic outcomes. Examining social capital can allow us some insights into the process of social reintegration that ex-combatants go through upon return to their communities. However, when looking at the complex social dynamics that ex-combatants experience we cannot draw meaningful insights without contextualizing these social dimensions with that of the community at large. Thus for optimal analytical value this section of the report should be read in conjunction with section 16 on social capital in the community dynamics annex of this report. 10.1 Networks and Sociability Across the GLR countries ex-combatants and community members are unlikely to be in many social groups – though community members are in slightly more. On average ex-combatants were in 0.46 social groups, while community members were in 0.63.157 Uganda stood apart from this overall GLR trend – community members averaging 0.93 social groups. Age showed a positive relationship to the average number of social groups among ex-combatants and community members alike, however this trend was much more pronounced among ex-combatants – in which young (18-30) ex-combatants have the lowest average number of social groups across all demographic groups (0.37). Across the GLR countries, 38.5% of ex-combatants said that the current number of social groups to which they are a member is greater than that of one year ago, 50.8% said the number is the same as one year prior and 10.7% said that their current number of social groups is less than it was one year ago. These proportions were reflected well within the GLR countries with the exception of Uganda, in which 85.3% said that their number of social groups had stayed the same in the last year. Across the GLR countries, female ex-combatants less frequently saw an increase in their number of social groups (23.1%) when compared to male ex-combatants (41.4%). Similarly, disabled ex-com- batants less frequently saw an increase in their number of social groups (24.4%) when compared to non-disabled ex-combatants (40.1%). Of ex-combatants, 32.6% were on a management or organizing committee for a local group or organization. Female ex-combatants were significantly less likely to be on a management committee (25.9%) compared to male ex-com- batants (34%). Ex-combatants over the age of 40 are most frequently on management or organizing committees (37.4%) compared to those 31-40 (36.8%) and those 18-30 (24.3%). The fact that older ex-combatants (over 40) have the most social groups on average and are most frequently on management committees is a broad indication of their social footing in the community. Inversely it flags younger ex-combatants as lagging behind in building a social foundation in the community. Generally speaking a large proportion of ex-combatants across the GLR countries had contact with their families (91.3%). However, DRC is a clear standout in this trend of high familial contact – only 62.1% reported having 157   Rwanda is excluded from findings regarding social networks due to lack of data. 75 Table 22: Ex-Combatant Frequency of Familial Contact Frequency of contact between community member and immediate family these days Daily Weekly Monthly Half yearly Once a year Occasionally Never Male 64.90% 13.10% 11.60% 1.20% 1.00% 7.70% 0.40% Female 76.50% 9.70% 4.20% 1.40% 1.50% 6.20% 0.50% Age 18-30 71.00% 10.00% 8.80% 1.50% 2.00% 5.90% 0.80% Age 31-40 59.90% 13.40% 13.60% 1.60% 0.30% 10.50% 0.60% Age Over 40 71.90% 13.20% 7.80% 0.30% 1.00% 5.80% 0.00% Disabled 88.20% 6.80% 1.10% 0.00% 0.00% 3.90% 0.00% Not Disabled 64.70% 13.20% 11.20% 1.40% 1.30% 7.90% 0.40% DRC 35.30% 20.00% 23.50% 1.10% 2.20% 17.90% 0.00% RoC 28.90% 24.30% 22.10% 3.00% 2.50% 18.00% 1.40% Uganda 92.70% 4.80% 1.80% 0.30% 0.30% 0.30% 0.00% GLR Average 67.00% 12.40% 10.20% 1.20% 1.10% 7.50% 0.50% contact with their families. What is even more notable though is that DRC is the only GLR Table 23: Ex-Combatant Desired Level of Familial Contact country in which ex-combatants have dramati- Is the current level of contact the maximum you wish cally more contact with their families than com- or could it be more frequent? munity members have with their own families Maximum Could be more frequent (62.1% vs. 31%). Male 71.00% 29.00% Female 80.40% 19.60% Of those ex-combatants who did have contact Age 18-30 71.10% 28.90% with their families, 67% across the GLR coun- Age 31-40 69.10% 30.90% tries had daily contact with their family, 12.4% Age Over 40 83.40% 16.60% had weekly contact, 10.2% had monthly contact, Disabled 83.00% 17.00% and the remainder had less frequent contact. 158 As visible in Table 22, Uganda stands out most Not Disabled 71.40% 28.60% clearly from the cross-country trend in this DRC 36.70% 63.30% instance, as 92.7% of ex-combatants in Uganda Republic of Congo 52.30% 47.70% who had contact with their family had daily Uganda 92.80% 7.20% contact with their family. In general female ex- GLR Average 72.80% 27.20% combatants had slightly higher levels of daily contact than male ex-combatants (76.5% vs. 64.9%). The only standout along gender demographic lines is in RoC, in which only 11.1% of female ex-combatants had daily contact with their family and 45.5% had contact with their family less frequently than monthly – as compared to male ex-combatants of whom 30.4% had daily contact and 22.2% had contact less frequently than monthly. Disabled ex-combatants were more likely to have daily contact with their families (88.2%) than non-disabled ex-combatants (64.7%). In all, ex-combatants across the GLR countries had daily contact with their families slightly more frequently than community members (67% vs. 63.9%). Across the GLR countries 27.2% of ex-combatants thought that contact with their family could be more frequent and 72.8% felt that their current level of contact with their family was the most they would prefer (see Table 23). 158   Rwanda and Burundi are absent from findings on levels and frequency of familial contact, as well as preferred levels of familial contact, due to lack of directly comparable data. 76 In DRC, where ex-combatants had the Table 24: Ex-Combatant Friend Group Demographics Summary lowest actual levels of familial contact, ex-combatants were dramatically more Frequency of contact between community member and immediate family these days likely than average to think that their frequency of familial contact could be Most friends Most friends Most friends are Most friends are more – even though those who did have are the same have the same the same age ex-combatants gender education level contact had it at a similar level to other GLR countries. Male 64.90% 13.10% 11.60% 1.20% Female 76.50% 9.70% 4.20% 1.40% Interestingly, ex-combatants across Age 18-30 71.00% 10.00% 8.80% 1.50% the GLR were more likely to think the Age 31-40 59.90% 13.40% 13.60% 1.60% level of contact that they had with their families was the maximum they would Age Over 40 71.90% 13.20% 7.80% 0.30% want than community members (72.8% Disabled 88.20% 6.80% 1.10% 0.00% vs.29.3%). Additionally, although in Not Disabled 64.70% 13.20% 11.20% 1.40% RoC female ex-combatants had notably GLR Average 35.30% 20.00% 23.50% 1.10% lower contact with their families than RoC 28.90% 24.30% 22.10% 3.00% male ex-combatants (as outlined Uganda 92.70% 4.80% 1.80% 0.30% above), they actually less frequently expressed that they thought they could GLR Average 67.00% 12.40% 10.20% 1.20% have more frequent contact with their family (41.9%) than their male ex-combatant counterparts (48.4%). In DRC and RoC, the GLR countries where ex-combatants most frequently thought they could have more contact with their families, 30.9% of those who thought they could see their family more frequently cited the distance of travel as the main reason they do not see their family more often, 20.4% cited lack of time, and 17.1% cited the cost of travel – flagging the geographic spread of families as a dimension to reintegration in these countries. Across the GLR countries, 49.1% of all ex-combatants had lots of friends, 30% had a few, good friends and 20.9% did not have many friends.159 This trend is well reflected within the GLR countries with the exception of Rwanda – where 23% had lots of friends, 44.2% had a few good friends and 32.8% did not have many friends. On average female ex-combatants slightly less frequently had lots of friends (44.3%) and more frequently had a few good friends (33.4%) or not many friends (22.3%) than male ex-combatants (49.9%, 29.5% and 20.6% respectively). Rwanda is the exception to this gender demographic trend in which female ex-combatants slightly more frequently than male ex-combatants had lots of friends (27.3%) or a few good friends (54.5%) than male ex-combatants (22.9% and 43.9% respectively). Across the GLR countries there were clear and consistent trends in terms of the age, gender, ex-combatant status and educational background of the friends of ex-combatants.160 The majority of ex-combatants across the GLR countries were likely to have friends within the same age and gender categories, but less likely to have friends who were ex- combatants or shared the same education level. These trends are displayed in Table 24. What is perhaps most noteworthy in these findings is that ex-combatants’ friend groups appear to be fairly diversi- 159   Burundi is excluded from findings regarding number of friends due to lack of directly comparable data. 160   Burundi is excluded here from findings regarding the proportion of other ex-combatants in ex-combatants’ social groups due to lack of data. Rwanda is excluded from findings on age, gender, and ex-combatant makeup of ex-combatant’s social groups due to lack of data. However, the case of ex-combatant status of ex-combatants’ social groups Rwanda is excluded from direct comparison due to a scaling issue in the data. In Rwanda 68.9% said that some of their friends were ex-combatants, 22.9% said that most of them were ex-combatants, 7.6% said none were ex-combatants and only 0.5% said that all their friends were ex-combatants. 77 fied, especially in terms of having friends who are ex-combatants. Indeed, only 26.7% of ex-combatants say that most of their friends are fellow ex-combatants, 23.6% say some, 36.4% say few and 13.3% say none. This suggests that ex- combatants are not becoming an isolated social group – only socializing with each other. Across the GLR countries, when asked who they would turn to for help if they were to encounter an economic problem 39.7% of ex-combatants responded that they would turn to their family; 30.9% responded that they would turn to a friend; 13.5% would turn to no one; and 10% would rely on a range of business, communal, or formal financial resources. Generally across the GLR countries older ex-combatants were more likely to rely on friends for economic help than younger ex-combatants – who were, themselves, more likely to rely on family than older ex- combatants. This lends evidence to the idea that, in general, ex-combatants’ primary pathway to economic assistance is through their families and extended social circles as opposed to formal institutional pathways. Indeed, while as a whole ex-combatants would turn to similar sources as their community member counterparts for economic help, community members were slightly more likely (7.5% vs. 3.3%) to rely on formal institutions. 10.2 Trust and Solidarity Drawing from Rwanda and Uganda we can see that ex-combatants have generally high levels of trust in their com- munities.161 Of the respondents, 58% said that they trust people in their community to a great extent, 31.2% said to neither a great nor small extent, and the remaining 10.8% said they trusted those in their community to a small extent. In Rwanda and Uganda female ex-combatants generally trusted less than male ex-combatants (18.5% vs. 9.5% trusted those in their community to a small extent). Age displayed a positive correlation to high trust in others in the community (47% of those 18-30, 62% of those 31-40, and 64.3% of those over 40). Overall ex-combatants displayed a similar level of trust, though slightly weaker than community members. Across the GLR countries, 18.3% of ex-combatants felt that if they were to Table 25: Ex-Combatant Perceptions of Change in Trust disagree with what everyone else in their area agreed on, they would not at all feel In the past year/ past two years has the level of trust in your area got better, worse, or stayed about the same? free to speak out, 60.2% felt that they would definitely feel free to speak out Better Same Worse and 19.5% felt that they would only feel Male 53.40% 36.90% 9.80% free to speak out on certain matters.162 Female 42.30% 48.50% 9.20% This trend was visible within each of Age 18-30 51.50% 37.90% 10.50% the GLR countries; only in Uganda was Age 31-40 54.30% 36.30% 9.40% willingness to speak out slightly higher – 9.6% feeling they would not speak Age Over 40 50.70% 40.40% 8.80% out, 71.3% feeling they would defi- Disabled 45.80% 43.60% 10.60% nitely speak out and 19.1% feeling that Not Disabled 52.70% 37.70% 9.60% they would only speak out on specific 62.70% 24.00% 13.20% Burundi matters. Community members were DRC 23.30% 67.70% 9.00% slightly more likely to feel they could definitely speak out. Republic of Congo 73.70% 19.70% 6.60% Rwanda 59.30% 28.00% 12.70% Across the GLR countries, 52% of ex- 43.60% 48.60% 7.70% Uganda combatants felt that in the last year / GLR Average 52.00% 38.30% 9.70% two years the level of trust between 161   This specific question regarding the overall extent of community trust was asked to ex-combatants only in Rwanda and Uganda. 162   Rwanda is excluded from findings regarding willingness to speak out due to lack of directly comparable data. 78 people in the area that they lived in had improved, 38.3% felt that trust had stayed the same and the remaining 9.7% felt that trust had deteriorated (displayed in Table 25).163 Female ex-combatants less frequently thought that trust had improved (42.3%) than male ex-combatants (53.4%). Disabled ex-combatants also slightly less frequently felt that trust had improved (45.8%) compared to non-disabled ex-combatants (52.7%). Community members were less likely to see trust as improved than ex-combatants (43.4% vs. 52%) but more likely to think it had stayed the same (47.9% vs. 38.3%). When those ex-combatants who felt that trust had improved were asked to explain why they thought it had improved the most common responses were: (i) 25.4% of ex-combatants across the GLR countries felt that peace in general was the reason for improved trust; and (ii) 22.5% thought communal living and growing understanding were the reasons for improved trust.164 It is important to note that while these trends give a general picture of the perceived drivers of trust across the GLR countries, within each country there were unique trends as well that deserve further investigation – for example in Burundi the charging of ex-combatants for their behavior was seen as the main driver of improvements in trust by 56% of the ex-combatants sample.165 One demographic trend that does endure across the GLR countries is that those ex-combatants aged 18-30 most frequently see the charging of ex-combatants for their behavior as the central driver of improved trust – 32.2% of those 18-30 compared to 15.8% of those 31-40 and 5% of those over the age of 40. When those ex-combatants who felt that trust had gotten worse were asked to explain why they thought trust had deteriorated, 27.8% cited dishonest people and 21.9% cited political problems or distrust in authorities. While the internal proportions of these two driving factors behind worsening perceptions of trust among ex-combatants varied within the individual GLR countries they were consistently the two most common explanations. 10.3 Social Cohesion and Inclusion When asked about the level of diversity in the area in which they live ex-combatants displayed a spread of respons- es across the GLR countries almost identical to community members.166 35.2% of ex-combatants described the people in the area in which they live in as characterized by many differences (diverse), 24% characterized them as having neither a great or small extent of differences, and 40.8% said there were few differences between people (not diverse). As visible in Table 26, Rwanda stood out from this trend – 61.4% saw high diversity, 17.8% average, and 20.8% low diversity.167 When asked whether or not the differences between people (level of diversity) were a source of problems such as disagreement, arguments, and disputes there were split results. In DRC and RoC, a low portion of ex-combatants 163   In Uganda and Rwanda this question was asked with reference to the last year, where as in DRC and RoC it was asked in reference to the last two years. This creates some issues with periodization and comparability. These figures should be treated with caution. Interestingly, though the question refers to a longer period of time in DRC, this does not appear to translate to greater perceptions of improved trust among community members. In the case of DRC this may be the product of continuing insecurity. 164   It is unclear whether ex-combatants being charged for their behavior is in formal or informal charging / accountability. Rwanda is excluded from findings regarding drivers of improved or depreciating levels of trust due to lack of directly comparable data. 165   Further information regarding trends within each of the countries can be found in each of the individual survey reports from the GLR countries. 166   Here the percpetion of diversity in consttuted but the perception of unsepcified differences among people in the community. Another way to phrase this would be the level of ”differentness” that ex-combatants percieve in their community. 167   It is possible that the perception of differences (or diversity) can have a varying range of meanings across the contexts of different GLR countries. For example DRC is a country with rich diversity along cultural, ethnic, and linguistic groups. However, the difficulty of movement in eastern DRC means that many such socio-linguistic groups live in isolation from each other. The community members may accurately perceive low diversity in their community, though at a national level diversity may be high. In contrast, the perception of differences (or diversity) may be high in Rwanda due to the centrality of the Hutu / Tutsi divide in the social history of the country and conflict there. Deciphering the role of perceived differences across the different GLR countries is a challenging task with few clear answers. 79 saw differences between people (diver- sity) as a problem (13.7% and 13.6%, Table 26: Ex-Combatant Perceptions of Community Diversity respectively), whereas in Burundi and To what extent do differences between people characterize Uganda these differences were much your community? more likely to be perceived as the source To a great extent, i.e. To a small extent, Neither great nor of problems (72.7% and 55.4%, respec- lots of differences small extent i.e. few differences tively). A similar split was seen in the between people between people community member sample. In DRC Male 36.20% 24.00% 39.80% and RoC, when questioned further as Female 29.10% 23.80% 47.10% to the type of problems that these dif- Age 18-30 35.20% 25.60% 39.20% ferences can cause, 27.8% of ex-combat- Age 31-40 37.60% 23.20% 39.20% ants said that envy, slander, and taunts were the most common problems, Age Over 40 32.30% 21.70% 46.10% 17.8% said misunderstandings were Disabled 36.50% 21.60% 42.00% the main problem, and 10.8% said that Not Disabled 35.10% 24.30% 40.60% mistrust was the result of differences Burundi 30.30% 25.20% 44.50% (diversity) between people in the area DRC 42.40% 37.10% 20.50% they live. Unfortunately there is no data available from Burundi and Uganda on Republic of Congo 28.80% 21.30% 49.90% the types of problems associated with Rwanda 61.40% 17.80% 20.80% diversity. Uganda 20.90% 16.40% 62.70% GLR Average 35.30% 24.00% 40.80% When questioned as to the level of to- getherness that ex-combatants feel with other people (unspecified who) in the area they live the response across the GLR countries was generally a high level of togetherness / closeness that was on par with community members. 76.6% felt close with others, 16.6% felt neither close nor distant and 6.8% felt distant from others in the area they lived – this trend was well reflected within the individual GLR countries – though in DRC, ex-combatants were slightly less likely to report high levels of togetherness (63.1%). Across demographic dimensions only age stood out – which showed a slight positive rela- tionship to the likelihood of feeling close to the community (75.3% of those 18-30, 77.6% of those 31-40, and 79.3% of those over 40). Across the GLR countries, 69.3% of ex-combatants had at some point in the past year worked with others in the place where they live to do something for the benefit of the community. Burundi and Rwanda stand out with even higher levels of working with the community – 79.3% in Burundi and 90.8% in Rwanda.168 Female ex-combatants less frequently took part in community projects (57.5%) when compared to male ex-combatants (71%) across the GLR countries – again, with the exception of Rwanda where female participation in community projects in the last year was absolute (100%), exceeding male ex-combatants (90.5%). When ex-combatants were asked whether there were any penalties for those who did not participate in community activities, 33.3% responded that penalties were very likely, 23.9% that they were somewhat likely, 16.9% that they were neither unlikely nor likely, 7% that they were somewhat unlikely, 14.3% that they were very unlikely and 4.7% that total social exclusion would be the result.169 168   In the case of Rwanda, higher levels of working for the benefit of the community is very likely a result of the institutionalized practice of Umuganda – a practice dating back to Rwanda’s colonial era in which on the last Saturday of every month all able bodied adults participate in unpaid communal labor – with enforced penalties for non-participation. In Burundi this trend is likely related to the similar policy of Travaux Communataires. 169   Rwanda and Uganda are excluded from these findings due to lack of directly comparable data. 80 10.4 Empowerment Empowerment is an important indicator of overall levels of social capital and is understood as a result of individu- als’ levels of social connection and their ability to leverage the benefits of these connections in the community and the larger context of society. Collectively, the extent of these benefits, and in turn the functions that they fulfill for individuals, play a role in the psychosocial concept of empowerment – the individual or collective ability to affect change in one’s life. In exploring issues around empowerment this report builds on survey data regarding: (i) the extent to which ex-combatants feel generally happy; (ii) the extent to which they perceive that they can make impor- tant decisions; (iii) the extent to which they have control over decisions in their daily life; (iv) the extent to which they feel valued by the community; and (v) the extent to which they engage in collective political action. Across the GLR countries when asked about their level of happiness 57.2% of all ex-combatants said that they were generally happy, 23.7% were neither happy nor unhappy, and 19.1% were generally unhappy. In this regard ex-com- batants were considerably less likely to report themselves as happy than community members (57.2% vs. 71.8%). This trend was well displayed with the individual GLR countries with the exception of Burundi in which the spread of responses from ex-combatants was much more even (31.6% happy, 39.7% neither happy nor unhappy, and 28.7% unhappy). Generally speaking, female ex-combatants were slightly less happy across the GLR countries than male ex-combatants in terms of happiness within the GLR countries with the exception of Rwanda where female ex- combatants were considerably more happy (90.9%) compared to male ex-combatants (61.6%). Across the GLR countries there was a slight positive relationship between age and happiness – 54% of those aged 18-30, 58.1% of those 31-40, and 60.7% of those over 40 identified as happy. When questioned about the extent to which they felt that they had the power to make important decisions that change the course of their lives 59.9% of all ex-combatants across the GLR countries responded that they felt that they had this power to a large extent, 25.7% to neither a large nor small extent, and 14.4% to a small extent. 170 This trend was consistently displayed within all of the GLR countries. In examining demographic subgroups both female ex-combatants and disabled ex-combatants showed considerably lower perceptions of power in shaping their lives (see Table 27). Overall, 44.1% of female ex-combatants compared to 62.5% of male ex-combatants felt they had the power to make important decisions in their lives to a large extent. Similarly, 49% of disabled ex-combatants compared to 61% of non-disabled ex-combatants felt they had the power to make important decisions in their lives to a large extent. While in the case of female ex-combatants the disparity with males was absorbed into both the categories “neither to a large nor small extent” and “to a small extent,” however in the case of disabled ex-combatants this difference with non-disabled ex-combatants was almost absolutely absorbed into the category “to a small extent” (28.1% of disabled vs. 16.1% of non-disabled). This may perhaps suggest that there is a more polarizing dynamic to the nature of empowerment for disabled ex-combatants than female ex-combatants, or any other demographic subgroup for that matter. Ex-combatants were asked the extent to which they felt they had the ability to make important decisions that change their lives.171 Across the GLR countries, 82.9% felt they were able to change their lives, 11.2% felt that they were neither able nor unable, and 5.9% felt that they were unable to make important decisions to change their lives. As with sense of power to change their lives, female and disabled ex-combatants less frequently reported having the ability to change their lives (73% and 68.8% respectively). When questioned as to the extent to which ex-combatants felt they had control over decisions that affect their 170   Rwanda is excluded from findings regarding ability to make large decisions due to lack of directly comparable data. 171   Rwanda and Burundi are absent from findings on ability to make important decisions in life due to lack of directly comparable data. 81 Table 27: Ex-Combatant Empowerment (Power, Ability, Control) Do you feel that you have the power to make important decisions that can change the course of your life? Large extent Medium extent Small extent Male 62.50% 24.60% 12.90% Female 44.10% 32.80% 23.10% Age 18-30 59.80% 23.20% 16.90% Age 31-40 66.10% 25.00% 9.00% Age Over 40 56.50% 29.90% 13.60% Disabled 49.00% 26.40% 24.60% Not Disabled 61.00% 25.50% 13.50% GLR Average 59.90% 25.70% 14.40% Do you feel that you have the ability to make important decisions that can change the course of your life? Able to change life Neither able nor unable Unable to change life Male 84.80% 10.40% 4.80% Female 73.00% 15.80% 11.30% Age 18-30 80.70% 13.30% 6.00% Age 31-40 85.40% 10.80% 3.80% Age Over 40 84.30% 8.20% 7.50% Disabled 68.80% 14.90% 16.30% Not Disabled 84.00% 11.00% 5.00% GLR Average 82.90% 11.20% 5.90% How much control do you feel you have over decisions that affect your everyday activities? Lots of Control Neither a lot nor a little control Little Control Male 73.40% 18.40% 8.20% Female 55.00% 28.90% 16.00% Age 18-30 68.90% 20.20% 10.90% Age 31-40 76.60% 16.80% 6.50% Age Over 40 70.30% 22.00% 7.70% Disabled 63.50% 21.10% 15.50% Not Disabled 72.10% 19.50% 8.50% GLR Total 71.10% 19.70% 9.20% everyday activities, 71.1% of all ex-combatants across the GLR countries expressed that they felt that they had a high level of control, 19.7% felt that they had neither a little nor a lot of control, and 9.2% felt that they had little control over decisions.172 This decreasing trend is present in all the GLR countries, however the peak is slightly shifted in Uganda where the curve is slightly different (52.1%, 31.4%, 16.5%). Female ex-combatants were consistently less likely to feel they had lots of control over decisions in their lives (55%%) when compared to male ex-combatants (73.4%). Disabled ex-combatants also were consistently less likely to feel they had a high level of control over deci- 172   This question regarding community members’ levels of control over everyday decisions has been recoded from a five point scale to a three point scale for increased comparability to the other two measures of empowerment (power and ability) presented here. 82 sions in their lives (63.5%) when compared to non-disabled ex-com- Table 28: Ex-Combatant Perception of Individual Impact on Community batants (70.3%). Overall ex-com- Do you personally have a positive or negative impact on the place batants felt slightly higher levels of that you live? empowerment than community Neither positive nor Positive impact Negative impact members in all three (power, ability, negative impact and control) measures.173 Male 59.90% 27.20% 12.90% Female 56.60% 29.90% 13.50% When asked to gage the impact that Age 18-30 52.40% 33.30% 14.30% they have on the place they live, 59.5% of ex-combatants across the Age 31-40 64.40% 24.30% 11.20% GLR countries feel that they have a Age Over 40 67.80% 21.00% 11.20% positive impact, 27.57% feel that they Disabled 74.10% 18.80% 7.20% have neither a positive nor negative Not Disabled 57.80% 28.60% 13.70% impact, and 13% feel that they have a Burundi 41.50% 42.80% 15.70% negative impact.174 This trend is well reflected in Burundi, DRC and RoC DRC 38.80% 37.90% 23.40% – however, in Uganda and Rwanda Republic of Congo 47.10% 32.60% 20.30% perceptions of positive impact were Rwanda 99.20% 0.00% 0.80% much more frequent (82.1% of ex- Uganda 82.10% 16.50% 1.50% combatants in Uganda perceived that GLR Average 59.50% 27.50% 13.00% they had a positive impact and 99.2% of those in Rwanda) – see Table 28. Across age demographics lines there was a positive relationship visible between age and ex-combatants perception of having a positive impact on the area in which they lived – 52.4% of those aged 18-30, 64.4% of those 31-40, and 67.8% of those over 40. Across the GLR countries, 72.1% of all ex-combatants felt that people in the area in which they live valued them, the remaining 27.9% did not feel valued. Uganda was the only country that departed slightly from the cross-coun- try trend, displaying higher levels of perceived value among ex-combatants (94.6%). Female ex-combatants were slightly less likely to feel valued (62.3%) compared to male ex-combatants (73.6%). Regarding collective political action, ex-combatants were asked how often they had joined with other people to express concerns to government officials or local leaders on issues concerning the community. Across the GLR countries 43% of all ex-combatants said that they had never done so in the last year, 12.7% that they had once, 18.1% that they had a few times (five or less), and 26.3% that they had many times (more than five times) – levels very similar to community members. Burundi to some extent, and Rwanda to a greater extent, broke from this trend and displayed higher levels of collective political action (visible in Table 29). In Burundi, 34.5% of ex-combatants had joined to address local leaders many times (more than 5) and 30.7% had a few times (less than five). In Rwanda, 91% of ex-combatants had joined to address local leaders many times in the last year.175 Across the GLR countries female ex-combatants were less likely to have gathered for collective political action then male ex-combatants, though at a 173   The analytical distinction between senses of empowerment in terms of power vs. ability is not clear. Interpreting any meaning to the disparity in levels of power and ability is therefore problematic and these data should be treated as a broad indicator of a positive sense of empowerment rather than as exact measures of different components of empowerment. 