53714 IndonesianSocialDevelopmentPapers Since1998,Indonesiahasbeenundergoingamomentouspoliticalandeconomictransition.The fall of the New Order, the economic crisis, and radical decentralization have changed the political, economic and social context. Within this new context, power relations are in flux, identities are being renegotiated, and institutions are changing. Changes in incentives, and in the role of formal and informal institutions at various levels, have altered the ways in which individualsandgroupsrelatetoeachotherandthestate.Understandingthisnewcontext,and the ways in which various actors (national and international) can promote progressive social changeisimportant. The Indonesian Social Development Papers series aims to further discussion on a range of issues relating to the current social and political context in Indonesia, and to help in the generation of ideas on how democratic and peaceful transition can be supported. The series willcoverarangeofissuesincludingconflict,development,corruption,governance,theroleof thesecuritysector,andsoon.Eachpaperpresentsresearchonaparticulardimensionofsocial developmentandofferspragmaticpolicysuggestions.Papersalsoattempttoassesstheimpact ofvariousinterventions--fromlocalandnationalactors,aswellasinternationaldevelopment institutions--onpreexistingcontextsandprocessesofchange. The papers in the series are works in progress. The emphasis is on generating discussion amongst different stakeholders--including government, civil society, and international institutions--rather than offering absolute conclusions. It is hoped that they will stimulate further discussions of the questions they seek to answer, the hypotheses they test, and the recommendationstheyprescribe. PatrickBarron(serieseditor)pbarron@worldbank.org CommunityBased Reintegration in Aceh AssessingtheImpactsofBRAKDP PatrickBarron MacartanHumphreys LauraPaler JeremyWeinstein December2009 IndonesianSocialDevelopmentPaperNo.12 PapersintheIndonesianSocialDevelopmentseriesarenotformalpublicationsoftheWorldBank.They arepublishedinformallyandcirculatedtoencouragediscussionandcommentbetweenthoseinterested inIndonesiandevelopmentissues.Thefindings,interpretations,judgments,andconclusionsexpressed in the paper are those of the authors and should not be attributed to: the World Bank and affiliated organizations; members of the World Bank's Board of Executive Directors or the governments they represent;oranyofthefundingagencies. ThefullrangeofpublicationsassociatedwiththebroaderstudyoflocalconflictinIndonesia(ofwhich thisreportisaproduct)isavailableonlineatwww.conflictanddevelopment.org. Emailaddressesforcorrespondence: pbarron@worldbank.org macartan@gmail.com lbp2106@columbia.edu jweinst@stanford.edu Copiesofthispaperareavailablefrom: PNPMSupportFacility JalanDiponegoroNo.72 Jakarta10310Indonesia Tel:+62(0)213148175 Fax:+62(0)2131903090 Preface TheendoftheconflictinAcehledtothearrivalofarangeofprogramsaimedat`reintegrating' formercombatantsandprovidingassistancetoconflictaffectedgroups.TheBRAKDPprogram wasaninnovativeattemptbylocalandnationalgovernmentstoemploylessonslearnedfrom successful past communitydevelopment work to postconflict Aceh. The program, the design andimplementationofwhichwassupportedbytheWorldBank,deliveredoverUS$20million to conflictaffected villages, aiming to support the welfare of conflict victims while building socialcohesionandtrustinthestate. Diditwork?Whatimpactsdidithaveandwhy?WhatcantheexperienceofBRAKDP,andthe broaderreintegrationeffortsinAceh,tellusabouthowgovernments,donorsandcivilsociety cansupportpeaceconsolidationinpostconflictarenas? This paper aims to contribute towards answers to these questions. It outlines the results of surveys of households, former combatants and village heads conducted across Aceh, which were designed to determine project impacts. It should be read in conjunction with a complementary paper, Delivering Assistance to ConflictAffected Communities: the BRAKDP Program in Aceh (Indonesian Social Development Paper number 13), which in addition to further analysis of the Aceh Reintegration Livelihood Surveys (ARLS) data set discussed here, provides evidence from the project's Monitoring Information Supervision (MIS) system, from supervision missions and from qualitative fieldwork. In addition to providing a deeper understanding of the BRAKDP programs and its impacts, the papers aim to contribute to a small but growing literature evaluating postconflict programs. We hope that the BRAKDP experience will serve as a laboratory through which broader processes of change can be understoodinasocietytransitioningfromconflict. PatrickBarron Conflict&Developmentteam WorldBank,Indonesia i ExecutiveSummary This paper describes the results of an evaluation of the CommunityBased Reintegration Assistance for Conflict Victims program (BRAKDP) funded by the Aceh PeaceReintegration Agency (Badan ReintegrasiDamai Aceh, or BRA) and implemented with support from the MinistryofHomeAffairsandtheWorldBank.BRAKDPwasdesignedtoassistconflictvictims acrossAcehaspartofthereintegrationprogramemergingfromtheHelsinkipeacedeal,which brought to an end a 30year conflict between the separatist GAM movement and the GovernmentofIndonesia.US$21.7millionoffundswerechanneledthroughtheGovernment's KecamatanDevelopmentProgram(KDP),whichhasbeenoperatinginAceh,andelsewherein Indonesia,since1998.Blockgrantsweredeliveredtocommunitieswhodecidedonhowfunds should be spent. As with other communitydriven development (CDD) projects, the program emphasized participation, local ownership, and transparency in giving communities control overthechoiceandimplementationofprojects. BydesigntheprogramwastobeimplementedthroughoutAcehintwophases.Intheevent, however,onlythefirstphasewasimplemented.ThisphasewasimplementedbetweenAugust 2006and2007,andtargeted1,724villages,aroundonethirdofthetotalinAceh,in67rural subdistricts. First phase subdistricts (kecamatan) were selected on the basis of a rule which took account of theextentto which areas wereaffectedbyconflict and how effectivelythey haddisbursedKDPmoniesinthepast;allvillagesinselectedsubdistrictsreceivedblockgrants. The program had both economic and social goals. It aimed to deliver quick assistance to conflictaffectedvillagersinordertoimprovematerialwellbeingintheshortterm.Inaddition, itsoughttopromotesocialcohesion,strengthenvillageleveldecisionmakinginstitutions,and cultivategreaterfaithingovernmentalinstitutionsintheaftermathoftheconflict. This evaluation uses data from a household and village head survey conducted after the programtoexaminetheextenttowhichthesegoalshavebeenmet.Theevaluationemploysa secondbestapproachtoestimatetheimpactofBRAKDPonmaterialwellbeing,socialcohesion, and trust in government. Because the program was not randomized, simple comparison of outcomes in project and control areas are unlikely to produce valid inferences of program effects. Indeed our data suggests that biases arising from selection effects would be considerable.Insteadtheevaluationreliesonaneconometricstrategythat(a)usesaformof matchingtoidentifysuitablesubdistrictsthatdidnotreceivetheprogramtoserveasacontrol group, (b) accounts for variables used in the selection process, and (c) uses an instrumental variablesapproachtodealwithissuesofnoncompliancewithtreatmentassignment. The evaluation finds that there were a large number of beneficiaries. An estimated 530,000 individuals live in households that directly received assistance. There was widespread participationintheprogram(over200,000peopleattendedBRAKDPmeetings)andpoorerand femaleheaded households were as likely to attend as others. The program was successful to some extent in reaching out to more marginalized groups in villages. Targeting of conflict victimswashoweverlimited.Overall24percentofconflictvictimsinthestudyareareceived benefits; within project areas, 44 percent of conflict victims were supported through the ii program,arateonlymarginallyhigherthanthatfornonvictims(40percent).Withinvillages, conflictvictimswerenomorelikelythannonvictimstoreceivebenefits.Howeverbecausethe designoftheprogramprovidedmorefundstoareaswithmoreexposuretoconflict,onaverage, conflict victims who received cash received 13 percent more than nonvictims. Particular categoriesofconflictvictimswerelikelytoreceivehigheramounts:the`mostconflictaffected' received19percentmorethannonvictims. Theprogramwasalsowellreceived.Peopleweremorelikelytobeawareoftheprogramthan regular KDP and reported it to be very popular. Ninetyfour percent of villagers in treatment areassaidtheprogramwashelpful,andthisfigureroseto96percentofconflictvictimsand97 percentforthemostconflictaffected. We find evidence that BRAKDP is associated with a strong set of welfare gains and improvements in perceptions of wellbeing. The participation of villages in the program is associatedwithan11pointdeclineintheshareclassifiedas"poor"asreportedbyvillageheads. Typically in programs of this form block, grants are used for community projects selected by villages. In this case, however, the majority of villages prioritized the provision of cash to individuals and groups, primarily for economic purposes. Our data suggest that these cash disbursements are associated with an increased ownership of assets (notably engines and motorcycles) among households in general and conflict victims in particular. The program is also associated with an increase in the farming of productive land (corresponding approximately to a doubling in land use for conflict victims). Conflict victims in areas that receivedtheprogramaresignificantlymorelikelytoreporttheirwelfarehasimprovedinthe pastyearthanarevictimsinvillagesthatdidnotgettheprogram.Thereis,however,nodirect evidenceofwelfareimpactsinotherareassuchasschoolattendance,healthandemployment levels. Unsurprisingly, given the small proportion of project funds used for public goods, the programisnotassociatedwithchangesinthelevelofcommunityinfrastructure. Theevidenceforimprovementsinsocialcohesionandstrongerrelationsbetweencitizensand governmentisweak.Levelsofsocialacceptanceofreturninggroups,reportedsocialtensions, divisionsandconflictandcommunityefficacyaresimilarbetweenthosevillagesthatreceived theprogramandthosethatdidnot.BRAKDPisassociatedwithanincreaseinparticipationin women's groups, but there is no evidence of an overall increase in associational activities. ThereissomeevidencethatBRAKDPresultsinlowerlevelsofacceptanceofexcombatantsby conflict victims. This could be because excombatants received funds that were meant for civilianconflictvictims,althoughthereisnoevidencethatthey`captured'theprogram.Itmay alsobearesultoffrustrationsfromexcombatantsthattheycouldnotbenefitmoreorbecause ofempowermentofcommunitiestostanduptoexcombatantdemands.Finally,thereisonly minimal evidence that exposure to BRAKDP resulted in higher levels of trust in village and higherlevelgovernments. While there is some evidence in the broader literature that CDD programs improve social cohesionandincreaseacommunity'scollectivecapacity,wedonotfindevidencethatBRAKDP had these effects. One reason may be that the program only ran for one year, limiting the iii extenttowhichsuchgains,whichtendtobuildovermultipleprogramcycles,couldeventuate. Anotherreasonmaybebecausecommunitydecisionstoemphasizeprivategoodslimitedthe opportunities for collective implementation. This latter account highlights a possible tension withintheCDDmodel.ManyofthegoalsofCDDmaydependuponprocessesthatarebrought into play conditional on particular types of activities (joint selection of projects, community oversight of implementation, etc.). Yet certain kinds of activities that communities might choose are less likely to encourage interaction, limiting some of the social gains that CDD projectsmightpurporttohave.CDDprogramscanalsoleadtoincreasedtensionsbypromoting competitionoverfiniteresources.Inthelongrun,thismayleadtoastrongerbasisforpeace, by empowering groups and building local institutions, but it can also create divisions in the shortrun.Inpostconflictcontexts,itisnecessarytoweighthese(potential)effects. iv TABLEOFCONTENTS Preface.............................................................................................................................................i ExecutiveSummary ........................................................................................................................ii TableofContents ...........................................................................................................................v TableofTables ............................................................................................................................. vii TableofFigures ............................................................................................................................. vi Acknowledgements ..................................................................................................................... viii 1 Introduction..............................................................................................................................1 1.1 ApproachestoReintegration.............................................................................................1 1.2 CommunityBasedReintegrationAssistanceforConflictVictimsinAceh.........................2 1.3 TheSelectionofProjectAreas...........................................................................................4 1.4 AnalyticalFrameworkandMethods..................................................................................6 Selectingappropriatecomparisonunitsexpost .....................................................................8 Datacollection ......................................................................................................................11 Estimationstrategy ...............................................................................................................12 2 ImplementingBRAKDP ..........................................................................................................14 2.1 WhoReceivedAssistancethroughBRAKDP? .................................................................14 Whoisaconflictvictim?........................................................................................................14 AreconflictvictimsconcentratedintheareasthatreceivedBRAKDP? ...............................16 Doconflictvictimsbenefitmorethanothersintreatmentareas?........................................19 Conclusionsontargeting.......................................................................................................21 2.2 HowWereBRAKDPFundsSpent? ..................................................................................22 Directbenefits .......................................................................................................................23 Projects..................................................................................................................................26 2.3 ParticipatinginBRAKDP .................................................................................................28 2.4 PerceptionsofandProblemswithBRAKDP ...................................................................29 3 ImpactsonWelfare ................................................................................................................34 3.1 PovertyProfile .................................................................................................................34 3.2 AssetIndex ......................................................................................................................36 3.3 HouseholdInfrastructure ................................................................................................38 3.4 LandUse ..........................................................................................................................39 3.5 EmploymentandWages..................................................................................................41 3.6 EducationandHealth ......................................................................................................42 3.7 PublicGoods ....................................................................................................................43 3.8 WelfarePerceptions ........................................................................................................44 3.9 ConclusionsonWelfare ...................................................................................................45 4 ImpactsonSocialCohesion ....................................................................................................46 4.1 SocialAcceptance ............................................................................................................46 4.2 SocialTensions.................................................................................................................50 4.3 ConflictResolution...........................................................................................................51 4.4 CollectiveEfficacy ............................................................................................................52 4.5 AssociationalLife .............................................................................................................53 4.6 ConclusionsonSocialCohesion.......................................................................................55 v 5 ImpactsonTrustinLocalGovernmentandStateSocietyRelations......................................56 5.1 TrustinCommunityDecisionMaking .............................................................................56 5.2 TrustinGovernment........................................................................................................58 Behavioralmeasures .............................................................................................................58 Attitudinalmeasures .............................................................................................................59 Knowledgeofgovernment ....................................................................................................60 5.3 AttitudesaboutGovernance ...........................................................................................61 5.4 ConclusionsonTrustinLocalGovernmentandStateSocietyRelations.........................62 6 CommunityDevelopmentandExCombatantReintegration.................................................63 6.1 ExCombatantCaptureofBRAKDPFunds?.....................................................................63 6.2 AlternativeExplanations..................................................................................................64 7 Conclusions.............................................................................................................................66 References....................................................................................................................................69 TABLEOFFIGURES FIGURE1.1:CONFLICTINTENSITYINACEH ............................................................................................................................. 