174   Rwanda is excluded from findings regarding ex-combatants’ perceived impact on their community due to lack of directly comparable data. 175   This high rate of public gathering to express concerns in Rwanda is likely another effect of Umuganda. While the main purpose of Umuganda is community work it also serves as a platform for leaders to communicate important news on a national and local level as well as for individuals and communities to express concerns and plan for future Umuganda. Further, every community has an ex-combatant representative who is responsible for relaying specific communication. 83 Table 29: Ex-Combatant Frequency of Public Gathering to Express Concerns In the past year, how often have you joined other people to express concerns to officials or local leaders on issues benefiting the community? Never Once A few times, five or less Many times, more than five Male 40.80% 12.70% 18.30% 28.30% Female 58.80% 12.80% 16.80% 11.60% Age 18-30 46.10% 13.10% 18.00% 22.70% Age 31-40 38.80% 10.70% 17.20% 33.30% Age Over 40 40.90% 14.20% 20.30% 24.60% Disabled 37.00% 7.40% 15.90% 39.70% Not Disabled 43.70% 13.30% 18.40% 24.70% Burundi 25.60% 9.20% 30.70% 34.50% DRC 64.10% 15.50% 12.80% 7.60% Republic of Congo 62.30% 17.00% 14.50% 6.10% Rwanda 1.90% 0.30% 6.80% 91.00% Uganda 49.60% 17.90% 22.10% 10.30% GLR Average 43.00% 12.70% 18.10% 26.30% very similar level to female community members– 58.8% of female ex-combatants having never gathered and 11.6% having gathered many times versus 40.8% and 28.3.2% respectively of male ex-combatants. When asked to what extent local government and leaders take into account the concerns voiced by their community when they make decisions, 17.6% of all ex-combatants across the GLR countries felt that local leaders took them into account a lot, 41.2% felt their voices were taken into account a little, and 41.2% felt that their concerns were not taken into account at all – nearly identical levels to those expressed by community members.176 Across gender and disability dimensions, ex-combatants’ responses were approximately even. However concerning age, older ex- combatants (aged over 40) were the most likely age demographic group (54.7%) to feel that their concerns were not taken into account, while younger ex-combatants (aged 18-30) were most likely of age demographic groups to feel that they were taken into account a lot (19.2%). This is likely related to older ex-combatants overall higher levels of social capital. 10.5 Social Change When asked about their outlook on the likelihood of their overall situation improving in the future, responses were quite polarized between those who thought that things would improve in a few years and those that thought that their situation would deteriorate in the future.177 Overall, only 1.2% of all ex-combatants across the GLR countries thought that their situation would improve within some weeks, 5.2% thought it would improve in some months, 43.7% thought that it would hopefully improve in some years, 8.7% thought that their situation would not improve in the future but stay the same, and 41.2% expressed that they thought that their situation would deteriorate in the future. The only GLR country that stepped away from this trend was Uganda in which 71.7% of ex-combatants were 176   Rwanda is excluded from findings regarding ex-combatants’ perceptions of whether leaders take their voices into account due to lack of directly comparable data. 177   Rwanda is excluded from findings regarding ex-combatants’ outlook on their future situation due to lack of directly comparable data. 84 Table 30: Ex-Combatant Social Change Ladder – One Year Ago and Today One Year Ago Consider a 9-step ladder where on the bottom, the first step, stand the SCHOOL FEES? CLOTHING? FINANCES? HOUSING? LEISURE? poorest people, and on the ninth step, HEALTH? FOOD? stand the richest - On which step were you one year ago in relation to: Male 3.17 2.97 3.20 2.70 3.39 3.23 3.33 Female 2.95 2.76 3.00 2.55 3.39 3.25 3.08 Age 18-30 3.02 2.89 3.09 2.57 3.25 3.09 3.17 Age 31-40 3.23 2.99 3.28 2.82 3.46 3.34 3.36 Age Over 40 3.18 2.93 3.17 2.62 3.54 3.37 3.43 Disabled 2.73 2.62 2.71 2.33 3.28 2.90 2.99 Not Disabled 3.18 2.97 3.22 2.71 3.40 3.26 3.32 Burundi 3.22 3.03 3.42 2.79 2.94 3.08 2.92 DRC 3.45 3.39 3.65 2.91 3.81 3.30 3.79 Republic of Congo 3.28 3.05 3.14 2.95 3.42 3.33 3.18 Uganda 2.60 2.28 2.47 2.06 XXX XXX XXX GLR Average 3.14 2.94 3.17 2.68 3.39 3.24 3.30 Today Consider a 9-step ladder where on the bottom, the first step, stand the SCHOOL FEES? CLOTHING? FINANCES? HOUSING? LEISURE? poorest people, and on the ninth step, HEALTH? FOOD? stand the richest - On which step are you on today relation to: Male 3.30 3.15 3.34 2.77 3.51 3.36 3.41 Female 2.99 2.80 3.25 2.62 3.38 3.32 3.07 Age 18-30 3.13 3.08 3.28 2.68 3.37 3.24 3.26 Age 31-40 3.40 3.15 3.41 2.90 3.56 3.42 3.43 Age Over 40 3.18 2.97 3.25 2.62 3.56 3.39 3.43 Disabled 2.76 2.64 2.77 2.34 3.30 3.01 2.95 Not Disabled 3.31 3.14 3.38 2.79 3.51 3.38 3.40 Burundi 3.40 3.25 3.56 2.88 3.04 3.20 2.98 DRC 3.38 3.46 3.61 2.84 3.82 3.34 3.79 Republic of Congo 3.51 3.22 3.38 3.14 3.65 3.54 3.35 Uganda 2.75 2.48 2.75 2.14 XXX XXX XXX GLR Average 3.26 3.10 3.33 2.75 3.50 3.36 3.37 hopeful that their situation would improve within a few years.178 In general, these findings suggest that ex-com- batants across the GLR countries have a good understanding of the time horizons of social change, but that a sig- 178   While there is no direct evidence for explaining why in this case ex-combatants in Uganda have more optimistic outlooks for their future, it is possible that this is linked to the relative stability of Northern Uganda and the overall pace of improvement away from a context of widespread displacement due to conflict and humanitarian intervention. 85 nificant proportion remains pessimistic for the future. Overall, ex-combatants had slightly less optimistic outlooks compared to community members. Interestingly, ex-combatants’ outlook on their economic situation (see section 9.1) was considerably better (73.7% reported seeing their economic situation improving in the future) than their overall outlook (a total of 58.8% report- ing expected improvement at various time scales). This evidence tacitly supports the idea that while ex-combatants can make improvements relatively quickly in economic terms, the diverse set of challenges that exist in the social sphere are slower to resolve. When asked about whether or not they were satisfied with their life in general up until then, 30.6% of all ex-com- batants across the GLR countries said they were satisfied, 7.5% that they were neither satisfied nor dissatisfied, and 61.9% were dissatisfied.179 This trend of the overwhelming majority of ex-combatants expressing dissatisfaction with their life was consistent in all the GLR countries except for Uganda where the spread of responses was much more even – 33.2% were satisfied, 30.7% were neither satisfied nor dissatisfied, and 36.1% were dissatisfied. These findings are perhaps not surprising, considering the heavy toll conflict can take on the lives of individuals. However, these effects are not isolated to ex-combatants – community member displayed a similar range of responses about life sat- isfaction. However, these findings stand in contrast to those on overall happiness (see section 10.4) in which nearly 60% of ex-combatants indicated that they were generally happy. Understanding the interplay between ex-combat- ants levels of happiness and their overall life satisfaction is a challenging task with no clear explanation in study. Ex-combatants were questioned using a 10-step ladder response prompt. Their responses are tabulated below in Table 12 by mean score.180 The lower the mean score is the closer the ex-combatant is to the bottom rung of the ladder – where the poorest people tend to be. Generally speaking, across and within the GLR countries ex-combat- ants consistently identified themselves in the poorest half of society (between steps 2 and 4). When looking across the GLR countries as a whole, there is a slight improvement in mean scores in all question categories from one year ago to the time of sampling. This trend is almost completely consistent within the indi- vidual GLR countries with the only exception being DRC – in which there was a slight decrease in mean scores from one year ago and time of sampling in the categories of finance, clothing and food. When focusing on demo- graphic subgroups, however, there are less consistent results. Three demographic subgroups standout in particular: female ex-combatants, ex-combatants aged 31-40, and disabled ex-combatants. Though female ex-combatants saw near unanimous improvements across all categories, leisure being the only exception, they consistently ranked a rung lower than male ex-combatants. Those aged 31-40 rank higher or equal than other age demographics across all categories. Similar to the trend of female ex-combatants, disabled ex-combatants saw improvements across all categories, leisure being the only exception. However disabled ex-combatants ranked consistently lower than non- disabled ex-combatants. 10.6 Summary The social-fabric of communities endures great detriment in the course of violent conflict. Indeed, it is no wonder that ex-combatants and community members alike struggle to mend their damaged social footing. However, con- sistent with analysis presented throughout much of this study, ex-combatants experience a range of additional chal- lenges in the process of social reintegration that collectively entail their disadvantage to community members. While collectively ex-combatants display a positive trajectory in terms of social reintegration, rebuilding social capital, and connecting into the social fabric of the community, the angle of this trajectory is considerably more shallow than 179   Rwanda is excluded from findings regarding life satisfaction due to lack of directly comparable data. 180   Rwanda is excluded from findings regarding ex-combatants’ perceptions of change within specific categories due to lack of directly comparable data. 86 in other dimensions of reintegration, such as economic – i.e. though ex-combatants are catching up to community members in terms of social indicators, the rate at which they are doing so is considerably slower than in other dimen- sions of reintegration. This evidence supports the idea that social reintegration is a slow process of social confronta- tion and atonement with no shortcuts. Though trust with community members may improve quickly, as outlined in section 17, ex-combatants still struggle to recover from the damage done to their social networks, solidarity with the community, their cohesion and inclusion in the community, as well as their overall sense of empowerment and positive social change. Ex-combatants have fewer social groups than community members and slightly less familial contact than commu- nity members overall. Though, ex-combatants who do have contact with their families have it much more frequently than community members indicating their heavy reliance on their immediate family for social support. Accordingly ex-combatants are more likely to turn to their family for economic help than friends or community / formal institu- tions. In terms of ex-combatants’ friend base there is a clear split in the GLR countries. In Uganda and Rwanda, on the one hand, ex-combatants have fewer friends than community members and thus an extremely focused social support network relying heavily on the family. However, by contrast, in DRC and RoC ex-combatants had larger friend bases than community members, indicating a good extended social support network – despite the clear presence of a range of social limitations in relation to community members. Collectively these findings indicate the extent to which ex-combatants’ social networks are more limited than those of community members and in turn the extent to which the functions of those social networks are limited as well – i.e. the psychosocial and economic value of social and familial networks. Despite the challenges that ex-combatants face in the process of rebuilding interpersonal social ties within the com- munity, they are generally well integrated and have a very similar understanding of the dynamics of their community. Ex-combatants and community members alike have generally high levels of trust – though ex-combatants perceive larger improvements in trust. The frequency at which ex-combatants work for the improvement of the community and feel an overall sense of togetherness is similar to community members. Ex-combatants generally feel similar if not stronger senses of empowerment to affect changes in the direction of their lives and control their everyday circumstance than community members. This is further evidenced in ex-combat- ants’ similar level of political engagement in community issues to community members. However, it is interesting to see that higher senses of empowerment among ex-combatants does not necessarily translate to higher levels of overall happiness or better perceptions of impact on the community. Indeed ex-combatants across the GLR coun- tries report being much less happy than community members and are less likely to view themselves as having a positive impact on the community. It is possible that ex-combatants’ overall happiness and senses of self-worth may be more tied to the personal psychological trauma ex-combatants carry with them in the wake of conflict than their absolute conditions (which while worse than community members in absolute terms, do display a clear positive trajectory) at the time of sampling. If this is so, it would lend considerable support to the idea of social reintegration as a slow, long-term process of interpersonal exchange and in turn intrapersonal betterment. Despite ex-combatants’ lower levels of happiness and sense of positive impact on the community, their outlook on the future and understanding of the temporal dynamics of social change are similar to community members. Ex- combatants and community members alike understand that positive change in their overall situation will happen on the scale of years – not weeks or months. This makes sense, as both community members and ex-combatants have seen slight improvement in their overall conditions in the past years / since demobilization, ex-combatants less so in absolute terms, but still consistently identify themselves in the worst off half of society. 10.6.1 Vulnerable Subgroups When examining social reintegration female ex-combatants continue to represent the most clearly and consistent- ly vulnerable demographic group among ex-combatants. Female ex-combatants have fewer and less diverse social networks, tending to rely even more exclusively on their immediate family than the rest of ex-combatants – who do 87 so to a greater extent than community members. In this sense, female ex-combatants face the highest risk of social isolation and marginalization across the GLR countries. This weak social capital in terms of the number and diver- sity of social groups corresponds to lower levels of trust, lower perceptions of improvement in trust, dramatically weaker senses of empowerment, and lower perceptions of their overall situation than the rest of ex-combatants. Despite the clear and consistent rage of vulnerabilities that female ex-combatants exhibit, their overall levels of hap- piness, life satisfaction and general outlook for the future are on par if not better than the rest of ex-combatants. Developing a clear understanding of the social and psychological coping strategies that female ex-combatants have developed to maintain even, if not more positive, senses of self worth, worth in the community, and outlook for the future – effectively mitigating against their heightened vulnerability across almost all social indicators – could prove relevant the development of future programming for female ex-combatants and male ex-combatants alike. Disabled ex-combatants exhibit a complex range of disadvantages in terms of social capital. In general disabled ex- combatants report far lower levels of personal empowerment and control of their lives. However, this is counter- balanced against their unexpectedly higher levels of political engagement in the community and stronger sense of positive impact on the community in comparison to non-disabled ex-combatants. In terms of social change disabled ex-combatants perceive a positive trajectory of social change over time across a broad range of categories. However despite this perceived positive trajectory, disabled ex-combatants consistently rank themselves a step below non- disabled ex-combatants. 10.6.2 Unique Country Trends Despite the many ways in which the individual GLR countries come together to represent a consistent collective narrative of the process of reintegration, there are also many ways in which they diverge – especially in terms of social reintegration. Here we can highlight a selection of unique country trends focused in DRC that represents an alternate narrative of reintegration than the one consistent across the other GLR countries. When examining the many dimensions of social reintegration across the GLR countries, DRC stands out most con- sistently and sharply. As presented above, the dominant narrative of social reintegration reflected across the GLR countries was one where ex-combatants had high levels contact with the family, though slightly less than community members, but stunted development in terms of social networks, friends, and connections to the broader community – in turn correlating to lower levels of happiness and perceptions of worth in the community. In DRC however, we see a distinctly different narrative emerge. Ex-combatants and community members alike in DRC have dramatically lower levels of familial contact than other GLR countries. DRC stands out even further in this regard because it is the only GLR country where ex-combatants are more likely to have contact with their family than community members – twice as much so. Further, those ex- combatants who do have contact with their families have it at a much lower frequency (split between daily, weekly and monthly contact) than in other GLR countries where daily contact was the norm (this divergence is also visible in RoC). As such, it makes sense when ex-combatants in DRC are the most likely to say that their contact with their family could be more frequent. When pared with findings on marriage rates in section 7.2, what emerges is an image of ex-combatants in DRC who are isolated from their immediate family though are on par with other GLR countries in terms of building new familial connections (marriage / cohabitation). However, as mentioned, when ex-combatants in DRC are compared to community members in DRC in terms of contact with the immediate family, community members are half as likely to be in contact with their immediate family. This is a perplexing trend to explain. While ex-combatants in DRC have weaker family networks than ex-combatants in other GLR countries, they are the only GLR country that has stronger familial connections than community members – which could simultaneously suggest that ex- combatants in DRC have been exceptionally successful in terms of rebuilding social capital relative to community 88 members; and that community members in DRC are a key vulnerable group across the GLR countries in terms of social capital. A hint to understanding the overall lower levels of familial contact in DRC is that of those few ex-combatants who did have familial contact and felt that this contact was the maximum that they would desire – the distance, time and cost of travel were all cited as reasons for not seeing their family more often. Indeed, the social geography of eastern DRC is particularly troubling. While countries across the GLR have experienced varying scales of war, and in turn levels of impact on society both economically and socially, the incessant insecurity in eastern DRC can perpetuate a series of dynamic forces that disperse pre-conflict social networks through displacement and migration. Persistent conflict and can trap individuals, due difficulty of travel due to zones continued insecurity. When these dynamic forces are coupled with static forces such as the mountainous topography of eastern DRC and heavy rains that can render roads impassible it is understandable that social networks are separated. Future study into this line of inquiry could prove valuable for explaining why ex-combatants and community members alike in DRC have considerably weaker familial and social networks than other GLR countries. It appears that though ex-combatants in DRC have strong connections to their immediate family relative to com- munity members this does not compensate for the overall lower levels of familial contact in absolute terms relative to other GLR countries in terms of overall social capital. Indeed ex-combatants in DRC are the most likely group among the GLR countries to turn to no one for economic help; have the weakest feeling of togetherness with the community; feel they have the least amount of power to make important decisions in their life; perceive the weakest ability to control their everyday activities; are the least likely to perceive that they make a positive impact on the community; are the least likely to gather to express political concerns; the least likely to feel their voice is taken into account by leaders; the most likely to think their overall situation will deteriorate in the future; and have the lowest level of life satisfaction across the GLR countries. In terms of social change, DRC is the only country where ex-combatants see drops in their perception of their situation relative to the rest of society in the last year in the cat- egories of food, clothing, and finance – though beyond weak social capital ongoing insecurity in eastern DRC likely plays a role in this. 89 11. DDR Experiences D DR processes across the GLR countries have taken place in a diverse range of contexts, as such the amount of validly comparable data on all phases of disarmament, demobilization, and reintegration is limited. As such, offering a comprehensive comparison of across the GLR countries of ex-combatants’ experiences of the process and dynamics of return, reception, demobilization, reinsertion, and reintegration is unfortunately not feasible in this study – primarily as a result of data validity challenges.181 However, drawing from the select data we can offer comparative insights on: (i) ex-combatants’ experiences of reinsertion process across the GLR countries, and (ii) a limited range of comparative insights regarding initial experiences with the community. 11.1 Reinsertion In examining ex-combatants’ attendance to a range of information sessions on various topics from general informa- tion to sessions on how to apply for credit or loans, and information on peace and reconciliation processes as part of the reinsertion phase of programming there were quite unique trends in each GLR country, though approximately even across all types of information sessions within each country.182 As a general indication, 96.3% of ex-combatants attended a general presentation of information related to the reinsertion process in Burundi, 85.7% in DRC, 68.5% in RoC, and 39.2% in Uganda. Looking at demographic subgroups, female ex-combatants and disabled ex-combatants were noticeably less likely to have attended information sessions – 63.5% of all female ex-combatants versus 74.8% of all male ex-combatants, and 59.5% of all disabled ex-combatants versus 74.5% of all non-disabled ex-combatants. Within each GLR country the disparities between demographic subgroups were very similar to those at a cross- country level, though fitting to the overall attendance level with each country. When asked whether or not they thought they had received enough information about the reinsertion package and its contents there was a clear correlation between the level of participation in information sessions within GLR countries and the perception of receiving sufficient information. In Burundi 81.4% of ex-combatants felt they received sufficient information regarding reinsertion, in DRC 65.2%, in RoC 45.8%, and in Uganda 19%. Further, it is perhaps not surprising then that female ex-combatants and disabled ex-combatants, the two demographic sub- groups least likely to attend information sessions on reinsertion package and process, were the most likely to feel that they received insufficient information surrounding the reinsertion process and package – 44.4% of female ex-com- batants versus 55.9% of male ex-combatants, and 41.7% of disabled ex-combatants versus 55.6% of non-disabled ex-combatants. Not only did the level of attendance to information sessions about the reinsertion package and process correlate to the perceived level of information sufficiency among ex-combatants, but also to the actual frequency at which they received reinsertion payments. The same descending trend can be observed again in Burundi, as 99.7% of 181   Essentially, the contextual differences between the different programming components of the entire DDR process in the various GLR coutnries are at at times great – thus the range of captured data on ex-combatants’s experiences of these processes is equally diverse. A valid systemmatic comparison of the different data from each context is judged as infeasible here. The few areas discussed here are those few in which data overlapped in each of the GLR countries. 182   Rwanda is excluded from all findings regarding the reinsertion process due to incompatible data. For a review of the key trends in Rwanda see the Rwanda comparative study report. 90 ex-combatants received payments as a part of reinsertion, in DRC 88.3%, in RoC 56.2%, and in Uganda 35.6%.183 Collectively these findings suggest that across the GLR countries attaining a sufficient level of information and sen- sitization regarding the reinsertion process is a key to reaping the benefits of reinsertion payments and support. Female ex-combatants also showed a visible correlation between information and sensitization exposure and actual reception of reinsertion payments at a cross-country level, though they were less likely to receive reinsertion as- sistance than male ex-combatants overall - 62.9% of female ex-combatants versus 73.2% of male ex-combatants received reinsertion payments. Interestingly, disabled ex-combatants – though they attended information sessions on reinsertion less frequently and were less satisfied with the information they received – were nearly evenly as likely (68.4%) compared to non-disabled ex-combatants (72%) to receive reinsertion payments at a cross country level. When those ex-combatants who did not receive reinsertion payments were asked why they thought that they had not received payments, a large number connected this to lack of information. In Uganda 44.4% of all ex-combatants identified lack of information at some level as the primary reason they did not receive a reinsertion payment. Female ex-combatants identified information more frequently than male ex-combatants in Uganda (55.1% vs. 43.7%). A similar trend along gender lines existed across Burundi, DRC and RoC, in which 35.5% of females identified infor- mation as the reason they did not receive reinsertion payments as compared to 13.7% of male ex-combatants. Both ex-combatants who had received reinsertion payments and those who had not were question about their levels of satisfaction with those payments. Surprisingly, there was no clear correlation between the rate at which ex- combatants received reinsertion payments and their satisfaction with those payments. Across the GLR countries, 32.3% of ex-combatants were satisfied with their reinsertion payments, 33.9% were neither satisfied nor dissatisfied, and 33.9% were dissatisfied. This trend was well reflected across the GLR countries with the exception of Uganda where levels of satisfaction were more clearly polarized (41.3%, 14.3%, and 44.3%, respectively). With the exception of RoC, female ex-combatants were generally more satisfied with reinsertion payments than male ex-combatants (41.8% vs. 30.8% at a cross-country level), even though the rate at which they actually received reinsertion payments was lower. Similarly, when questioned further to their overall level of satisfaction with the totality of the reinsertion package contents ex-combatants were generally satisfied. Across the GLR countries, 47.1% of ex-combatants said that they were satisfied, 29.2% said that they were neither satisfied nor dissatisfied, and 23.7% were dissatisfied. This trend towards general satisfaction is reflected within all of the GLR countries. Again, this is somewhat puzzling, as one would expect some correlation between overall satisfaction of reinsertion packages and the rate at which they are actually received. Female ex-combatants are more likely to be satisfied at a cross-country level (54.1%) when compared to male ex-combatants (46%) – this is reflected in all GLR countries with the exception of DRC. Examin- ing age demographics reveals some interesting contrasts. In Uganda and Burundi, there is a clear trend that as age increases likelihood of satisfaction with the overall contents of the reinsertion package decreases. However this trend was reversed in DRC, as age increased likelihood of being satisfied increased as well. A comprehensive analysis of the uses of reinsertion payments across the GLR countries is challenging, however we can extract several general observations here.184 Generally speaking the most consistently cited use of reinser- tion payments was meeting immediate food and subsistence needs, suggesting that perhaps food security is among 183   The especially low reception of reinsertion payment in Uganda warrents some contextualization. Most ex-combatants in Uganda receive amnesty and reinsertion / reintegration assistance retroactively after returning to their communities. This ”trickle in” model in the context of Uganda means that many ex-combatants demobilize informally and thus the bottleneck for information and sensitization that formal demobilization processes represent is largely absent – making information and sensitization a key programming challenge. In addition, this data does not necessarily mean that those ex-combatants that have not received reinsertion payments have not received amnesty – as of 2011 (the time of sampling) the UgDRP still had a considerable backlog of upaid reinsertion assistance. 184   The main barrier to a comprehensive analysis lay in the different scales used to capture data on the use of reinsertion payments across the GLR countries. Rwanda is excluded from general trends due to lack of data. 91 the most pressing needs for ex-combatants at the time of demobilization. Assistance to family, parents, spouse or partner was also among the most common uses of the reinsertion payments. Additionally, investment of some sort, for example for business or in livestock, ranked high among the uses of reinsertion payments. 11.2 Experiences of Return Drawing from DRC, RoC, and Uganda we can see that the vast majority of ex-combatants (90.1%) report being welcomed home by their families immediately after demobilization.185 This proportion was high in both DRC in Uganda, but notably lower (81.4%) in RoC. This finding supports analysis across this report that ex-combatants generally experience high levels of acceptance and support from their immediate families. In accordance, after receiving reinsertion packages the majority (76.5%) of ex-combatants reported that they had no problems with their families, however again in RoC this percentage was slightly lower – 63% of ex-combatants had no problems with their family after reinsertion packages. In RoC female ex-combatants and disabled ex-combat- ants were especially more likely to encounter problems with their families – 48.1% of female ex-combatants having problems with their families after reinsertion payments versus 35.2% of male ex-combatants, and 53.6% of disabled ex-combatants versus 35.7% of non-disabled ex-combatants. When those ex-combatants who did encounter problems with their families after receiving reinsertion packages were asked to explain the specific nature of these problems a distinct range of answers was given. Although there is little data that is comparable across the GLR countries on this we can look at Uganda for a precursory survey of the kinds of problems that ex-combatants may face. The range of explanations of the problems that ex-combatants face with their families after receiving reinsertion packages in Uganda often reflected a perceived sense of animosity from families and communities towards ex-combatants. Common explanations included: (i) family wanted to take reinsertion money (19%), (ii) accusation of unfairness of payments to ex-combatants (14.3%); (iii) undermined and ridiculed by community (9.5%); (iv) accused of seeking government handouts (9.5%); and (v) attacked by neighbor for being an ex-combatant (9.5%). Drawing from data on DRC and RoC we can observe that in general ex-combatants feel that most people in their community treat them the same as they do everyone else – though this was slightly more so in DRC. This trend was reflected in reference to a range of different social categories, e.g. elders, male peers, female peers, work colleagues, people in authority, youth, and strangers. Younger ex-combatants (aged 18-30) were slightly less likely across almost all categories to feel that people in their community treated them the same as other non-ex-combatants in the com- munity. 11.3 Summary This limited examination of the DDR experiences of ex-combatants across the GLR countries reveals two key findings of substantial analytical value related to: (i) the importance of information and sensitization campaigns; and (ii) the considerably different levels of acceptance and welcome that ex-combatants perceive from family members versus the wider community upon initial return. In terms of information sensitization, ex-combatants’ levels of participation in various information sessions about the reinsertion and reintegration process has a clear correlation to ex-combatants’ levels of satisfaction with the level of information they receive. This in itself is perhaps not surprising, but what is more so is that ex-combatants’ levels of participation in information and sensitization sessions have an equally clear correlation to the actual rate at which ex-combatants receive reinsertion payments. Indeed, those ex-combatants who did not receive payment most 185   Rwanda and Burundi are absent from all findings on immediate reintegration experiences with the family and community due to lack of directly comparable data. 92 commonly cited lack of information as the reason why. The majority of ex-combatants are using reinsertion funds for their intended purpose of meeting immediate subsistence needs. As such, It appears as though effective informa- tion and sensitization campaigns, reaching a large proportion of ex-combatants, can play a key role in assuring that ex-combatants do indeed receive reinsertion funds to meet their immediate subsistence needs upon return to their communities – mitigating the economic burden on the families and wider communities that must absorb them. Throughout this report there has been considerable evidence to show that ex-combatants experience very high levels of acceptance and support from their immediate families upon return, in both social and economic dimen- sions. However, by contrast, there is also considerable evidence to suggest that they face only a relatively moderate level of acceptance from the broader community upon their immediate return. As outlined in section 17.2 of the community dynamics portion of this survey, community members hold high levels of fear surrounding the return of ex-combatants and the range of negative behaviors associated with them before ex-combatants’ return. However, community members’ fear surrounding ex-combatants all but disappears in the time before their arrival in com- munities to the time of sampling (4.05 years on average across the GLR countries). These findings are part of a dispersed range of evidence that suggest that while the vast majority of ex-combatants are quick to reach accep- tance and reintegrate into the family, they experience a slower, though positive, trajectory towards acceptance in the broader community. 11.3.1 Vulnerable Subgroups In terms of the limited range of DDR experiences explored here, female ex-combatants and disabled ex-combatants experience a clear and continuous range of disadvantages. Female and disabled ex-combatants are the demographic groups that are least likely to attend all varieties of information sessions about the reinsertion and reintegration processes. In turn female and disabled ex-combatants are not only the groups least satisfied with the level informa- tion that they received, but also the least likely groups to actually receive reinsertion payments – destabilizing their ability to meet their immediate subsistence needs. 11.3.2 Unique Country Trends As outlined repeatedly in the analysis in this section, the varying levels of ex-combatants participation in information and sensitization sessions related to reinsertion and reintegration processes displayed a clear correlation to the rate at which ex-combatants actually received reinsertion funds and presumably affected their ability to meet immediate subsistence needs. With this we can ask: how were more ex-combatants exposed to information and sensitization in Burundi and fewer in Uganda (the rest of the GLR countries falling in between)? What strategies were successful in some instances and unsuccessful in others? A more explicit understanding of the considerable variation in informa- tion and sensitization exposure across the GLR countries and its relationship to actual reinsertion payment recep- tion and meeting of subsistence needs then would hold considerable programming value for the future. 93 12. Conclusions T his study has found that across numerous dimensions ex-combatants in the GLR countries have been largely successful in reintegrating with community members. While across the range of core social and economic indicators explored in this study ex-combatants collectively represent a disadvantaged group, they show a clear trajectory towards reaching economic and social parity with community members – in many cases having already reached equal footing or occasionally exceeding community members’ performance across core indi- cators of reintegration processes. This study has found that among ex-combatants across the GLR countries, female ex-combatants and young ex-combatants (18-30) both male and female encounter a distinct range of additional challenges in reintegration processes and in this represent key vulnerable groups. While young ex-combatants lag behind their older peers, their overall trajectory is indeed positive. There is evidence to suggest that, however, for female ex-combatants across the GLR countries there is a consistent range of structural barriers that at the very least could slow down the processes of reintegration further, and at the very worst could leave them locked out of certain economic and social processes – at a high risk for economic marginalization and social isolation. There is no one driver or determinant of reintegration. Instead reintegration is understood here as embodied by a diverse range of simultaneous and overlapping processes (e.g. social, psychological, political, economic) that dy- namically interact with one another. In viewing the sum of these multiple reintegration processes and their interac- tion we can grasp the overall trajectory of reintegration that ex-combatants hold in their return to and interaction with the community. Violent conflict throughout the Great Lakes Region has damaged the social and economic fabric of society; dis- rupting economies, disintegrating families, and fragmenting social networks for ex-combatants and community members alike. Thus understanding the challenges of reintegration must in part be understood in the context of larger post-conflict peacebuilding and development processes. However the challenges that ex-combatants face in rebuilding and reintegrating into the damaged social and economic fabric of society are immediate and acute. It is ex-combatants’ ability to re-enter and make functional the familial unit and larger social networks in the community, in turn the social and economic functions these social units play, that constitute evidence of successful reintegration processes. With this in mind it appears that ex-combatants have been successful in reintegrating into the family unit. Ex-com- batants’ families have been open and accepting, serving the core function of the social and economic support while ex-combatants gain footing. While families appear to have played an especially important role in the immediate return of ex-combatants, the process of confrontation and exchange with the broader community appears to have progressed much more slowly. Rebuilding social networks is not only essential for acceptance and participation in the community, but for economic opportunity. In this, ex-combatants lag behind community members in their broader social footing and economic security – remaining especially reliant on the familial unit. It is with the support of the familial unit, and through their positive trajectory in terms of access to marriage, that ex- combatants have reached parity with community members in terms of housing, access to land, and upward mobility in land access. Though, ex-combatants continue to face challenges in terms of household hunger and nutrition. Despite ex-combatants’ positive trajectory, they perceive themselves as worse off than others in the community and see overcoming stigma and distrust in the community as the primary barrier to reintegrating with the commu- nity, followed by education and qualification barriers that may exist as a result of time lost in conflict. Community 94 members corroborate ex-combatants’ perceptions, explaining a range of fears and stigma associated with ex-com- batants upon their immediate return that, however, dissipate quickly over time leaving key barriers to ex-combat- ants’ reintegration as revolving around making up time for missed education / skills qualification attainment and a broader social and economic track record in the community. Even with the diverse range of social and economic challenges that ex-combatants face, they have strong senses of empowerment to shape their situation going forward (with the exception of females). 12.1 Ex-Combatants and Economic Reintegration This study has found that ex-combatants across the GLR countries show a positive trajectory towards gaining self- employment in agriculture or small business – though there is still considerable improvement that ex-combatants must make to reach parity with community members. The context of economic reintegration across the GLR countries is one of severe overall development challenges. As such, ex-combatants and community members alike identify their primary barrier to gaining employment as lack of opportunity in general. Ex-combatants, however, face a range of additional barriers related to: (i) closing literacy, education and skill gaps with community members; (ii) establishing an economic track record; in order to (iii) access credit and other financial institutions; and (iv) to erode stigma and distrust through the slow process of confrontation that social reintegration entails. Collectively the unique barriers that ex-combatants face are a product of their overall stunted economic networks – leading to an overall higher exposure to economic insecurity and reliance on the familial unit compared to com- munity members – who have more diversified economic networks and tend to be more integrated into formal and community based economic institutions. 12.2 Ex-Combatants and Social Reintegration This study has found that ex-combatants across the GLR countries exhibit a positive, but shallow, trajectory of social reintegration. While ex-combatants are quick to reintegrate with their immediate family and to breakdown trust barriers with the wider community, their progress from there forward is slow – owing to their stunted social networks and track record in the community. In this sense the social and economic reintegration of ex-combatants are strongly interrelated – their long-term success revolving around rebuilding the networks that are instrumental to social and economic security. Mending the damaged social networks and in many cases building new ones from scratch, with the additional barriers of residual stigma and lack of social track record, is a slow process of confronta- tion and atonement. Though ex-combatants have remarkably high levels of agency in terms of social empowerment, often surpassing community members, they understand that the process of social reintegration has no shortcuts and will take place in the scale of years. With this ex-combatants remain significantly less happy and with lower levels of self worth than community members – leaving them exposed to risk of marginalization and social isolation from the wider com- munity. 12.3 Female Ex-Combatant Subgroup Throughout the analysis of the reintegration processes of ex-combatants across the Great Lakes Region presented in this study female ex-combatants have stood out as the most clear and consistent vulnerable subgroup. While young ex-combatants (aged 18-30) and disabled ex-combatants display a range of disadvantages related to a lack of social and economic track record and to health, respectively, they do not depart significantly from the overall positive trajectory of ex-combatant reintegration in the Great Lakes Region. Female ex-combatants, however, encounter an extensive range of disadvantages that collectively paint a picture of the structural barriers they face in reintegration processes in terms of: (i) familial networks; (ii) economic networks; and (iii) broader social networks in the com- munity. These structural barriers force a distinctly different trajectory of reintegration. 95 As discussed above, rebuilding damaged social and economic networks in the community is a key dimension to the overall process of reintegration. Like male ex-combatants, female ex-combatants have done well to reintegrate into their immediate families. However, unlike male ex-combatants, female ex-combatants have been largely unsuc- cessful in building new familial connections through marriage – remaining the least marrying demographic group across ex-combatants and community members alike. While male ex-combatants have seen a sharp rise in marriage rates since demobilization, female ex-combatants have shown a very shallow, and ultimately marginal, trajectory of increased marriage. Attitudinal indicators reveal that female ex-combatants have the smallest proportion of the pop- ulation, both ex-combatant and community member, that is open to marrying them due primarily to stigma related issues – lending some explanation for growing disparity in marriage rates between male and female ex-combatants. Both male and female ex-combatants experience stigma, however in the case of marriage it appears to likely be a key structural barrier to building new familial networks and in turn to accessing the social and economic resources that they represent. As discussed above, the context of economic reintegration in the Great Lakes Region is one of severe development challenges. The primary pathway to economic stability for ex-combatants and community members alike is through self-employment in agriculture – this is even more so for female ex-combatants. As such, access to arable land is an important indicator of economic stability – in turn growth in access to arable land as an indicator of a positive economic trajectory. In this female ex-combatants lag behind male ex-combatants with slightly lower levels of both land access and improvement in access to land – which when combined indicate female ex-combatants’ shallower trajectory of economic improvement – despite their slightly lower unemployment rate. There appear to be three structural barriers to land access mobility for female ex-combatants: (i) capital; (ii) inheritance; and (iii) marriage. First, both male and female ex-combatants alike identify access to capital as the largest barrier to increased land access. However, for female ex-combatants, who have considerably lower literacy and educational achievement levels, the challenges to accumulating capital through bountiful agricultural production are acute. Indeed, females clearly identify lack of education and skills as among their key barriers to economic stability. Second, the challenges to capital accumulation that female ex-combatants face are amplified when inheritance dynamics are taken into account. Females who do experience increases in land access are much less likely than male ex-combatants to cite inheritance – indicating that this is a pathway to land access, and in turn a positive economic trajectory, that females are not accessing at the same level. Thirdly, marriage is an important pathway to increased land access for male ex-combatants. However, as outlined above, female ex-combatants experience a set of distinct structural barriers to accessing marriage. When these three dimensions interact, the result is a dynamic structural barrier that female ex-combatants face in terms of building a positive economic trajectory and the economic networks associated with them. By effect of their structurally hindered familial and economic networks, female ex-combatants face challenges in building social capital and broader networks in the community. This weak social capital in the community has con- sequences for female ex-combatants in terms of lower levels of trust, lower perceptions of improvement in trust, dramatically weaker senses of empowerment, and lower perceptions of their overall situation than the rest of ex-com- batants. These factors collectively interact to put females at risk of marginalization and isolation with the community – in turn potentially reinforcing the structural restraints that shape their weak familial and economic networks. It is the dynamic interaction of the familial, economic and broader social structural dimensions that shape the overall shallower trajectory of reintegration for female ex-combatants across the GLR countries and constitute them as a distinctly disadvantaged group. Looking at the structural challenges that female ex-combatants face reveals much about the overlapping, interrelated, and simultaneous nature of reintegration processes – an insight that is not only relevant to female ex-combatants, but to all ex-combatants across the Great Lakes Region. 96 Annex II - Great Lakes Region Community Dynamics: Comparative Survey Results and Analysis 97 13. Demographics T he following is a capture of the community member sample for this comparative study. The demographics presented here are not representative of the overall community member populations of each of the five GLR countries of study, but instead reflect a range of purposive sampling biases. For more information about the specific sampling methods and decisions in each of the GLR countries please see the individual survey studies in each of the five GLR countries.186 For a brief introduction to the reintegration programming context in each of the GLR countries seesection 5.1. The total unweighted sample of community members from across the five GLR countries amounts to 3,380 respon- dents which, when combined with the ex-combatant sample of 6,475 respondents, represents 34.3% of the total GLR sample. The total unweighted community member sample contributions from each of the five GLR countries are as follows: Burundi comprises 15.1% (n=510) of the total Community Member Sample, DRC 21.4% (n=722), RoC 43.1% (n=1456), Rwanda 15.1% (n=510), and Uganda 5.4% (n=182). However, in an effort to create valid cross-country analysis of community members across the GLR, and especially for comparison to the ex-combatant sample, which contains proportionally different sample contributions from the five GLR countries, the raw sample contributions from each country have been weighted evenly. Further, for reasons explained below the valid sample used for analysis in this study are often notably lower than the total sample of 3,380 community members. Though Burundi does contribute 510 respondents to the total GLR Community Member sample no age, gender or disability details were collected for respondents as a part of the Third Beneficiary Assessment in 2011 – thus a systematic analysis of the Burundi portion of the total GLR community member sample along demographic lines is not possible. In addition, little data was collected in Burundi that is directly comparable to the rest of the GLR data anyways. In effect, with the exception of some short sections, data from Burundi will be absent from the analysis here thus leaving the unweighted valid sample of community members at n=2870. Unless explicitly stated other- wise, all cross country statistics from here forward refer to the valid sample excluding Burundi. In addition, as discussed in more detail in Annex I, integration of the full range of data from the Rwanda Survey has proved challenging in this study. The evolving format for the individual GLR country surveys has been a continual process of learning and iterative refinement. The Rwanda survey format is the starting point from which surveys evolved in RoC, Burundi, Uganda, and DRC. So, while data content in the Rwanda surveys is very much in line with the rest of the GLR countries, much of the specific question formatting is often different enough that a direct comparison of data is not feasible. Such instances are explained in footnotes. Data along health and disability demographics also presents challenges in the total GLR community member sample. Health and disability data for community members were only collected in Rwanda and Uganda – absent from Burundi, DRC and RoC. However, even the data from Rwanda and Uganda is limited as only n=58 disabled community members were sampled (n=49 from Rwanda and n=9 from Uganda). Thus, drawing valid comparisons between these two samples of 49 and 9 disabled community members is judged as infeasible – furthermore, com- 186   For Burundi see (2011) Troisieme Enquete des Beneficiarires du PDRT; for DRC see (2011) Rapport d’Evaluation des Bénéficiaires du PNDDR Réintégration des hommes soldats et ex-combattants démobilisés; for RoC see (2011) Réintégration Socio-Economique des ex-Combattants Région du Pool, République du Congo; for Rwanda see (2012) Comparative report: Ex-combatant and community study on the socio-economic reintegration; and for Uganda see (2011) Reporter Re-integration and Community Dynamics Survey Report Reporter Re-integration and Community Dynamics Survey Report. 98 Table 31: Community Members - GLR Country Demographics Country DRC Republic of Congo Rwanda Uganda GLR Total Male 52.2% 54.8% 68.8% 68.7% 61.2% Female 47.8% 45.2% 31.2% 31.3% 38.8% Subtotal 100.0% 100.0% 100.0% 100.0% 100.0% Age 18-30 34.2% 20.3% 57.0% 46.7% 40.0% Age 31-40 26.7% 22.4% 25.9% 22.2% 24.3% Age Over 40 39.1% 57.3% 17.1% 31.1% 35.7% Subtotal 100.0% 100.0% 100.0% 100.0% 100.0% paring these 58 disabled community members to the 454 disabled ex-combatants in Annex I of this study presents further issues for validity. For these reasons analysis of community members along the lines of disability will be absent from this section of the study. Collectively the data restrictions present in this study of community members across the GLR countries mean that the task of this study is to present a mosaic of findings. Up close, the pieces of the picture are not always complete and data is not always congruent. However, there are clear data trends, nonetheless, that represent a distinct narra- tive of community dynamics across the GLR countries. Across the GLR countries, 61.2% of community members were male while 38.8% were female. In DRC and RoC there was a fairly close split between male and female community members within the sample, while Rwanda and Uganda were closer to the cross-country average split in gender. Table 31 above presents a cross-tabulated break- down of age and gender demographics for the community member sample of each of the GLR countries. Of the total sample of community members, 40% were between the ages of 18 and 30, 24.3% were between that ages of 31 and 40, and 35.7% were over the age of 40.187 The within-country age splits of each of the GLR countries do not necessarily follow cross-country trend. As is visible in Table 31 above, Rwanda and Uganda community members between the ages of 18 and 30 are most dominantly represented while in RoC those over 40 are most represented and DRC falls closer to the cross-country split. The dimensions of the lives of community members explored in the following sections are key indicators of com- munity dynamics and furthermore relate to the basic units and processes in society: the family unit, and the process of marriage, in the community. The value of this section of the study is not just as a control group for which ex- combatant progress can be studied, but also as a key measure of the overall levels of social and economic stability of the core units of reintegration across society in the Great Lakes Region. 13.1 Marriage and Household Marriage dynamics are an important indicator of community members’ basic social standing. Indeed, marriage dynamics can tell us much about community members’ ability to leverage familial, economic, and social networks towards the attainment of marriage and in turn their ability redouble their engagement in these social structures through marriage – all indicators of a strong footing in the community. 187   Across the total sample of community members from the GLR countries there were 26 respondents under the age of 18. For purposes of consistency in sample delimitation and comparative validity these 26 (3 from DRC and 23 from RoC) have been omitted from the sample for analysis here. 99 Table 32: Community Member Marital Status Marital Status at Sampling Single/Never Married Living together Divorced or Separated Widowed married Male 55.40% 16.20% 2.70% 2.00% 23.70% Female 38.10% 18.00% 11.10% 12.80% 20.10% Age 18-30 35.80% 13.50% 2.60% 0.30% 47.70% Age 31-40 60.40% 19.10% 6.00% 3.30% 11.20% Age Over 40 57.30% 17.10% 9.60% 14.40% 1.60% DRC 66.10% 7.10% 7.30% 7.10% 12.50% RoC 18.50% 53.50% 12.30% 10.90% 4.90% Rwanda 46.90% 4.90% 0.40% 2.60% 45.30% Uganda 62.60% 2.70% 3.80% 4.40% 26.40% GLR Average 48.60% 16.90% 5.90% 6.20% 22.30% Across the GLR countries the most common groupings for marital status of community members are as follows: 48.6% are married, 16.9% are living with a partner but are not married, 22.3% are single and have never been married, 5.9% are separated or divorced, and the remaining 6.2% are widowed. These figures are very much an average as within each of the GLR countries community members displayed a more unique distribution of marital statuses (summarized in Table 32 below). In DRC and Uganda “married” is the most common marital status for community members – at over 60% in both countries. While in Rwanda “married” is still the most common marital status, it is almost evenly split with “single / never married”. The country that differs most from the general trend is RoC, in which “living together” but not married is the most common marital status. Across the GLR countries female community members were less likely to be married than male community members (38.1% vs. 55.4%). In DRC and Uganda this trend was exaggerated and the gap between male and female commu- nity members with the marital status “married” was as much as 38 percentage points (in Uganda). At a cross-country level the lower representation of married female community members was effectively absorbed into the categories of separated or divorced and widowed (11.1% and 12.8% respectively) which were much less common among male community members (2.7% and 2.0% respectively). These findings should flag divorce, separation, and widowing as key dimensions to female community members’ absolute disadvantage in marriage rates across the GLR countries compared to male community members. There was a visible positive relationship between age and likelihood of being separated, divorced, or widowed. Of community members 18-30 years of age, 2.6% were separated or divorced and 0.3% were widowed, compared to 6% and 3.3% (respectively) of those aged 31-40, and 9.6% and 14.4% (respectively) of those aged over 40. Drawing exclusively from Rwanda and Uganda we can observe that only 5.8% of community members had a spouse (married or unmarried) that was an ex-combatant, the remaining 94.2% having civilian spouses – though this does not necessarily imply that their spouse was a combatant / ex-combatant at the time of marriage.188 On average female community members were more likely to have an ex-combatant or combatant spouse (11.1%) compared to male community members (3.9%). This could serve as at least a partial explanation for female community members’ higher levels of widowed marital status – while anecdotally this makes sense further study would be needed to confirm this relationship 188   Rwanda and Uganda were the only GLR countries where surveys included questions on community members’ spousal ex-combatant status. 100 Across the GLR countries 74.7% of community members report that they would not consider marrying an ex- combatant, with the remaining 25.2% saying that they would consider marrying an ex-combatant. This trend was generally reflected within the individual GLR countries with the exception of Uganda, where 56.8% of community members said that they would consider marrying an ex-combatant and 43.2% would not.189 Across demographic lines age showed a positive relationship with unwillingness to consider marrying an ex-combatant. Of those over 40 years of age, 85% were not willing to marry an ex-combatant, compared to 74% of those aged 31-40, and 66.7% of those aged 18-30. The most common explanations for why community members were unwilling to marry an ex- combatant revolved around various forms of stigma or fear. Drawing exclusively from DRC and RoC, we can see that when community members observed other marriages in the community in which one member is an ex-combatant, 44.9% of community members perceived these mar- riages as having a harder time than those without an ex-combatant.190 When asked to explain further as to why they thought these marriages were more difficult the most common explanations were as follows: (i) 35% of community members cited misunderstandings; (ii) 22% cited brutality and fighting; and (iii) 8.5% cited bad habits of ex-com- batants acquired during combat (including drug use). Across the GLR countries 43.6% of community members see themselves alone as responsible for the financial and food need of the family, 16.8% see their spouse or partner as responsible, and 25.5% see food and financial needs as the shared responsibility of both themselves and their spouse or partner. The remaining 14.1% indicated that household food and finance responsibility were dispersed among various other family members.191 Male communi- ty members were significantly more likely to see household finance and food provision as solely their responsibility (56.6%) compared to female community members (21.8%). Inversely, female community members were dramati- cally more likely to see household finance and food provision as the sole responsibility of their spouse or partner (36.8%) compared to male community members (4.7%). This gendered trend was especially exaggerated in DRC (61.3 vs. 16.7% and 48.7 vs. 6.7 respectively). As age increases community members are more likely to see themselves as solely responsible for household finance and food provision (32.7% of those age 18-30, 44.4% of those 31-40, and 51.5% of those over 40) and less likely to see their spouse or partner as solely responsible (22.7% of those age 18-30, 18.8% of those 31-40, and 10.3% of those over 40). It appears that this age-based trend may be primarily descriptive of female community members. When looking at age trends in community members’ perceptions of household finance and food provision further subdivided by gender there are distinct trends. Male community members see themselves as primarily responsible for their household finance and food provision at even levels across all age categories (55.3% of those 18.30, 59.4% of those 31-40, and 55.8% of those over 40). In contrast, as age increases for females so too does the likelihood of seeing oneself as solely responsible for household finance and food (9.2% of those 18-30, 20.1% of those 31-40, and 40.2% of those over 40). As discussed above, as age increase so too does the likelihood of being separated, divorced, or widowed. As such, those female community members who are separated, divorced, or widowed are highly likely to see household finance and food provision as solely their responsibility (62.1% of those who are widowed and 70.1% of those who are separated or divorced). These findings should flag female-headed households as exposed to particular economic instability.192 189   It is possible that the greater openess in Uganda is related to the nature of mobilization and reurn in which many ex-combatants were abducted or forcibly recruited into conflict – upon return being simultaniously understood as victims and perpatrators. This dyanmic plus the widespread employment of traditional reconciliation ceremonies in Northern Uganda (though not necessarily as a part of reintegration programming) may hold some explanitory wieght. 