5 FIGURE1.2:PROPENSITYSCORESFORTREATMENTANDCOMPARISONUNITS ............................................................................... 9 FIGURE1.3:CHARACTERISTICSOFSUBDISTRICTSSELECTEDINTOSTUDY ................................................................................... 10 FIGURE1.4:LOCATIONOFSUBDISTRICTSSELECTEDFORTHESTUDY......................................................................................... 11 FIGURE2.1:NUMBEROFCONFLICTVICTIMSINPROJECTANDCOMPARISONVILLAGES ................................................................. 18 FIGURE2.2:DISTRIBUTIONOFCONFLICTVICTIMSANDBENEFICIARIESWITHINVILLAGES .............................................................. 20 FIGURE3.1:SELECTIONINTOBRAKDPANDEFFECTSONPOVERTY ......................................................................................... 35 FIGURE3.2:MOTORCYCLEHOLDINGSANDSELECTIONINTOBRAKDP..................................................................................... 38 FIGURE3.3:LANDUSE .................................................................................................................................................... 40 FIGURE4.1:CONFLICTVICTIMSACCEPTANCEOFEXCOMBATANTS .......................................................................................... 48 FIGURE4.2:VILLAGEHEADACCEPTANCEOFEXCOMBATANTS ................................................................................................ 48 vi TABLEOFTABLES TABLE1.1:VILLAGEBLOCKGRANTALLOCATIONS ................................................................................................................... 3 TABLE2.1:CORRELATIONBETWEENSUBJECTIVEANDOBJECTIVEMEASURESOFVICTIMHOOD ........................................................ 16 TABLE2.2:CONFLICTVICTIMPRIORITIZATION ...................................................................................................................... 17 TABLE2.3:BRAKDPBENEFICIARIES ................................................................................................................................. 21 TABLE2.4:USEOFFUNDS(MISDATA) .............................................................................................................................. 22 TABLE2.5:BRAKDPGOODS(TREATMENTSAMPLEONLY) ................................................................................................... 24 TABLE2.6:CASHBENEFITSBYCATEGORYOFCONFLICTVICTIM ................................................................................................ 25 TABLE2.7:HOWWEREBENEFITSUSED(TREATMENTSAMPLEONLY)....................................................................................... 26 TABLE2.8:PROJECTSAPPROVEDANDSUPPORTED ........................................................................................................... 27 TABLE2.9:PROJECTSSUPPORTED ...................................................................................................................................... 28 TABLE2.10:BRAKDPAWARENESS&PARTICIPATIONI(PROJECTAREASONLY)........................................................................ 28 TABLE2.11:BRAKDPAWARENESSANDPARTICIPATIONII(PROJECTAREASONLY)................................................................... 29 TABLE2.12:BRAKDPCONDUCT(TREATMENTAREASONLY)................................................................................................. 30 TABLE2.13:AWARENESSOFDEVELOPMENTPROJECT ........................................................................................................... 31 TABLE2.14:PROBLEMSINDEVELOPMENTPROJECTS ............................................................................................................. 32 TABLE2.15:HARMFUL/HELPFUL ....................................................................................................................................... 33 TABLE3.1:AGGREGATEMEASURESOFCOMMUNITYWELLBEING(BYVILLAGEHEADS) ................................................................ 34 TABLE3.2:ASSETINDEX .................................................................................................................................................. 36 TABLE3.3:ASSETSBYCATEGORY ....................................................................................................................................... 37 TABLE3.4:QUALITYOFHOUSING ...................................................................................................................................... 39 TABLE3.5:WATERSOURCE .............................................................................................................................................. 39 TABLE3.6:LANDUSE ...................................................................................................................................................... 40 TABLE3.7:EMPLOYMENT................................................................................................................................................. 41 TABLE3.8:AVERAGEDAILYWAGESOFLABORERS ................................................................................................................ 42 TABLE3.9:SICKNESS ....................................................................................................................................................... 42 TABLE3.10:INSCHOOL(INDIVIDUALS25YEARSOLD) ........................................................................................................ 43 TABLE3.11:COMMUNITYPUBLICGOODS ........................................................................................................................... 44 TABLE3.12:SUBJECTIVEPERCEPTIONSOFWELLBEING .......................................................................................................... 45 TABLE4.1:SOCIALACCEPTANCE ........................................................................................................................................ 47 TABLE4.2:GROUPSTHATBENEFITMORETHANOTHERS ........................................................................................................ 49 TABLE4.3:SOCIALTENSIONS ............................................................................................................................................ 50 TABLE4.4:ESCALATINGTOVIOLENCE ................................................................................................................................. 51 TABLE4.5:CONFLICTRESOLUTION ..................................................................................................................................... 52 TABLE4.6:COLLECTIVEEFFICACY ....................................................................................................................................... 52 TABLE4.7:COMMUNITYLEADERSHIPINPUBLICGOODSPRODUCTION ...................................................................................... 53 TABLE4.8:ASSOCIATIONALLIFE(BYVILLAGEHEADS) ............................................................................................................ 54 TABLE4.9:INVOLVEMENTINASSOCIATIONALLIFE ................................................................................................................ 55 TABLE5.1:SATISFIEDWITHDECISIONS................................................................................................................................ 56 TABLE5.2:VILLAGERS'ROLEINDECISIONMAKING .............................................................................................................. 57 TABLE5.3:POLITICALEFFICACY ......................................................................................................................................... 58 TABLE5.4:TRUSTINDISTRICTGOVERNMENT ...................................................................................................................... 59 TABLE5.5:TRUSTINVILLAGEAPPARATUS ........................................................................................................................... 59 TABLE5.6:CONFIDENCEINEFFECTIVENESSOFSUBDISTRICT,DISTRICT,ANDPROVINCIALGOVERNMENT ........................................ 60 TABLE5.7:AWARENESSOFGOVERNMENT .......................................................................................................................... 61 TABLE5.8:SUPPORTFORDEMOCRACY ............................................................................................................................... 61 TABLE6.1:SHAREOFEXCOMBATANTSANDCIVILIANSRECEIVINGBENEFITS(PROJECTAREASONLY) ............................................. 64 vii ACKNOWLEDGEMENTS Theworkonthispaperwasfundedfrommultiplesources.AgrantfromtheUKDepartmentfor InternationalDevelopment(DFID)paidforthemainsurveysused.Analysiswassupportedbya grantfromtheWorldBank'sPostConflictFundandfromtheEmbassyoftheNetherlands.The Swedish government, through the Folke Bernadotte Academy, provided generous financial support, which made possible the participation of Humphreys, Paler and Weinstein in this research. Many other individuals and institutions have provided support. Makiko Watanabe, Adrian Morel and Yuhki Tajima helped develop the evaluation framework, the surveys and ensured theireffectiveimplementation.NeilsenIndonesiaoversawthesurveys'implementation.Cecilia Mohelpedinthedevelopmentandpilottestingoftheinstruments.MuslahuddinDaud,Susan Wong and Rob Wrobel provided critical feedback throughout. Scott Guggenheim provided useful comments on a draft of this paper. Pak Rusli (and his team from KDP) played an important role in supporting the evaluation. Key staff from the Ministry of Home Affairs, Bappenas and the KDP's National Management Committee were also supportive. Wawan HerwandiandRajyasriGayatrihelpedwithresearchandlogisticalassistance. TheauthorswouldalsoliketothankBRA'sleadership,bothPakIslahuddin(theagency'sformer head)whowasinstrumentalinimplementingtheprogram,andcurrentdirectorPakNurDjuli. viii 1. INTRODUCTION This paper describes the impact of Aceh's CommunityBased Assistance for Reintegration of Conflict Victims program (BRAKDP). The program, implemented between August 2006 and August 2007, sought to address the needs of conflict victims in a way that would not only generate improvements in welfare, but also increase community cohesion and strengthen relations between citizens and government. This evaluation assesses whether BRAKDP was successfulinachievingtheseobjectives. 1.1 ApproachestoReintegration As international donors accumulate lessons from nearly two decades of work in postconflict settings,thereisastrongconsensusthateffortstodisarm,demobilize,andreintegrateformer fighters are essential to the creation of a stable peace. But an initial, exclusive focus on the needsoffightersinpostconflictprogramshasgraduallygivenwaytoabroadervisionofthe reintegrationprocess--onethatemphasizestheneedsofconflictvictimsaswell. To better reflect the needs and interests of those who must accept fighters back into their communities and rebuild their own lives, governments and donors have begun to support communitybased approaches to facilitate reintegration as a complement to traditional DDR programs. 1 These communitybased processes are designed to engage perpetrators and victims alike in the process of community rebuilding in the hope that such efforts aid in the consolidationofpeaceandthepromotionofsocialcohesion.Suchprogramscanbeeffectivein reconstructing conflictaffected infrastructure, while strengthening the wartorn social fabric (Cliffe, Guggenheim and Kostner 2003). The conception and design of BRAKDP reflected this newemphasisoncommunitybasedapproachestoreintegration,withaparticularfocusonthe welfareofcivilianconflictvictims. Yet,despitethenewconsensusontheimportanceofcommunitybasedprocesses,andaswith programstargetedatformercombatants,therehavebeenfewattemptstoassessempirically theefficacyoftheseprogramsortotestthehypothesesimplicitinthisapproach(Humphreys andWeinstein2007;Muggah2008).2Studieshavetendedtotaketheformoflessonslearned fromindividualcasestudies.Withoutvariationonthekeyindependentvariable(i.e.,whether there was a program), analysts have been poorly positioned to say much about programs' causalimpacts.Hereweseektoundertakeasystematicinvestigationofjustthis,toexamine theimpactsofBRAKDPandprovideadeeperanalysisofwhatworked,forwhom,andwhy. 1 SeeSwedishMinistryofForeignAffairs(2006).TherecentCartagenaCongress,hostedbytheGovernmentof Colombia,broughttogethermorethan1,500academics,practitionersandpolicymakerstodiscusshow communitybasedapproachestoreintegrationcanbestbedesignedandimplemented. 2 MansuriandRao(2004)arguethatevidenceontheimpactsofcommunitydrivendevelopment(orCDD)projects lagsbehindtheextenttowhichtheyarebeingimplemented. 1 1.2 CommunityBasedReintegrationAssistanceforConflictVictimsinAceh Foralmost30years,separatistrebelswiththeFreeAcehMovement(GAM)andGovernmentof Indonesia(GoI)securityforcesengagedinamilitaryconfrontationinAceh.Whiletheconflict occurred in several stages during that period, civilians frequently suffered the brunt of hostilities. GoI forces adopted a strategy of trying to undercut popular support for GAM by terrorizingsuspectedciviliansupporters.GAM,too,killedorintimidatedsomewhorefusedto support the movement. This resulted in an unconfirmed number of instances of murder, torture,rape,displacementandpropertydestruction,from1990onwards.3 TheHelsinkipeaceagreementsignedbyGAMforcesandtheIndonesiangovernmentinAugust 2005 contained provisions for reintegration assistance to former combatants, pardoned political prisoners and conflict victims. The central government provided substantial sums to support this. A government body, the Aceh Peace Reintegration Agency (Badan Reintegrasi DamaiAcehorBRA)wasestablishedtoadministergovernmentfundsandtocoordinatewith donors providing assistance. After an initial failed attempt to manage individuallysubmitted proposals from victims, BRA opted for a communitybased approach to supporting conflict victims. BRA channeled its funds through the existing Government of Indonesia (and World Bank supported), Kecamatan Development Program (KDP), which had been operating across theprovincesince1998--producingBRAKDP.4AswithregularKDP,theprogramhadan`open menu'withonlyaverylimitednumberoftypesofactivitiesineligibleforfunding.Communities could choose to use their funds on private or public goods, or a mix. The KDP program was adaptedtoprovidefundsdirectlytovillages(KDPnormallyprovidesgrantstosubdistricts,with villages competing over funds) and facilitators were tasked with helping communities to identifyconflictvictimswhocouldbenefit.BRAKDPallocatedapproximatelyUS$21.7million to1,724villages,aroundonethirdofthetotalinAceh,in67subdistrictsinall17ruraldistricts inAceh,atargetareawithapopulationof1.1million. Accordingtoprogramdocumentation: TheprimaryfocusofBRAKDPistoassistconflictaffectedcommunitiesinimprovingtheirliving conditionsthroughprovisionofsmallprojectsthataccordwiththeirneeds.Italsoencourages peopletoovercomemistrustofgovernmentthatisaresultoftheconflictbydeliveringtangible outputsthatfitwithcommunities'priorities.Equallyimportantistheprocessbywhichvillagers identify, prioritize and implement their projects. A World Bank study on the efficacy of KDP suggests that by applying the principles of participation, transparency, local choice and accountability, communitydriven development programs help improve intergroup and state societyrelations,helpingareastobemoreimmunetoviolentconflicts.BRAKDPisanattempt to apply the communitydriven development approach for reintegration with the hope that it will improve relations between different groups such as exGAM combatants, IDPs [internally 3 SeeIOMandHarvardUniversity(2007),whichshowslevelsofabuseandtraumainAcehofasimilarscaleto AfghanistanandBosnia. 4 In2007,theGovernmentchangedthenameofKDPtotheNationalProgramforCommunityEmpowermentin RuralAreas(PNPMRural)andadaptedittobecomeitsflagshipcommunitypovertyreductionprogram.Funding forPNPMRuralisnowUS$1.2billion.Asof2009,PNPMRuralcoversall6,408ruralsubdistrictsinIndonesia. 2 displaced persons], antiseparatist front members and communities, and to improve relations between these groups and other villagers and the local state (Terms of Reference, BRAKDP EvaluationandGAMLivelihoodsStudy). Targetingwastobeachievedinthreeways.Firstsubdistrictswithhigherlevelsofconflictwere prioritized(asecondcriterionwasalsoused:agoodperformancerecordinpriorKDPgrants). Second,thesizeofthegrantallocatedtoavillagedependedonestimatesofsubdistrictconflict exposureandvillagepopulationsize(seeTable1.1).Finally,itwashopedthatintheirdecisions regarding how to deploy funds villagers would take special account of the needs of conflict victims. It was originally envisioned that a second round of BRAKDP would follow, providing assistance to remaining villages, but BRA decided to revert to their initial program of directly targetingindividuals.5 TABLE1.1:VILLAGEBLOCKGRANTALLOCATIONS Population Large Medium Small (>=700persons) (300699persons) (<299persons) Confl High 170,000,000 150,000,000 120,000,000 6 ict (US$19,000) (US$17,000) (US$13,000) Inten Medium 120,000,000 100,000,000 80,000,000 sity (US$13,000) (US$11,000) (US$8,900) Low 80,000,000 70,000,000 60,000,000 (US$8,900) (US$7,800) (US$6,700) Morethansimplyofferingfunds,participationintheBRAKDPprogramrequiredvillagestohold aseriesofcommunitymeetingsinwhichvillagersthemselvesdecidedhowthemoneyshould be allocated.7Communities were required to hold at least four meetings, run by trained KDP facilitators.Thefirstmeetingoccurredatthesubdistrictlevelandwasprimarilyinformational, emphasizingthatconflictvictimswereintendedtobetheprimaryrecipientsandoutliningbasic program procedures. Subsequent meetings took place at the village level. In the second meeting,villagersestablishedcriteriaforidentifyingconflictaffectedvillagersandtodetermine who in that village should be considered a conflict victim. Proposals were put forward and voted on by the community at the third meeting. Any villager could submit a proposal for consideration. Proposals had to be for a `project', outlining how funds would be used, by individual households, groups, or (for public goods) the wider community. The fourth `accountability' meeting took place after the funds had been spent, with village officials and facilitators reporting on how the funds had been spent. BRAKDP thus emphasized local ownership,puttingdecisionmakingauthorityandoversightinthehandsofvillagers,including conflictvictims. 5 SeeBarronandBurke(2008)andICG(2007)forbackground. 6 USdollarvaluesarethoseatthetimetheprogramwasimplemented. 7 SeeMorel,WatanabeandWrobel(2009)forafulleroutlineofhowtheprogramworked. 3 1.3 TheSelectionofProjectAreas Everyvillageinselectedsubdistrictsreceivedtheprogram.TheruleemployedbyBRAtoselect subdistricts into the program was complex but still relatively welldefined (see the full description in Box 1). For all rural subdistricts, BRA assigned subdistricts to receive the programonthebasisoftwovariables:conflictintensityandspendingcapacity.Theirgoalwas totreatatargetnumberofthemostconflictaffectedsubdistrictsinadistrict,conditionalona subdistrictsurpassingaspendingcapacitythreshold. When assignment is not done randomly, understanding the selection rule by which communitiesareassignedtoreceiveaprogramisimportantsincedifferenttypesofselection can produce different sorts of bias. This in turn affects how we can interpret differences in outcomesbetweenprojectandcomparisonareas.Selectingthe`neediest'places,forexample, canresultintreatmentareasthatareexanteworseoffthancontrolareas,whichcouldleadto an underestimation of program effects if selection is not accounted for. Selecting the most `capable' places, on the other hand, can result in treatment areas that are ex ante better off thancontrolareas,whichcouldleadtoanoverestimationofprogrameffectsifselectionisnot accounted for. In the case of BRAKDP, one selection variable (subdistrict conflict exposure) reflectsa`need'criterion.Theotherselectionvariable(spendingcapacity)reflectsa`capacity' criterion.Theneteffectofthesetwopotentiallyoffsettingcriteriaonbiasisambiguous. Forthefirstselectionvariable,BRAusedaWorldBankproducedmeasureofconflictintensity, generatedthroughafactoranalysisofnineindicators.Indicatorsincludeddataonthenumber of conflict victims in each of the three years preceding the end of hostilities; the number of reported clashes between GAM and GoI forces; and perceptions of conflict intensity from surveydata.Factoranalysiscreatesacontinuousmeasureofconflictintensity,whichBRAused todividesubdistrictsintolow,mediumandhighconflictintensitygroupssothateachgroup hadapproximatelythesamenumberofsubdistricts.Thisdivisionwasthenusedtoselectthe total number of subdistricts to treat in each district, as well as to rank subdistricts within a districtbytheirlevelofconflictaffectedness.Figure1.1showsthegeographiclocationofsub districtsinAcehbyconflictintensity.Highconflictintensitysubdistrictsclusterprimarilyonthe coast,especiallyneartheoilandgasrichregionsonthenortherncoast,whilethelowintensity regionsareprimarilyinthehighlandareastothesouth. Thesecondassignmentvariable,spendingcapacity,requiredsubdistrictstohavespentatleast 60percentoftheir2005KDPfundsatthetimeoftreatmentassignmentweredeemedeligible toparticipateintheprogram.Thisrenderedineligiblethosesubdistrictsthatwerenotableto handleeffectivelytheinflowoffundsfromBRAKDP. 4 FIGURE1.1:CONFLICTINTENSITYINACEH Source:Conflict&Developmentprogram,WorldBank BRAaimedtousethesetwovariablestoprioritizethemostconflictaffectedsubdistrictsina district(withaminimumofoneperdistrict),conditionalonsubdistrictsmeetingthespending capacity threshold. This was done by using an assignment `rule' that combines the conflict intensityandspendingcapacityvariablesinanonlinearwaytoselectsubdistrictsforBRAKDP. First, BRA selected a target number of subdistricts to treat in each district. That target was determinedbythenumberofhighconflictaffectedsubdistrictsinadistrict.Iftherewereno high conflict affected subdistricts, then the target was set equal to the number of medium conflictaffectedsubdistricts.Iftherewerenohighormediumaffectedsubdistricts,however, the target was set to one low conflictaffected subdistrict. This was in order to meet the requirementthatatleastonesubdistrictbetreatedineachdistrict.Afterthetargetnumber wasselected,subdistrictswererankedineachdistrictbyconflictintensity,fromhightolow. Subdistricts were then selected for treatment up to the target, conditional on their meeting the60percentspendingcriterion.Ifnosubdistrictinadistrictmetthespendingcriterion,then the highest spending capacity subdistrict was selected.8Box 1.1 summarizes the assignment ruleBRAused. 8 ThisonlyoccurredinthedistrictofSimeulue. 5 Box1.1:TheTreatmentAssignmentRule 1. Setatargetnumberofsubdistrictstobetreatedineachdistrictequalthenumberofhigh conflictaffected subdistricts in the district. If there are no high conflict affected, set the target equal the number of medium conflict affected districts; if there are no high or medium,letthetargetequalone. 2. Select the most conflict affected subdistricts in each district up to the target in each district,conditionalonasubdistrictmeetingthe60percentspendingcriterion. 3. If insufficient subdistricts meet the spending criterion, select subdistricts with the best spendingperformance. Using the data originally employed by BRA, we applied this rule to reproduce the treatment assignmentprocess.Thisexercisecorrectlyclassifies207(or92percent)ofall225ruralsub districtseligibleforthisstudy. Theclassificationrateisreassuringlyhigh.Yet,aremainingeightpercentdidnot`comply'with thetreatmentassignmentinthattheyshouldhavereceivedtheprogram(accordingtotherule) anddidnotreceiveit,ortheyshouldnothavereceiveditandtheydid.Inparticular,tensub districtsthatshouldhavebeenassignedtotreatmentdidnotreceiveit,andeightthatshould not have been assigned to treatment did receive it. This noncompliance could result from humanerrorinapplicationoftherule,oritcouldreflectonthespotdecisionsthathadtobe madeduringtherolloutprocess,oritcouldreflectotherelementsoftheassignmentrulethat have not been as clearly codified.9The danger of nonrandom noncompliance is that it can introduce bias in the form of factors that we cannot account for but nonetheless impact the estimationofprogramimpacts.SincewecannotbeconfidentthatnoncomplianceinBRAKDP israndom,wetakestepstoaddressthisissueinouranalysis. 1.4 AnalyticalFrameworkandMethods BRAKDP is hypothesized to impact three families of outcomes: material wellbeing, social cohesion, and trust in government. In this paper, we focus on nine explicit hypotheses about theimpactofBRAKDP.ThesearedetailedinBox1.2. 9 Whileweusethelanguageof`compliance'toatreatment,thereisnoimplicationthatthedecisiontotakepart ornotwasmadebysubdistricts.Indeedthemostlikelyreasonfornoncomplianceisthatfurthercriteriawere usedbyprogramofficerswhenmakingselections.Inparticularfromcorrespondence,weknowthatinsomecases forbudgetaryreasons,someformallynoneligiblesubdistrictsreplacedthosethatmettheselectioncriteriain ordertoensurethatblockgrantsmatchedwiththeoverallprogrambudget.Wedonothoweverhavedetailson thispartoftheassignment. 6 Box1.2:Hypotheses 1Welfare 1.1 SocioeconomicwelfarewillbehigherandwelfareimprovementswillbegreaterinBRA KDPtargetvillagesthaninvillageswithoutBRAKDP. 1.2 In particular, socioeconomic welfare levels and gains among conflict victims will be higherinBRAKDPtargetvillagesthaninvillageswithoutBRAKDP. 2.SocialCohesion 2.1 Social, economic, and political reintegration of exGAM combatants, militias, and IDPs willbegreaterinvillagesinwhichBRAKDPhasestablishedprograms. 2.2 CommunitieswillbelessresentfulofbenefitstargetedatexGAMcombatants,militias, and IDPs in villages in which BRAKDP has worked and more accepting of their participationinthesocial,economic,andpoliticallifeofthevillage. 2.3 Disputes will be less likely to escalate in villages in which BRAKDP has implemented programs. 2.4 Communities will be better able to solve local collective action problems in villages in whichBRAKDPprogramshavebeenestablished. 2.5 AssociationallifewillbemoredevelopedinvillagesinwhichBRAKDPhasoperated. 3.TrustinGovernment 3.1 TrustinthedecisionmakingprocessesofvillagegovernmentswillbehigherinBRAKDP communities. 3.2 Trust in the ability of local government to deliver services/benefits will be greater in villagesinwhichBRAKDPhasbeeninoperation. Source:Programdocuments We use a second best strategy to estimate the causal effects of the programs on these objectives.Anoptimalapproachwouldbetorelyonsomeformofrandomizationinorderto generate comparison (`control') groups that are every way identical to program (`treatment') groups.Becauseoftheneedtoprioritizesubdistrictsintherolloutoftheprogram,BRAKDP did not use a randomized design. This results in treatment areas that are systematically different(moreconflictaffectedandmoreefficientatdisbursingKDP)thancomparisonareas. Nevertheless,theselectionofareasintotheprogramgenerallyfollowedthewelldefinedand transparentprocessjustdescribed.Bysystematicallyaccountingforthisprocess,weareableto produceacontrolgroupassimilartothetreatmentgroupaspossibleandtoestimateprogram effects,eventhoughcomparisonareasexhibitsystematicdifferenceswithcontrolareas. Ourempiricalstrategyhasthreemaincomponents: a) astrategyforidentifyinganappropriatecomparison(control)groupofsubdistrictsex post; b) aprotocolfordatacollection;and c) an estimation strategy to account for systematic differences between the comparison and treatment groups and the fact that not all communities that should have been assignedtotreatmentwereandviceversa. Wedescribethesethreecomponentsnext. 7 Selectingappropriatecomparisonunitsexpost Therewasnosetof`control'unitsidentifiedbeforeprogramimplementation;inaddition,the treatment units were selected according to a deterministic criterion and subdistricts that received the program differ systematically from those that did not. These features make it difficulttoidentifyanappropriatecomparisongroupofsubdistricts.Identifyinganappropriate comparisongroup,however,isessentialifwearetoestimatewhatmighthavehappenedinthe absenceoftheprogramandinthiswaytoascertaintheprogram'simpact. Ourstrategywastoidentifyapoolofsubdistrictsthatweremostsimilartothosethatreceived the program at the time the treatment assignment decision was made. Conceptually, our strategy is similar to determining the `propensity' for assignment to treatment, or the probabilitythatanygivensubdistrictwouldhavereceivedthetreatmentgiventherule.10For example,inBRAKDPhighconflictandhighcapacitysubdistrictsshouldhavehadaveryhigh probability of receiving treatment, whereas low conflict and low capacity subdistricts would not.Wethenselectcomparisonunitsthathavepropensitiessimilartothoseofunitsthatwere infactassignedtotreatmentundertheassignmentrule. Thedifficultyweface,however,isthattheactualassignmentruleproducesbinarypropensities. In other words, under the rule and conditional upon the data, all potential comparison sub districts either were or were not assigned to treatment. Thus on the surface it appears that some subdistricts had 0 percent probability of receiving the program and some had a 100 percentprobabilityofreceivingtheprogram.Toselectcontrolsubdistricts,weneedinsteada continuous measure of propensity. To do this, we draw on the knowledge that, since continuousmeasureswereusedtoassigntreatment,someuntreatedunitsare,infact,`closer' to being treated than others. Our challenge was to translate these notions of proximity into usablenotionsofassignmentprobability. An obvious approach is to generate a propensity score by a logistic or probit regression of actualtreatmentonthetwoassignmentvariables.Whilethisapproachgeneratesacontinuous propensity score, it does not correctly capture the known assignment process. A measure of treatment propensity should model not only the assignment variables but also the known selectionruleinascertainingtheproximityofcontrolstobetreated. Instead we use an approach in which we generate a continuous measure of treatment propensity under the assumption that a deterministic rule is applied to data that is fundamentally`noisy'(i.e.theassignmentvariables--likemostmeasures--havesomerandom error in them). The advantage of this approach is that it allows us to make full use of the assignmentruleanditprovidesanaturalwaytoincorporatecomplexdependenciesbetween unitsintheassignmentprocess(forexampleifvaluesforonesubdistrictdeterminethetarget numberofsubdistrictsinadistrict). 10 RosenbaumandRubin(1983,1984)haveshownthat,ifassignmentismadeonthebasisofobservablevariables andallassignmentvariablesareaccountedfor,selectingtreatmentandcontrolgroupswithsimilarpropensity scorescreatescomparablegroupsforcausalinference.SeealsoMorganandWinship(2007:99). 8 FIGURE1.2:PROPENSITYSCORESFORTREATMENTANDCOMPARISONUNITS Source:Authors'calculations In practice, we simultaneously apply a small independent shock (distributed with mean zero andvarianceequaltohalfthestandarddeviationofthatvariable)toboththeconflictintensity and spending capacity assignment variables in every subdistrict. This slightly changed values forthesevariablesineachsubdistrict.Thenweappliedthetreatmentassignmentruletothis `perturbeddata',whichinturnchangedthesetofsubdistrictsselectedintotreatment.Wedid this10,000timesandtooktheaveragenumberoftimesasubdistrictwasselectedintoBRA KDPof10,000,producingacontinuousmeasureofpropensity.11Inthisway,foreachunit,we estimate the probability that it would have been selected were the underlying data slightly differentfromtherecordeddata. Figure 1.2 shows the actual distribution of treatment and control subdistricts according to continuous propensityscoresgeneratedusingthismethod.Asisclearfromthefigure,there are notable differences between the assignment probabilities for treatment and control sub districts.Asonewouldexpect,thesubdistrictsthatwereactuallytreatedhadhighestimated propensitiesoftreatmentwhilethesubdistrictsthatwerenottreatedhadlowerestimated propensities. Excluded areas had uniformly lower propensities. Nevertheless there was substantial overlap and in particular some nontreated subdistricts had a much higher propensitytobeselectedthanothers,accordingtothemeasure. 11 Toprovideamoreconcreteexample,considerahighconflictaffectedsubdistrictwitha59percentspending capacity.BytheBRA'srule,thatsubdistrictwouldhavebeenineligiblebecauseitdidnotmeetthe60percent spendingcapacity.Nonetheless,thatsubdistrictwas`close'tohavingreceivedtreatment.Saythatafterapplying 10,000smallshockstothatsubdistrict'sscoreonbothvariables,itsconflictintensityscorealmostalwaysremains inthehighcategorybutitsspendingcapacityscorecrossesthe60percentthresholdfortreatment3,500of10,000 times.Thiswouldgiveitapropensityscoreof35percent. 9 These propensities were then used to select 67 (highest propensity) areas to serve as a comparisongrouptothe67treatedsubdistricts.Figure1.3showsthedistributionoftreatment and comparison units over the two main assignment variables, broken down according to whethertheywereselectedintothestudy(rightpanel)ornot(leftpanel)andFigure1.4shows the geographic distribution of treatment and control areas. We see from the figures that control areas are `close' to treated areas both geographically and in terms of the assignment criteria. FIGURE1.3:CHARACTERISTICSOFSUBDISTRICTSSELECTEDINTOSTUDY Not selected for study Selected for study 8 C Conflict Exposure 6 T T T C 4 C C T T T T T T T T T T T T TT T T T T T C C T C T TT C TT T TT TT T T C C T T TT CC C C C C C CT TC 2 C C T T C C T T C C C TT TC C C C CCC C C T T C C C C C C T C CC C C T C T CC C C C C C C C C C C C C CC C C TC T CC C C C C CC C T C C T TC TT C C C C C C CC C C T C C CC T C C C CC T CT C C C CT CC C C C C C C C C C CCC C CC C CC C C C C C C CC CC C T C C C C CC C C C C C T 0 0 50 100 0 50 100 Disbursement record Graphs by select3 Theleftpanelshowsthesetofsubdistrictsnotselectedintothestudyandthevaluesof theseontwooftheassignmentcriteria.Therightpanelshowsthoseselectedinthestudy, markersindicatewhethersubdistrictsweretreated(T)ornot(C). Source:Authors'calculations 10 FIGURE1.4:LOCATIONOFSUBDISTRICTSSELECTEDFORTHESTUDY Source:Conflict&Developmentprogram,WorldBank Datacollection Having identified a sample of 67 BRAKDP and 67 comparison subdistricts, we designed and implemented the Aceh Reintegration Livelihood Surveys (ARLS), a largescale household and villageheadsurveytogathermeasuresthatcouldbeusedtoassesstheimpactoftheprogram. Givenasetofsubdistricts,oneineveryeightvillageswasrandomlyselectedforenumeration. Strata(subdistrictandpopulation)wereusedtoensurebalancetotheextentpossible.Within villages,fivehouseholdswererandomlyselectedtoserveasrespondents.Withinhouseholds, one individual was randomly selected from among all household members aged between 18 and65.Ultimately,thehouseholdsurveywasadministeredto2,315householdsintotal,1,090 of which resided in areas that received BRAKDP. Households were sampled in 67 treatment subdistrictsand68controlsubdistricts,covering17ruraldistrictsand461villagesoverall. The survey instrument asked a set of questions designed to measure outcomes including household and community welfare; individual level behavior and attitudes; communitywide collectiveaction;andperspectivesontheBRAKDPprogramitself.Inaddition,itincludedone behavioralmeasuredesignedtocapturetrustinlocalgovernment.12 A parallel survey of village heads was conducted focusing on communitylevel outcomes includingmaterialwellbeingandcollectiveaction.TheARLSalsosurveyedarandomsampleof excombatants,whichprovidessomedatausedinthisstudy. 