190   DRC and RoC were the only countries where community members were asked about their perception of marriages in which one person was an ex-combatant. 191   Rwanda is excluded from findings on household finance and food responsibility due to lack of directly comparable data. 192   A slightly exaggeraged version of this age trend in female community members existed for female ex-combatants as well. 101 13.2 Literacy, Education, and Vocational Training Levels of literacy, educational achievement, and vocational training are important indicators of community members’ basic ability to engage with educational and vocational structures, to the extent they exist in the different GLR country contexts, and further to leverage the dividends of this engagement towards further economic and social opportunities – in the end solidifying their footing in the community. Literacy was generally high among community members Table 33: Community Member Literacy across the GLR countries, and it Marital Status at Sampling was slightly higher than for ex- Neither Read nor combatants; 74.3% of commu- Read only Read and write Write nity members could both read Male 12.90% 4.10% 83.00% and write, 4.6% could only read, Female 32.40% 5.20% 62.40% and the remaining 21.1% were Age 18-30 13.50% 3.50% 83.10% illiterate (compared to 71.6%, 8.3% and 20.1% respectively in Age 31-40 17.60% 5.50% 76.90% ex-combatants). Notably, RoC 193 Age Over 40 28.80% 4.30% 66.80% had the lowest literacy levels DRC 16.30% 4.30% 79.40% across the GLR countries. Female RoC 27.70% 7.30% 65.00% community members are notably less likely to be able to read and Uganda 19.30% 2.20% 78.50% write (62.4%) compared to male GLR Average 21.10% 4.60% 74.30% community members (83%), and GLR Average 48.60% 16.90% 5.90% more likely to be illiterate (32.4% vs. 12.9%). After female community members, those aged over 40 are the second most likely group to be illiterate (28.8%). These trends are displayed in Table 33. In regards to educational achievement, community members most commonly had either some secondary education (31%) or had completed secondary education (18.2); followed by some primary education (17.1%), completed primary education (11.2%), and no education (11.1%).194 As is visible in Table 34, in Uganda education levels were skewed lower overall. Across demographic lines there are a few interesting trends to extract. Female community members are the most likely group to have no education (18.3% vs. 6.1% of male community members) followed by those aged over 40 (17.3% vs. 8.4% of those aged 31-40 and 5.6% of those aged 18-30). Further, both female com- munity members and those over 40 had educational achievement levels skewed lower overall. Across the GLR, overall levels of educational achievement for community members were skewed higher than for ex- combatants – with higher levels of partial secondary education (31% vs. 23.4%), full secondary completion (18.2% vs. 5.2%), and partial or full higher education (8.5% vs. 2%). It is worth noting however that in DRC and RoC ex-combatants displayed much higher levels of professional educational achievement than community members (34.3% and 20.3% vs. 2.4% and 1.8% respectively). Most community members reflected an understanding of the educational achievement gap between community members and ex-combatants – (58.4%) of community members reported that they believe that ex-combatants have 193   Rwanda is excluded from findings on literacy due to lack of directly comparable data. 194   Rwanda is excluded from findings on education achievement levels due to lack of directly comparable data 102 Table 34: Community Member Educational Achievement Levels Education Level at Sampling higher education Some secondary Islamic or reli- Some primary Some higher Professional Completed Completed Completed secondary education primary gious None Male 6.10% 1.60% 16.00% 11.20% 33.90% 18.10% 4.40% 6.90% 1.70% Female 18.30% 1.50% 18.70% 11.20% 26.90% 18.00% 1.80% 2.80% 0.90% Age 18-30 5.60% 0.00% 17.50% 10.30% 39.20% 17.90% 3.30% 5.10% 1.10% Age 31-40 8.40% 2.40% 15.30% 8.60% 32.90% 19.80% 4.70% 6.90% 0.90% Age Over 40 17.30% 2.20% 18.00% 13.30% 22.80% 17.00% 2.70% 4.80% 1.90% DRC 11.80% 0.40% 13.00% 8.00% 30.30% 22.60% 4.70% 6.80% 2.40% RoC 14.60% 0.70% 12.90% 17.10% 20.00% 28.00% 3.00% 1.90% 1.80% Uganda 7.30% 3.40% 25.10% 8.90% 41.90% 4.50% 2.20% 6.70% 0.00% GLR Average 11.10% 1.50% 17.10% 11.20% 31.00% 18.20% 3.30% 5.20% 1.40% lower levels of education than other people in the area in which they live.195 Of the remaining community members, 40.3% believe that ex-combatants and civilians have the same level of education and only 1.3% perceives that ex- combatants have higher levels of education. In addition, 76.9% of community members said that the perceived difference in levels of education between community members and ex-combatants was a problem. When asked to explain further in Uganda community members most commonly pointed out that (i) ex-combatants wouldn’t be able to gain employment and thus look after their families (33.9%) and that (ii) low literacy was a problem. In DRC and RoC the most common responses from community members as to why ex-combatants’ lower education levels were a problem were: (i) Irresponsible behavior (36.9%) and (ii) misunderstandings that lead to arguments (31.4%). Across the GLR countries, 20.8% of community members received vocational training in the last years.196 Male com- munity members more frequently (22.9%) received vocational training compared to female community members (17.6%). Age also showed a clear relationship to vocational training – the higher the age of community members the less likely they were to have received vocational training (26.2% of those aged 18-30, 24.7% of those aged 31-40, and 14.6% of those over the age of 40). Of those 20.8% of community members who had received vocational training, 78.3% said that they were currently using the vocational skills that they had been trained in.197 When the 21.7% of community members who were not using their vocational training were asked to explain further the most common reasons cited were: (i) lack of tools or work facilities (27%); (ii) still completing training; and (iii) no opportunity. Interestingly, while ex-combatants have 195   Rwanda is excluded from findings on community members’ relative perceptions of ex-combatants’ education levels. 196   In Uganda this question referred to in the last five years, while in DRC and RoC it referred to only the last year. It makes some sense then that Uganda displays the highest rate of community member vocational training across the GLR countries (29.6%). Rwanda is excluded from findings on vocational training due to lack of data. 197   Regardless of having received vocational training or not, 38.8% of community members across the GLR countries reported that they were currently working in their “field of skills”. Male community members were slightly more likely to be working in their field (40.1%) when compared to female community members (35.7%). 103 received vocational training on average at twice the rate that community members do, in DRC as much four times the rate and in Uganda at a quite similar rate, they are still less likely to be utilizing that vocational training than com- munity members (62.7% vs. 78.3%). This could be an indication that while ex-combatants display a positive trajec- tory in terms of closing education and skills gaps with community members, there may be additional barriers they face to reaching parity that, at least in part, may revolve around access to social and economic networks or possibly problems in reintegration programming design.198 13.3 Summary The analysis of community member demographics and core indicators presented in this section are useful not only as a backdrop against which to contextualize ex-combatant reintegration in the Great Lakes Region, but more gen- erally as a baseline by which to understand the overall levels of societal stability and functionality of communities across the Great Lakes Region in the wake of violent conflict. Across the indicators explored in this section, community members consistently perform better than ex-combatants. Community members are more likely to be married than ex-combatants – a fact that may help explain why com- munity members are less likely than ex-combatants to see themselves alone as responsible for the food and finances of their household. Community members are married to ex-combatants at half the rate that ex-combatants are and attitudinally remain largely closed to the idea – citing stigma as core reason.199 Indeed, when community members observe marriages in which one member is an ex-combatant they commonly describe these marriages as problematic. These findings have two core implications: (i) stigma is a core barrier to community member / ex-combatant intermarriage across the GLR countries and (ii) beyond actual marriage rates the pool of partners who are attitudinally open to marriage with ex-combatants is largest among other ex-combatants. If stigma shapes a portion of ex-combatants marriage pathway as to one with only other ex-combatants this could have consequences for these familial units’ ability to interact with the community – in a sense possibly facing compounded stigma barriers. The evolving nature of community member and ex-combatant intermarriage should be flagged as a key issue for future studies in the GLR. Community members across the GLR have education levels skewed significantly higher than ex-combatants – likely a result of ex-combatants’ time lost while mobilized in conflict. Yet, community members are far less likely than ex-combatants to receive vocational training as a part of reintegration related programming. Despite this, however, community members are actually more likely to be currently utilizing their vocational training – suggesting that there may be additional barriers to utilizing vocational training for ex-combatants including programming flaws. Developing a stronger understanding of this dynamic should be flagged for future studies in the region. 13.3.1 Vulnerable Subgroups In the demographics analyses of the community member sample for this study the most vulnerable sub-group that emerges, as with the ex-combatant sample, is that of females. Generally speaking, female community members have a similar range of disadvantages to male community members as female ex-combatants do to male ex-combatants. However, the general gap between community member and ex-combatant samples is such that female community members (and male community members for that matter) are almost always significantly better off than their ex- 198   Example of problems with vocational training components of reintegration programming design can include, for example, that vocational training paths offered are not based on market analysis, in turn creating an oversupply of a particular set of skills in one area. This phenomenon is well documented in a number of DDR programming contexts. See for example: Jennings, K. M. (2007). The struggle to satisfy: DDR through the eyes of ex-combatants in Liberia. International Peacekeeping, 14(2), 204-218. 199   It is important to note that these findings do not differentiate between those community memebers who married combatants or ex-combatants and those who married a civilians who later became combatants / ex-combatants. 104 combatant counterparts. Female community members are less likely to be married than male community members (who are themselves similarly likely to be married as male ex-combatants), though they are slightly more likely to be married than their female ex-combatant counterparts. Female community members are three times more likely than male community members to be separated or divorced and six times more likely to be widowed – though it is unknown whether the male or female initiated the divorce or separation. Regarding marriage to ex-combatants, female community members are three times more likely than their male counterparts to be married to an ex-combatant – though it is unknown whether these marriages pre-exist the combatant / ex-combatant status of their spouse. However, female ex-combatants are four times more likely than female community members to be married to an ex-combatant. Similar to the ex-combatant sample, female community members had significantly lower literacy levels than male community members. Both male and female community members displayed slightly higher levels of literacy than ex-combatants – in accord, community members education levels were skewed higher overall than ex-combatants. However, female community members’ educational achievement was skewed below that of male community members. 13.3.2 Unique Country Trends There are several important marriage-related trends that stand out in individual GLR countries that merit further ex- amination. Rwanda stands out as the only country where community members marry less frequently than ex-com- batants. However, we can add considerable contextual detail here. In Rwanda males are required to have access to adequate housing in order to get married. However, the formal regulations for what qualifies as adequate housing in Rwanda are somewhat narrowly defined under the policy of imidugudu - a large scale body of housing policy aimed at consolidating dispersed housing in an overall effort toward villagization. The result has been inflation in adequate housing prices and in turn a severe crisis in the availability of adequate housing overall that in effect is locking many Rwandans out of official marriage – though they may cohabitate without formalized marital status.200 In contrast to community members, most ex-combatants are returning to Rwanda from Eastern DRC where they have been away for an average of nine years. In this time some ex-combatants have married and when returning to Rwanda bring their spouse with them. The legal status of these marriages in Rwanda is unclear, however it is possible that some ex-combatants unwittingly navigate past the formal barriers to marriage that community members face – in turn accounting for their slightly higher marriage rates. While it is likely that the interaction of housing policy, marriage, and dynamics of return are key in understanding why ex-combatants marry more than community members in Rwanda this exact narrative must be treated as conjecture. These topics should be flagged for future analysis on reintegration processes in Rwanda. Republic of Congo also stands out with unique marriage trends. In RoC marriage rates among community members, and ex-combatants, are a fraction of those in other GLR countries. Instead, cohabitation with a spouse without formal marriage is the primary marital status – even when disaggregated across age and gender groups. These findings are confounding and go without clear explanation in this study. It is possible that: (i) there has been an unbeknownst error in data capture and coding that produces these findings or (ii) that there is an unknown regional dynamic af- fecting marriage for community members and ex-combatants alike in RoC. Future study on reintegration processes in RoC should flag marriage as an area of special interest to further triangulate or refute these findings. Lastly, in Uganda community members are more than twice as likely as the GLR average to report willingness to 200   This narrative of the interrelated nature of housing policy and marriage in Rwanda is well documented in Sommers (2012) Stuck: Rwandan Young and the Struggle for Adulthood. 105 marrying an ex-combatant in the future. Though there is no direct explanation it is possible that the specific dynamics of combatant mobilization in Uganda may play a role in this trend. In Uganda abduction was a well-known tactic of mobilization, especially by the LRA. Though abductees may have committed violent acts against their communities, often forcibly, there is evidence that ex-combatants are simultaneously understood as victims (due to abduction and forced recruiting) and perpetrators (due to the violence committed as soldiers) by community members – a factor that has reportedly contributed to a general willingness to accept returning ex-combatants back into communi- ties.201 This dynamic may contribute to community members in Uganda’s openness to marriage with ex-combatants. Futures studies could flag this conjecture for further analysis. 201   See for example: Finnegan, A. C. (2010), Forging Forgiveness: Collective Efforts Amidst War in Northern Uganda. Sociological Inquiry, 80: 424–447 106 14. Housing, Land, Livestock and Food Security T he context of communities in the Great Lakes Region is overwhelmingly one of severe development chal- lenges where small-scale agriculture is instrumental to individual and familial well-being in terms of both economic security and food security. As such, understanding the pathways to land access among communi- ty members is a key contextual element for understanding the overall economic situation for community members in the Great Lakes Region and their capacity to absorb returning ex-combatants. The following is an examination of (i) the household characteristics of community members including issues of dwelling ownership and tenure; and (ii) the food security of community members including their levels of access to land for cultivation. Across the GLR countries community members are most likely to live with: (i) the same family as before conflict (38.3%); (ii) with a family but different to that from before conflict (27.3%), or (iii) with a spouse or partner (21.5%). These three categories were the most common across all GLR countries, but varied some in their distribution from country to country. Uganda, where 51.7% of community members live with the same family; 25% live with a dif- ferent family; and 22.2% live with a spouse or partner, displays the most exaggerated version of this cross-country trend. As is visible in Table 35 below, Rwanda stands out as the Table 35: Community Member Household Membership only GLR country with a notable Who Do You Live With? portion (15%) of community With same family as members who live with friends.202 With family but dif- ferent to that from before the conflict before the conflict Family (unidenti- With spouse / With friends partner Female community members Alone Other fied) were slightly more likely to be living with a different family than the one before conflict (29.8%) Male 5.60% 23.60% 38.50% 25.70% 5.20% 0.20% 1.10% and slightly more likely to live alone (7.3%) than their male Female 7.30% 18.10% 38.20% 29.80% 3.30% 0.90% 2.40% counterparts (24.3% and 5.6% Age 18-30 6.70% 18.40% 37.30% 27.90% 8.60% 0.30% 0.80% respectively). In addition, female Age 31-40 3.60% 23.60% 34.80% 32.90% 3.20% 0.40% 1.60% community members were less Age Over 40 7.20% 23.00% 43.10% 22.90% 1.10% 0.60% 2.10% likely to be living with a spouse DRC 7.50% 29.40% 43.80% 17.50% 1.90% 0.00% 0.00% or partner than male community members (18.1% vs. 23.6%). RoC 10.00% 33.80% 20.80% 26.60% 0.60% 1.80% 6.40% Rwanda 7.10% XXX 37.00% 40.70% 15.00% 0.00% 0.20% In regards to housing, across Uganda 0.60% 22.20% 51.70% 25.00% 0.60% 0.00% 0.00% the GLR countries community members were most commonly GLR Average 6.30% 21.50% 38.30% 27.30% 4.50% 0.50% 1.70% living in a house (42.3%), followed The use of XXX indicates that respondents in Rwanda were not given the option to respond that they live with by a hut or tent (25.9%). This a partner or spouse – the responses that would have been in the feild are likely absorbed into the categories of those who live with a family either the same or different from the one before conflict. trend is generally reflected across 202   This trend in Rwanda is possibly related to housing shortages as a product of Imidigudu and overall urban migration. 107 Table 36: Community Member Housing Ownership Who Owns The Housing In Which You Currently Live? Family I live Spouse, with such as Non-family My relatives, I own it Joint owners My friend Other partner son, mother, member family parents Male 54.50% 2.00% 4.30% 9.50% 0.00% 1.00% 8.40% 20.30% Female 25.10% 20.90% 5.50% 11.80% 0.00% 1.10% 12.40% 23.00% Age 18-30 23.20% 7.80% 2.90% 21.70% 0.00% 1.20% 12.70% 30.50% Age 31-40 44.70% 10.30% 6.10% 4.10% 0.00% 0.50% 10.70% 23.60% Age Over 40 64.60% 10.20% 5.90% 2.60% 0.00% 1.10% 5.70% 9.90% DRC 37.50% 11.10% 7.60% 7.20% 0.00% 0.80% 12.60% 23.10% RoC 38.20% 14.40% 6.90% 2.00% 0.00% 1.80% 19.20% 17.50% Rwanda 35.30% 4.10% 0.00% 14.50% 0.00% 1.60% 4.10% 40.40% Uganda 61.70% 7.80% 4.40% 17.80% 0.00% 0.00% 4.40% 3.90% GLR Average 43.20% 9.30% 4.70% 10.40% 0.00% 1.00% 10.10% 21.30% the GLR countries with the clear exception of Uganda, where 67% live in a hut or tent, 18.1% live in a daub or wattle, and 14.3% live in a house.203 Rwanda also stands out in that 68% of community members live in a house and only 1.8% live in a hut or tent. Across the GLR (including Uganda) female community members are more likely to live in a house (50.2%) than male community members (37.2%). Marital status does not show a directly discernable relationship to housing type. Age also shows a clear positive relationship to living in a house – 27.7% of those com- munity members aged 18-30 live in a house, 45.1% of those aged 31-40, and 55.1% of those over the age of 40.204 Turning to patterns of ownership for housing among community members we can observe that most community members either: (i) own the property they live on (43.2%); (ii) their relatives or parents own the property they live on (20.5%); or (iii) their spouse / partner owns the property that they live on (9.3%). This trend was remark- ably durable across the GLR countries, though in Uganda there were much higher levels of self-ownership (61.7%) and in Rwanda renting was on par with self-ownership (36.9%).205 Despite the differences in housing ownership among community members in Uganda and Rwanda there are very clear trends across gender and age demograph- ics. As visible in Table 36 Female community members are less than half as likely to own the property they live on compared to male community members (25.1% vs. 54.5%), and dramatically more likely to have their spouse / partner own the property they live on (20.9% vs. 2%). In regards to age demographics, as the age of community members increases the likelihood that relatives or family own their housing decreases (34.4% of those community members aged 18-30, 14.8% of those aged 31-40, and 8.3% of those over the age of 40) and the likelihood of self-ownership increases (23.2% of those aged 18-30, 44.7% of those aged 31-40, and 64.4% of those over the age of 40). 203   It is possible that the dominance of semi-permanent housing in Uganda is an indication of the overall development level in Northern Uganda. It is also possible that there is merely a reflection of traditional housing style preferences. 204   This age demographic trend is not reflected in Uganda – where hut / tent was the most common housing type across all demographics. For a more in depth discussion of community member housing types see the Uganda Reporter Re-integration and Community Dynamics Survey Report (2011). 205   The higher rate of ownership in Uganda is likely a result of the lower barriers to owning the dominant housing type in Uganda: a hut or tent. Inversely, in Rwanda housing shortages as a product of Imidugudu have increased the barriers to housing ownership. 108 Community members across the GLR countries generally see themselves as equally as well off as their neighbors (57.3%), though 21.2% see themselves as worse off and 21.3% see themselves as better off. All of the GLR countries reflect this trend of the bulk of community members seeing themselves as on equal footing to their neighbors. In most countries the distribution is skewed towards seeing themselves as slightly better off, though RoC is the only country where this skewing goes the other direction. Female community members were more likely to see them- selves as worse off than their neighbors compared to male community members (25.2% vs. 18.6%). In DRC and RoC, community members were asked how they perceived their living situation relative to two years prior. The majority of community members (65.9%) see their situation as the same, while 18.5% see it as better and 13.8% see it as worse. In Uganda the same question was asked, but instead of being asked about their current situ- ation relative to two years ago, community members were asked to rate their current living situation relative to five years ago.206 In the case of Uganda, 62.6% of community members saw their current living situation as better than five years prior, 16.8% saw it as the same, and 20.7% saw it as worse.207 14.1 Land Access and Food Security Access to land for cultivation among community members across the GLR countries is generally high – 89.8% report that they have access to land.208 This figure is characteristic of all the GLR countries except for RoC in which access to land among community members was considerably lower (55.9%) in addition to being the only GLR country where land access among community members was lower than for ex-combatants (94.2%) – though the reason behind these findings are unclear. Female community members were slightly less likely to have access to land for cultivation than male community members (87.3% vs. 91.3%). When those community members who did not have access to land for cultivation were asked to explain why in DRC and RoC the most common replies were: (i) all land was occupied (29.5%); (ii) fear for the return of conflict (19.7%); and (iii) bad memories associated with their land and they did not want to return (14.2%).209 Female com- munity members were less likely to cite land occupation as the reason for their lack of access to land for cultivation than male community members (21.5% vs. 37.6%), and more likely to cite fear of conflict (25% vs. 14.3%) and lack of capital (12.5% vs. 5.3%). When community members who did have access to land for cultivation across the GLR countries were asked whether they had more than two years prior, 44.5% said that they did have more access. This level of increased access to land for cultivation is very much an averaged figure. There is a sharp split between DRC and RoC where increased land access was high (69.1% and 71.6% respectively), and Uganda in which only 15% reported increased access.210 However, a closer inspection of community members in Uganda with a more finely grained scaled shows that while only 15% had increased access to land for cultivation, 49.4% had the same level of access as two years prior and 36.5% had less access. Of those community members who had less access to land than two years prior across the GLR countries there were a range of explanations given. In Uganda the most common explanations were that there had been regulated 206   This question about community members’ perception of their current living situation relative to a previous point in time was not asked in Rwanda. 207   These findings are particularly subject to periodization issues and should be treated cautiously. 208   Rwanda is absent from findings related to land access, change in land access, and reasons for positive and negative changes in land access due to lack of directly comparable data. 209   This question about why community members did not have access to land for cultivation was only asked in DRC and ROC. 210   One possible explanation that has been put forth, anecdotally, for the relatively high increases in access to land in DRC and RoC is that access to land for cultivation in these countries has been so low to start with, due to displacement as a part of prolonged conflict and insecurity, that the consolidation of a relative peace in DRC and RoC has exposed a larger proportion of the population to the possibility of increases (effectively having started with no land access) than in other GLR countries where some land ownership may have persisted through conflict. 109 Table 37: Community Member Explanations for Lack of Livestock Country Country Explain the reason that your household does not have any livestock – excluding poultry DRC Republic of Congo Uganda Subtotal Insecurity due to conflict 12.0% 34.7% 3.6% 19.4% Lack of access to suitable land 23.0% 26.6% 29.1% 25.4% Ill health 16.6% 4.6% 5.5% 10.1% No male adult to look after livestock 0.0% 0.0% 1.8% .3% Experiences of crime (livestock stolen, etc.) 9.7% 31.8% 18.2% 19.7% Other 0.0% 0.0% 3.6% .6% Used for dowry 0.0% 0.0% 3.6% .6% Poverty (insufficient resources, etc.) 28.8% 1.2% 32.7% 18.8% Poor management 0.0% 0.0% 1.8% .3% Not interested / Not a Breeder 7.8% 1.1% 0.0% 3.9% Sale of livestock .4% 0.0% 0.0% .2% Moves and migration 1.8% 0.0% 0.0% .8% division of land by their family (47.6%), followed by unregulated division of land such as grabbing, etc. (23.8%). In DRC and RoC lack of resources (25.8%), land infertility (20.8%), and land sale (20.8%) were the most common explanations. Across the GLR countries when community members were asked to explain why they had more access to arable land for cultivation than two years prior there was a range of explanations. In Uganda the most common explanation was that a household member had purchased more land for investment (40%) – this answer was especially promi- nent for female community members (50%) compared to male community members (38.1%). In DRC and RoC, inheritance was also an important pathway to increased access to land for cultivation (29.1%), especially for female community members (33.4%) compared to males (26.7%). Across the GLR countries livestock ownership, excluding poultry, is generally low among community members (35.5%), with the exception of Uganda where livestock ownership is significantly higher (66.3%).211 Despite these differences female community members were the least likely demographic group to own livestock (25.4%), espe- cially when compared to male community members (42.7%). When those who had no livestock were asked to explain further the four most common answers from community members across the GLR countries were as follows: (i) 25.4% cited lack of access to suitable land; (ii) 19.7% cited crime; (iii) 19.4% cited insecurity due to conflict (39.1% crime and insecurity collectively); and (iv) 18.8% cited poverty. While instructive of general trends, these cross-country figures do not fully depict the intricacies of the range of explanations given in each GLR country. For example, in DRC poverty was the most common explanation (28.8%) and livestock theft was less frequently cited (9.7%), while in RoC poverty was infrequently cited (1.9%) and livestock theft was much more frequent (31.8%). A summary of the range of explanations for lack of livestock among community members can be found above in Table 37. 211   Rwanda is absent from findings on livestock ownership, change in livestock ownership over time, and reasons for positive and negative changes in livestock ownership over time due to lack of directly comparable data. 110 Looking to overall change in the quantity of livestock in the last two years, 54.2% Table 38: Community Member Household Hunger of community members across the GLR How Frequently Do People in Your Household Go Hungry? countries had seen an increase in their overall quantity of livestock – 8.5% stayed Always Often Seldom Never the same and 37.3% saw a decrease in live- Male 12.0% 34.7% 3.6% 19.4% stock. Female community members were less likely in general to see an increase in Female 23.0% 26.6% 29.1% 25.4% their livestock (46.3%) and more likely Age 18-30 16.6% 4.6% 5.5% 10.1% to see a decrease (42.9%) than their male Age 31-40 0.0% 0.0% 1.8% .3% community member counterparts (57.5% Age Over 40 9.7% 31.8% 18.2% 19.7% and 35% respectively). DRC 0.0% 0.0% 3.6% .6% The pervasive development challenges RoC 0.0% 0.0% 3.6% .6% that characterize the GLR countries mean Uganda 28.8% 1.2% 32.7% 18.8% that food security is a key issue. As such, understanding the relationship between GLR Average 0.0% 0.0% 1.8% .3% access to land for cultivation, in addition to livestock ownership and household hunger and nutrition, as core indicators of food security for community members, is important for understanding the overall development context of GLR countries. Across the GLR countries, 27.5% of community members reported that people in their household never went hungry, 35.4% seldom, 28.8% often, and 8.3% always.212 Uganda stood out from these dominant trends – the entire distribution of community members being shifted towards less household hunger (35.9% never, 51.4% seldom, 11.6% often, and 11.1% always). Female community members across the GLR countries were slightly less likely to never go hungry than males (24.9% vs. 29.3%). Age showed a distinct relationship to household hunger - as age in- creases the likelihood of belonging to households where people always or often go hungry increases. In accordance, as age increases the likelihood of coming from a household that seldom or never goes hungry decreases (as visible in Table 38). Household nutrition and nourishment has largely been unchanged (41.9%) for community members over the last two years. Of those who have seen a change in household nutrition, 32.5% have seen improvements and 25.6% have seen deterioration. Again, Uganda stood out from this trend with higher levels of nutritional improvement (55.2%), and less unchanged nutrition (19.9%) as well as deterioration of nutrition (24.9%). Across GLR countries female community members were slightly less likely to see improved nutrition and more likely to see worsening nutrition (28.8% and 28.3% respectively) when compared to male community members (35.2% and 23.6% respectively). Age showed a negative relationship to improved nutrition (38.6% of those 18-30 years of age, 37.9% of those 31-40 years of age, and 24.8% of those over 40 years of age) and, in accord, a positive relationship to worsening nutrition (20% of those 18-30 years of age, 20.5% of those 31-40 years of age, and 32.8% of those over 40 years of age). 14.2 Summary Community members across the GLR countries display a range of unique trends in regards to their patterns of housing, property ownership, land access and livestock. Indeed, focused analysis reveals that there is much further variation along regional and factional lines within each individual GLR country. However, despite this variation there is a core set of trends that emerge. Community members are most likely to be living in some form of family structure – whether it is the same family that they lived with before conflict or a different one. Community members 212   Rwanda is absent from findings on household hunger and nutritional change due to lack of directly comparable data. 111 are most likely to be living in a house that they or someone in their family owns. This is a very similar picture to that of ex-combatants across the GLR countries. Access to land for cultivation is generally very high for community members as is the stability of this access – i.e. while there is variation from country to country as to increases in land access, there are relatively few commu- nity members who have seen decreases in land access since the years before sampling. Those who do see increased access to land do so most commonly through household land purchase with capital from high agricultural yields or through inheritance. In contrast to ex-combatants, few community members cited marriage as a key pathway to increased land access. Livestock ownership across the GLR countries is generally low – though slightly higher than among ex-combatants. Community members cite lack of access to resources such as capital or suitable grazing land as a key reason for low livestock ownership in addition to overarching concerns about crime and general insecurity. Food security remains an important concern for community members across the GLR countries. While community members generally face a significantly lower level of food insecurity compared to ex-combatants across the GLR countries, there is still a sizable portion that often or always experience household hunger. Collectively, country to country variation in levels of access to land for cultivation and livestock among community members appears to show little relationship to core indicators of food security – which may be as much related to the overall economic situation in each of the GLR countries (e.g. through inflated prices of food stocks) than to land access and livestock ownership outright. 14.2.1 Vulnerable Subgroups Female community members across the GLR countries display a range of differences in terms of the housing, land access, and food security that collectively may not necessarily entail a disadvantage, but do at least indicate a slightly altered narrative. Females are more likely to live with a different family than before conflict – possibly in part because of their high rate of being widowed compared to male community members. Female community members are more likely to live in a house, but dramatically less likely to own their house – instead commonly citing spousal or familial ownership. Female community members have slightly less access to land for cultivation than male community members in addition to being less likely to owning livestock. However, female community members fare better overall in indicators of food security across the GLR countries. Female community members overall perform slightly better than female ex-combatants across the GLR countries in terms of housing, land, livestock, and food security. The core difference in the narrative that female community members and female ex-combatants experience across the GLR countries is that female community members are generally more likely to be integrated into a familial unit and reap the benefits this extended support network. 14.2.2 Unique Country Trends As outlined above, land access among community members in RoC is considerably lower than on average across the rest of the GLR countries. It is possible that this trend is a product of the sample from RoC being captured ex- clusively in the Pool region of the country. Pool was the region of RoC in which the low level insurgency prevailed in the early 2000s and persisted longer than in other parts of the country. Due to displacement, insecurity and laggard recovery, land access has become a prevalent issue. Indeed, while community members in RoC have seen the greatest increases in their levels of land access of the GLR countries this is likely because community members in RoC started with very low land access at the end of conflict in the first place – and despite large improvements continue to have the least access to land across the GLR countries. Land access in RoC is also likely tied to the lowest levels of livestock ownership across the GLR countries. 112 15. Economic Issues T he following is an analysis of the economic status of community members and their relationship to that of ex-combatants across the GLR countries. The analysis proceeds in five main parts: (i) an examination of community members’ employment statuses and general outlooks on employment; (ii) an examination of the barriers that non-economically active community members face to gaining a stable economic status; (iii) an ex- amination of female community members’ specific economic issues; (iv) an examination of community members’ levels of income, savings, and access to credit as indicators of their general economic stability and ability to leverage economic opportunities; and (v) an examination of community members’ level of engagement with economic as- sociations as an extended support / opportunity network. The structured analysis here follows in close parity with that of the ex-combatant sample, serving as a comparison for charting the degree to which ex-combatants achieve economic reintegration and economic stability – the two not necessarily being the same thing. Indeed, conventional wisdom is that economic reintegration is essential for the process of building peace and security – however the economic context across the Great Lakes Region is often one of severe development challenges posed to both community members and ex-combatants alike. One way to view this is that ex-combatants reaching parity with community members along economic lines may end up meaning reaching a state of economic instability equal to community members. 15.1 Economic Status and History Employment status was very consistent across the GLR countries for community members through time (prior to conflict, at the end of conflict, and at the time of sampling).213 As can be seen in Table 39 the most common em- ployment status for community members across the GLR countries was self-employed in agriculture (36.9% prior to conflict, 42.5% at demobilization, and 38.3% at sampling). It appears that the slight spike in those community members self-employed in agriculture at the end in conflict coincides with drops in the number of unemployed and, more notably, drops in the number of community members studying or training compared to before conflict (19% prior to conflict, 11.8% at the end of conflict, and 7.9% at sampling). Uganda displays an exaggerated version of this trend, as prior to conflict most community members in Uganda were studying or training (43.3%) followed by self-employed in agriculture (33.9%), however by the time of sampling the employment statuses of community members in Uganda fell very much in line with the cross-country average. Female community members were slightly more likely to be unemployed, self-employed in retail, or a housewife working in the home at all time periods when compared to male community members (see Table 39). Over time female community members have seen a slight drop in the rate of being self-employed in agriculture accompanied by slight increases in public and private sector employment as well as unemployment. When those community members across the GLR countries who were not employed at the time of sampling were asked to explain why they were not working, there were several dominant responses: (i) 43% said they were not working due to lack of opportunity; (ii) 16.8% cited financial problems (including lack of access to credit); (iii) 213   Due to the state of continued conflict in DRC, employment status was surveyed in reference to 5 years ago, 3 years ago, and today. Data on community member employment status in Rwanda is only available for the time of sampling. 113 Table 39: Community Member Economic Status at Three Time Points215 Who Owns The Housing In Which You Currently Live? Hustle or involved in or reliant Self-employed manufacturing through subsistence activities Retired, but not economically Housewife or Working in the Self-employed agriculture on the informal economy employer, private sector Supplementing income employer, public sector Self-employed services Employed working for Employed working for Employed working for employer, agriculture Studying or Training Self-employed retail Unemployed active home Employment Status Prior to Conflict Male 12.90% 0.70% 6.90% 9.20% 37.00% 6.00% 2.80% 0.80% 0.70% 0.00% 20.10% 0.20% 2.50% Female 18.20% 0.30% 2.80% 3.90% 36.40% 5.10% 6.20% 1.30% 0.00% 0.00% 17.60% 8.10% 0.10% Age 18-30 18.90% 0.80% 4.50% 2.60% 18.40% 4.10% 3.00% 0.60% 0.60% 0.00% 42.60% 3.80% 0.00% Age 31-40 14.20% 0.60% 6.30% 6.10% 34.70% 8.20% 6.20% 0.90% 0.80% 0.00% 17.90% 4.00% 0.10% Age Over 40 12.70% 0.30% 5.40% 11.30% 51.10% 5.30% 4.30% 0.90% 0.00% 0.00% 2.00% 3.20% 3.60% DRC 19.90% 1.10% 9.00% 13.30% 18.10% 8.90% 8.10% 1.30% 0.00% 0.00% 11.40% 8.20% 0.80% RoC 15.60% 0.60% 2.90% 3.30% 59.60% 4.80% 2.60% 1.70% 0.00% 0.00% 2.90% 2.00% 3.90% Uganda 9.40% 0.00% 3.50% 4.10% 33.90% 2.90% 1.80% 0.00% 1.20% 0.00% 43.30% 0.00% 0.00% GLR Average 15.00% 0.60% 5.20% 7.00% 36.90% 5.60% 4.20% 1.00% 0.40% 0.00% 19.00% 3.50% 1.60% Employment Status at the End of Conflict Male 11.70% 1.00% 7.00% 7.70% 44.50% 6.60% 3.70% 0.80% 0.30% 0.60% 13.00% 0.60% 2.30% Female 14.70% 0.50% 3.70% 4.80% 39.50% 5.50% 7.40% 1.20% 1.80% 0.90% 10.10% 9.80% 0.20% Age 18-30 14.00% 0.70% 6.90% 3.10% 28.40% 5.30% 4.30% 0.60% 2.30% 1.20% 28.10% 5.30% 0.00% Age 31-40 12.70% 1.50% 6.60% 7.10% 39.00% 8.70% 8.20% 0.80% 0.00% 0.80% 9.10% 5.40% 0.10% Age Over 40 12.10% 0.30% 4.40% 9.30% 54.90% 5.50% 4.40% 0.90% 0.50% 0.50% 0.50% 3.20% 3.40% DRC 17.50% 1.10% 8.50% 12.80% 20.40% 9.40% 8.90% 1.40% 0.00% 0.00% 8.80% 10.30% 1.00% RoC 15.10% 0.60% 2.00% 3.10% 62.00% 5.00% 3.10% 1.50% 0.00% 0.00% 1.50% 2.60% 3.60% Uganda 6.20% 0.60% 6.20% 3.40% 46.30% 4.00% 3.40% 0.00% 2.80% 2.30% 24.90% 0.00% 0.00% GLR Average 12.90% 0.80% 5.60% 6.50% 42.50% 6.20% 5.20% 0.90% 0.90% 0.80% 11.80% 4.40% 1.50% Employment Status at Sampling Male 16.30% 1.80% 6.80% 8.40% 38.30% 8.60% 4.10% 1.10% 0.80% 1.50% 9.30% 0.60% 2.50% Female 18.20% 0.80% 4.30% 5.70% 37.90% 5.30% 7.90% 1.50% 2.30% 1.10% 5.70% 8.90% 0.30% Age 18-30 20.20% 1.30% 8.40% 5.30% 25.00% 7.70% 5.30% 0.80% 2.10% 2.50% 17.70% 3.80% 0.00% Age 31-40 15.60% 1.70% 5.90% 9.60% 39.20% 8.20% 6.30% 1.80% 1.40% 0.60% 4.00% 5.20% 0.30% Age Over 40 14.70% 1.40% 3.20% 8.50% 50.90% 6.40% 5.40% 1.20% 0.30% 0.50% 0.00% 2.90% 4.60% DRC 19.50% 1.00% 6.80% 12.20% 20.50% 9.40% 9.70% 2.20% 0.00% 0.00% 6.80% 10.20% 1.70% RoC 14.40% 0.40% 1.70% 2.60% 64.70% 5.10% 2.60% 1.50% 0.00% 0.00% 0.90% 1.90% 4.10% Rwanda 28.80% 3.60% 8.90% 8.10% 13.70% 6.20% 5.20% 1.20% 1.60% 3.60% 16.50% 2.20% 0.60% Uganda 5.10% 0.60% 5.60% 6.20% 55.40% 8.50% 4.50% 0.00% 4.00% 1.70% 7.30% 0.60% 0.60% GLR Average 17.00% 1.40% 5.80% 7.30% 38.30% 7.30% 5.50% 1.20% 1.40% 1.30% 7.90% 3.80% 1.70% The use of the phrase “at three time points” indicates that respondents were surveyed at one time point with questions regarding three different time points. Because of the inclusion of Rwanda, cross-country community member employment status at the time of sampling is biased compared to data for before conflict and at the time of demobilization programming. For example, the cross-country GLR figure for unemployment at the time of sampling without Rwanda in the sample is only 13% – compared to 17%. These cross-country figures should be approached cautiously and with an eye for detail. 114 11.1% cited lack of sponsor; and (iv) 11.8% pointed to lack of marketable skills.214 Uganda stood out from this general trend in reasons for not working – instead the majority (44.4%) said they were not working because they were a student (matching the findings on Uganda’s higher rate of studying and training above). Female community members across the GLR countries were less likely to cite lack of opportunity and lack of sponsor as the reason for their unemployment (37.6% and 8.6%, respectively) than male community members (48.2% and 13.1%, respectively). Inversely, female community members were more likely to cite a lack of market- able skills and financial problems including lack of credit (18.6% and 19.4%, respectively) compared to their male community member counterparts (5.8% and 14.1%, respectively). Along age demographic categories, those com- munity members aged over 40 were less likely to cite lack of opportunity as a reason for their unemployment (36.5% vs. 51.9% of those 31-40 and 45.4% of those 18-30) and more likely to cite financial problems including lack of credit (23.3% vs. 13.9% of those 31-40 and 8.5% of those 18-30). Across the GLR countries 31.9% of community members reported having more than one job / income earning ac- tivity.215 When asked to explain further in Uganda there were several key responses: (i) 31.1% cited income supple- mentation for general survival as the reason for having more than one income earning activity; (ii) 19.7% explained that they subsistence farmed in addition to having a small business; and (iii) 19.7% said they worked more than one income generating activity for the general betterment of their economic situation. Females in Uganda were less likely to subsistence farm on the side of another job (14.3%) and more likely to work more than one job to meet basic needs (35.7%) than male community members (21.3% and 29.8%, respectively). In DRC and RoC the range of explanations was similar though differently distributed, 40.7% farmed on the side of another income generating activity and 19.1% worked more than one job to meet basic needs. Of community members across the GLR countries, 57.3% would be willing to consider moving to another part of their country for better job opportunities.216 This figure reflects the trends in DRC and RoC well, however in Uganda the proportion of community members willing to consider moving for improved job opportunities was tipped the opposite direction (66.9% reporting that they would not consider moving). Across the GLR countries female community members were slightly less willing to consider moving for better job prospects compared to male community members (53% vs. 60.3%). Also, community members over the age of 40 were less likely to be willing to consider moving (49.2% vs. 64.5% of those 31-40 and 61.7% of those 18-30) – sacrificing their overall stronger social footing in the community. When community members who were willing to consider moving for improved job prospects in Uganda were asked to explain further the most common responses were: (i) to improve their standard of living (33.9% total – 46.7% of females vs. 29.3% of males) and (ii) to seek out new opportunities and experiences (21.4% total – 6.7% of females vs. 26.8% of males).217 Again in Uganda, when those community members who were not willing to consider moving to another part of the country for improved job prospects were asked to explain further the most common responses were: (i) lack of education / still being a student prevents work (31.9% total – 42.5% of females vs. 26% of males) and (ii) family responsibilities (30.1% total – 35% of females vs. 27.4% of males). Drawing from Uganda only, 59.4% of community members reported that they believe that it is harder for ex-com- batants than others to find a job (compared to 78.7% of ex-combatants in Uganda and 64.6% of ex-combatants 214   Rwanda is absent from findings on explanations for unemployment among community members due to lack of directly comparable data. The term ”sponsor” here refers to the apprentice / master relationship. Community members are saying that they do not have anyone to apprentice under. 215   Rwanda is absent from findings on number of community members who participate in more than one economic activity and the reason why due to lack of directly comparable data. 216   Rwanda is absent from findings on community member willingness to migrate for better economic opportunities due to lack of directly comparable data. 217   Questions regarding community members’ explanations for willingness to migrate for economic opportunities were only asked in Uganda. 115 overall).218 Female community members in Uganda were slightly more likely to perceive that ex-combatants had a harder time finding jobs than others when compared to male community members (64.2% vs. 57.4%). In addition, age showed a negative relationship to the perception that ex-combatants have a harder time finding a job (67.5% of those 18-30, 57.9% of those 31-40, and 48.1% of those over 40). Collectively community members’ economic trajectory and understanding of the dynamics surrounding this tra- jectory come together in community members’ overall outlook on their future. The vast majority of community members (77.9%) see their economic situation improving in the future. Both male and female community members across the GLR displayed a very similar positive outlook (79.3% vs. 76.8% respectively). Age displayed a negative re- lationship to the frequency of reporting a positive outlook on one’s economic future with a steep threshold for those aged over 40 (83.2% of those 18-30, 81.7% of those 31-40, and 65.3% of those over 40) which, as outlined below, is especially tied to health related issues. The range of explanations given from community members for a positive outlook on economic prospects in the future were diverse across the GLR countries, making a meaningful cross-country comparison difficult, however a certain range of responses were more common: (i) improved agricultural performance was seen as key in Uganda (23.3%) however less important in DRC and RoC (11.6% and 8.6% respectively); (ii) gaining employment was important in all countries (15.8% in Uganda, 16.8% in DRC, and 12.9% in RoC); (iii) personal effort or hard work was especially important in RoC but much less so in Uganda and DRC (35.8% in RoC, 13% in Uganda, and 17% in DRC); (iv) religious faith / grace of god was central in DRC and RoC (25% and 20.5% respectively) and completely absent (0%) in Uganda.219 Of those community members who had a negative outlook on their economic prospects in the future there were also a range of common answers: (i) ill health was a common response in Uganda and DRC (34% and 22.2%, re- spectively) – this was an especially prevalent response among community members over the age of 40; (ii) poor agricultural yield was a common explanation in Uganda and RoC (21.3% and 23.2%, respectively); and (iii) general economic decline / lack of opportunities was a frequent explanation as well (14.9% in Uganda, 24.2% in DRC, and 34.3% in RoC). As a final point, looking only to DRC and RoC we can observe that 56.1% of community members work for pay 12 months of the year, the remainder working for paid labor closer to the average 9.47 months a year. This makes sense, as 72.3% of all community members work 1-3 months for unpaid labor in addition to their paid labor. Female com- munity members worked slightly longer on average for both paid and unpaid labor compared to make community members (9.59 months paid labor vs. 9.36 months, and 3.23 months unpaid labor vs. 3.05 months). Community members over the age of 40 had the largest average period of the year spent in paid labor (9.78 months) and the lowest average period spent in unpaid labor (3.07 months). It appears as though in DRC and RoC, community members most commonly fall into one of three categories: (i) working 12 months a year for paid labor; (ii) working 12 months a year for paid labor and for 1-3 months of the year (farming season) working for unpaid labor; and (iii) working for around nine months of the year for paid labor and spending the remaining three months (farming season) in unpaid labor. 218   Again, Uganda is discussed exclusively here because questions on the perception of relative difficulty that ex-combatants have finding jobs were only asked to community members in Uganda, whereas in ex-combatant surveys the same question was asked across all GLR countries excluding Rwanda. 219   These findings should not play down the role of religion in Northern Ugandan culture. For example, the Acholi people have a rich tradition of belief blending indigenous and Christian religious customs. Rwanda is excluded from findings on reasons for community members’ positive and negative outlooks on their overall economic situation due to lack of directly comparable data. 116 15.2 Non-Economically Active Community Members on Employment Issues Those community members across the GLR countries who were not economically active explained a range of coping mechanisms to get by financially without an income. Most commonly, community members reported that they relied on cash contributions from family (28%), borrowed money from unspecified sources (19.3%), used past savings (15.8%), or got help from public sources such as the community or a church (14.1%).220 Female community members were less likely to use savings (11.7%) compared to male community members (19.5%) and more likely to rely on cash contributions from family than male community members (33.9% vs. 23%), supporting the idea that females (ex-combatants and community members alike) are especially reliant on the family and have fewer economic support networks overall. Age showed a positive relationship to the likelihood of using savings to get by (8% of those 18-30, 17.8% of those 31-40, and 19.8% of those over 40). Looking specifically to Uganda, a large proportion of non-economically active community members (77.8%) believe that that they have a harder time finding a job than other people.221 Further, 100% of non-economically active female community members thought they had a harder time finding a job compared to 66.7% of their male counterparts. When asked to explain why they thought they had a harder time, non-economically active community members in Uganda most commonly cited incomplete studies / still a student (28.6%), low or no education (28.6%), and dis- ability (28.6%). Female community members cited lack of education or qualifications at a higher rate (50%). This reinforces the evidence that community members, at least in Uganda, see education as a key pathway to gaining employment. Across the GLR countries, non-econom- ically active community members gen- Table 40: Non-Economically Active Community erally held quite polarized outlooks on Member Outlook on Employment their prospects of gaining employment What are your chances of getting a job in the near future? in the near future. Of those community members, 48.9% reported that they think Good Poor they have a good chance of getting a job 50.13% 49.23% Male in the near future, 1.4% that they have Female 48.70% 51.30% neither a good or bad chance, and 49.7% that they have a poor chance. This clear po- Age 18-30 54.15% 45.85% larization was characteristic most clearly Age 31-40 52.60% 47.40% in DRC, however in RoC a larger propor- Age Over 40 38.60% 61.40% tion has positive outlooks (66.6%) and in DRC 47.60% 52.40% Rwanda a lower proportion had positive Republic of Congo 66.60% 33.40% outlooks (33.1%). These trends are visible in Table 40. Age showed a clear positive Rwanda 33.10% 66.90% relationship to the likelihood of having a Uganda 50.00% 50.00% poor outlook on gaining employment in GLR Average 49.60% 50.40% the future (44.1% of those 18-30, 47.4% of those 31-40, and 61.4% of those over 40). In Uganda non-economically active community members were questioned further as to the reasons behind their 220   Rwanda is absent from findings on non-economically active community members’ coping strategies due to lack of directly comparable data. 221   This question on whether non-economically active community members perceive having a harder time finding a job than others was only asked to community members in Uganda, while for ex-combatants it was asked in DRC, RoC, Uganda, and Burundi. 117 answers.222 Those who had a positive outlook explained their optimism with one of three answers: (i) that they hold qualifications and have papers (50%), (ii) that they are currently pursuing studies, and (iii) that they are bright, capable, hard working and motivated. Of non-economically active female community members in Uganda with a positive outlook on gaining employment in the near future, 100% explained their optimism as tied to their current studying. Of those non-economically active community members in Uganda who had a negative outlook on gaining employ- ment in the near future there were also three common explanations: (i) that they were disabled (50%); (ii) that they had low or no qualifications (25%); or (iii) that corrupt officials made gaining employment unlikely (25%). We can take away that, at least in Uganda, non-economically active community members perceive the attainment of education as among the key pathways to gaining employment in the near future – and inversely the lack there of as a key barrier. 15.3 Female Community Members on Employment Issues Non-economically active female community members across the GLR countries generally did not feel discrimi- nated against as a female (83.2% did not feel discrimination and 16.8% did). Uganda was the only GLR country that stood out from this trend – where 50% of non-economically active female community members felt discrimi- nated against on the basis of being female. In Uganda, 100% of those non-economically active female community members identified male bosses or employers as the ones discriminating against them on the basis of gender. 223 When economically active female community members across the GLR countries were asked whether they per- ceived being discriminated against on the basis of being a female the proportion which felt discriminated against as a female was similar, but slightly lower (14.6%) than with non-economically active females. In Rwanda discrimination was perceived on the lowest level (4.8%) across the GLR countries, while it was highest in Uganda (23.3%). Still looking at Uganda, of those economically active female community members who did feel discriminated against on the basis of gender 57.2% identified that discrimination as coming from co-workers (50% of those specified male co-workers, 25% female co-workers, and 25% all co-workers). When economically active female community members across DRC and RoC were asked whether they perceived female ex-combatants as having a harder time, 36% responded yes (the other 64% replying no). When those 36% that did think that female ex-combatants had a harder time than others were asked to explain, the most common explanations were the brutality and misconduct of ex-combatants (34.6%), the poor reputation of ex-combatants (32.1%), and distrust of ex-combatants (12.6%). 15.4 Income, Savings and Access to Credit In the context of the severe development challenges that characterize most of the Great Lakes Region, levels of community member economic activity are a good starting point for understanding basic individual and household economic stability. However, it is through more closely examining community members’ income, savings and access to credit that we can begin to reveal some about their ability, or in some cases lack thereof, to move beyond mere subsistence by leveraging economic opportunities and in turn exhibiting economic mobility. Across the GLR countries, 49.7% of community members reported that they were the sole breadwinner in their household, the remaining 50.3% reporting that others assist them. Despite this cross-country average, communi- ty members within the individual GLR countries responded in different proportions as to whether they were the 222   Questions on the reasons behind non-economically active community members’ economic outlooks were only asked in Uganda. 223   Questions on both non-economically active and economically active female community members’ perceived sources of discrimination were only asked in Uganda. 