12 Thesurveyinstrumentsareavailableatwww.conflictanddevelopment.org 11 Estimationstrategy Our selection of subdistricts into the study helped to ensure that we collected data on the most `similar' comparison subdistricts as well as the treated subdistricts. However simple differencesbetweenoutcomesinthesetwogroupsshouldstillnotbetakenasestimatesofthe causal impact of the BRAKDP program. The reason, as is clear from our earlier discussion, is that even though we selected most similar comparison areas, there are still systematic differences between the treatment areas and the comparison areas that are not due to the impactsoftheprogram,butratherreflecttheassignmentruleoftheprogram. We use two strategies to account for these further differences. As noted above, the vast majority of subdistricts `complied' with their assignment, meaning that those who were selectedtoreceiveBRAKDPdidindeedreceiveit,andthosewhowerenotselectedtoreceive itdidnot.Forthosecasesinwhichthereisthiskindofcompliancetotherule,weareinthe fortunate position in which we know the exact variables that were used for assignment into treatmentandcantakeaccountofthesevariablesinouranalysis.Asdescribedabovethefull assignment rule is complex, but the core substantive criteria make use of only two variables: spendingcapacityandexposuretoconflict.13 Under the twin assumptions that (a) there is a constant treatment effect and (b) treatment assignmentis`unconfounded'conditionalonspendingcapacityandexposuretoconflict(i.e.we have accounted for all variables not related to program impacts that might affect measured outcomes),weestimatetreatmenteffectsbyregressingoutcomesontreatmentalongsidefirst, secondandthirdorderpolynomialsofthesevariablesaswellastheirinteraction.Inessence, thisapproachseekstoaccountforthosefactorsthatdeterminedassignmenttoBRAKDPand thenestimatethecausaleffectofBRAKDPitself,independentoftheassignmentprocess.Both assumptions(a)and(b)arenecessaryforthevalidityofourestimates. There is, however, one further complication. As discussed above, noncompliance presents complications for the analysis by introducing the possibility of unobserved factors that bias outcomes. To account for this, we employ an approach in which we instrument actual treatment (whether a subdistrict participated in BRAKDP) with assignment to treatment (whetherasubdistrictwasassignedtoparticipateinBRAKDP).Inessence,thisenablesusto estimate the effect of BRAKDP for only those areas that `complied' with their treatment assignment and takes into account the potential bias caused by having noncompliers in the study.14Theresultingestimateofthetreatmenteffectis,webelieve,amorereliableestimate ofthecausaleffectoftheprogramthanisthesimpledifferenceinmeans. 13 Indeed,conditionalonoursamplewecancorrectlyclassify87percentofsubdistrictsintoassignment categoriesusinginformationondisbursementandconflictintensityonly;inparticular,byemployingamodelin whichthesetermsraisedtothefirst,secondandthirdpowersareenteredalongwiththeirinteraction. 14 Inaregressiondiscontinuityframeworkthisinstrumentalvariablesapproachcorrespondstoa`fuzzyRD' approach,andinanaveragetreatmenteffectsframeworkitproducesthe`localaveragetreatmenteffect'. 12 In addition to this core strategy we undertake a series of checks of robustness to model specification and report when different approaches yield substantially different outcomes.15 Weillustrateanumberoftheresultsbyshowingtheestimatedintentiontotreateffectsthat result from employing a `regression discontinuity' model in which we create a single running variablethatdeterminesassignmenttotreatmentandestimatetheeffectoftreatmentatthe cutoffpointonthisvariable.Theapproachusedforthesefiguresisdescribedingreaterdetail below. Inthetablesthatfollowthatcomparetreatmentandcontrolgroups,webeginineachcaseby reporting results for the control group (column 1) and the treatment group (column 2). The third column presents the simple difference in means in outcomes across these groups. This difference in means has a very transparent interpretation and reflects well the differences in thesituationsinthedifferentareas.Butitdoesnotnecessarilyprovideagoodestimateofthe causal impacts of the program. The fourth and final column presents the estimated causal effectsaccountingforalltheselectionandnoncomplianceissuesdiscussedabove.Finally,all results reported here take account of the characteristics of our sample (sampling weights, strataandclusters)andclusterstandarderrorsatthesubdistrictlevel,thelevelatwhichthe BRAKDPtreatmentwasassigned. 15 Wefocusonwhatwethinkisthebestestimationstrategygiventhedatastructureexaminedhere.Indifferent robustnesscheckswe:(a)conditiononourestimatedpropensityscores;(b)implementaregressiondiscontinuity designbyconditioningonasingle"runningvariable"thatcombinesthetwokeyvariables;and(c)examinedirect insteadofinstrumentedeffects.Otherapproachestoestimatingcausaleffectsarepossiblehowever.Twostand out.Inone,wecoulduseourestimateofthepropensitytoreceivetreatmentasaconditioningvariableto establish`unconfoundedness'orasbasistocreatematchestoestimateaveragetreatmenteffects.Second,we couldapproximateamorestandardregressiondiscontinuitymodelbyseeking(endogenous)breakpointswithin eachdistrict,andseekingtoestimatethetreatmenteffectatthe(twodimensional)cutoffbetweentreatment andcontrolunits;again,inthiscase,wewouldneedtoinstrumenttoaccountfornoncompliance. 13 2. IMPLEMENTINGBRAKDP 2.1 WhoReceivedAssistancethroughBRAKDP? A key aim of the BRAKDP program was to ensure that BRAKDP funds, while benefiting communitiesingeneral,wereespeciallyeffectiveatreachingconflictvictimsinparticular.The difficultyofachievingthisgoalderivesfromatleastthreechallenges:(a)thelackofconsensus overa clear definition of what constitutes a conflict victim; (b) uncertainty in knowing where conflictvictimsarelocatedandthepracticaldifficultyoftargetingindividualsusingaprogram that, by its nature, works at a higher level of aggregation (BRAKDP is implemented at the villagelevel,buttheprogramisassignedatthesubdistrictlevel,withallvillagesinachosen subdistrictreceivingit);and(c)thefactthattheultimatedecisionsabouthowtotargetfunds restedwithvillagerswhohadfreedomtochoosehowandtowhomfundswouldbeallocated. Weconsidereachofthesethreeaspectsinturn. Whoisaconflictvictim? Havingacleardefinitionofconflictvictimsisrendereddifficultbecauseofthemanydifferent waysanddegreesinwhichindividualscanbeaffectedbyconflict.Theconflictimpactedalmost everyone in Aceh, directly or indirectly. By some accounts everyone was a conflict victim. Moreover,individualperceptionsoftenmatterasmuchaswhatexperiencespeoplemayhave endured. The program opted for subjective assessments of victim status; while guideline categorieswereprovided,itwaslefttocommunitiestodefinewhoisandisnotaconflictvictim. Forthisreason,webaseourcoreanalysesonameasurethatreflectssubjectiveperceptionsof victimstatus.Inaddition,however,wegenerateanobjectivemeasurebasedonselfreported exposure to conflict. In practice these two measures are very closely related, as described below.16 The subjective measure of conflict victim status simply captures whether an individual respondedaffirmativelywhenaskedinthesurveyiftheyconsiderthemselvesaconflictvictim. In order to associate this selfreported status with actual experience of conflict, we also gathereddataonwhyindividuals considerthemselvesvictims.Victimscouldprovidemultiple reasons for victimhood, ranging from death of a family member to internal displacement to mental illness. Using this data we generate a finer category which aims to get at those who weremostaffectedbyconflictamongthosewhodeclarethemselvestobeconflictvictims.We codeanindividualas`mostseverelyconflictaffected'ifeither(a)afamilymemberwaskilledor disappeared,(b)theywerephysicallyinjured,(c)theirhousewasdestroyed,or(d)theywere 16 Thereareanumberofargumentsthatcanbemadeinfavorofusingoneorothermeasure.Thesubjective measurehastheadvantageofcorrespondingmorecloselywiththeprojectapproach;inadditionitcaptures featuressuchasselfreportedmentalillnessesforwhichwehavenoobjectivemeasures.Apotentiallyimportant shortcomingofthesubjectivemeasuresisthatsinceBRAKDPareasweresocializedonwhataconflictvictimis,it ispossiblethatindividualslivinginBRAKDPareasweremorelikelytothinkofthemselvesasvictimsthaninnon BRAKDPareas;thatisthesubjectivemeasureprovidesaposttreatmentmeasureofstratum.However,thefact thatthesetwomeasuresarelargelyconsistentwitheachotherisreassuringonthispoint. 14 displaced by the conflict.17Note that by construction, this `most severely affected' category, thoughtiedto`facts',isnestedwithinthesubjectivecategory. To complement this measure, we also constructed an objective measure that classifies an individualasavictimbasedontheirreportedexposuretoconflict,regardlessofwhetherthey selfidentifyasavictim.Thismeasureexplicitlytakesaccountoftheconflictexperienceofthe individual'shouseholdandnotsimplyoftheirownexperience,usingdataonfamilymember related deaths, disappearances and injuries. We also collect data on homes and workplaces destroyedduetoconflict.Thesubcomponentsofthesubjectiveandobjectivemeasuresrelate broadly to each other although on two items--family member kidnapped and missing body part/physical disability--the two measures record somewhat different ideas 18 and for one item--mentalillness--wehaveasubjectivebutnoobjectivemeasure. Table2.1belowpresentsthemaincategoriesofvictimhood,thelevelatwhicheachvariableis measured, and the correlation between objective and subjective measures of victimhood. Overall,thecorrelationsarestrikinglyhigh,especiallyforthecriteriaformostconflictaffected. Theonlytwoitemsforwhichtherearelowcorrelations(familymemberkidnappedandmissing bodypart/physicaldisability)arethosetwoforwhichthewordingsofthemeasuresdiffer. These high correlations are reassuring and suggest that findings presented here are likely robust to the choice of measure employed. While the main emphasis is on the subjective measurehereandthroughoutthepaper,inthenextsectionwealsoreportboththeobjective measure and a combined measure. The combined measure captures whether either the subjectiveorobjectivecriteriaapply,andcanbeinterpretedastheupperboundonvictimhood. 17 ThisdefinitionofmostconflictaffectedfollowssuggestionsofWorldBankstaff. 18 Formissingbodyparts,thesubjectivemeasureasksrespondentsiftheyhave"missingbodypartsorpermanent physicaldisabilityduetoconflict"whereastheobjectivemeasureasksiftheywere"injuredormaimed(resulting inhospitalizationorinabilitytofunctionnormallyforatleastonemonth)asaresultoftheconflict."Thesubjective measureforkidnappedasksindividualsifa"familymemberdisappeared/waskidnapped/ordetaineddueto conflict".Theobjectivemeasureasks(onlyformembersofthe1998household)whyanindividualinthemain respondent's1998householdisnolongerinthe2008household.Householdmemberswho"disappeared/were takenawaybecauseoftheconflict"werecodedfortheobjectivecriteria. 15 TABLE2.1:CORRELATIONBETWEENSUBJECTIVEANDOBJECTIVEMEASURESOFVICTIMHOOD Levelatwhichsubjective Levelatwhich Correlation itemdefined objectiveitemis between defined measures Nonvictims .67 Conflictvictims(includingmostaffected) .67 Mostconflictaffected .86 Familymemberkilled Household Household .92 Familymemberkidnapped/detained Household Household .12 Missingbodypart/physicaldisability Individual Individual .23 Housedamaged/destroyed Household Household .78 Primarylivelihooddamaged/destroyed Household Household .61 Internallydisplaced Individual Household .93 Mentalillness(selforfamilymember) Individualorhousehold Physicalillness(selforfamilymember) Individualorhousehold Household .29 Source:ARLS AreconflictvictimsconcentratedintheareasthatreceivedBRAKDP? Table 2.2 explores how well BRAKDP reached conflict victims. Focusing on the subjective measure,weseethat42percentofallindividualsinthestudypopulationareconflictvictims; of these 28 percent meet the criteria for mostconflictaffected. (If the combined measure is employed, 45 percent are victims and 30 percent of the population are `mostaffected'). We estimatethatbetween617,000and657,000peopleinareasthatreceivedBRAKDParevictims. The greatest share of conflict victims were internallydisplaced, followed by those suffering frommentalillnesses. 16 TABLE2.2:CONFLICTVICTIMPRIORITIZATION Panel I II III IV Shareofstudy ShareofpopulationinShareof...thatareinEstimatednumber populationthatare... treatmentareasthat treatmentareas(%) of...intreatment (%) are...(%) areas('000) Sub Ob Comb Sub Ob Comb Sub Ob Comb Sub Ob Comb Nonvictims 58 68 55 51 62 48 41 42 40 643 780 603 Conflictvictims 42 32 45 49 38 52 54 55 54 617 480 657 Mostconflictaffected 28 28 30 34 33 37 57 56 56 429 420 462 Shareamongconflict Shareamongconflict Ofall...thatarein Estimatednumber victimsinstudy victimsintreatment treatmentareas(%) ofthosewith...in populationwith...(%) areaswith...(%) treatmentareas ('000) Sub Ob Comb Sub Ob Comb Sub Ob Comb Sub Ob Comb Familymemberkilled 4 7 5 5 8 6 70 66 66 30 38 38 Familymemberdetained 6 1 6 7 1 7 63 56 62 43 5 48 Physicaldisability 4 8 8 5 10 9 68 62 63 28 46 61 Housedamaged/destroyed 19 33 24 19 34 26 55 56 56 117 161 168 Primarylivelihooddamaged 19 18 21 18 20 22 50 61 56 110 95 144 Internallydisplaced 50 65 49 54 68 52 58 56 56 335 337 346 Mentalillness(selforfamily) 32 32 28 28 47 47 175 175 Physicalillness(selforfamily) 19 21 28 17 22 27 45 56 52 102 105 177 Tablereportsestimatedpopulationmeans. Source:ARLS Withintheareasthatreceivedtheprogram,49percentofthepopulationconsiderthemselves conflictvictims;incontrolareas,36percentdoso(notshown).Forthemostconflictaffected, 22 percent meet the criterion in comparison areas, compared to 34 percent in areas that receivedtheprogram.Thesedifferencesarebothsignificantatthe99percentlevel. DifferencesinthedistributionofconflictvictimsatthevillagelevelareillustratedinFigure2.1; thefigureshowsthedistributionoftheshareofthefiverespondentsineachvillagethatwere conflictvictims.Whilethedistributionisskewedtotheleftforcomparisonareasweseethatit is relatively uniform in project areas; there are almost as many villages for example with no victims reporting as there are with all five reporting. The differences between these two distributions is significant: There were approximately twice as many villages in comparison areasthathadzerooffiverespondentsreportingasconflictvictimsandabouttwiceasmany villagesinprojectareaswithallfiverespondingpositivelycomparedtocomparisonareas. 17 FIGURE2.1:NUMBEROFCONFLICTVICTIMSINPROJECTANDCOMPARISON VILLAGES BRA-KDP control group BRA-KDP treatment group 60 Number of villages 40 20 0 0 1 2 3 4 5 0 1 2 3 4 5 Number of conflict victim respondents per village (maximum=5) Note:Figureshowsthedistributionoftheshareofthe5respondentsineachvillagethatwere conflictvictims.Forexample,0of5respondentsclassifiedasconflictvictimsinaboutoneinfour comparisonareavillages,butsuchlowreportswereonlyseeninoneineightprojectarea villages. Source:ARLS Nevertheless,itisclearaswellthattheoverallabilityoftheprogramtoreachconflictvictims was limited, in part because the program was not continued into its planned second round. Althoughconflictcriteriawereusedtoselectprogramareas,amongconflictvictims,onlyabare majority,54percent(seeTable2.2,PanelIII)ofallthoseinthestudyarea,livedinsubdistricts that received BRAKDP; the corresponding numbers for nonconflict victims and the most conflictaffectedare41percentand57percent,respectively.Thesenumbersaremuchhigher fordeathsandphysicaldisability(70percentand68percentoftheseareintreatmentareas) although for some categories (mental and physical illness) the estimated shares are actually higherinthegeneralpopulation. Thissuggestsrealdifficultiesintargetingconflictaffectedindividualsthroughprogramsthatare administered atmoreaggregatelevels(i.e.subdistrict). Thedifficulty reflectsseveralfactors. The conflict in the Aceh was widespread, which will make any effort of this form imperfect. Furthermore, the correlation between the share of victims and the measure of conflict exposureavailabletoBRAinassigningtheprogramisstatisticallystrongbutsubstantivelyweak (0.36).Inotherwords,someareaswithmanyconflictvictimsscoredquitelowonthemeasure ofconflictintensity(examplesincludeBendaharaandSyiahUtama)Italsoreflectsthetradeoff inherentinassigningtreatmentonthebasisofspendingcapacity:someareasthatscoredhigh on conflict measures were not treated because they fell short on the disbursement criterion (PeureulakTimurisonesuchcase).Ofcourse,thesewouldnothavebeenissuesiftheprogram hadbeenextendedintoallremainingsubdistrictsinitssecondyear,aswasoriginallyplanned. 18 Finally, we note that the broad patterns we describe above hold for both the objective and subjectivemeasuresofconflictaffectedness. Doconflictvictimsbenefitmorethanothersintreatmentareas? Conflict victims can benefit from BRAKDP through both private transfers and public investments. As we will see in the next section, the large majority of BRAKDP funding went towards direct private transfers (for example, to support small businesses); moreover, the scope for targeting victims is likely to be greater for private transfers compared to public investments.Inthissection,wefocusonthesetransfersandexaminetheextenttowhichthese weresuccessfullytargetedatconflictvictims. The ARLS data asks respondents whether they benefited directly from BRAKDP.19As seen in Table2.3,thesetransfersreachedwidesegmentsofthepopulation:about530,000individuals live in households that benefited from these transfers, of whom approximately 308,000 are adultsaged1865(confidenceintervalfortheadultsbenefitingdirectlyis230,000­386,000). Of total recipients, approximately 273,000 were conflict victims by the subjective measure or 287,000usingthecombinedmeasure. Since BRAKDP aimed to direct assistance to conflict victims through a community decision making process, it is worth looking at how conflict victims fared as recipients visàvis non victimsinBRAKDPareas.Wefindthatanestimated44percentofallconflictvictimsinBRA KDPareasreceivedgoodsfromtheproject,comparedto40percentofnonvictims(usingthe subjective measure). While this suggests that conflict victims did marginally better than non victims,thisfourpercentagepointdifferenceisnotsignificantatconventionallevels.Inother words,withinBRAKDPareas,selfreportedvictimsandnonvictimsappearaboutequallylikely tobenefitfromBRAKDP. Asimilarresultobtainsifwelookatdistributionswithinvillagesinsteadofwithinthetreatment areas grouped together: within villages conflict victims are not, on average, more likely to receivesupportthannonvictims.Figure2.2illustratesthepointbyshowingthattheshareof beneficiaries that are conflict victims is on average the same as the share of all respondents thatareconflictvictims. Together,theanalysissuggeststhatwhileconflictvictimsweremorelikelythannonvictimsto benefit overall, this is largely because there were more conflict victims in BRAKDP areas. Overall, about 24 percent of conflict victims in the study areas (both treatment and control) received direct benefits whereas only 16 percent of nonconflict victim households reported receivingthissupport.Thisdifferenceislargeandstatisticallysignificant.Itisdriven,however, almostentirelybytheselectionofsubdistrictsintotheprogramandnotbytheallocationof fundswithinthevillages. 19 Itasks,"DidyouoryourhouseholddirectlyreceiveanymoneyorgoodsfromBRAKDP?" 19 Share of Beneficiaries that are Conflict Victims FIGURE2.2:DISTRIBUTIONOFCONFLICTVICTIMSANDBENEFICIARIESWITHINVILLAGES 1 .8 .6 .4 .2 0 0 .2 .4 .6 .8 1 Share of respondents that are Conflict Victims Note:Figureshowstheshareofdirectbeneficiariesthatareconflictvictimswithinavillagetotheshare (of5)respondentswithinavillagethatareconflictvictims.Dataisconditionalupontherebeingatleast onereportedbeneficiary.Sizesofcirclesrepresentthenumberofvillagesthattakeagivenvalue. Source:ARLS 20 TABLE2.3:BRAKDPBENEFICIARIES Shareofall... Shareofall... Shareof Estimatednumber Estimated Estimated thatreceived intreatment beneficiariesin ofindividualsthat numberofadults numberof benefits(%) areasthat treatment are...in thatare...in householdsthat received areasthat householdsthat householdsthat receiveddirect benefits(%) are...(%) receiveddirect receiveddirect benefit benefits benefits ('000) ('000) ('000) Su Sub Ob Comb b Ob Comb Sub Ob Comb Sub Ob Comb Sub Ob Comb Sub Ob Comb Nonvictims 16 18 16 40 43 40 48 63 46 257 336 243 152 197 144 46 60 43 Conflictvictims 24 22 24 44 40 44 52 37 54 273 194 287 155 111 164 50 36 53 Mostconflictaffected 24 23 24 41 41 42 34 33 37 178 174 194 101 100 111 32 32 36 Shareofall Shareofall Shareofvictim Estimatednumber Estimated Estimated victimswith... victimswith... beneficiariesin ofindividualsin numberofadults number thatreceived intreatment treatment householdswith... inhouseholds households benefits(%) areasthat areaswith...() whoreceived with...who with...who received directbenefits receiveddirect receiveddirect benefits(%) ('000) benefits benefits ('000) ('000) Su Sub Ob Comb b Ob Comb Sub Ob Comb Sub Ob Comb Sub Ob Comb Sub Ob Comb Familymemberkilled 22 29 26 32 41 39 4 5 5 10 15 15 6 9 9 2 3 3 Familymemberdetained 28 22 30 45 47 48 7 1 8 20 4 24 12 2 14 3 1 4 Physicaldisability 13 27 23 20 43 36 2 7 8 6 20 22 3 10 12 1 4 4 House damaged/destroyed 25 25 24 46 42 42 20 23 24 54 66 70 30 39 41 9 12 12 Primarylivelihood damaged 21 24 25 40 38 43 17 11 22 44 37 62 23 20 32 8 7 11 Internallydisplaced 23 23 23 39 40 40 48 67 47 130 135 139 72 76 78 24 25 26 Mentalillness(selfor family) 19 19 40 25 25 70 70 41 41 14 14 Physicalillness(selfor family) 25 22 22 55 38 38 20 13 13 56 41 84 31 23 46 10 9 16 Question:DidyouoryourhouseholdreceiveanymoneyorgoodsfromBRAKDP?Tablereportsestimatedpopulationmeans.Source:ARLS Thedataalsoprovidesinformationonhowwellcertaintypesofconflictvictimsdidcompared to nonvictims. For instance, only 3239 percent of victims in program areas with family members killed and 2036 percent with physical injuries received goods from BRAKDP. This couldbeduetothefactthatthesegroupsweresomehowdisadvantagedinapplyingforBRA KDP funds, or because they were already viewed as having benefited from other conflict assistance,amongotherreasons.Incontrast,4548percentofthoseinprogramareaswhohad afamilymemberdetailed,4246percentofthosewhosehousewasdamaged,andaround40 percentofthosewhohadbeendisplacedreceivedsupportthroughtheprogram. Conclusionsontargeting Overall,thisanalysissupportstwoconclusions.First,largenumbersofdirectbeneficiarieswere reached,includingbothconflictvictimsandothers.Theinterruptionoftheprogrammeantthat many victims (in comparison areas) were not reached; nevertheless the geographic 21 prioritization procedure used helped to offset the effect of the interruption, with conflict victimsmorelikelythanotherstohaveaccesstotheprogram.AcrossAceh,alargerproportion of conflict victims than nonvictims received support. Second, a large proportion, but by no meansall,oftheconflictvictimswithintreatmentareaswerereached.Withinvillagetargeting wasrelativelyblunt,however,inthatconflictvictimstendednottodomuchbetterthannon victims.Thisresultshouldhoweverbeconsideredinacontextinwhichtheseindividualsmight ordinarilyhavedoneworse,withconflictvictimspotentiallymoremarginalizedthanothersin theircommunities,andhencepotentiallylesslikelytobeabletoaccessfundsallocatedthrough acollectivedecisionmakingprocess. 2.2 HowWereBRAKDPFundsSpent? Under BRAKDP, individual villages had the freedom to decide how best to spend funds from BRAKDP.Wecanassesswhatdecisionsweremadeusingbothofficialprojectmonitoringdata (MISdata)20anddatacollectedthroughtheARLS. Table 2.4 provides a summary of official data from BRAKDP regarding how funds were allocated.Thedatasuggestthatthevastmajorityoffunds(Rp.182billion,whichcorresponds to approximately US$ 20 million at the time the project was implemented) were spent on `economic'projects,deliveredintheformofdirecttransfers.21 Abouthalfofthisamountwasallocatedtoprojectsthatpurchasedcattle,andanotherquarter for other types of agriculture; most of the remainder was allocated to trading and business development. Official numbers state that there were 233,000 beneficiaries from these economicinvestmentsforapercapitavalueofRp.780,000perbeneficiary. Infrastructure accounted for less than 10 percent of expenditures. Half of all infrastructure expenditurewenttothebuildingofmeunasah(communitycenters);thenextlargestcategory (18percent)wasinvestmentinmosques,followedbyroads(12percent).Allothercategories accountedforlessthan4percentofinfrastructuralexpenditures.Thesenumbersarebroadly consistent with the survey data in many respects, both for direct benefits and for project support. TABLE2.4:USEOFFUNDS(MISDATA) Cost Share Estimatednumberof Activity (Rp.billion) (%) beneficiaries Economy 182 83.48 233,115 Infrastructure 21 9.83 194,408 Education 0.1 0.03 121 Others 1 0.49 14,497 Operationalfund 13 6.16 Total 218 100 442,141 Source:BRAKDPMISdata.Wenotethatitisnotclearthatthebeneficiariesfromeconomic projectsarenecessarilydifferentfromthebeneficiariesfromotherprojects. 20 TheMIS(MonitoringInformationSystem)dataisexploredfurtherinMorel,WatanabeandWrobel(2009). 21 Thisis83percentofallfundsand89percentofnonoperationalfunds. 22 Directbenefits Aswehaveseen,manyhouseholdsreportedreceivingdirectbenefitsfromBRAKDP.Ofthose thevastmajority,94percent,receivedcash.Thesectiononeconomicoutcomesexploresmore indepthhowthesefundswerespentusingthesurveydata. On average, households that received money received Rp. 606,000 (approximately US$ 60; confidence interval is Rp. 507,000 ­ 704,000). This figure is significantly below the MIS data estimateofRp.780,000.Howeverthepercapitadatafromtheproject'sMISalsoincludescases wherefundswereprovidedinkind(e.g.wheregoodssuchasagriculturalinputswereprocured at a level above that of the household). Given that 6 percent of those who report receiving economic assistance say support was inkind (Table 2.5) the amounts suggested by ARLS and the MIS data are fairly close. The average amount of cash received by nonvictims was somewhat lower than that received by victims, and the mostaffected received the greatest amountonaverage(althoughthedifferencesarenotstatisticallysignificant).22 The formal project data breaks allocations down by category (cattle, farming, etc), but the respondents typically report simply receiving cash. The reason for this stems from BRAKDP's design. Unlike many other livelihoodstype programs, project staff members normally do not procuregoods.Rather,communities,villagers,orgroupsofvillagersreceivecashandthenuse ittobuythegoodstheyneedfortheireconomicactivity.Beforeindividualsorgroups(orfor public goods, villages) receive funds they prepare a project proposal, which outlines how money received will be spent. As seen in Table 2.5, around 6 percent report receiving other pastoral or agricultural inputs. For these households, it is likely that goods were procured by beneficiarygroupleaders. This analysis provides stronger evidence that more funds reached conflict victims. Whereas before we found that in project areas, conflict victims were not significantly more likely to receivedirectbenefitsthannonvictims,herewefindthatthoseconflictvictimsthatreceived cashreceivedapproximately13percentmorethannonconflictvictims.Themostaffectedgot around19percentmoreonaverage. 22 Fortherestofthepaper,weusethesubjectivecriteriaforbreakdowns. 23 TABLE2.5:BRAKDPGOODS(TREATMENTSAMPLEONLY) Only Differencebetween Differencebetween Allnon All All most victimsandnon mostaffectedand victims victims affected victims nonvictims Share, among those receiving benefitsthatreceived: 0.04 0.04 Money/cash 0.94 0.92 0.96 0.96 (0.03) (0.03) 0.01 0.02 Poultry/goats/cows 0.02 0.03 0.02 0.01 (0.01) (0.02) 0.03 0.02 Fertilizer/Seeds/rice 0.03 0.05 0.02 0.02 (0.03) (0.03) Avg.cashamountreceivedby 73,215 106,147* thosehouseholdsthatreceived 605,919 568,405 640,662 673,595 (53,617) (63,412) somecash Avg.cashamountreceivedall 58,582*** 63,265*** 109,278 84,715 143,297 147,980 (treatmentandcomparisonareas) (17,346) (19,067) N(allrecipients) 524 254 270 174 "ForthosewhoreceivedassistancefromBRAKDP,whatwasthemostimportantthingyoureceived?"Tablereportsestimated populationmeans,standarderrorsandsampleN's.Estimatesofaveragecashamountreceivedwerecalculateddroppingwith outliersdropped.Source:ARLS This feature results largely from the fact that the BRAKDP budgets were larger in areas with more conflict victims and is not a result of withinvillage allocations being made disproportionately toconflictvictims.Whenweconditionontheallocationmadeinavillage, conflict victims received on average Rp. 13,000 more than others, and the most conflict affectedreceivedanaverageofRp.2,000lessthanotherconflictvictims(resultsnotshown). Both of these numbers are small and neither is statistically significant. Nevertheless, because more conflictaffected subdistricts received larger allocations, on average the mostconflict affectedhouseholdsreceivedalmost50percentmoreindirectbenefitsthananaveragenon conflict affected household. As we found before, those who had a family member kidnapped andthosewithaphysicalinjuryreceivedthemostonaverage(Table2.6). 24 TABLE2.6:CASHBENEFITSBYCATEGORYOFCONFLICTVICTIM Averageamountreceived(Rp.) (se) 676,000 Familymemberkilled (140,000) 702,000 Familymemberkidnapped/detained (187,000) 609,000 Missingbodypart/physicaldisability (193,000) 662,000 Housedamaged/destroyed (85,000) 524,000 Primarylivelihooddamaged/destroyed (95,000) 650,000 Internallydisplaced (70,000) 643,000 Mentalillness(selforfamilymember) (80,000) 743,000 Physicalillness(selforfamilymember) (108,000) Source:ARLS Althoughouraccountofaprogrammaticfocusonprivatebenefitsisconsistentacrossofficial MIS and ARLS data, there are difficulties reconciling the total amounts spent. The question asked in ARLS probes allocations made to individuals or their households. Under the assumptionthatindividualsfullyreportedthetransfers,theestimatedtotalallocationofcash benefitsisRp. 54billion (confidenceinterval:Rp.3970billion), lessthan athird ofthevalue reportedbyMIS.Wenote,however,thatwhilerespondentswereinstructedtoreportfortheir households, it is possible that individuals reported only benefits received by themselves and may not have known or reported the total transfers to their households. Such a tendency would have to be very strong however for it to account for the shortfall, and even then beneficiarynumberswouldbeinconsistentbetweenthetwodatasources.23 23 Inparticular,ifallindividualsreportedonlyownreceiptsthentheestimatedtotalexpenditurewouldbeRp.174 billion(confidenceinterval:Rp.124billion­Rp.224billion).Thiscorrespondsbroadlytoofficialexpenditure figuresalthoughitsuggestsabeneficiarypopulationgreaterthanthatreportedbyMIS(308,000ratherthan 230,000). 25 TABLE2.7:HOWWEREBENEFITSUSED(TREATMENTSAMPLEONLY) All Allnon All Onlymost Difference Differencebetween victims Victims affected betweenvictims mostaffectedand tononvictims nonvictims (se) (se) Withinonemonth: 0.06 0.03 Production 0.45 0.48 0.42 0.43 (0.06) (0.07) 0.02 0.02 Soldandinvestedproceeds 0.02 0.01 0.03 0.01 (0.03) (0.02) 0.06 0.05 Retainedbutnotused 0.15 0.18 0.12 0.12 (0.05) (0.05) 0.03 0.02 Consumed 0.22 0.20 0.23 0.20 (0.05) (0.07) 0.00 0.02 Gaveaway/takenaway 0.01 0.01 0.01 0.00 (.) (0.02) Subsequently: 0.12 0.07 Usedforproductionbutnowgone 0.56 0.51 0.63 0.61 (0.07)* (0.09) 0.06 0.03 Beingused 0.16 0.19 0.13 0.14 (0.06) (0.06) 0.00 0.00 Beingsaved 0.01 0.01 0.01 0.01 (0.01) (0.01) 0.06 0.04 Sincegivenaway/consumed/taken 0.27 0.29 0.24 0.24 (0.07) (0.08) ***Significantat99%;**Significantat95%;*Significantat90%.Tablereportsestimatedpopulationmeansandstandarderrors. Whichofthefollowingbestdescribeswhatyoudidwiththesegoodswithinonemonthofreceivingthem? Ifusedforproduction/soldandinvested/retainedbutnotused/soldandsaved:Whichofthefollowingbestdescribeswhat youhavedonesincewiththesegoods?Source:ARLS The ARLS data also allow for a deeper examination of the uses of direct funds. As shown in Table2.7,45percentofrecipientsusedthefundsforproductionwithinthefirstmonth.In22 percentofcases,fundswereconsumedwhilein15percentofcasestheyweresaved(therest fallintoothercategories,thelargestofwhichis"usedforthehouseholdeconomy").Fiftysix percent report that the goods have since been exhausted while 27 percent report they have sincebeenconsumedorotherwiselost.Thegoodsarestillbeingusedforabout16percentof households.24Therearefewnoticeabledifferencesacrossgroupsalthoughconflictvictimsare morelikelytoreportthatthegoodsareexhausted. Projects Table2.8,describestheprojectsthatwereapprovedinvillagesaswellastheshareofvillagers approvingdifferentprojects.ConsistentwiththeMISdata,thegreatestshareofvillagesopted for economic activities although the data confirm that these typically took the form of cash 24 Notethatgoodsbeingexhausteddoesnotmeantheydonothavestillhaveproductiveimpacts.Forexample,if fundswereusedtopayforlabortoclearland--somethingthatappearstohavebeencommongivenreported resultsbelow--beneficiariesmaystillbereapingeconomicgains. 26 payments (`bagi rata' or equal cash disbursement 25 accounted for 67 percent of projects accordingtoreportswhileagriculturalprojectsaccountedforafurther7percent).Amongother projects, the most common were improvements to village buildings, including meunasah and mosques(14percent),androads(5percent).Therearenosignificantdifferencesacrossgroups intermsofsupportfordifferentprojects. TABLE2.8:PROJECTSAPPROVEDANDSUPPORTED Share Sharenon Share Sharemost Difference Difference whosaid victims victims affected betweenvictims betweenmost their supporting supporting supporting andnonvictims affectedandnon village (se) victims received (se) ... Improvevillage buildings(including 0.03 0.02 0.14 0.14 0.10 0.11 meunasahand (0.03) (0.03) mosques) Improveroadsand 0.03 0.03 0.05 0.02 0.06 0.06 bridges (0.03) (0.04) Improveaccessto 0.03 0.03 0.04 0.02 0.05 0.05 water (0.01)* (0.02) 0.01 0.02 Supportforagriculture 0.07 0.06 0.04 0.04 (0.02) (0.02) 0.02 0.04 Bagirata 0.78 0.67 0.69 0.71 (0.05) (0.07) ***Significantat99%;**Significantat95%;*Significantat90%.Tablereportsestimatedpopulationmeansandstandard errors.Note,percentagesdonotaddto1asvillagesmayhavereceivedmultiplekindsofsupportfromBRAKDP.Reportingfor onlyprojectsthatwereatleast4percentoftotalapproved.Source:ARLS The table also provides information regarding who supported different types of projects, averaging over all villages. In general, support for cash transfers is stronger among conflict victims, while village infrastructure projects were more popular among nonvictims. Non victimsmayhavefeltthattheywouldbemorelikelytobenefitfrominvestmentsthatbenefited the whole village, than from private goods where victims would be more likely to be prioritized.26 Althoughthenumbersselectingdifferentprojectstrackthenumberofprojectssupportedfairly well on average, this does not mean that in most cases the projects that people chose were implemented.Infact,asreportedinTable2.9only65percentofindividualsreportedthattheir preferredprojectwasapproved.Thisisprobablybecausetheprogramencouragedcompetition within villages over funds: villagers are expected to have viable project ideas if they are to receivefunds. 