118 sole breadwinners of their household – these Table 41: Community Member Sole Household findings are presented in Table 41. Concerning Breadwinner Proportions demographic categories, age shows a positive relationship to the likelihood of being a sole Are you the sole, or only, breadwinner or do others in your household also earn an income? breadwinner (42.9% of those 18-30, 54.2% of those 31-40, and 55.7% of those over 40. Sole Others assist In addition, across the GLR countries female Male 57.50% 42.50% community members were less likely to be the Female 32.80% 67.20% sole breadwinner of their household compared Age 18-30 42.90% 57.10% to male community members (32.8% vs. 57.5%) – contributing to the idea that female Age 31-40 54.20% 45.80% community members generally have fewer Age Over 40 55.70% 44.30% economic networks than males. DRC 80.00% 20.00% Of those community members across the Republic of Congo 36.00% 64.00% GLR countries who are the sole breadwin- Rwanda 61.30% 38.70% ners of their household only 29.3% report that Uganda 44.80% 55.20% they break-even in meeting their household 49.70% 50.30% GLR Average expenses each month while the remaining 42% usually have to borrow money, 21% rely on money transfers from family, and 4.5% use past savings.224 Only 3.2% of community members usually have money left over after meeting their monthly expenses. Within each of the GLR countries the variation across these responses is displayed in Table 42. Female community members were less likely to break even in their monthly expenses than males (22.9% vs. 32.5%) and more likely to rely on family money transfers (24.4% vs. 19.3%) and borrowing in general (49.8% vs. 38.2%). Regarding age demographics, those 31-40 are the most likely age group to break even on their monthly expenses (38.70%) and the least likely to rely on family money transfers (10.7%) – whereas those 18-30 and over 40 were less likely to break even (25.1% and 29.5%, respectively) and more likely to rely on family money transfers (24.9% and 23.7%, respectively). However, in terms of borrowing more generally those aged 31-40 were on par with other Table 42: Community Member Sole Breadwinner Meeting Monthly Expenses At the end of each month, do you meet your household expenses? Usually have money Usually have to use Rely on family money Usually have to Usually break-even left over past savings transfers borrow Male 4.50% 32.50% 5.50% 19.30% 38.20% Female 0.40% 22.90% 2.40% 24.40% 49.80% Age 18-30 4.40% 25.10% 3.10% 24.90% 42.50% Age 31-40 0.30% 38.70% 7.90% 10.70% 42.50% Age Over 40 3.60% 29.50% 4.00% 23.70% 39.20% DRC 0.00% 40.00% 0.00% 0.00% 60.00% RoC 9.10% 34.10% 13.60% 18.20% 25.00% Uganda 3.00% 29.10% 4.20% 21.20% 42.40% GLR Average 3.20% 29.30% 4.50% 21.00% 42.00% 224   Rwanda is absent from findings on community member sole breadwinners’ meeting monthly expenses, surplus percentages, and deficit percentages due to lack of directly comparable data. 119 age demographics. These trends could suggest that those aged 31-40 are in a period where they are financially inde- pendent from the familial unit in which they were raised, though not yet having established their own familial unit to such a level that it can serve as a extended support network – though to confirm this speculation would require triangulation in future studies. Of those 65.4% community members across the GLR countries who are sole breadwinners and have a shortage of income for meeting their monthly expenses, they are on average short by 46% of their required income. Of those 4.2% of community members across the GLR who have a surplus of income after meeting monthly expenses, they have on average a surplus of 32% of their income. Within-country averages for monthly income deficit and surplus are displayed in Table 43. Female community members on average have larger income shortages (mean 53%) and slightly smaller surpluses (mean 30%) than male community members (mean 41% and 34% respectively). Age showed a negative relationship to the average income shortage among community members (mean 51% of those 18-30, 42% of those 31-40, and 40% of those over 40). However in terms of average income surplus Table 43: Community Member Sole Breadwinner those 31-40 have the smallest average surplus (22%) Average Monthly Income Shortages and Surpluses. while those 18-30 and over 40 have somewhat larger Average Monthly Average Monthly surpluses (28% and 38% respectively). Income Shortage Income Surplus Male 41% 34% Those community members across the GLR coun- Female 53% 30% tries who are not sole breadwinners contribute 40% on average of their total household income. The Age 18-30 51% 28% variation in average income contribution within each Age 31-40 42% 22% of the GLR countries is displayed in Table 44. On Age Over 40 40% 38% average non-sole breadwinner female community DRC 30% 13% members contribute less than males (mean 37% vs. 43%). Those aged 31-40 contribute the largest pro- RoC 42% 35% portion of household income on average compared Uganda 46% 31% to other age demographic groups (mean 52% vs. 33% GLR Average 46% 32% of those 18-30 and 44% of those over 40). In the two years prior to sampling 31.7% of commu- Table 44: Community Member Average Non-Sole Bread- nity members across the GLR countries have had to winner Household Income Contribution borrow money to help meet their day to day needs, Average Non-Sole Breadwinner Household Income the remaining 68.3% not having needed to borrow, Contribution Percentage though there is a split between DRC and RoC where Male 43% 23.1% and 20.3% (respectively) had to borrow, and Female 37% Uganda where 52.3% had to borrow.225 In DRC and Age 18-30 33% RoC the most common expenses that borrowed Age 31-40 52% money was used for were: (i) to assist family (26.7%); (ii) as a means of subsistence (21.6%); and (iii) for Age Over 40 44% a business investment (18.5%). In addition in DRC DRC 42% and RoC the most common borrowing source was RoC 51% friends (53%) followed by family (20%). Rwanda 31% Across the GLR countries, only 13.3% of community Uganda 43% members had ever applied for micro-credit from a 40% GLR Average 225   Rwanda is absent from all findings on borrowing to meet monthly expenses due to lack of directly comparable data. Further, questions regarding the sources and uses of borrowed money were only asked to community members in DRC and RoC. 120 financial institution, the remaining 86.7% never having applied (possibly due to lack of access). This cross-coun- try figure is very much an average in that there was a clear split between DRC and RoC, on the one hand, where micro-credit application rates were very low (5.3% in DRC and 3.4% in RoC), against Rwanda and Uganda, on the other hand, where higher proportions of community members had applied for micro-credit (24.6% and 26.4% respectively). This division may be a product of the overall levels of development in the sampled areas of DRC and RoC (especially eastern DRC) – financial institutions as such being nearly non-existent. Generally speaking female community members were just as likely to have applied for micro-credit across the GLR countries when compared to male community members (13.1% vs. 13.4%). In Uganda there was a gendered trend visible in which 37.1% of female community members versus 19.6% of male community members had applied for micro-credit. Of those community members who had applied for micro-credit most had successful applications (90.4%). At a cross-country level female community members reported slightly lower success rates in micro-credit applica- tions (84.4% vs. 94.5%) – at a within-country level there is further nuance to examine. In Uganda, male commu- nity members were 100% successful in their micro-credit applications compared to 69.2% of female community members. Inversely, in RoC 85% of female community members had successful applications compared to 60% of male community members. The explanatory factor behind these opposing gender trends in micro-credit application rates is challenging to identify, but could prove a useful direction of inquiry in future studies. 15.5 Economic Associations Across the GLR countries, just over half of community members (53.7%) Table 45: Community Member Economic Association Membership are currently involved in micro-eco- nomic activities – though in DRC In the past two years have you ever been a member this was a notably lower proportion of an Economic Association (34.2%). Yes, have been a No, have never been a Yes, am currently member previously – member of an economic a member now Across the GLR countries 21.8% of but not now association community members were currently 7.50% 23.30% 69.20% Male a member of an economic association, Female 7.70% 19.70% 72.60% 7.6% were previously a member but were no longer, and 70.6% had never Age 18-30 7.00% 21.00% 72.00% been a member of an economic as- Age 31-40 7.10% 27.10% 65.80% sociation. However, in Rwanda and Age Over 40 8.10% 20.10% 71.80% Uganda there were notably higher pro- DRC 4.80% 9.80% 85.30% portions of community members who Republic of Congo 4.40% 9.40% 86.20% were currently members of economic associations (25.2% and 42.9% respec- Rwanda 10.30% 25.20% 64.40% tively) – these trends are displayed in Uganda 10.90% 42.90% 46.30% Table 45. Across the GLR countries, 7.60% 21.80% 70.60% GLR Average female community members were slightly less likely to currently be a member of an economic association when compared to male community members (19.7% vs. 23.3%). However, Uganda stands out from this otherwise durable trend – female community members were actually more likely to currently be in an economic association than their male counterparts (54.5% vs. 37.5%). Looking only at Rwanda and Uganda, we can see that the most common form of economic association for commu- nity members is local savings and credit cooperatives, in which 40.3% of those who were currently a member of an 121 Table 46: Community Member Economic Association Members Breakdown Who are the members that comprise this Economic Association? reporters and non- Mixed but mostly Specified profes- Mixed with both Only Ex-Combat- Mainly disabled Civilians, no Ex- non-Ex-Combat- Mix but mostly sionals such as Ex-Combatants Ex-Combatants Combatants teachers Military people Other ants ants DRC 1.6% 1.6% 0.0% 58.7% 30.2% 0.0% 1.6% 0.0% 6.3% RoC 1.9% 5.7% 43.4% 46.2% 2.8% 0.0% 0.0% 0.0% 0.0% Rwanda 0.0% 9.2% 6.4% 84.4% 0.0% 0.0% 0.0% 0.0% 0.0% Uganda 0.0% 36.4% 49.4% 0.0% 11.7% 1.3% 0.0% 1.3% 0.0% GLR Average .3% 22.5% 31.9% 33.3% 9.8% .7% .2% .7% .7% economic association in Uganda belonged to and 61.1% of those in Rwanda belonged to.226 The next most common form of economic association among community members was farmers associations (8.3%) – of which female com- munity members were less likely to be a member of when compared to males (3.3% vs. 11.9%). When asked about the primary benefits they gain from membership to their economic association there were diverse trends across the GLR countries. In Rwanda and Uganda, community members most commonly identify financial support as a key benefit of their membership to an economic association (38.9% and 69.2% respectively). However, for community members in DRC and RoC, social networking (30.2% in DRC and 23.9% in RoC) and economic networking (39.7% in DRC and 46.9% in RoC), and moral support (27% in DRC and 19.5% in RoC ) were the key benefits to membership in an economic association. Though not the most common reply, social and economic networking were also perceived as benefits to economic associations in Rwanda (14.2% and 20.4%) but not in as much so in Uganda (3.1% and 4.6%). Age appears to play a role in the value of economic associations in at least two ways. First, as age increases, commu- nity members are less likely to see financial support as the main benefit of being in an economic association (52.3% of those 18-30, 46.9% of those 31-40, and 37.5% of those over 40) – most relevant in Rwanda and Uganda. Second, as age increases, community members are more likely to see economic networking as the chief benefit (14% of those 18-30, 19.1% of those 31-40, and 20% of those over 40) – especially relevant in DRC and RoC.227 Across the GLR countries when community members were asked about the membership of their economic as- sociations there was considerable variation across countries. The variation within each GLR country for commu- nity members’ perception of the membership of their economic association is presented in Table 46. Community members in DRC and Rwanda, and to a lesser extent RoC, were most heavily involved in economic associations without ex-combatant members, while community members in Uganda were more commonly members of economic association that also had ex-combatant members. It unclear this is a reflection of programmatic design or the result of social forces such as stigma. 226   Questions regarding the type of economic associations to which community members belonged were only asked in Rwanda and Uganda. In Rwanda these local savings and credit cooperatives commonly existed under the name VSLA – Village Savings and Loan Association. 227   However, this age related trend in DRC and RoC may be a product of the sampling biases in these countries towards community members over the age of 40. 122 15.6 Summary Community members across the GLR countries show a relatively stable trajectory of economic status over time – the average unemployment rate varying as little as 2%. The majority of community members are self-employed in small-scale agriculture. Indeed, through time this remains the most important economic activity for community members across the GLR. Though the proportion of community members self-employed in small-scale agriculture peaks at the end of conflict and then drops some at the time of sampling, these drops, along with a continuous drop in the number of community members studying, are absorbed most notably into employment in public and private sector as well as self-employment in non-agricultural service and retail – indicative of the initial onset of improved stability and security in the wake of peace. Though, in DRC these is a nearly static state of employment among com- munity members through time – likely a result of the continuing state of conflict in Eastern DRC. Like ex-combatants, community members see their primary barrier to gaining a stable economic status as the lack of opportunities available to them but are also more likely to cite economic problems (such as access to credit) and lack of marketable skills as barriers than ex-combatants. In further contrast to ex-combatants, few community members were willing to migrate for better economic opportunities, likely an indicator of their stronger social and economic footing in the community. Non-economically active community members relied heavily on borrowing from family and friends to get by. However, overall community members had a positive outlook on their economic prospects in the future. Community members across the GLR displayed a near even split between household sole breadwinner and non-sole breadwinner status. While on average non-sole breadwinners contributed less than half of their total monthly house- hold income this was supplemented against the support of the rest of their household members’ income contribu- tions, placing them at a clear advantage to sole breadwinners. Over half of sole breadwinners had to borrow money from family or friends to meet their household expenses on a monthly basis. Sole breadwinners between the ages of 31 and 40 were the most likely to meet their monthly expenses without borrowing, and when they did borrow they were much more likely to do so from friends instead of family. Further, those non-sole breadwinners aged 31-40 also contributed the most to their total household income on average. These elements combined suggest that com- munity members aged 31-40 are at their economic prime and among the most capable at meeting their household economic responsibilities. The number of community members who have access to micro-credit is low across the GLR countries, and few were members of economic associations – though some community members came into economic associations with ex- combatants as a part of reintegration programming in some countries (e.g. DRC and RoC). The primary value that community members identified to economic associations (most commonly local credit and savings), was largely reflective of programmatic dimensions in each country – e.g. in DRC and RoC social and economic networking were the primary value of economic associations that community members identified. 15.6.1 Vulnerable Subgroups As consistent with the analysis presented throughout this survey of community dynamics in the Great Lakes Region, female community members exhibit a distinct range of characteristics that collectively paint a picture of a weaker platform of economic stability. Female community members are more likely than males to be unemployed through time, and vastly more likely to work taking care of the household – to an extent this can be expected as a result of traditional gender roles in the GLR countries. Female community members are less likely to cite lack of opportunity as the primary barrier to gaining productive economic status, though still the primary, and more likely to cite lack of education or skills than male community members. Across the GLR, household sole breadwinner status was generally an indicator of economic instability. In this sense females are at an advantage to males, being less likely to be a sole breadwinner. However, those female community 123 members who were non-sole breadwinners contributed less to the household on average than males. Further, those female community members who were sole breadwinners were less likely to meet their expenses and more likely to borrow from family or friends to meet household expenses on a monthly basis than male community members – their income shortages were larger and their surpluses were smaller. So while female community members are less likely to be exposed to the economic vulnerability of sole breadwinner status, when they are, this vulnerability is more accentuated than for male community members. These findings flag female headed households as particularly vulnerable. At first glace young community members (18-30) also appear to have some disadvantages to other age demographic groups. Young community members are the most likely to be unemployed at any time point, are less likely to meet monthly expenses, and have larger income shortages than their older peers. However, young community members are also the least likely age group to identify themselves as a household head and receive the support of their familial/ household unit – in a sense insulating them from the weight of their employment and income disadvantages. It may be that the disadvantages that young community members face are simply an indication of their life stage in estab- lishing an income source and building economic networks. 15.6.2 Unique Country Trends In the analysis of economic issues presented in this section of the study, community members in DRC stand out subtly. First, across the GLR countries, self-employment in agriculture is the dominant economic status at all time points. While this is still the case in DRC, the overall proportion of community members self-employed in agricul- ture at all time points (before conflict, at start of demobilization programming, at the time of sampling) is consider- ably lower than the cross-country average. The difference is explained in part by the community members in DRC’s higher levels of unemployment (highest at all time points across the GLR), employment in the public sector, and self-employment in services or retail. Overall this could suggest that community members in DRC have a harder time gaining employment, specifically self-employment in agriculture, than in other GLR countries and as a result participate more deeply in a range of alternative income activities. In addition, community members in DRC are the most likely across the GLR to be sole breadwinners – and thus subject to greater household economic instability. Indeed, while sole breadwinners in DRC are the most likely across the GLR countries to meet their monthly expenses, less than half do so – instead relying on borrowing from friends and family on a regular basis. Though sole breadwinners in DRC have smaller monthly income shortages than those in other GLR countries, they also have smaller surpluses. Essentially, while sole breadwinner community members in DRC are slightly better off than those in other GLR countries, community members in DRC are also more likely to be a sole breadwinner – still an indicator of greater exposure to economic instability. Collectively these two points, community members in DRC as the most likely to be unemployed and the most likely to be a sole breadwinner, cement the economic conditions for community members in DRC as the weakest in the GLR countries. 124 16. Social Capital T he following section provides a discussion and analysis of the many facets of social capital in the commu- nity member sample. The concept of social capital essentially revolves around the idea that social networks have value, both tangible and intangible, for individuals and communities and are a key indicator of the overall social health of communities – in turn, their ability to leverage this social capital towards social and economic outcomes. As such the analysis of social capital for community members across the Great Lakes Region presented here is comprised of five core components: (i) an examination of the size of community members’ social networks and their levels of sociability; (ii) an examination of individual community members’ levels of trust and solidarity with others in their community; (iii) an examination of community members levels of social cohesion and inclusion in the community; (iv) in turn, an examination of how these factors come together in community members’ overall sense of empowerment; and (v) their perception of social change over time. Beyond serving as a key backdrop for understanding ex-combatants’ position relative to community members, the analysis here represents a look into the core social dynamics present in communities across the GLR countries. 16.1 Networks and Sociability Across the GLR countries, community members are unlikely to be in many social groups – though they are still in more social groups than ex-combatants on average. Community members across the GLR countries are in an average of 0.63 social groups. This average is reflected across the GLR countries with the exception of Uganda, where community members had more (0.93) social groups on average. Female community members had slightly fewer social groups than male community members on average (0.56 vs. 0.67). In terms of change in number of social groups, 40.7% of community members across the GLR countries reported that the number of social groups that they belonged to at the time of sampling was more than that of one year prior, 45.8% the same number as one year prior, and 13.5% reported their current number of social groups was less than one year prior.228 However, there is a sharp division between DRC and RoC on the one hand and Uganda on the other. In DRC and RoC, 77.9% of community members had seen an improvement in their number of social groups. Uganda stood in contrast from DRC and RoC as only 9.8% of community members were currently in more social groups than one year ago, 73.6% were in the same number, and 16.6% were in fewer groups. So while in Uganda community members had more social groups than in DRC and Roc in absolute terms, those community members in DRC and RoC had seen considerably more improvement.229 Across the GLR countries, female community members were more likely to be in more social groups than one year ago compared to male community members (46.4% vs. 36.9%). In addition age displayed a positive relationship to the likelihood of being in more social groups than one year ago (31.3% of those 18-30, 45.5% of those 31-40, and 46.1% of those over 40). Only 39.6% of community members across the GLR countries were on a management or organizational committee 228   Rwanda is absent from findings on change in number of social groups due to lack of directly comparable data. 229   It is possible that this gap between DRC/RoC and Uganda could be an indicator of the health of the overall social fabric in these countries, however it is also possible that it could be a product of periodization issues – e.g. a longer amount of time passed between the start of DDR programing and the time of sampling in Uganda than DRC in which time the overall security situation has improved considerably. 125 Table 47: Community Member Frequency of Familial Contact Frequency of contact between community member and immediate family these days Daily Weekly Monthly Half yearly Once a year Occasionally Never Male 69.40% 11.90% 8.20% 1.60% 0.60% 7.90% 0.20% Female 55.10% 11.30% 14.60% 1.30% 1.50% 15.90% 0.40% Age 18-30 78.00% 9.30% 5.20% 1.10% 0.60% 5.70% 0.10% Age 31-40 63.50% 13.40% 11.50% 1.60% 1.40% 8.50% 0.20% Age Over 40 52.90% 12.70% 14.40% 1.50% 0.90% 16.80% 0.80% DRC 33.70% 17.40% 17.90% 0.50% 2.20% 28.30% 0.00% RoC 23.30% 22.40% 25.00% 3.80% 2.20% 22.00% 0.90% Uganda 93.20% 4.50% 1.10% 0.60% 0.00% 0.60% 0.00% GLR Average 63.90% 11.70% 10.60% 1.50% 1.00% 10.90% 0.30% for a local group or organization – another indicator of social interaction and overall engagement in the communi- ty.230 Female community members were less likely to be on a committee than male community members (30.4% vs. 45.7%). Those between the ages of 31-40 were the age segment that most commonly was on a committee (45.7%), compared to those 18-30 (33.1%), and those over 40 (41.5%). The majority of community members (73%) across the GLR countries have contact with their immediate family.231 However this cross-country figure masks some nuance in the trends within each GLR country. For example in RoC the proportion of community members who had contact with their families was absolute (100%) and in Uganda nearly so (97.3%). However, in DRC only 31% of community members had contact with their immediate family – likely a product of the extreme difficulty of travel and overall dynamics of displacement in eastern DRC. Female community members across the GLR were slightly less likely than male community members to have contact with their immediate family (68.2% vs. 76.2%). Of those community members across the GLR countries who did have contact with their immediate family, they most frequently had daily contact (63.9%), though in Uganda this proportion was much larger (93.2%). A cross- country summary of community members’ frequency of familial contact is displayed in Table 47. Age showed a negative relationship to the frequency which community members reported having daily contact with their immedi- ate family (53.1% of those 18-30, 40.1% of those 31-40, and 25.9% of those over 40). When community members across the GLR countries who did have contact with immediate family were asked whether the current level of their contact was the maximum they would desire, 49.3% responded yes – the remain- ing 50.7% responding no. Again, there is some nuance to be added here. In Uganda where daily contact was much higher, 87% responded that their current level of familial contact was the maximum they would desire. In contrast, in DRC, where the number of those who had contact with their family at all was much lower, only 11.9% felt they currently had the maximum level of contact with their families that they would desire – these trends are displayed in Table 48. 230   Rwanda is absent from findings on membership to management or organizational committees due to lack of directly comparable data. 231   Rwanda is absent from findings on familial contact, frequency of familial contact, reasons for levels of familial contact, and desired levels of familial contact due to lack of directly comparable data. 126 When in DRC and RoC community members Table 48: Community Member Desired Level of were asked to explain why they do not see their Familial Contact families more often, the most common respons- Is the current level of contact the maximum you wish or es were: (i) distance of travel (37.1%), (ii) not could it be more frequent? enough time (15.5%), and (iii) the cost of travel Maximum Could be more frequent (19.2%) – largely corroborating the assertion above that familial contact, especially in eastern Male 54.20% 45.80% DRC, is a product of the difficulty of travel due Female 42.30% 57.70% to weak road infrastructure, mountainous terrain, Age 18-30 54.60% 45.40% strong seasonal rains, and continuing regional in- Age 31-40 45.20% 54.80% security.232 It is also likely that regional dynamics of forced displacement and migration may further Age Over 40 47.10% 52.90% damage social capital in DRC, RoC, and the GRL DRC 11.90% 88.10% more broadly. Republic of 35.10% 64.90% Congo Across the GLR countries, when community Uganda 87.80% 12.20% members were asked to describe the number of GLR Average 49.30% 50.70% friends they had the majority reported that they have lots of friends (48.5%), followed by a few good friends (30.9%), and not many friends (20.6%). Uganda stands out from this trend with 74.3% of community members reporting having lots of friends. Female community members across the GLR countries were less likely to describe having lots of friends and more likely to describe having not many friends compared to male community members (39% vs. 54.3% and 17.8% vs. 16.1%, respectively). Across the GLR countries, when community members were asked to think about the age, gender, and educational background of their friends,233 the majority of community members reported that their friends mostly shared the same age (57.7%) and gender (62.1%), while few (25.7%) shared the same educational background. These trends are durable across the GLR countries and are displayed in Table 49. In terms of demographic groups it is worth noting that female community members were slightly less likely than male community members to have most of their friends of the same age (54.4% vs. 60.1%) or educational background (24.9% vs. 26.3%), but slightly more likely to have them of the same gender (64.5% vs. 60.4%). Across age demographics, those over 40 were consistently the least likely to have most of their friends in the same age, gender, or education background group (only marginally less than those 31-40, see Table 49) – an indicator of older community members slightly more diverse social groups and overall stronger social footing. When community members across the GLR countries were asked to whom they would turn to for help if they were to encounter an economic problem the most common responses were (i) family (39.9%) and (ii) friends (33.4%) – the remaining 26.7% said they would turn to no one (10.3%), to formal institutions such as local saving and credit associations (7.5%) or a range of other sources including the church (8.9%). Female community members were less likely to turn to friends and more likely to turn to family compared to male community members (27.2% vs. 37.2% and 44.9 vs. 36.8%, respectively). Age showed a negative relationship to the frequency at which community members reported that they would turn to family for economic support (50.4% of those 18-30, 35.8% of those 31-40, and 31.8% of those over 40). In addition, age showed a positive relationship to the frequency which commu- nity members reported turning to no one (6.3% of those 18-30, 9.2% of those 31-40, and 15.4% of those over 40). 232   Questions regarding the reasons for less familial contact than desired were only asked in DRC and RoC. 233   Rwanda is absent from findings of the demographic background of community members’ friends due to lack of directly comparable data. 127 Table 49: Community Member Friend Group Demographics Thinking about your friends, are most of them of the same age? Most Some Few None Male 60.10% 26.70% 10.20% 3.00% Female 54.40% 27.50% 12.70% 5.50% Age 18-30 62.30% 22.50% 12.50% 2.70% Age 31-40 55.60% 32.10% 8.70% 3.60% Age Over 40 55.20% 27.90% 11.70% 5.20% DRC 60.20% 24.60% 8.80% 6.50% Republic of Congo 54.20% 29.40% 10.60% 5.80% Uganda 58.20% 27.50% 14.30% 0.00% GLR Average 57.70% 27.10% 11.30% 4.00% Thinking about your friends, are most of them of the same gender? Most Some Few None Male 60.40% 27.20% 10.20% 2.20% Female 64.50% 24.70% 7.30% 3.50% Age 18-30 64.20% 24.10% 9.10% 2.60% Age 31-40 62.60% 26.30% 8.40% 2.70% Age Over 40 59.60% 27.80% 9.50% 3.10% DRC 60.20% 25.10% 9.30% 5.40% Republic of Congo 62.50% 26.30% 8.30% 2.90% Uganda 63.50% 27.10% 9.40% 0.00% GLR Average 62.10% 26.20% 9.00% 2.70% Thinking about your friends, are most of them of the same educational background or level? Most Some Few None Male 26.30% 32.90% 31.30% 9.50% Female 24.90% 28.00% 29.80% 17.30% Age 18-30 32.30% 30.40% 26.60% 10.70% Age 31-40 22.90% 30.50% 32.60% 13.90% Age Over 40 22.40% 31.30% 32.90% 13.40% DRC 21.40% 35.40% 23.90% 19.30% Republic of Congo 21.50% 26.10% 35.10% 17.20% Uganda 33.50% 30.80% 33.50% 2.20% GLR Average 25.70% 30.90% 30.70% 12.70% Generally speaking, across the GLR countries community members did not socialize often in public – meeting with people in a public place either to talk or have food or drinks. On average community members across the GLR countries met in public to socialize 1.33 times a week – though females met less often than males (0.95 vs. 1.57 times a week). An interesting note is that in DRC and RoC when community members were asked how often they met to discuss community issues with others over food or drinks, as opposed to just for socialization, the response rates were notably higher (mean = 2.12 times a week). With this, the majority of community members (57.4%) in DRC and RoC think that community issues have created the space by which they can more generally meet people and socialize. 