25 Forsurveypurposesbagiratawasdefinedasequalcashdisbursement.However,fieldsupervision(ofthe programandthesurvey)showedawiderangeofunderstandingsoftheterm,withmanypeoplereportingbagi rataforanytypeofcashdisbursement. 26 Indeed,programsupervisionmissionssuggestthatthiswasoftenthecase.SeeMorel,WatanabeandWrobel (2009). 27 Thisfigureis6percentagepointslowerforconflictvictimsand10percentagepointslowerfor themostconflictaffectedsuggestingthatconflictvictims,ingeneral,hadamoredifficulttime ensuringthattheprojectstheypreferredwereimplemented.Again,thissuggeststhattargeting wasimperfectalthoughweemphasize(apointwereturntolater)thatthesedifferencecould reflect possibly large preexisting power differentials between conflict victims and others. The differencescouldalsobeduetothefactthatconflictvictimspresentedlowerqualityproposals andcouldhavebenefitedfrommoreassistanceindraftingandpresentingproposals. TABLE2.9:PROJECTSSUPPORTED All Nonvictims Conflict Most Difference Difference victims affected betweenvictims betweenmost tononvictims affectedandnon (se) victims(se) Probabilitythat 0.65 0.68 0.62 0.58 0.06 0.10 preferredprojectwas (0.04) (0.04) (0.05) (0.05) (0.05) (0.06)* implemented ***Significantat99%;**Significantat95%;*Significantat90%.Tablereportsestimatedpopulationmeansandstandard errors.Thistableshowstheprobabilitythataprojectwasimplementedconditionaluponitbeingsupportedbydifferent categoriesofrespondent.Reportingforonlyprojectsthatwereatleast4percentoftotalapproved.Source:ARLS 2.3 ParticipatinginBRAKDP WasBRAKDPsuccessfulatengagingpopulationsindecisionmakingovervillageexpenditures? We answer this question by examining awareness of and participation in BRAKDP across a number of categories of interest. As shown in Table 2.10, approximately 57 percent of individualsinprogramareashadheardofBRAKDP.Therearenosignificantdifferencesacross groups. Thirtyseven percent of adults were aware of the meetings. Twenty percent participated,over200,000peopleacrossAceh. TABLE2.10:BRAKDPAWARENESS&PARTICIPATIONI(PROJECTAREASONLY) Sharewho... Only Differencebetween Differencebetweenmost Allnon All Allvictims most victimsandnon affected victims affected victims(se) andnonvictims(se) HaveheardofBRA 0.01 0.03 0.57 0.57 0.57 0.55 KDP (0.05) (0.05) Areawareof 0.02 0.07 0.37 0.38 0.36 0.32 meetings (0.04) (0.04)* 0.02 0.03 Attendedmeetings 0.20 0.19 0.21 0.18 (0.03) (0.03) ***Significantat99%;**Significantat95%;*Significantat90%.Tablereportsestimatedpopulationmeansandstandard errors.Source:ARLS InTable2.11weexaminehowthesenumbersdifferfordifferentcohorts.Men,wefind,were morelikelythanwomentohaveheardofmeetingsandattended;12percentofwomeninBRA KDPareasattendedmeetingscomparedwith27percentofmen.Headsofhouseholdwerealso muchmorelikelytoattend,withonly11percentofnonheadsofhouseholdattending.There arenodiscernibledifferencesbetweenpoorerpeopleandwealthierpeople,orbetweenmale and femaleheaded households. This is surprising given that poorer and femaleheaded households tend to be less powerful within villages, and hence have less incentive to attend 28 given--allelsebeingequal--theywouldbelesslikelytobenefit.OtherresearchonregularKDP inIndonesiahasshownthatitoftenhasdifficultiesinreachinghighlyvulnerablegroupssuchas femaleheaded households with such groups often not attending program meetings (McLaughlin, Satu and Hoppe 2007). In contrast, the evidence here shows that BRAKDP was successfultosomeextentinreachingouttomoremarginalizedgroupsinvillages. TABLE2.11:BRAKDPAWARENESSANDPARTICIPATIONII(PROJECTAREASONLY) Female Difference Maleheaded Difference Sharewho... Men Women headed (se) households (se) households HaveheardofBRA 0.62 0.52 -0.10 0.57 0.56 -0.01 KDP (0.04)** (0.05) Areawareof 0.40 0.34 -0.06 0.37 0.36 -0.01 meetings (0.04) (0.05) Attendedmeetings 0.27 0.12 -0.15 0.20 0.17 -0.03 (0.04)*** (0.03) Poorest Difference Headof Difference Others Others third (se) household (se) HaveheardofBRA 0.57 0.56 -0.01 0.63 0.51 0-.11 KDP (0.04) (0.04)*** Areawareof 0.39 0.34 -0.05 0.42 0.32 -0.10 meetings (0.04) (0.05)** Attendedmeetings 0.20 0.19 -0.01 0.29 0.11 -0.17 (0.03) (0.04)*** ***Significantat99%;**Significantat95%;*Significantat90%.Tablereportsestimatedpopulationmeansandstandard errors.Source:ARLS Note:Reportsresultsformainrespondentswhoresideinmaleorfemaleheadedhouseholds. 2.4 PerceptionsofandProblemswithBRAKDP WeclosethissectionbyconsideringhowBRAKDPwasperceivedbyrespondents.BRAKDPwas verypopularamongstpeopleinareasthatreceivedtheprogram.Ninetyfourpercentofpeople intreatmentareassaidtheythoughtheprogramwashelpful;thisfigurerisesto96percentfor victimsand97percentforthemostaffected,althoughdifferencesarenotsignificant. Complaints about the program were also low. Table 2.12 provides the share of respondents that agreedwith a set of complaints about theimplementation of BRAKDP,brokendown by victim status. Across all measures we see that the level of complaints is relatively low; the greatestcomplaint,madeby14percentofrespondents,wasofdiversionofmoney27andthat themostrelevantprojectswerenotselected.Conflictvictimswerelesslikelytosharethese concerns;forthesegroupsthegreatestcomplaintwasthattheprogrambenefitedothergroups toomuch--notablyexGAM,PETA28andIDPs. 27 Thesurveydidnotgiveadefinitionofwhatdiversionofmoneymeans.Fromthesupervisionmissionsand qualitativefieldwork,itappearsarangeofissuesmayhavebeencapturedunderthiscategoryincludingmoney beingspentonprojectsthatsomevillagersdidnotprioritize. 28 PETAaremembersofformermilitiagroupsthatwereformedtofightagainstGAMduringtheconflict. 29 TABLE2.12:BRAKDPCONDUCT(TREATMENTAREASONLY) Groups ComparingGroups Share saying they `agree' or `strongly All Allnon All Most Victimsto Mostaffected agree'(forthosewhoheardofBRAKDP) victims Victims affected nonvictims tononvictims n=704 n=347 n=357 n=471 diff diff se Se 0.14 0.16 0.11 0.11 0.05 0.05 Activitiesnotmostimportant (0.04) (0.04) 0.10 0.09 0.11 0.08 0.01 0.01 Didnotbenefitenoughpeople (0.04) (0.04) 0.12 0.12 0.12 0.10 0.00 0.02 Didnotbenefitconflictvictims (0.04) (0.04) 0.11 0.10 0.13 0.15 0.02 0.05 BenefittedexGAM/PETA/IDPstoomuch (0.03) (0.04) 0.09 0.11 0.07 0.07 0.03 0.04 Disagreementsnotwellhandled (0.03) (0.03) 0.14 0.16 0.11 0.09 0.05 0.07 Diversionsofmoney (0.03) (0.04)* 0.09 0.11 0.06 0.03 0.05 0.09 Extortion (0.03) (0.03)** 0.10 0.13 0.07 0.07 0.06 0.06 Allocationacrossvillagesunfair (0.04) (0.04) Share that feel the program was generally 0.04 0.05 0.94 0.92 0.96 0.97 helpful (0.03) (0.03) ***Significantat99%;**Significantat95%;*Significantat90%.Tablereportsestimatedpopulationmeans,standarderrors andsampleN's.Source:ARLS To put these responses in context, we examine how BRAKDP fared compared to another prominent development project in nonBRAKDP locations. In nonBRAKDP locations, the villageheadwasaskedtoprovidethenameofthe"mostimportantdevelopmentprojectinthe village,intermsofmoneyinvested."In64percentofvillages,theheadreportedthatKDP(as distinct from BRAKDP) was the most important other project. KDP has been active in Aceh since1998,andfundsweredeliveredthroughtheprograminparalleltotheBRAKDPfunds.For theanalysisinTable2.13,wefocusonsuchcasesforcontrolcommunities. 30 TABLE2.13:AWARENESSOFDEVELOPMENTPROJECT HaveyouheardoftheBRAKDP Individualsincontrol Individualsintreatment Simple Difference (treatmentcommunities)/KDP communities communities difference accounting (controlcommunities) (KDP) (BRAKDP) (se) forselection developmentproject? (N) (N) (se) All 0.42 0.59 0.17*** 0.17* (911) (965) (0.06) (0.09) Conflictvictims 0.46 0.60 0.14* 0.25* (331) (466) (0.07) (0.13) Mostconflictaffected 0.45 0.59 0.14 0.07 (208) (321) (0.09) (0.18) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Source:ARLS Webeginwiththemostbasicmeasureofprojectvisibility,whethertheindividualisawareof the project in their village. Table 2.13 reports whether individuals in BRAKDP locations had heard about BRAKDP, compared to whether individuals in control communities had heard abouttheKDPprojectintheirvillages.ThereisevidencethatBRAKDPhadahigherprofilethan KDP. The difference is particularly large for conflict victims, although less so for the most affected.Thesefindingsarerobusttomultiple(butnotall)alternatespecifications. 31 TABLE2.14:PROBLEMSINDEVELOPMENTPROJECTS Lookingbackattheimplementation Individualsin Individualsin Simple Difference ofBRAKDP(treatment)/KDP control treatment difference accountingfor (control)doyouagreewiththe communities communities (se) selection followingstatements? (N) (N) (se) Activities selected were not the mostimportantones All 0.12 0.12 0.01 0.1 (434) (646) (0.03) (0.09) Conflictaffected 0.11 0.09 0.01 0.23* (163) (321) (0.04) (0.12) Mostaffected 0.1 0.1 0 0.43** (106) (215) (0.05) (0.17) Activities selected did not benefit 0.14 0.1 0.04 0.05 enoughpeopleinthevillage (433) (646) (0.04) (0.09) Activities selected did not benefit conflictvictims All 0.12 0.12 0 0.06 (427) (641) (0.03) (0.09) Conflictaffected 0.16 0.11 0.05 0.09 (162) (319) (0.04) (0.18) Mostaffected 0.07 0.1 0.03 0.14 (105) (214) (0.05) (0.23) Activities benefitted exGAM, PETA 0.1 0.11 0.02 0.04 orIDPStoomuch (419) (635) (0.03) (0.07) Disagreements in village not well 0.13 0.08 0.05 0.09 handled (421) (632) (0.03) (0.07) Obviousdiversionsofmoney All 0.16 0.12 0.04 0.15 (383) (606) (0.03) (0.12) Conflictaffected 0.14 0.1 0.04 0.21 (149) (299) (0.05) (0.17) Mostaffected 0.08 0.09 0.01 0.30* (95) (200) (0.04) (0.17) Moneywasextorted 0.11 0.08 0.03 0.03 (373) (603) (0.04) (0.11) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communities using least squares and instrumental variable regressions. IV regressions control for conflict and spending capacity,theirquadraticandcubedterms,andtheirinteraction.Source:ARLS The survey also collected evidence on what individuals think the major problems are in BRA KDPand(thecontrol)KDP(Table2.14).Theshareofindividualsreportingproblemsislow.Only 12 percent felt that the projects selected were not the most important ones, a share that is higheramongconflictvictimsforBRAKDPcomparedtoKDPareas.Respondentsdidnotclaim 32 that BRAKDP benefited conflict victims any more or less than regular KDP, surprising given BRAKDP's specific focus on conflict victims.29The most common complaint overall was that there were diversions of money due to collusion, corruption or nepotism, although these complaintswerelowerforBRAKDPthanforKDP(16percentforKDP,12percentforBRAKDP). Aswesawbefore,thevastmajorityofindividuals inBRAKDPlocationsandcontrollocations felt that the development project was helpful. Table 2.15 reveals that this number is no differentthanthatfoundforthestandardKDPprogram. TABLE2.15:HARMFUL/HELPFUL Individualsin Individualsin Simple Difference control treatment difference accounting Agree that BRAKDP/KDP was communities communities (se) forselection typicallyhelpfulforthevillage (N) (N) (se) All 0.96 0.94 0.02 0.03 (436) (649) (0.02) (0.07) Conflictvictims 0.94 0.96 0.02 0.15 (163) (322) (0.03) (0.12) Mostconflictaffected 0.98 0.96 0.01 0.08 (106) (216) (0.02) (0.16) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Ingeneral,whichofthefollowingstatements wouldyousaybestcharacterizestheworkofBRAKDP/KDPinthisvillage?Source:ARLS 29 ThismaybebecauseKDPfocusesonassistingthepoorandvulnerablewithincommunities,manyofwhomare presumablyalsovictimsoftheconflict. 33 3. IMPACTSONWELFARE HowdidBRAKDPaffectthesocioeconomicconditionsofthosewhoreceivedtheprogramand especially conflict victims? There is some evidence that KDP elsewhere in Indonesia has impacted different dimensions of welfare. Alatas (2005) and Voss (2008) both find gains in consumption among beneficiaries. In postconflict settings, severely conflictaffected communities often have needs for immediate livelihoods support such as capital, fishing or farming equipment, and income generation activities. Indeed, previous research in Aceh suggeststhatthetoppriorityformanycommunitiesinAcehwaslivelihoodssupportandmany villagersnotedthatprovisionofcapitalwouldbestenablethemtorebuildtheircommunities after the losses they suffered from the conflict (World Bank 2006). Enhancing welfare, especiallyforconflictvictims,wasthusakeyobjectiveoftheprogram. BRAKDP communities had tremendous flexibility in deciding what projects to fund, how to targetthebenefits,andwhethertofinanceinvestmentsinpublicgoodsortodisbursecapitalto households. As a result, improvements in welfare may be broad or narrowly distributed and theymaybereflectedinpublicgoodsorprivategoods.Wefocusfirstonwelfareimpactsthat might be the result of cash disbursement and then turn to welfare impacts that might be caused by investments in public goods provision or the downstream benefits of greater householdwealth. 3.1 PovertyProfile Webeginouranalysiswithafocusontheaggregatepovertyprofileofthevillages.Itappears thatBRAKDPhasresultedinasubstantialdecreaseinpovertylevels. Ineachvillage,weaskedleadersasetofquestionsdesignedtotapintocomparativelevelsof welfare. Table 3.1 presents the results from a question in which village heads are asked estimatetheshareofvillagehouseholdsthatshouldbeclassifiedas`poor'. TABLE3.1:AGGREGATEMEASURESOFCOMMUNITYWELLBEING(BYVILLAGEHEADS) Individualsincontrol Individualsin Simpledifference Difference communities treatment (se) accountingfor (N) communities selection (N) (se) Shareofhouseholds 0.69 0.69 0.00 0.11** classifiedaspoor (242) (217) (0.02) (0.05) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standarderrors andsampleN's,aswellasthedifferenceforpopulationsintreatmentandcontrolcommunitiesusingleastsquaresand instrumentalvariableregressions.IVregressionscontrolforconflictandspendingcapacity,theirquadraticandcubedterms,and theirinteraction.Source:ARLS Thetableshowsthatasimplecomparisonofresponsesinprojectandcomparisonareasreveals no difference. However, once we account for selection effects into the program we find that BRAKDPisassociatedwithan11percentagepointdropinthereportedshareofhouseholds that are poor. These results are robust to alternative specifications with similar estimated 34 magnitudes.Theresultssuggestthattheprogramhashadamajorimpactinreducingpoverty inconflictaffectedcommunities,atleastaccordingtotheclassificationsofvillageheads. ThislogicisillustratedinFigure3.1.Beforeinterpreting,wetakeamomenttoexplainwhatis conveyedinthisfigureandsubsequentfigureslikeit.Onthehorizontalaxisthereisameasure ofassignmenttotreatment.Thiswasmadebycollapsingthetwoassignmentvariables(conflict intensity and spending capacity) into one continuous measure.30The cutoff for assignment to treatmentisatthezeropoint,withallvillagestotherightofthecutoffinsubdistrictsassigned totreatmentandalltotheleftofthatcutoffassignedtothecontrol.Rightinthevicinityofthe cutoff,wecanassumethatsubdistrictsoneithersidearehighlysimilar;asyoumovefurther away from the cutoff in either direction the subdistricts become increasingly dissimilar and hence treatment and controls become less comparable. Hence, the estimation of treatment effectscomesfromcomparingthesizeofthejumpordropintheverticalaxis(otherwisecalled a`discontinuity')atthecutoff.31Insum,therearetwomainthingstonoteinthesefigures.The firstistheslopeofthelines,whichshowtheselectioneffects--ortherelationshipbetweenthe probabilityofbeingselectedforBRAKDPandtheoutcome.Theseconditemofinterestisthe sizeofthegapatthecutoff,whichindicatesthesizeofthetreatmenteffectestimatedatthe threshold. FIGURE3.1:SELECTIONINTOBRAKDPANDEFFECTSONPOVERTY 1 .8 Share Poor .6 .4 .2 0 -2 -1 0 1 2 Distance from Selection Threshold Point (Areas > 0 Receive BRA-KDP) Source:ARLS Figure3.1thusillustratesthedifferenceintheestimatedeffectsonreportedpovertywhenwe doanddonotaccountforselection.ThoseplacesthataremorelikelytobeselectedforBRA KDP are in general also more likely to have more poor households in the absence of the program.Thusapositiveeffectoftheprogramcouldbemaskedbyanegativeselectioneffect. 30 Sincetherearetwodimensionsatworkandassignmentdependedonreachingathresholdonbothdimensions simultaneouslywegeneratedameasureof(Leontief)distancefromthethresholdfrontier.Thefrontieriscodedas 0,pointstothenortheastofthefrontierarepositiveandpointstothesouthwestarenegative. 31 Note,sincethisfigureusesassignmenttotreatmentitdoesnotdealwithissuesofnoncompliancediscussed earlier.Hence,thesizeofthediscontinuityreflectsanintentiontotreat(ITT)effectandconditionsonlyonan aggregatedmeasureofdistancefromeachofthethresholdswithineachdistrictandnotontheindividualvaluesof theselectioncriteria.Thisisthereforedifferentfromthelocalaveragetreatmenteffectpresentedinthetables. 35 Thesolidlinesinthefigureshowtheaveragelevelofpovertyforeachvalueonthedistance fromtreatmentscale(xaxis).Twofeaturesstandout.Thefirstisthattheselinesareupward sloping.Thissimplyreflectsthatfactthatthoseareasmorelikelytobeselectedwerealsomore likelytohavemorepoorhouseholds. Thisistrueforbothprojectandcomparisonareas.The secondfeatureisthatatthethresholdpoint,0,thelefthandlineliesabovetherighthandline. Thedifferencebetweenthesetwolinesispreciselytheestimatedtreatmenteffectatthecutoff point.Thisshowsthattheprogramhadaresultinreducingpovertyasreportedbyvillageheads. 3.2 AssetIndex Wenow consider moredirectmeasuresofhousehold welfareinanefforttocorroboratethe impressionsofvillageleaders.Wefocusfirstonassetholdings,whichareplausiblyinfluenced bythedisbursementofcashthroughBRAKDP.Respondentswereaskedabouttheirhousehold asset holdings with respect to 16 different types of assets that can be used for consumption andinvestment(transport,agriculturalequipmentandlivestock/fowl). Table3.2presentsestimatesoftheimpactofBRAKDPonanaggregateassetindex.32Thedata suggeststhat,takingaccountofselectioneffects,theprogramisassociatedwithgainsonthe order of onethird of a standard deviation on overall levels of assets especially for conflict victims.Theresultsarenot,however,robustacrossalternativespecifications.Table3.3breaks downtheassetsonebyone.Thesimplecomparisonofprojectandcomparisonareassuggests lowerassetholdingsingeneralinprojectareas;thiseffectis,however,inpartduetoselection. TABLE3.2:ASSETINDEX Individualsin Individualsin Difference Difference control treatment (se) Accounting Indexof2008assetholdings communities communities forSelection (N) (N) (se) All 0.22 0.07 0.16** 0.04 (1225) (1090) (0.07) (0.12) Conflictvictims 0.04 0.01 0.05 0.34** (455) (528) (0.09) (0.17) Mostconflictaffected 0.13 0.04 0.09 0.43 (282) (269) (0.12) (0.26) *** Significant at 99%; ** Significant at 95%; * Significant at 90%. The table reports estimated population means, standard errors and sample N's (where total sample size is 2,315), as well as the difference for populations in treatment and control communities using least squares and instrumental variable regressions. All regressions control for conflict and spending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Howmanyofthefollowingthingsdoyouormember ofyourhouseholdpossess?Source:ARLS Broken down asset by asset, the significant gains related to the program are in engine/motorcycleholdingsandagriculturalmachinery(Table3.3).Theseresultsarerobustin only one of two alternative specifications. In most instances the gains are modest and not 32 Theassetindexwasformedbyafactoranalysisofseveralvariablesmeasuringthequantityofassetsowned, rangingfromchickensandlivestocktolargescaleagriculturalmachinery. 36 statistically significant. However for conflict victims, in particular, there is evidence for higher levels of household stove ownership in BRAKDP areas and there is very strong evidence on engine/motorbike holdings (these engines are often vital for taking produce or goods to markets). Indeed, the increase in motorcycle engine holdings for conflict victims is dramatic. Once outliers are removedfrom the dataset, the effectsize drops from .6 to .48 motorbikes, whichstillsuggeststhatparticipationinBRAKDPisassociatedwithgainsinholdingsofthese largeassetsforoneintwohouseholdswithconflictvictims.Giventheuseoffundsreportedin theboththeprogram'sMISsystemandinthesurvey,itisunlikelythatalargeshareofthese extraengineswereboughtwithBRAKDPfunds.Rather,itmaybethatprogrambeneficiaries areusingmoneytheyearnfromeconomicactivitiestobuymotorbikes. TABLE3.3:ASSETSBYCATEGORY Individualsin Individualsin Simple Difference Difference control treatment difference accounting accountingfor communities communities (se) forselection selectionConflict (N) (N) (se) victims(se) Stove 1.19 0.96 0.23*** 0.17 0.49* (1,225) (1,090) (0.08) (0.16) (0.29) Radio/taperecorder/video 0.5 0.49 0.01 0.03 0.02 (1,225) (1,090) (0.04) (0.07) (0.11) Television 0.69 0.61 0.09** 0.05 0.15 (1,225) (1,090) (0.04) (0.06) (0.10) Parabolaantenna 0.44 0.35 0.09* 0.08 0.04 (1,225) (1,090) (0.05) (0.08) (0.10) Ornamentalsideboard/buffet 1.43 1.32 0.11 0.05 0.13 (1,225) (1,090) (0.09) (0.14) (0.19) Refrigerator 0.31 0.28 0.04 0.02 0.02 (1,225) (1,090) (0.03) (0.06) (0.09) Bicycle/rowboat 0.74 0.69 0.05 0.12 0.06 (1,225) (1,090) (0.07) (0.11) (0.17) Motorcycle/portableengine 0.68 0.68 0.00 0.20* 0.60*** (1,225) (1,090) (0.06) (0.12) (0.19) Car/motorizedboat 0.06 0.08 0.03 0.02 0.07 (1,225) (1,090) (0.02) (0.03) (0.06) Telephone/cellularphone 0.99 0.83 0.16 0.03 0.3 (1,225) (1,090) (0.10) (0.19) (0.26) Chicken/fowl 4.7 5.4 0.71 0.08 0.58 (1,225) (1,090) (0.58) (0.79) (1.24) Goats/sheep 0.56 0.56 0 0.08 0.36 (1,225) (1,090) (0.12) (0.21) (0.42) Waterbuffalo/cows/horses 0.22 0.37 0.15 0.43 1.07 (1,224) (1,090) (0.13) (0.37) (0.97) Store/kiosk 0.11 0.11 0 0.03 0.01 (1,225) (1,090) (0.03) (0.04) (0.07) Largeagriculturalmachinery 0.01 0.02 0.01 0.02* 0.04 (1,225) (1,090) (0.01) (0.01) (0.03) Nonagriculturalmachinery 0.09 0.11 0.02 0.01 0.04 (1,225) (1,090) (0.03) (0.05) (0.08) *** Significant at 99%; ** Significant at 95%; * Significant at 90%. The table reports estimated population means, standard errors and sampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrolcommunitiesusingleast squaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspendingcapacity,theirquadraticandcubedterms, andtheirinteraction.Question:Howmanyofthefollowingthingsdoyouoramemberofyourhouseholdpossess?Source:ARLS 37 Figure3.2providesanotherwayofexaminingtheeffect;thefigureshowstheexpectednumber ofmotorcycles/enginesheldbyconflictvictimhouseholdsasafunctionoftreatmentstatusand `distance'fromthetreatmentcutoff.Thefigureagainhighlightstheimportanceoftheselection effects: the higher the score on the selection variables the lower the holdings; however holdings are considerably greater for project households than for comparison households aroundthecutoffpoint. FIGURE3.2:MOTORCYCLEHOLDINGSANDSELECTIONINTOBRAKDP Source:ARLS 3.3 HouseholdInfrastructure Wealsocollecteddataonaseriesoflargerhouseholdassets,suchashouseconstructionand access to water. Table 3.4 presents information about the share of households that have constructed homes out of concrete. Only about onethird of households live in high quality, concrete housing. There is no evidence that individuals are more likely to live in more permanent,higherqualityhousingasaconsequenceofBRAKDP. 38 TABLE3.4:QUALITYOFHOUSING Individualsin Individualsin Simple Difference control treatment difference accountingfor Share of those whose houses are communities communities (se) selection madeofconcrete (N) (N) (se) All 0.29 0.30 0.01 0.03 (1,225) (1,090) (0.04) (0.08) Conflictvictims 0.25 0.24 0.01 0.07 (455) (528) (0.05) (0.11) Mostconflictaffected 0.25 0.22 0.03 0.07 (282) (369) (0.05) (0.13) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.Allregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Whatisthematerialusedmostinyourhouse walls?Source:ARLS Table3.5showsthedistributionofaccesstowaterfromcleanorprotectedsources.Inall,less thantwothirdsofrespondentshaveaccesstocleanwateraccordingtothismeasure.Aswith other welfare measures, conflict victims have less access to clean water than nonconflict victims.Overall,however,thereisnodifferencebetweentreatmentandcontrolcommunities. TABLE3.5:WATERSOURCE Individualsin Individualsin Simpledifference Difference control treatment (se) accounting Access to water from a clear or communities communities for protectedsource (N) (N) selection (se) All 0.63 0.57 0.06 0.12 (1,225) (1,090) (0.06) (0.11) Conflictvictims 0.55 0.50 0.05 0.20 (455) (528) (0.07) (0.13) Mostconflictaffected 0.56 0.51 0.05 0.12 (282) (369) (0.08) (0.16) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.Allregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Whatisthishousehold'smostimportantsourceof water?Source:ARLS 3.4 LandUse As agriculture is a major source of income for village households in Aceh, we asked about accesstolandandtheextentoffarminghouseholdsareundertaking.Thereisstrongevidence thatBRAKDPisassociatedwithlargeincreasesintheamountoflandbeingfarmed.AsTable 3.6shows,theseresultsareespeciallypowerfulforconflictvictims. 39 TABLE3.6:LANDUSE Individualsin Individualsin Simpledifference Difference control treatment (se) accounting 2 m oflandthatisfarmedby communities communities forselection household (N) (N) (se) All 7,740 9,438 1,697 12,201 (644) (617) (2855.43) (7940.23) Conflictvictims 6,906 7,044 138 7,591*** (245) (297) (1114.56) (2174.80) Mostconflictaffected 8,215 7,607 608 7,382*** (152) (200) (1614.86) (2774.70) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.Allregressionscontrolforconflictandspending 2 capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Howmanym oflandisbeingfarmedbythis household?Source:ARLS Theseresultsarequiterobusttoalternativespecifications.Onaverageconflictvictimsseethe landtheyfarmdoubleasaresultoftheprogram.Thesupervisionmissionsshowedthatmany beneficiariesoftheprojectusedtheBRAKDPmoneytheyreceivedforagriculturalinputsand toclearlandthathadbecomeovergrownduringtheconflict. Figure 3.3 presents another view of the data, showing the average level of land use for treatmentandcontrolhouseholdsasafunctionofthepropensityofavillagetoreceiveBRA KDP.Thefigurehighlightsthefactthatvillagesmorelikelytobeselectedintotheprogramare alsomorelikelytohavelimitedlanduse,butthatconditionalonselectioncriteria,entrytothe programhasstrongpositiveeffects. 15000 FIGURE3.3:LANDUSE 10000 Land Farmed 5000 0 -2 -1 0 1 2 Distance from Selection Threshold Point (Areas > 0 Receive BRA-KDP) Source:ARLS 40 3.5 EmploymentandWages BRAKDP does not appear to have had a significant impact on employment levels. Data collected on entire households through the survey permit us to generate estimates of employment and unemployment. It is possible that BRAKDP generates higher levels of employment either through community investments in public goods provision or indirectly throughgreatereconomicactivityresultingfromtheinfusionofcapital. The measure of employment we use includes individuals that are not actively seeking employment in the denominator, a decision that results in lower overall employment rates. Nevertheless,employmentrates,asreportedinarehigh.Althoughtheyaremarginallyhigher overall in treatment communities, this difference is not significant. There is no evidence of differencesforconflictvictimsinparticular(Table3.7). Informationontheaveragedailywagesoflaborers(bothmaleandfemale)isgiveninTable3.8. Wefindnoevidencethattheprogramaffectedthecostoflabor,atleastasestimatedbyvillage leaders.Notethatwagefiguresdonotcorrespondtoincomesincetheydonottakeaccountof employmentratesorofnonwageincome,whichislikelytobehighgiventhelargeamountof people working in agriculture who are essentially selfemployed. Indeed, the poverty figures above would suggest that at the lower ends of the spectrum, there have been increases in income. TABLE3.7:EMPLOYMENT Individualsin Individualsin Simple Difference control treatment difference accountingfor Shareofnonstudentsof communities communities (se) selection workingagewithconsistentor (N) (N) (se) fulltimeemployment All 0.81 0.83 0.02 0.03 (3,309) (2,966) (0.02) (0.03) Conflictvictims 0.79 0.85 0.05** 0.01 (1,231) (1,411) (0.02) (0.05) Mostconflictaffected 0.79 0.84 0.05* 0.01 (757) (1,005) (0.03) (0.06) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.Allregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Whichoptionbestdescribes[...]semployment situation?Source:ARLS 41 TABLE3.8:AVERAGEDAILYWAGESOFLABORERS Individualsin Individualsin Simple Difference control treatment difference accounting communities communities (se) forselection (N) (N) (se) Averagedailywageofafemale 29,954 29,225 729 1,429 farmworkerordaylaborer (237) (216) (867) (1608) (controllingforseason­Rupiah) Averagedailywageofamalefarm 41,748 41,631 117 856 workerordaylaborer(controlling (242) (217) (1027) (2098) forseason­Rupiah) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Source:ARLS 3.6 EducationandHealth Table 3.9 and Table 3.10 employ further data from the entire household to estimate welfare outcomes in terms of health and education. We focus on the incidence of sickness for the former and school enrollment rates for the latter. Again, on these measures there is no evidence that BRAKDP communities fare better or worse than those villages that did not receivetheprogram. TABLE3.9:SICKNESS Individualsin Individualsin Simpledifference Difference control treatment (se) accounting Sharesickinlastmonth communities communities for (N) (N) selection (se) All 0.06 0.07 0.01 0.01 (5,619) (5,106) (0.01) (0.02) Conflictvictims 0.08 0.09 0.01 0.03 (2,151) (2,507) (0.02) (0.03) Mostconflictaffected 0.08 0.10 0.02 0.01 (1,345) (1,752) (0.03) (0.04) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.Allregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Did[...]sufferfromanysicknessthatprevented him/herfromworkingorgoingtoschoolinthepastmonth?Source:ARLS 42 TABLE3.10:INSCHOOL(INDIVIDUALS25YEARSOLD) Individualsin Individualsin Simpledifference Difference control treatment (se) accounting Shareofthoseaged5­25 communities communities for attendingschool (N) (N) selection (se) All 0.65 0.65 0.00 0.01 (2,596) (2,373) (0.02) (0.03) Conflictvictims 0.65 0.65 0.01 0.06 (1,033) (1,192) (0.03) (0.06) Mostconflictaffected 0.67 0.62 0.05 0.05 (650) (824) (0.04) (0.08) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.Allregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Is[...]currentlyinschool?Source:ARLS 3.7 PublicGoods WealsoexploreddataonapotentialdirectoutputoftheBRAKDPprogram--levelsofpublic goodsinvillagesinAceh.Welookedfirstattheaveragenumberofavarietyofdifferenttypes of public goods at the time of the village head survey in 2008. As Table 3.11 shows, there is weak evidence that BRAKDP is associated with higher levels of public goods. It appears that treatmentcommunitiesarelikelytohavemoreTPAschoolsandmosques,buttheseresultsare notstatisticallystrong.Forothertypesofpublicgoods,thereisnostrongevidenceofapositive or adverse program impact. This is unsurprising given the relatively small proportion of BRA KDPfundsthatcommunitieschosetospendonpublicgoods. 43 TABLE3.11:COMMUNITYPUBLICGOODS Sharerespondentsreportingthere Individualsin Individualsin Simple Difference is...intheirvillage control treatment difference accountingfor communities communities (se) selection (N) (N) (se) Elementaryschool 0.51 0.44 0.07 0.07 (243) (220) (0.06) (0.10) Junior/seniorhighschool 0.20 0.12 0.08** 0.13 (243) (220) (0.04) (0.09) TPA(AlQuraneducation) 0.81 1.15 0.34*** 0.31 (243) (220) (0.13) (0.21) Madrasah(Islamichighschool) 0.12 0.15 0.03 0.01 (243) (220) (0.04) (0.07) Pesantren(Islamicboardingschool) 0.28 0.38 0.10 0.08 (243) (220) (0.06) (0.11) Mosque/church 0.71 0.63 0.08 0.39 (243) (220) (0.09) (0.21) Villagemeetinghall 0.27 0.22 0.05 0.06 (243) (220) (0.05) (0.10) Puskesmas(healthcenter) 0.40 0.46 0.06 0.09 (243) (220) (0.06) (0.10) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Howmanycompleted[...]arethereinthisvillage now?Source:ARLS Overall,thedatasuggestthatBRAKDPhadlittleeffectonthelevelofpublicgoodsprovisionin treatment communities. While in principle BRAKDP could have had an important impact on theseoutcomes,ifcommunitieshadchosetospendtheirmoneyinthisway,theresultshere are consistent with the fact that in practice BRAKDP money was largely used for private economicactivities. 3.8 WelfarePerceptions We close our discussion of welfare effects by examining perceptions of welfare (Table 3.12). The evidence suggests that a large share (about onethird) of individuals in treatment communitiesismorelikelytoclaimthattheirlivingconditionsaresubstantiallybetterthanthe year before. These results are especially strong for conflict victims. However, they are not robustacrossalternativespecifications. 44 TABLE3.12:SUBJECTIVEPERCEPTIONSOFWELLBEING Shareofindividualswhoratetheir Individualsin Individualsin Simple Difference livingconditionsas"better"or control treatment difference(se) accounting "muchbetter"than12months communities communities forselection earlier (N) (N) (se) All 0.33 0.34 0.01 0.07 (1225) (1090) (0.03) (0.06) 0.3 0.36 0.06 0.18* Conflictvictims (455) (528) (0.04) (0.09) Mostconflictaffected 0.32 0.35 0.03 0.16 (282) (369) (0.05) (0.11) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Lookingback,howdoyourateyourliving conditionsnowcomparedtotwelvemonthsago?Source:ARLS 3.9 ConclusionsonWelfare Overall, we find strong evidence of positive program impacts of BRAKDP with respect to the welfare of beneficiaries and target villages. The data suggests that BRAKDP has resulted in a significant drop in poverty, a large increase in the assets held by conflict victims, and strong perceptions of improved welfare by conflict victims. Land use for program recipients has increasedsignificantlycomparedtothatincontrolareas.Wefindlesseffectoftheprogramon infrastructure, reflecting the decisions made by communities regarding how best to use BRA KDPfunds.Fundswereusedlargelytopromoteeconomicactivitiesthroughprivatetransfersof cash to individuals. On other welfare measures, such as health and education, there is less evidenceofgainsinprojectareas. 45 4. IMPACTSONSOCIALCOHESION Sustaining peace in Aceh will require the reintegration of former combatants and internally displacedpersons(IDPs)intoruralvillages.Wherereintegrationissmooth,socialcohesioncan berebuiltprovidingabasisforavirtuousreinforcingcycleofdevelopmentandsecurity. OnemajormotivationforthecommunitybasedBRAKDPprogramisthatitmayincreasesocial cohesion. Policymakers and practitioners have placed considerable hope in the potential for participatory development projects to improve a community's conflict management capacity (Chopra and Hohe 2004). Indeed, previous research on KDP in other parts of Indonesia has shownstrongimprovementsinsocialrelations,increasesinparticipationinlocalciviclife,and improvements in local conflict resolution capacity in project areas (Barron, Diprose and Woolcock 2006). Indeed, unlike the vast majority of development projects, KDP remained in Aceh during the conflict, suggesting some degree of robustness to violence. By introducing inclusiveandcollectivedecisionmakingandproblemsolving,BRAKDPmaypositivelyenhance the reintegration of different community members such as exGAM combatants, PETA (anti separatist groups), and IDPs. Conversely, BRAKDP may create new tensions or exacerbate existingconflictsortensionsbyintroducingcompetitionoverlimitedresources.Iftheprocessis seenasunfairandnontransparent,favoringcertaingroupsofpeople,itcouldstrengthenthe divide among different groups and potentially create conflicts. Does participation in BRAKDP strengthenorweakensocialcohesion? 