128 Community members across the GLR countries indicate that their level of public socialization is most commonly the same as two years ago (50.4%), followed by more often (28.1%), and less often (21.5%). Female community members were notably less likely to see improvements in their level of public socialization in the two years prior to sampling than male community members (23.3% vs. 31.5%). 16.2 Trust and Solidarity Drawing from Rwanda and Uganda, we can observe that trust among community members towards others in their community is generally high.234 The majority of community members (63.3%) believe that people in their com- munity can be trusted to a great extent, followed by to neither a great nor small extent (22.8%), and lastly to a small extent (13.9%). Female community members on average were less trusting of others in the community than their male counterparts – 58% of female community members had high trust compared to 65.7% of male com- munity members and 18.8% of female community members had low trust compared to 11.7% of male community members. As a further indication of this general level of trust, across the GLR countries 18.8% of community members felt that if they were to disagree with something that everyone else in their community agreed on they would not at all feel free to speak out, 63.5% reported they would definitely feel free to speak out, and 17.7% that they would feel free to speak out but only on certain matters.235 Female community members were slightly more likely to feel they could not speak out at all and less likely to feel they could definitely speak out when compared to male community members (22.8% vs. 15.9% and 57.6% vs. 67.8%, respectively). It is possible that local gender based social norms play a role in these findings. When asked whether or not they felt that the level of trust had Table 50: Community Member Perceptions of Change in Trust improved in the last year / two years in the community, 43.4% In the past year / two years, has the level of trust in your area got better, worse, or stayed about the same? of community members across the GLR countries felt that it had Better Same Worse improved, 47.9% that it was the Male 45.90% 45.50% 8.50% same, and only 8.7% that trust had Female 39.40% 51.70% 9.00% deteriorated.236 This cross-country Age 18-30 41.60% 47.30% 11.10% figure however is very much an Age 31-40 41.30% 51.90% 6.80% average as within the individual GLR countries there were distinct Age Over 40 45.30% 46.90% 7.80% trends – for example, in Uganda DRC 15.30% 73.60% 11.10% a clear majority (63.3%) felt that Republic of 56.10% 39.50% 4.40% trust had improved while in DRC Congo an even larger majority (73.6%) Rwanda 39.30% 49.40% 11.30% felt that trust had stayed the same. 63.30% 28.90% 7.80% Uganda These within-country trends are GLR Average 43.40% 47.90% 8.70% displayed in Table 50. A consistent 234   This specific question about general trust levels in the community was only asked in Uganda and Rwanda. 235   Rwanda is absent from findings on comfort of speaking out in disagreement with their community due to lack of directly comparable data. 236   In Uganda and Rwanda this question was asked with reference to the last year, where as in DRC and RoC it was asked in reference to the last two years. This creates some issues with periodization and comparability. These figures should be treated with caution. Interestingly, though the question refers to a longer period of time in DRC and RoC, this does not appear to translate to greater perceptions of improved trust among community members. In the case of DRC this may be the product of continuing insecurity. 129 trend along gender demographic lines does, however, exist. Across and within the GLR countries, female commu- nity members are consistently less likely than male community members to see trust as improved (39.4% vs. 45.9% at a cross-country level) and more likely to see it as the same (51.7% vs. 45.5% at a cross-country level). When those community members across the GLR countries who thought that trust had improved in the last two years were asked to explain further, the majority (42.2%) cited improved safety and security as the main reason for improved trust – this answer was particularly prevalent among female community members (56.6% vs. 34.7% of male community members). A notable portion of community members (44.5%) also expressed a range of ex- planations that related to improved collaboration, cooperation and understanding due to communal living – a key component of the confrontational process of social reintegration. Looking the other direction, when those community members across the GLR countries who thought that trust had deteriorated in the last two years were asked to explain further, responses were diverse – however, the most common were as follows: (I) dishonesty in general (23.4%), (ii) dishonest authorities (19.6%), and (iii) insecurity (11.9%) 16.3 Social Cohesion and Inclusion When reflecting on the level of diversity among the people they live around, 36.7% of community members de- scribed their community as diverse (characterized by lots of differences between people), 25.9% as neither particu- larly diverse or homogenous (neither a great nor small extent of differences between people), and the remaining 37.4% described their community as fairly homogenous (characterized by few differences between people). This relatively even distribution across the GLR countries can be nuanced with a closer look within each of the countries – for example, in Uganda and DRC, community homogeneity (low diversity) was perceived as considerably higher (57.5% and 56.7%, respectively) while in Rwanda, community members’ perceived high levels of diversity (62.1%). These specific within-country trends are displayed in Table 51. At a cross-country level, age showed a distinct relationship to the perception of community diversity among com- munity members. As age increased the likelihood of perceiving high diversity decreased (42.4% of those 18-30, 36.9% of those 31-40, and 30.7% of those over 40) and accordingly the likelihood of perceiving low diversity in- creased (33.5% of those 18-30, 34.5% of those 31-40, and 42.9% of those over 40). When community members Table 51: Community Member Perception of Community Diversity across the GLR countries were asked whether differences To what extent do differences between people characterize your community? between people in their commu- To a great extent, i.e. To a small extent, i.e. nity caused problems such as dis- lots of differences Neither great nor few differences agreement, arguments or disputes small extent between people between people the majority (68.5%) replied no Male 37.50% 24.70% 37.80% (31.5% responding yes). Only Female 35.50% 27.80% 36.70% Uganda breaks significantly from Age 18-30 42.40% 24.00% 33.50% this trend – 69.1% of community members did think that differ- Age 31-40 36.90% 28.70% 34.50% ences caused problems in their Age Over 40 30.70% 26.40% 42.90% community. Congruent to the DRC 34.50% 47.00% 18.40% age demographic trend above in Republic of 25.10% 18.20% 56.70% regards to the perception of di- Congo versity, as age increases among 62.10% 20.40% 17.60% Rwanda community members across the Uganda 24.90% 17.70% 57.50% GLR countries they are less likely to see differences between people GLR Average 36.70% 25.90% 37.40% 130 as a source of problems (43.9% of those 18-30, 29.5% of those 31-40, and 24.3% of those over 40) and more likely to not see them as a source of problems (56.1% of those 18-30, 70.5% of those 31-40, and 75.7% of those over 40). To recap, older community members are both less likely to see differences between people, and less likely to see these differences as a source of problems. Though female community members identified levels of diversity in their com- munities on a level similar to male community members, they were less likely to think that diversity was a source of problems (28.7% vs. 33.5%). Looking specifically at DRC and RoC, community members were asked to further explain the nature of the kinds of problems they encounter. Community members most commonly described the problems as revolving around (i) envy, slander or taunts (29.8%); (ii) misunderstanding (20.5%); or (iii) unspecified accusations made towards ex-combatants (12.7%). Despite varying levels of perceived diversity and their association with problems in the community, across the GLR countries the majority of community members (75.3%) report that they feel a high level of togetherness and close- ness with their community (19.6% feel neither distant nor close and 5.1% feel distant). Across demographic lines, this level is very even as well. However, DRC stands out from the trend as the country with the lowest proportion of community members who feel a high level of closeness and togetherness (61.4%). In addition, while in other GLR countries there is little variation along demographic lines, in DRC female community members are less likely to feel close to their community (57.4% vs. 65.1%) and more likely to feel neither distant nor close (32.2% vs. 27.5%) or distant (10.4% vs. 7.5%) compared to male community members. In alignment with overall feelings of togetherness with the community, across the GLR countries 69.5% of com- munity members had in the last year worked with others in the area they live to do something for the benefit of their community (the remaining 30.5% not having done so). Despite this cross-country figure there is a clear polarization between DRC and RoC, on the one hand, where the rate of participation was lower (53.2% and 54.9%, respectively) and Rwanda and Uganda, on the other, where participation was higher (92.9% and 76.9%, respectively). There is a dispersed range of information regarding the perceived importance of community participation from community members. For example, in Uganda 25.3% of community members cite lack of participation in commu- nity activities as the cause of marginalization in the area that they live. In DRC and RoC an average of 53.3% of com- munity members reported that there were penalties, both formal (such as a ticket or fee) or informal (such as social resentment or exclusion), for those who didn’t participate in community activities – though within each country the figures were almost perfectly inverse (In DRC, 39.7% said that there were penalties and 60.3% said there were not, while in RoC the distributions were 57.6% and 42.4%, respectively). 16.4 Empowerment Empowerment is an important indicator of overall levels of social capital and is understood as a result of individuals’ levels of social connection and their ability to leverage the benefits of these connections and the community and the larger context of society. Collectively, the extent of these benefits and in turn the functions that they fulfill for individuals play a role in the psychosocial concept of empowerment – the individual or collective ability to affect change in one’s life. When asked to reflect on their general level of happiness, 71.8% described themselves as happy, 17.4% described themselves as neither happy nor unhappy, and 10.8% reported that they were unhappy. Community members in DRC were the least likely to report being happy (65.2%), while community members in Uganda were the most likely to report so (80.8%). Across the GLR countries, female community members were slightly less likely to be happy and more likely to be unhappy than male community members (67% vs. 70.6% and 15% vs. 11.4%, respec- tively). When asked to what extent they felt they had the power to make important decisions that affect the course of their 131 Table 52: GLR Community Member Empowerment (Power, Ability, and Control) Do you feel that you have the power to make important decisions that can change the course of your life? Large extent Medium extent Small extent None Male 67.0% 24.7% 8.3% 3.00% Female 35.6% 40.3% 24.1% 5.50% Age 18-30 46.5% 34.4% 19.1% 2.70% Age 31-40 55.6% 31.6% 12.8% 3.60% Age Over 40 55.1% 30.6% 14.2% 5.20% GLR Average 51.3% 33.3% 15.4% 6.50% Do you feel that you have the ability to make important decisions that can change the course of your life? Able to change life Neither able nor unable Unable to change life None Male 86.0% 10.1% 3.9% 2.20% Female 67.9% 22.7% 9.3% 3.50% Age 18-30 75.0% 18.1% 6.9% 2.60% Age 31-40 80.7% 13.7% 5.5% 2.70% Age Over 40 78.0% 15.6% 6.3% 3.10% GLR Average 78.8% 14.9% 6.7% 5.40% How much control do you feel you have over decisions that affect your everyday activities? Neither a lot nor a little Lots of Control Little Control None control Male 71.6 % 21.6% 6.8% 9.50% Female 53.1 % 31.9% 15.1% 17.30% Age 18-30 61.1 % 26.7% 12.2% 10.70% Age 31-40 68.3 % 26.3% 5.4% 13.90% Age Over 40 65.7 % 23.5% 10.8% 13.40% GLR Average 64.4 % 25.6% 10.0% 19.30% lives, community members across the GLR countries most commonly reported that they felt that they had such power to a large extent (51.3%), followed with decreasing frequency by to a medium extent (33.3%) and to a small extent (15.4%).237 These overall perceptions of power were remarkably durable within each of the GLR countries – however there are distinct demographic trends in regard to gender. Female community members across the GLR countries were nearly half as likely as their male counterparts to report having a large extent of power to make deci- sions in their lives (35.6% vs. 62.3%) while more likely to perceive power to a medium extent (41.8% vs. 27.3%) and more than twice as likely to have it to a small extent (22.6% vs. 10.4%). Interestingly, when community members were asked a very similar question as to what extent they felt they had the ability (as opposed to power) to make important decisions that affect the course of their lives, perceptions of empowerment were considerably higher – 78.8% reporting that they were able to make changes, 14.9% that they neither were able or unable to make changes, and 6.7% that they were unable to make changes in their life.238 Very 237   Rwanda is absent from findings on sense of empowerment in terms of power to make important decisions and ability to make important decisions due to lack of directly comparable data. 238   The analytical distinction between senses of empowerment in terms of power versus ability is not clear. Interpreting any meaning to the disparity in levels of power and ability is therefore problematic and these data should be treated as a broad indicator of a positive sense of empowerment rather than as exact measures of different components of empowerment. 132 similar to the question above on Table 53: Community Member Perception of Individual power to make change, females Impact on Community also considerably less frequently reported having the ability to Do you personally have a positive or negative impact on the place you live? make changes and more fre- Positive impact Neither positive nor Negative impact quently neither being able nor negative impact unable as well as just unable (as is Male 70.10% 20.70% 9.20% displayed in Table 52). Female 51.00% 28.80% 20.20% Age 18-30 68.30% 19.50% 12.20% When asked yet another question about perceived levels of empow- Age 31-40 64.30% 23.70% 12.10% erment, but this time scaled in Age Over 40 57.20% 28.00% 14.80% reference to the extent that com- DRC 35.80% 37.30% 26.90% munity members feel control Republic of 39.20% 36.80% 23.90% over decisions that have an Congo effect of their everyday activi- Rwanda 93.70% 4.90% 1.40% ties, the trends are remarkably Uganda 81.30% 16.50% 2.20% similar to the previous two ques- tions above.239 Of community GLR Average 62.60% 23.90% 13.60% members questioned across the GLR, 64.4% perceive that they control most decisions that affect their everyday lives, 25.6% perceive that they control some decisions, and 10.1% few decisions. A very similar gender-based demographic trend was exhibited here as well – as is visible in Table 52. When community members across the GLR countries were asked about whether or not they had a positive impact on the community they live in, there was a clear polarization between DRC and RoC on the one hand, and Rwanda and Uganda on the other. As is visible in Table 53 in DRC and RoC there were relatively even distributions of community members’ responses to having a positive impact, neither a positive nor negative impact, and a negative impact. In Rwanda and Uganda, by contrast, the frequency of community members having the perception of having a positive impact on their community was high – in the case of Rwanda, almost absolute. In regards to gender, in DRC and RoC female community members much less frequently than male community members reported having a positive impact on their community (27.9% vs. 46.1%), while in Rwanda the extent of the gap between female and male community members is approximately half of that in DRC and RoC (87.9% vs. 96.3%). In Uganda, female community members even have a slightly higher likelihood of perceiving a positive impact than males (82.5% vs. 80.8%). Turning to age demographics, in DRC and RoC age showed a positive relationship to the likelihood of the percep- tion of having a positive impact on the community (26.3% of those 18-30, 39.1% of those 31-40, and 43.6% of those over 40). While there was no linear trend in regards to age visible in Rwanda; in Uganda, age showed a slight negative relationship to the likelihood of the perception of having a positive impact on the community (85.7% of those 18-30, 82.5% of those 31-40, and 73.2% of those over 40). Certain parts of this trend of polarity between DRC/RoC and Uganda/Rwanda continue when community members are asked to what extent they feel valued by others in the area they live. On average across the GLR countries, 70.3% of community members felt valued by others in their community. However DRC showed smaller proportions of 239   This question regarding community members’ levels of control over everyday decisions has been re-coded from a five point scale to a three point scale for increased comparability to the other two measures of empowerment (power and ability) presented here. 133 community members who felt valued Table 54: Community Member Frequency of Public Gathering to Express Concerns (64.4%), while in Uganda almost all (98.3%) community members felt In the past year, how often have you joined other people valued. Female community members to express concerns to officials or local leaders on issues were notably less likely to feel valued benefiting the community? by their community compared to Never Once A few times, Many times, male community members (66.5% vs. five or less more than five 72.9%). Male 34.60% 13.40% 24.10% 27.90% When asked how often in the past year Female 60.70% 8.70% 16.30% 14.30% they had joined with other people to Age 18-30 43.30% 9.90% 20.20% 26.50% express concerns to the government or Age 31-40 39.40% 11.60% 23.20% 25.90% local leaders for the benefit of the com- 49.30% 12.80% 21.00% 17.00% munity, 44.7% of community members Age Over 40 across the GLR countries had never DRC 69.60% 11.00% 13.60% 5.80% done so, 11.6% had once done so, 21.1% Republic of Congo 70.40% 13.70% 14.40% 1.50% had done so a few times (five or less), Rwanda 5.90% 5.90% 25.70% 62.40% and 22.6% had done so many times 34.10% 15.90% 30.20% 19.80% (five or more). This cross-country trend Uganda in which the large majority of commu- GLR Average 44.70% 11.60% 21.10% 22.60% nity members have never gathered to express community concerns is charac- teristic of DRC, RoC and Uganda. However, in Rwanda the frequency of gathering was most commonly many times (62.4%) – as is displayed in Table 54.240 Female community members were more likely to have never participated in voicing community issues when compared to male community members (60.7% vs. 34.6%) and less likely than males to have participated once (8.7% vs. 13.4%), a few times (16.3% vs. 24.1%), or many times (14.3% vs. 27.9%). Age shows a slight negative re- lationship to the likelihood of gathering many times for political participation in the last year (26.5% of those 18-30, 25.9% of those 31-40, and 17% of those over 40).241 When questioned further as to the extent that they thought that local government and leaders take into account those concerns voiced by the community when they make important decisions that affect the community, 17.4% of community members across the GLR countries felt that leaders took their concerns into account a lot, 41.7% a little, and 40.9% not at all.242 Female community members were less likely than male community members to feel that leaders took their concerns into account either a lot or a little (15.8% and 37.5% vs. 17.6% and 44.7%, respectively) and more likely to feel that leaders did not take their concerns into account at all (46.7% vs. 36.8%). 16.5 Social Change Similar to trends in the ex-combatant sample, across the GLR countries community members generally were polar- ized in their outlook on the likelihood of their overall situation improving in the future between those that thought 240   This high rate of public gathering to express concerns in Rwanda is likely another effect of Umuganda. While the main purpose of Umuganda is community work it also serves as a platform for leaders to communicate important news on a national and local level as well as for individuals and communities to express concerns and plan for future Umuganda. 241   This age related trend in the likelihood of public gathering may in part be related to the heavy sampling bias in RoC, the country where community public gathering was lowest, towards community members over 40. 242   Rwanda is absent from findings on the extent to which community members feel leaders take their concerns into account due to lack of directly comparable data. 134 Table 55: Community Member Cross-Category Social Change 245 Consider a 9-step ladder where on the bottom (the first step) stand the poorest people, and on the ninth step FOOD? HOUSING? CLOTHING? FINANCES? SCHOOL FEES? HEALTH? LEISURE? stand the richest – On which step were you one year ago in relation to: Male 3.49 3.39 3.53 2.93 3.97 3.71 3.69 Female 3.57 3.60 3.62 3.20 4.04 3.77 3.66 Age 18-30 3.44 3.35 3.51 2.95 4.30 3.69 3.94 Age 31-40 3.68 3.59 3.67 3.15 4.00 3.86 3.83 Age Over 40 3.50 3.52 3.54 3.04 3.82 3.69 3.45 Burundi 3.39 3.18 3.57 2.91 2.77 3.30 XXX DRC 4.06 4.19 4.23 3.49 4.37 3.89 4.40 RoC 3.58 3.46 3.47 3.17 3.60 3.59 2.92 Uganda 2.94 2.77 2.99 2.46 XXX XXX XXX GLR Average 3.49 3.40 3.57 3.01 3.58 3.59 3.67 Consider a 9-step ladder where on the bottom (the first step) stand the poorest people, and on the ninth step FOOD? HOUSING? CLOTHING? FINANCES? SCHOOL FEES? HEALTH? LEISURE? stand the richest – On which step are you today in relation to: Male 3.73 3.60 3.89 3.16 4.06 3.81 3.80 Female 3.75 3.78 3.85 3.31 4.19 3.84 3.77 Age 18-30 3.85 3.65 3.98 3.28 4.35 3.88 4.07 Age 31-40 3.91 3.85 3.96 3.41 4.18 3.97 3.96 Age Over 40 3.54 3.61 3.73 3.07 3.94 3.69 3.54 Burundi 3.61 3.45 3.72 2.94 3.03 3.44 XXX DRC 3.96 4.17 4.18 3.41 4.31 3.81 4.36 Roc 3.93 3.70 3.94 3.42 3.91 3.84 3.18 Uganda 3.34 3.16 3.48 2.84 XXX XXX XXX GLR Average 3.70 3.62 3.83 3.15 3.74 3.69 3.78 The use of XXX signifies a field where no data is available. that it would improve in a few years and those that thought that their situation would deteriorate in the future. Overall, only 1.5% of community members thought that their situation would improve in a few weeks, 4.9% thought it would improve in the coming months, 50% that it would improve in a few years, 6.4% that it would remain the same, and 37.3% reporting that they foresee their overall situation deteriorating in the future.243 As in the ex-combat- ant sample, only Uganda stood apart from this trend – 79.3% of community members reporting that they thought their overall situation would improve in a few years. These findings may suggest that while in general communi- 243   Rwanda is absent from findings on community members’ overall outlook on their situation for the future and their overall level of satisfaction with their life up until sampling due to lack of directly comparable data. 135 ty members have a polarized outlook for their future, those who do have a positive outlook understand the time horizons of social change – occurring in the scale of years rather than days, weeks, or even months. Female community members across the GLR were less likely to report that their situation would improve in the next few years compared to male community member (43% vs. 54.9%) and more likely to think that their overall situation would deteriorate in the future (44.7% vs. 32%). Age as well held a clear relationship to polarized response between these two outcomes. As age increased community members were less likely to see their overall situation improving in a few years (61.4% of those 18-30, 54.4% of those 31-40, and 40.3% of those over 40) and more likely to see it deteriorating (24.8% of those 18-30, 37% of those 31-40, and 46.9% of those over 40). When questioned whether they are satisfied with the way that their life has been to date, across the GLR countries 32.5% of community members reported that they were satisfied, 8.2% that they were neither satisfied nor dissat- isfied, and the remaining 59.3% that they were dissatisfied. However, this cross-country figure fails to depict the nuance between GLR countries as there was a clear split between Uganda on the one hand, and DRC / RoC on the other. In Uganda 43.3% of community members being satisfied, 24.4% neither satisfied nor dissatisfied, and 32.2% being dissatisfied. In contrast, in DRC and RoC 73% of community members were dissatisfied with their life to date and only 27.1% were satisfied.244 Female community members were slightly more likely to be dissatisfied with their life to date than male community members (62.9% vs. 55.9%). Community members were questioned about their perceptions of their own position in society across a range of fields at the time of sampling and a year prior using a nine-step ladder response prompt.245 Their responses are tabulated below in Table 55 by mean score. The lower the mean score is the closer the community member is to the bottom rung of the ladder – where the poorest people tend to be. Generally speaking, across and within the GLR countries community members, as with ex-combatants, consistently identify themselves in the poorest half of society – between steps two and four. However community members rank themselves slightly better than ex- combatants on average across all categories. Looking at the GLR countries as a whole there is a slight increase in the mean scores for community members across all categories. This trend is reflected within each country with the exception of DRC – where on average, scores were higher across all categories, but had declined across all categories from a year prior (with the exception of leisure). A closer look at DRC reveals that the only demographic group that saw average improvements across any catego- ries was those aged 18-30 (who improved across all categories with the exceptions of finance and school fees). At a cross-country level, all gender and age demographic categories see improvements across all categories (with the exception of the health category for those over 40 which stays the same over time). Interestingly, despite the range of economic and social disadvantages that females hold, they perceive themselves as slightly better off than males across all categories except for clothing and leisure at the time of sampling and one year prior. 16.6 Summary Overall, community members across the GLR countries show positive levels of social capital and a general trajec- tory of improvement. Community members have a growing number of social groups and high levels of contact with their families, forming a broad social platform that can serve as a fallback position in times of hardship or a spring- board in moments of opportunity. While community members have diverse friend groups who they can often turn to for support, the family unit is still the core of their social support network. 244   It should be noted that in DRC and RoC, community members were not given the option of replying that they were neither satisfied nor dissatisfied with their life up to the time of sampling – this scaling issue may have inadvertently inflated the number of community members who expressed being dissatisfied with their life. 245   Rwanda is absent from these social change findings due to lack of directly comparable data, however this is the only section in the community dynamics annex of this study where Burundi is included. 136 With these generally strong social networks, community members in turn display a high level of trust in the com- munity and show a continued positive trajectory in this field – also aided by increased stability and security in the end of conflict. These factors have set the context in which community members feel a strong sense of togetherness and meet to work together for the betterment of their communities. Further, community members report being generally happy and describe a broadly positive sense of empowerment in their lives (though they are simultane- ously dissatisfied with their lives to date in general). While community members rank themselves consistently in the poorest half of society across a range of categories, they also display a shallow trajectory of improvement over time. Indeed, while community members are polarized in their general outlooks for the future, those with a positive outlook express that they understand that social change does not occur over night, but rather in the scale of years. Very generally speaking, it appears that the social dynamics of communities across the GLR countries (with the exception of DRC) and provide a context for which ex-combatants can return to communities and strive towards reintegration into an already stable community setting in terms of social capital. However, this general ability of the communities in the GLR absorb ex-combatants and serve as a setting which they can reintegrate into should not mask the realities of the post-conflict social landscape. Families, communities and broader networks in the GLR countries have been affected severe violence and displacement - to the great detriment of trust, solidarity, and social cohesion across the broader social fabric of society. In this sense to long-term project of rebuilding society is one that ex-combatants and community members face together. 16.6.1 Vulnerable Subgroups As consistent with the analysis presented throughout this report, female community members fare worse off than male community members in terms of most indicators of social capital and are thus further solidified as a vulnerable group. Female community members have weaker social networks and less family contact that subtracts overall from their ability to leverage the value of these social connections – leaving female community members in a position of relative social isolation. Though there is some variation from country to country, female community members feel less trust with the com- munity, less togetherness with the community, are less happy personally, are less likely to feel they have a positive impact on the community, and feel less empowered to affect change in their lives. However, despite this broad range of disadvantages in terms of social capital female community members consistently perceive themselves as slightly better off relative to the rest of society than male community members across a broad range of categories including food, housing, finances, and health. Many of the social disadvantages that female community members display may be the result of traditional gender structures and their resulting gender-based inequalities. Understanding these disadvantages is important in the examination of community dynamics themselves, but also carries weight for the return of female ex-combatants. What this means for female ex-combatants is that social reintegration (in terms of reaching parity with commu- nity members) may inadvertently mean reintegrating back into basic gender inequalities – possibly with the added dimension of stigma as an amplifying force to these disadvantages. With this in mind it is important to recognize the importance of reintegration programming that not only addresses the specific disadvantages that female ex- combatants face, but to fit in as part of and effort towards affecting a larger collective shift towards gender equality in post-conflict and development settings. 16.6.2 Unique Country Trends Overall, community members in DRC rank lower than community members in the rest of the GLR countries across a broad range of social capital indicators. Collectively, the core weaknesses of community members in DRC in terms of social capital can be characterized along three dimensions: (i) weak family connections; (ii) weak community connections; and (iii) weak personal self-worth and empowerment. 137 Access to family networks is an important inroad for building further social and economic networks and in turn leveraging the tangible and intangible value of these networks. In terms of family, community members in DRC have the lowest levels of contact with their families, those who do have contact with their families have it the least frequently, and in line with this community members in DRC are the least likely to be satisfied with their level of familial contact. The weak state of familial networks that are characteristic of community members in the DRC are likely a product of the social geography of eastern DRC. Many community members have been displaced or migrated and continuing instability coupled with the mountainous landscape, near non-existent road infrastructure, and heavy seasonal rains keep family networks effectively fractured – isolated by social and physical barriers. The weakness in family connections in DRC corresponds to a distinct weakness in community connections among community members as well. Community members in DRC have a low number of social groups on average, reported weakest levels of improvement in trust in the community, the lowest sense of togetherness, and were the least likely to work with others for the betterment of their community compared to community members in other GLR countries. These indicators of weak social capital for community members in terms of family connections and community networks correspond to the overall weaker economic situation of community members highlighted in section 15.6.2 of this annex. Further, these broad weaknesses in community members in DRC’s familial and community networks correspond to their low senses of self-worth and empowerment. Community members in DRC are the least likely to feel they have a positive impact on the community, the least likely to feel valued by others in the community, and the most likely to be dissatisfied with their life compared to community members in other GLR countries. In addition while community members in DRC perceive themselves as slightly better off compared to the rest of society across a range of categories than community members in other GLR countries, they are the only group who see a decrease in their perceived standing over time – possibly a result of continuing instability in the region. Violent conflict has damaged the social fabric of individuals and communities across the GLR. However, it appears that the continued insecurity in eastern DRC coupled with the intense geographic landscape in the region has con- tributed to a fragmented social geography in which familial and communal networks are fractured and cannot be leveraged for their value by community members – leaving them particularly exposed to social and economic isola- tion. Future studies on social capital in the region could flag the interaction of social capital and social geography as a field for further analysis. 138 17. Reintegration Experiences T he following is an analysis of community member experiences of the reinsertion and reintegration of ex- combatants. Most importantly, the analysis here highlights the changes in community perspectives towards ex-combatants since the reintegration process began. For the greatest analytical value this chapter should be read in conjunction with section 11 on ex-combatants’ DDR experiences. Owing to data constraints, this section of the study draws exclusively from DRC, RoC, and Uganda. 17.1 Community Sensitization and Preparedness Across the GLR countries, community members most commonly received information, though not necessarily official information, about ex-combatants coming to the area they live in to reintegrate through: (i) word of mouth (41.2%); (ii) radio (27.3%); or (iii) a community meeting (11.1%). In Uganda, though the three most common mediums by which community members received information about returning ex-combatants were the same, radio was the most common medium (30.2%), followed by word of mouth (22.5%), and community meetings (14.8%). Across the GLR countries, female community members were more likely to get information about ex-combatants from word of mouth than male community members (49% vs. 35.9%) and less likely to get it from radio (23.8% vs. 29.8%). Those aged 18-30 were particularly likely to have received information through radio compared to other age demographic groups (35.2% of those 18-30, 27.1% of those 31-40, and 22.1% of those over 40). The vast majority of community members across the GLR countries (70.5%) reported that they were given no help in understanding how reintegration was going to take place, the remainder reporting receiving some help (20%) or reporting receiving lots of help (9.5%).246 Uganda stands out from the cross-country trend with a less unipolar distribution (47.2% no help, 34.8% some help, 18% lots of help). In regards to gender, female community members were more likely to report receiving no help compared to male community members (77.3% vs. 65.8%). While this gendered trend continued in Uganda there was an additional dimension – female community members were also more likely to have received lots of help on understanding how reintegration would take place compared to male community members (23.2% vs. 15.6%). When asked further whether they thought they should have been informed or given more help before ex-combat- ants were reintegrated into their community, there was a near even split across the GLR countries in community members responses – 52.1% reporting that yes they should have been given info and help and 47.9% replying no. A closer look at each of the individual countries shows that in RoC and Uganda there was an approximate 60/40 split between those who responded yes and no. Interestingly, in DRC this split in responses was reversed 40/60. This is interesting because DRC was the country where community members most frequently (87.6%) reported receiving no help on understanding how reintegration would take place. When asked by what medium they would have liked to have received information about the reintegration process the three most common replies are the same as the three most common mediums by which community members actually did receive information – though with distinctly different distributions between these responses – 44.1% of community members wanted to receive information about reintegration in community meetings, 29.3% preferred 246   At least in eastern DRC, the geographic challenges of face-to-face sensitization can play a role in these figures. 139 radio, and 9.4% by word of mouth.247 As dis- Table 56: Community Member Information Sources on Reintegration250 played in Table 56 though the most common medium by which community members How were you informed across the GLR countries received informa- How should about Ex-Combatants coming to reintegrate into the area you have been tion about reintegration was word of mouth informed? the most preferred was clearly community you live? Community meeting 11.1% 44.1% meeting. Female community members were slightly less likely to prefer radio as an infor- Word of mouth 41.2% 9.4% mation medium compared to male commu- Radio 27.3% 29.3% nity members (26.5% vs. 30.7%) and more Church or mosque 1.5% 4.7% likely to prefer word of mouth (16.5% vs. NGOs or charities 1.4% 2.2% 5.6%). Government ministries 1.9% 6.5% 17.2 Community Perspectives on Newspaper .8% 2.3% Ex-Combatant Reintegration and Phone call/ Megaphone/ Public Announcement .0% .2% Fear National Commission 0.0% .1% Across the GLR countries, community Letter .2% members had only a moderate level of .1% Door-to-door personal interaction with returning ex-com- .1% .1% announcement batants – 35% had lots of direct contact, Other 3.6% .3% 21.2% had a little direct contact, and 43.9% Was not informed 11.0% XXX had no contact. However, this cross-country figure masks the diversity in levels of commu- All Means XXX .8% nity member contact within the GLR coun- Subtotal 100.0% 100.0% tries – In DRC, the levels of contact were The use of XXX signifies a field with no available data. drastically lower than average (10.1% lots of contact, 11.1% some contact, and 78.8% no contact) while in Uganda, contact levels were generally higher than average (63.7% lots of contact, 23.1% some contact, and 13.2% no contact).248 RoC fell closest to the cross-country average with 30.1% lots of contact, 31.2% some contact, and 38.8% no contact. In DRC and RoC, where community member contact with returning ex-combatants was lower, female community members were less likely to respond that they had lots of direct contact than male community members (7.8% vs. 12.2% in DRC and 23.4% vs. 35.9% in RoC) and more likely to respond that they had no contact (82.6% vs. 75.3% in DRC and 43.4% vs. 34.8%) while in Uganda, where contact levels were generally higher, the trends were reversed – female community members were more likely than male community members to respond that they had lots of direct contact than male community members (73.7% vs. 59.2%) and less likely to respond that they had no contact (8.8% vs. 15.2%).249 Across the GLR countries, age showed a negative relationship to the likelihood of reporting having lots of contact with returning ex-combatants (44.2% of those 18-30, 31.9% of those 31-40, and 287% of those over 40) and, inversely, a positive relationship to the likelihood of having a little contact (13.7% of those 18-30, 247   Unfortunately there is no data available regarding community members’ perspectives on the content of the information and sensitization they did receive. 248   In DRC, especially eastern DRC, these lower levels of community member contact with returning ex-combatants may be a product of the difficulty of travel and continued insecurity as a part of the dynamics of return. 249   This gendered trend is likely a product of the fact that female community members who had a spouse were more than twice as likely as male community members to have a spouse who was an ex-combatant. 140 23.5% of those 31-40, and 25.9% of those over 40). Table 57: Community Member Fear of Ex-Combatants Drawing specifically from DRC When ex-combatants first came to live Country and RoC, the majority of commu- in your community, did you fear the Republic nity members (64.3%) described listed reporter group: DRC of Congo Uganda Subtotal their contact with ex-combatants as positive, while 25% described Male Yes, I feared them 97.1% 95.6% 83.3% 91.1% their contact as neither positive nor Ex-Combatants No, I did not fear them 2.9% 4.4% 16.7% 8.9% negative, and the remaining 10.7% Yes, I feared them 60.4% 62.2% 29.4% 47.4% Female as negative.250 Female community Ex-Combatants No, I did not fear them 39.6% 37.8% 70.6% 52.6% members were slightly less likely Yes, I feared them 63.6% 63.9% 25.5% 46.8% than male community members Child Ex-Combatants to describe their contact with ex- No, I did not fear them 36.4% 36.1% 74.5% 53.2% combatants as positive (60.5% vs. Disabled Yes, I feared them 63.0% 57.0% 19.6% 42.2% 66.7%) and more likely to describe Ex-Combatants No, I did not fear them 37.0% 43.0% 80.4% 57.8% it as neither positive nor negative (26.7% vs. 23.9%) or just negative (12.9% vs. 9.4%). Community members over the age of 40 were the least likely demographic subgroup to describe their interactions as positive (60.1%) and the most likely to see interactions as neither negative nor positive (27.1). Interestingly, the two demographic subgroups with the highest frequencies of describing their contact with ex-combatants as either negative or neither positive nor negative, females and those over 40, were also those that reported the lowest levels of direct contact with ex-combatants as described above.251 When asked to reflect on when ex-combatants first came to live in their community, just over half of community members (51.5%) reported that they had fears about their presence – the remaining 48.5% reporting that they had no fears. This near even split is fairly durable across the GLR countries. Female community members are slightly more likely than male community members to report having fears about ex-combatant presence in the community (53.5% vs. 50%). When asked about which specific groups of ex-combatants they feared, community members across the GLR coun- tries gave a consistent message: community members reported fearing male ex-combatants to a very high level (91.1%) and female, child and disabled ex-combatants to a considerably lower level (47.4%, 46.8% and 42.2%, re- spectively). As is visible in Table 57, across all categories Uganda showed lower levels of overall fear – especially in regards to female, child, and disabled ex-combatants.252 In regards to community member demographic trends, female community members were slightly more likely to report fearing ex-combatants across all categories and age showed a positive relationship to the likelihood of fearing ex-combatants across all categories. In Uganda, community members were asked to outline what kinds of specific fears they held about different kinds of ex-combatants. As is visible in Table 58, the most common fear that community members held in regards to the 250   Questions regarding the positive or negative nature of contact with returning ex-combatants were only asked in DRC and RoC. 251   It is difficult to decipher the relationship between levels of community member contact with returning ex-combatants and perceptions about the positive or negative character of those interactions – if there is one at all. One could postulate that lower levels of contact with returning ex-combatants provides a limited base on which for community members to break down stereotypes and stigma. Or, one could just as well propose that precisely because of negative experiences with returning ex-combatants community members have minimized contact. 252   One explanation for the lower levels of fear of child ex-combatants in Uganda could be related to the dynamics of mobilization and return. Abduction is a known recruitment tactic of the LRA in northern Uganda. In terms of dynamics of return this has created a sentiment among community members in which they view child ex-combatants simultaneously as victims and perpetrators and have displayed accepting attitudes of their return. It is also possible that the long period of time between informal and formal demobilizations and the overall trickle-in model of demobilization in Uganda may play some role in the slightly lower overall levels of fear surrounding the return of ex-combatants. 141 Table 58: Community Member Specific Fears of Ex-Combatants in Uganda Disabled Ex- Combatants Combatants Combatants Combatants Female Ex- Child Ex- Male Ex- When ex-combatants first came to live in your community, describe what fears you had of the listed reporter group: Possibility of ex-combatants being a perpetrator of violent activity or crime such as 38.6% 28.6% 38.5% 31.6% murder, rape Fear due to ex-combatants carrying firearms and weapons 14.5% 7.1% 11.5% 10.5% Lack of trust in ex-combatants or fear of ex-combatants 7.2% 3.6% 0.0% 0.0% Fear due to possibility of resumption of rebel activity by ex-combatants, or retalia- 4.8% 3.6% 3.8% 15.8% tion, or resurgence of rebel activities Intelligence gathering, spying 1.2% 10.7% 11.5% 10.5% Inability to stay or coexist with community 1.2% 0.0% 3.8% 0.0% Interruption of the community, cause problems in the community, cause insecurity 3.6% 7.1% 7.7% 10.5% in the community Uncontrollable, badly behaved, drinking, unsociable habits 13.3% 10.7% 7.7% 5.3% No mercy or sympathy or empathy shown, bad character of ex-combatants 1.2% 14.3% 7.7% 5.3% Ex-combatants’ appearance 1.2% 3.6% 3.8% 5.3% Interpersonal conflicts with other people 1.2% 0.0% 0.0% 0.0% Ex-combatants being bitter and unforgiving 0.0% 0.0% 0.0% 5.3% Psychological problems, such as they quickly change moods and become hostile 12.0% 10.7% 3.8% 0.0% return of all types of ex-combatants was the possibility of ex-combatants being a perpetrator of violent crime such as murder or rape. Interestingly when community members are asked about the fears they have about the presence of ex-combatants in their community today now that ex-combatants have been there for some period of time, 93.1% report that they have no fears – the remaining 6.9% still holding some fears. This denotes a dramatic improvement in the commu- nity’s ability to absorb ex-combatants since their initial return and a key hint for understanding the process of social reintegration in the GLR countries. Across the GLR countries, 31.3% of community members believe that ex-combatants should have behaved dif- ferently since coming to the community (the remaining 68.7% responding that ex-combatants should not have behaved differently), a figure that is very consistent within the individual GLR countries and across demographic subgroups. Similarly, when community members were asked whether they thought the community should have behaved differently since the arrival of ex-combatants, 27.5% thought that the community should have behaved dif- ferently. When asked about whether or not there was any resentment in the community about the support that ex- combatants received, 27.9% thought that there was resentment, though it is unclear how this resentment is related to the ways in which community members think returning ex-combatants and community members should have behaved differently. 17.3 Positive and Negative Perceptions of Ex-Combatants Across the GLR countries, 29.8% of community members believe that there are negative dimensions to having ex- combatants in the community – the remaining 70.2% responding that there are no negative dimensions. However Uganda stood apart from this trend, instead community members less frequently identified ex-combatant presence 142 as having negative dimensions (18.1% yes negative factors, 81.9% no negative factors). When asked to outline the types of negative dimensions related to having ex-combatants in the community the most notable responses were: (i) that having ex-combatants in the community increased the risk for violent crime (54.4%); (ii) that ex-combat- ants have generally bad or brutal behavior (18.3%); or (iii) that ex-combatants can bully, intimidate, or threaten others (10.7%). When asked whether there were positive aspects to having ex-combatants in the community, across the GLR coun- tries, 67.7% of community members responded that there are distinct positive dimensions to having ex-combatants in the community – a higher proportion than identified negative aspects. Again, Uganda stands apart with 95% of community members identifying that there are distinct positive dimensions to having ex-combatants in the com- munity. When asked to outline the main positive dimensions to having ex-combatants in the community, notable responses from community members were: (i) that ex-combatants give sound advice to other people and serve as good role models (23.9%); (ii) that ex-combatants make positive contributions to the economic fabric of the com- munity (23.7%); and (iii) that ex-combatants handle all village security issues (7.8%). 17.4 Summary There are several key findings to take away from this section. In terms of information and sensitization: (i) across the GLR countries community members most commonly received information and sensitization about the return and reintegration of ex-combatants through word of mouth; (ii) community members across the GLR countries would most dominantly have preferred to receive information and sensitization about the return and reintegration of ex- combatants in a community meeting forum. Turning to community members’ levels of fear surrounding returning ex-combatants there are also several key points: (i) community members across the GLR countries had generally high levels of fear, particularly in regards to violent crime, associated with the return of ex-combatants – especially male ex-combatants before their return; (ii) after ex-combatants have returned to communities the level of fear that community members hold towards ex- combatants dropped drastically – though some resentment remained; and (iii) after ex-combatants have returned to communities, community members more commonly identify a range of positive aspects to having ex-combatants than negative. Collectively this narrative of high community member fear, exposure to ex-combatants, followed by low fear with a mostly positive perception of ex-combatants is a positive indication of communities’ ability to absorb returning ex-combatants. Further, this narrative gives support to the idea that much of the social dimension of reintegra- tion is constituted by a process of confrontation and atonement – eroding distrust and stigma. While it appears as though initial trust barriers may fall quickly the longer road to reaching social and economic parity for ex-combat- ants remains. 143 18. Conclusions C onflict across the Great Lakes Region has carried enormous weight in affecting the lives of ex-combatants and community members alike. Though conflict-affected countries in the GLR are generally characterized by severe economic development challenges and a deteriorated social fabric, this study has revealed that in the wake of peace, communities across the GLR have reached a level of relative social and economic stability. It is this stability that constitutes communities’ capacity to play a positive role in accepting and absorbing returning ex- combatants into their social and economic fabric. Indeed, without a relatively stable social and economic base in the community the idea of the “reintegration” of ex-combatants would lose much meaning – as ex-combatants would reintegrate into economic instability and social marginalization. Thus, understanding the state of communities and their social and economic dynamics is an essential backdrop for understanding ex-combatants’ position and trajec- tory on the path to reintegration – gaining social and economic parity with community members. 18.1 The Community and Economic Reintegration The analyses of the community member sample presented in this study have shown that community members across the GLR display a stable economic trajectory over time. The majority of community members are engaged in self- employment in small-scale agriculture and as such land access for cultivation and grazing is a key issue. In addition, community members show some diversification into self-employment in service or retail related activities. Overall community members’ employment statuses are stable over time and unemployment varies little on average. Like ex-combatants, community members see the primary barrier to improving their economic situation as revolv- ing around lack of opportunities. Beyond this, community members cite lack of access to capital and credit as among the additional barriers to leveraging what opportunities do come towards their economic betterment – and indeed their access to capital and credit in terms of the reception of micro-loans or membership in economic associations such as local savings and credit organizations is low. In the context of the severe development challenges that characterize the GLR countries, community members’ core strength lies in their relative economic stability. The vast majority of community members meet their monthly household expenses alone, or with the help of others in their household. Only a minority is locked into patterns of borrowing from family and friend networks to meet their basic needs. It is this economic context of relative stability that provides the context in which ex-combatants can return to communities and strive towards parity in a meaning- ful sense – the longer term process of upward economic mobility occurring outside the bounds of reintegration. 18.2 The Community and Social Reintegration The analysis presented in the community member sample presented in this study shows that across the GLR coun- tries community members have a generally positive level of social capital, and further a positive trajectory over time – as the social fabric of communities is mended in the wake of improved peace and security. The core of social capital revolve around social networks, be they familial, communal, interpersonal friendships, or strictly economic. Networks have value both in the sense that they serve as a platform for social and economic support within commu- nities, but also can be leveraged to create new social and economic opportunities. Community members across the GLR countries show that they have connection to those around them in terms of social groups, diverse friends, and economic networks. In this sense older community members (over 40) have perhaps the highest social capital and a 144 solid footing in the community – often rank highest on core indicators. However the core of community members’ social capital, and gateway to accessing broader social networks, is their solid grounding in the family unit accessed through marriage. Indeed, marriage rates are a powerful indicator of overall community social capital – correlating to larger social and economic networks on average. As community members marry they expand their social networks and the overlap of these individual networks grows – in a very literal sense weaving together to constitute the social fabric of com- munities and societies. Community members’ rates of marriage are entangled with their number of social groups in general and contribute to their overall engagement in the community in terms of trust, solidarity, social cohesion, and inclusion – in turn feeding back into network building. It is this dynamic interaction of community members’ networks and their collective benefits that feed back to the individual as well – cementing their personal sense of empowerment and understanding of their place in society. Understanding the dynamism of social networks, the family core among them, as fabric connecting individuals into communities is core to understanding the contexts which ex-combatants approach in the process of social reintegration. Essentially social reintegration means that ex- combatants must find a way to connect into this social fabric – perhaps most meaningfully through marriage. The analysis presented in this annex suggests that though issues of stigma and distrust towards returning ex-com- batants may exist in many contexts across the GLR, these barriers break down fairly quickly. It is the presence of an underlying social fabric, in terms of individual social capital, that exists throughout communities across the GLR countries, with the notable exception of DRC – discussed below, that can serve as the necessary condition for ex- combatants’ embankment on a path towards social reintegration. 18.3 Female Community Member Sub-Group Throughout the analysis of community dynamics presented in this annex female community members have consis- tently displayed a range of disadvantages across nearly all core demographic, economic, and social indicators that collectively paint a narrative of gender inequality across the GLR countries. Female community members have lower literacy and educational achievement levels than male community members – this, in part, affects their higher likelihood of unemployment through time. Female community members under- stand this connection between education and unemployment – being more likely to cite lack of education and skills as a barrier to gaining a productive economic status. Furthermore, perhaps not surprisingly, female community members are more likely than males to work in the household fulfilling traditional gender roles. Female community members are less likely to be a sole household breadwinner, an advantage, though when they are they fare consider- ably worse off than male community members in terms of meeting monthly expenses. Beyond their weaker overall economic position, female community members also face considerable disadvantages in terms of social capital. Female community members have smaller social networks in terms of levels of familial contact and number of social groups; in turn, they are less integrated into the social fabric of communities – leverag- ing the value of their networks in terms of support and opportunities. Female community members feel less trust in the community, less togetherness with the community, feel they have less of a positive impact on the community, and are less happy and empowered overall. Collectively the range of disadvantages that female community members face across the GLR countries is likely a product of traditional gender inequalities. In this sense these disadvantages are both structurally ingrained and culturally reproduced. Acknowledging the social-structural disadvantages that female community members face across the GLR countries is not only an important dimension of understanding community dynamics, but also the prospects that female ex-combatants face as they approach the process of reintegration. If female ex-combatants are to gain parity with female community members, issues of stigma will no doubt serve as a barrier to entering the com- munity, but if female ex-combatants are rather to reach parity with male community members, a much deeper set of 145 social-structural barriers stand in their way – barriers that they and their female community member counterparts face together. In this sense reintegration programming is poised to serve not only the needs of female ex-combatants, but also represents an opportunity to encourage a larger community-wide transformation. 18.4 DRC – A Splintered Society While throughout the analysis of community members presented in this annex each of the GLR countries has varied considerably in terms of specific contextual trends, only DRC displays a truly divergent narrative of community dynamics. As outlined in section 16.6.2, community members in DRC stand out from the rest of the GLR countries with the weakest levels of social capital across a broad range of indicators. When female community members across the GLR countries display disadvantages, these disadvantages are often exaggerated in DRC. Though the exact reasons for these trends are unclear, it is likely that this weak social capital at the individual level, and weak social fabric at the community level, are related to ongoing instability in Eastern DRC coupled with the harsh social geog- raphy in the region – keeping families, social groups, and networks separated by physical barriers. The analysis pre- sented paints DRC as a splintered society where community members have weak familial and communal networks – missing the opportunity to leverage their value. This narrative has considerable weight for understanding the community dynamics in DRC itself, but is also essential for understanding the prospects for meaningful ex-combatant reintegration in DRC. If reintegration means reaching parity with community members then ex-combatants appear to have done well in reentering this splintered society with weak social fabric – though this is not to suggest that ex-combatants in DRC do not face significant barriers to reentering communities. However, if social reintegration is understood as going beyond mere parity, to a process of building social networks and in turn leveraging their value then this is a challenge that community members and ex-combatants alike will face in DRC. With this in mind, it may be that in the context of DRC, or perhaps settings of long-lasting or continuing conflict in general, community based approaches to reintegration focused on benefit- ing the community could prove particularly impactful. However, as always, reintegration programming must be grounded in the context that is meant to affect. In DRC, or elsewhere, meaningful reintegration programming must be anchored in complexities of the local context – a challenging endeavor indeed. 146 147 Transitional Demobilization and Reintegration Program World Bank 1818 H Street NW Washington, DC 20433 USA www.TDRP.net 148