4.1 SocialAcceptance ToexaminewhetherparticipationinBRAKDPincreasedthesocialacceptanceofmarginalized groups,wegeneratedastandardversionofasocialdistancescale,askingaboutarespondent's comfort level with particular groups as members of the village, participants in community associations, asleadersofthecommunity,asclosefriends,andaskinbymarriage.Wefocus ourattentionontwomarginalizedgroupsinparticular:excombatantsandinternallydisplaced people(IDPs). AsonecanseeinTable4.1,expressedlevelsofcomfortwithbothgroupsareuniformlyhighin bothtreatmentandcontrolcommunities.Thetablereportstheshareofindividualsexpressing comfortwithmembersofthemarginalizedgroupinalloftherolesmentionedabove. Even with levels of acceptance very high overall, there are some differences between individualsintreatmentandcontrolcommunities.Nearlyalloftheestimatedprogrameffects are negative, suggesting lower levels of acceptance of these marginalized groups in communities exposed to BRAKDP. The results are particularly powerful for excombatants: conflictvictimsintreatmentcommunitiesarelessacceptingofexcombatantswheretheBRA KDPprogramwasimplementedthantheywouldlikelyhavebeeniftheyhadnotreceivedthe program.Thisistruedespitethefactthatvictimsintreatmentcommunitiesaregenerallymore acceptingofexcombatantsoverall;itsuggeststhatbeforetheprogramsuchindividualswere evenmoreacceptingofexcombatantsthantheyaretoday(seeFigure4.1).Theresultsonhow conflict victims perceive excombatants are, however, not robust across all alternate 46 specifications.Weexplorepotentialreasonsfordecreasedacceptanceofformercombatantsin Section6. TABLE4.1:SOCIALACCEPTANCE Individualsin Individualsin Simple Difference control treatment difference accounting communities communities (se) forselection (N) (N) (se) Sharethatreportfullwillingnesstoacceptexcombatantsinallroles All 0.77 0.86 0.09* 0.08 (1211) (1076) (0.05) (0.06) Conflictvictims 0.83 0.89 0.06 0.18** (452) (522) (0.04) (0.09) Mostconflictaffected 0.80 0.89 0.08* 0.19* (279) (364) (0.05) (0.11) Villageheads 0.80 0.87 0.07* 0.19*** (239) (217) (0.04) (0.07) SharethatreportfullwillingnesstoacceptIDPsinallroles All 0.68 0.63 0.05 0.08 (1223) (1088) (0.04) (0.07) Conflictvictims 0.69 0.65 0.03 0.11 (455) (528) (0.06) (0.11) Mostconflictaffected 0.65 0.66 0.02 0.03 (282) (369) (0.07) (0.13) Villageheads 0.73 0.77 0.04 0.06 (242) (220) (0.04) (0.08) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Iwouldnowliketoaskyousomequestionsabout yourfeelingstowarddifferentcategoriesofpeople.Should[...]befullywelcomedinthisvillage?Allowedmembershipin communityassociations?Allowedtobeamongtheleadersofthevillage?Amongyourclosefriends?Welcomedintoyour familythroughmarriage?Source:ARLS Alsoconsistentwiththeresultsfromoursampleofhouseholds,thereisevidencethatvillage heads in treatment communities are less accepting of excombatants, although levels of acceptance are very high in both project and nonproject areas. This suggests an adverse impactoftheprogramonthewillingnessofcommunityleaderstowelcomeexcombatants.We illustrate the relation for one component of this index--the willingness to accept ex combatants into one's family through marriage--in Figure 4.2. These negative results are robustacrossalternativespecifications. 47 FIGURE4.1:CONFLICTVICTIMSACCEPTANCEOFEXCOMBATANTS Note:Thefigureshowstheaveragewillingnesstoacceptanexcombatantintoone's communityasafunctionofselectioncriteria.Villageswithvaluesonorabove0onthe horizontalaxiswereselectedtoparticipateinBRAKDP.Source:ARLS FIGURE4.2:VILLAGEHEADACCEPTANCEOFEXCOMBATANTS Note:Areastotherightof0onthehorizontalaxiswereselectedtoreceive treatmentthoseontheleftwerenot.Acceptancelevelsaretypicallyhigherinareas thataremorelikelytogainaccesstotheprogram;butconditioningonthisselection, acceptanceislowerinprogramareas.Source:ARLS We also asked about the way in which different groups are treated in the context of communitydecision making processes: do some groups benefit more than others when decisionsaremadeabouthowtoallocateresources?Table4.2suggeststhatthepoor,conflict affected,elderly,andIDPstendtodobetterthanothergroupsinthevillage.Incontrast,people whoarewellconnectedwiththevillagegovernmentandwithKPA(theorganizationsetupby exGAMcombatantstohelpthemtransitionintoapoliticalmovement)dorelativelypoorly. 48 InexaminingtheimpactofBRAKDP,thereisstrongevidencethatthosemostaffectedbythe conflict are perceived as benefiting more from community decisionmaking in areas that receivedtheprogramthanincontrolcommunities.Thisresultisapparentinsimpledifferences andismorepowerfulonceweaccountforselectioneffect.ItsuggeststhatBRAKDPmayhave helpedconflictvictimsplayalargerroleinvillagedecisionmaking.Theseresultsarerobustin oneofthetwoalternatespecifications. Strikingly, there is no evidence that excombatants are perceived as benefiting disproportionatelyfromvillagemeetingsinBRAKDPcommunities.TheadverseimpactofBRA KDP on acceptance of excombatants is therefore likely not the result of former fighters exertingundueinfluenceintheprocessandcouldinsteadsimplyreflectalowertolerancefor formerfighters(discussedfurtherinSection6). TABLE4.2:GROUPSTHATBENEFITMORETHANOTHERS Individualsin Individualsin Simple Difference control treatment difference accounting communities communities (se) forselection Sharesayingthat(...)benefitmore (N) (N) (se) Thosemostaffectedbyconflict 0.44 0.56 0.11*** 0.22** (1,213) (1,086) (0.05) (0.09) Conflictvictimssubsample 0.42 0.57 0.15** 0.21* (453) (527) (0.06) (0.12) Therelativelypoor 0.58 0.55 0.03 0.10 (1,220) (1,086) (0.04) (0.08) Conflictvictimssubsample 0.54 0.55 0.02 0.24** (454) (527) (0.06) (0.12) OlderPeople 0.41 0.45 0.04 0.11 (1,219) (1,086) (0.04) (0.07) Friendsandfamilyofthevillageleader 0.22 0.21 0.00 0.01 (1,211) (1,075) (0.03) (0.06) People that are wellconnected with 0.17 0.17 0.00 0.02 localgovernment (1,208) (1,072) (0.04) (0.08) People that are well connected with 0.13 0.15 0.02 0.02 KDPfacilitators (1,206) (1,070) (0.04) (0.08) People that are wellconnected with 0.12 0.14 0.03 0.01 KPA (1,194) (1,071) (0.03) (0.07) ExGAMcombatants,PETAmembers 0.17 0.19 0.02 0.04 (1,197) (1,071) (0.04) (0.07) IDPs 0.38 0.43 0.06 0.09 (1,205) (1,077) (0.04) (0.07) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending 49 capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Whenthecommunityhastomakeadecisionabout howtoallocateresourcesinthevillage,sometimessomegroupsbenefitmorethanothers.Generally,doyouthinkthefollowing peopledoespeciallywellorespeciallybadlyrelativetootherpeopleinthegroup?Source:ARLS 4.2 SocialTensions By emphasizing transparency and participatory decisionmaking, BRAKDP aimed to provide communities with mechanisms for handling community level tensions. We asked a series of questionstorespondentsabouttheextentofsocialdivisionintheirvillage. TABLE4.3:SOCIALTENSIONS Individualsin Individualsin Simple Difference control treatment difference accounting Towhatextentdothefollowing communities communities (se) forselection differencestendtodividepeoplein (N) (N) (se) yourtown? Receivedspecialassistancefrom 0.45 0.41 0.04 0.02 government (1,224) (1,090) (0.04) (0.07) Betweenrichandpoor All 0.21 0.22 0.01 0.02 (1,224) (1,090) (0.03) (0.05) Conflictvictims 0.20 0.26 0.05 0.05 (455) (528) (0.04) (0.09) Mostaffected 0.21 0.23 0.03 0.07 (282) (369) (0.04) (0.10) Menandwomen 0.06 0.08 0.02 0.03 (1,224) (1,090) (0.03) (0.05) Youngerandoldergenerations 0.04 0.05 0.01 0.00 (1,224) (1,090) (0.02) (0.03) Returnees/IDPsandvillagers 0.02 0.02 0.00 0.02 (1,224) (1,090) (0.01) (0.02) Newmigrantsandvillagers 0.03 0.03 0.00 0.03* (1,225) (1,089) (0.01) (0.02) Excombatantsandvillagers 0.03 0.04 0.01 0.02 (1,224) (1,088) (0.01) (0.02) Differentethnicgroups 0.03 0.04 0.01 0.05** (1,225) (1,090) (0.02) (0.02) Villageandneighboringvillage All 0.06 0.06 0.00 0.09*** (1,225) (1090) (0.02) (0.03) ConflictVictims 0.03 0.05 0.02 0.11** (455) (528) (0.02) (0.04) Mostaffected 0.04 0.05 0.02 0.16** (282) (369) (0.02) (0.06) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Source:ARLS The most prevalent division we find in the data is between those who have received special assistancefromgovernmentandthosewhohavenot.Thesecondgreatestdivisionisbetween 50 richandpoor.Thereissomeevidencethatdivisionsbetweennewmigrantsandvillagersand amongethnicgroupsarehigherinBRAKDPcommunities(thesefindingsarerobustinsomebut not all alternative specifications), although very small proportions of the population reports these divisions. While conflict between the village and neighboring villages is not a major sourceofdivisioningeneral,thereisevidencethatitisneverthelessagreatersourceofdivision in BRAKDP treatment villages. Again, the results are not very robust to alternative specifications. 4.3 ConflictResolution In strengthening cooperation and collective decisionmaking, BRAKDP is hypothesized to increase the ability of communities to resolve tensions and increase satisfaction with how problemsinthevillageareresolved. Table4.4looksatwhetherthedivisionsoutlinedabovehaveescalatedtophysicalviolencein the past six months. It is hypothesized that BRAKDP reduces the escalation of tension to violence by strengthening the capacity of individuals to resolve conflict through formal mechanisms.Itisworthnotingfirstthatdivisionsrarelyescalatetoviolenceinbothtreatment and control communities. Nevertheless, there is no evidence that exposure to BRAKDP strengthenedtheabilityofcommunitiestopreventescalation. TABLE4.4:ESCALATINGTOVIOLENCE Individualsin Individualsin Simple Difference control treatment difference accounting communities communities (se) forselection Share of divisions that escalated to (N) (N) (se) violenceinthepastsixmonths Received special assistance from 0.07 0.12 0.05 0.04 government (553) (420) (0.05) (0.08) Betweenrichandpoor 0.09 0.13 0.04 0.01 (237) (200) (0.07) (0.09) Menandwomen 0.15 0.11 0.05 0.12 (55) (55) (0.10) (0.16) Youngerandoldergenerations 0.12 0.09 0.03 0.02 (54) (45) (0.06) (0.08) Returnees/IDPsandvillagers 0.08 0.00 0.08 0.18 (17) (14) (0.08) (0.15) Newmigrantsandvillagers 0.09 0.00 0.09 0.24 (35) (24) (0.06) (0.15) Excombatantsandvillagers 0.11 0.07 0.04 0.11 (35) (37) (0.08) (0.16) Differentethnicgroups 0.16 0.06 0.10 0.06 (32) (29) (0.09) (0.27) Villageandneighboringvillage 0.04 0.02 0.02 0.02 (61) (45) (0.03) (0.05) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Source:ARLS 51 Table 4.5 presents evidence of whether individuals feel that problems in the village are normallyresolvedsatisfactorilyorwhethertheytendtoendure.Overall,thedatapointtohigh levels of satisfaction with problemsolving. But there is no evidence that communities that receivedBRAKDPexhibitgreatersatisfactionthanthosethatdidnot. TABLE4.5:CONFLICTRESOLUTION Individualsin Individualsin Simple Difference Shareagreeingthatproblemsinthe control treatment difference accounting villagearenormallyresolved communities communities (se) forselection satisfactorily (N) (N) (se) All 0.81 0.81 0.00 0.05 (1,208) (1,077) (0.03) (0.05) Conflictvictims 0.82 0.81 0.01 0.09 (449) (522) (0.04) (0.07) Mostconflictaffected 0.80 0.83 0.04 0.02 (277) (366) (0.04) (0.10) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Source:ARLS 4.4 CollectiveEfficacy Inseekingtoeasetensionsamonggroups,onegoalofBRAKDPistoimprovethecapacityof villages to act collectively. By strengthening collective action, BRAKDP communities could be moreeffectiveininitiatingorsecuringotherprojectsforthebenefitofthecommunity. TABLE4.6:COLLECTIVEEFFICACY Sharereportingthatinthepastsix Individualsin Individualsin Simple Difference months,therehasbeena(nonBRA control treatment difference accounting KDP)projectinvolvingthe communities communities (se) forselection communityto: (N) (N) (se) Buildorrebuildaschool 0.32 0.29 0.03 0.03 (1,221) (1,089) (0.05) (0.08) Buildorrepairaroad 0.44 0.49 0.04 0.08 (1,223) (1,089) (0.04) (0.08) Digorrepairawell 0.19 0.18 0.01 0.05 (1,221) (1,090) (0.04) (0.07) Organizesecurity 0.20 0.11 0.08** 0.00 (1,224) (1,088) (0.04) (0.06) Increaseagriculturalproductivity 0.25 0.26 0.02 0.07 (1,219) (1,086) (0.03) (0.07) Build or rebuild a meeting hall or 0.67 0.73 0.07* 0.08 mosque (1,223) (1,089) (0.04) (0.07) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Source:ARLS 52 Table 4.6 presents evidence of whether BRAKDP communities had more village projects ongoing than nonBRAKDP communities. Simple differences suggest that BRAKDP communitieswerelesslikelytoorganizeinitiativestoincreasesecurityandmorelikelytobuild newmosques,buttheseresultsdonotsurviveonceselectioneffectsaretakenintoaccount. Table 4.7 offers another way of presenting the results. Respondents were asked about the number of nonBRAKDPprojects initiated in their community in theprevious six months; for thosewhereaprojectwasinitiatedtheywerefurtheraskedtospecifywhetheritwasinitiated bythecommunity,bythegovernmentorbyaninternationalorganization.Toexaminewhether BRAKDPinducedhigherlevelsofcommunitypublicgoodsprovision,theanalysisinTable4.7 focuses on the share of communityinitiated projects. Efforts to provide security and build a mosqueormeetinghallarethemostcommonactivitiesthatarecommunityinitiated.Thedata suggest,however,thatBRAKDPcommunitieswerelesslikelytoinitiateaprojecttorebuilda meetinghallormosquethancontrolcommunitiesinthepastsixmonths(althoughthisresultis notrobusttoalternativespecifications). TABLE4.7:COMMUNITYLEADERSHIPINPUBLICGOODSPRODUCTION Shareofprojectsthatwere Individualsin Individualsin Simple Difference communityinitiated(forthose control treatment difference accounting whereanonBRAKDPprojectwas communities communities (se) forselection initiatedinthepast6months): (N) (N) (se) Buildorrebuildaschool 0.18 0.17 0.02 0.02 (304) (247) (0.07) (0.10) Buildorrepairaroad 0.21 0.25 0.05 0.03 (528) (499) (0.04) (0.08) Digorrepairawell 0.25 0.27 0.02 0.14 (210) (163) (0.08) (0.12) Organizesecurity 0.81 0.76 0.06 0.05 (205) (137) (0.07) (0.14) Increaseagriculturalproductivity 0.16 0.21 0.05 0.11 (349) (258) (0.05) (0.08) Build or rebuild a meeting hall or 0.81 0.74 0.07 0.17** mosque (775) (756) (0.04) (0.07) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Source:ARLS 4.5 AssociationalLife AfinalapproachtoassessingtheimpactofBRAKDPoncollectiveefficacyexplorestherichness ofassociationallifeinAcehnesevillages.IsitthecasethatexposuretoBRAKDPhasledtothe sproutingofneworganizationsorincreasedtheinvolvementofvillagersinexistingones? Table4.8reportsontheexistenceofassociationsofdifferenttypesintreatmentcommunities, as described by village heads. Most villages have associations that focus on farming or other productiveactivities,religion,youthandsports,andwomen. 53 TABLE4.8:ASSOCIATIONALLIFE(BYVILLAGEHEADS) Individualsin Individualsin Simple Difference control treatment difference accountingfor communities communities (se) selection (N) (N) IV (se) Farmer's group / professional / trader'sassociation/union 0.75 0.79 0.03 0.06 (243) (220) (0.04) (0.08) Credit/financegroup 0.21 0.29 0.08* 0.08 (243) (220) (0.04) (0.08) Communitydevelopment/selfhelp 0.06 0.04 0.03 0.03 (243) (220) (0.02) (0.04) Religiousgroup 0.93 0.97 0.04** 0.03 (243) (220) (0.02) (0.03) Cultural/ethnicassociation 0.21 0.28 0.07* 0.09 (243) (220) (0.04) (0.08) Politicalgroup 0.15 0.20 0.05 0.07 (243) (220) (0.04) (0.06) Youthorsportsgroup 0.91 0.93 0.02 0.01 (243) (220) (0.03) (0.05) Women'sgroup 0.94 0.92 0.02 0.12** (243) (220) (0.02) (0.05) KPA 0.03 0.12 0.09*** 0.06 (243) (220) (0.02) (0.04) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Areanyofthefollowingtypesofassociations activeinyourvillage?Listincludesfarmer'sgroup/traders'association/union/professionalassociation,credit/financegroup, communitydevelopment,religiousgroup,cultural/ethnicassociation,politicalgroup,youth/sportsgroup,women'sgroup. Source:ARLS Thereissomeevidenceofagreaterdiversityofassociationallifeincommunitiesthatreceived BRAKDP, although only the result on women's groups survives accounting for selection. Althoughthelikelihoodthatvillageshavewomen'sgroupsisveryhighoverall,thereisstrong evidence that BRAKDP increases this likelihood (a result that is robust to alternative specifications). Finally we examine the extent of participation in associational life by community members. Table4.9reportstheshareofexistingassociationsinwhichindividualsparticipateasmembers. There is no evidence that individuals are more actively engaged in association life in communitiesthatbenefitedfromBRAKDP. 54 TABLE4.9:INVOLVEMENTINASSOCIATIONALLIFE Individualsin Individualsin Simple Difference control treatment difference accounting Share of existing associations in communities communities (se) forselection whichindividualsparticipate (N) (N) (se) All 0.40 0.42 0.02 0.04 (1,221) (1,084) (0.02) (0.04) Conflictvictims 0.47 0.44 0.03 0.03 (453) (523) (0.03) (0.06) Mostconflictaffected 0.47 0.44 0.03 0.04 (280) (364) (0.03) (0.08) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Ifoneexists,areyouamemberofanyofthe followingtypesofassociations?Listincludesfarmer'sgroup/traders'association/union/professionalassociation, credit/financegroup,communitydevelopment,religiousgroup,cultural/ethnicassociation,politicalgroup,youth/sports group,women'sgroup.Source:ARLS 4.6 ConclusionsonSocialCohesion Overall,acceptanceofconflictrelatedgroups,especiallyexcombatantsandIDPsishighoverall in both treatment and control locations. Notably, however, areas that received BRAKDP demonstratedsignificantlylessacceptanceofexcombatants,especiallyamongconflictvictims and community leaders. Interestingly, there is no evidence, however that excombatants are perceivedasbenefittingdisproportionatelyfromvillagemeetingsinBRAKDPcommunities.The adverseimpactofBRAKDPonacceptanceofexcombatantsisthereforelikelynottheresultof formerfightersexertingundueinfluenceintheprocessandcouldinsteadsimplyreflectalower toleranceforformerfighters,atopicwereturntoinSection6. BRAKDPhadsomepositiveimpactonsomemeasuresofsocialcohesion,suchastheexistence ofwomen'sgroups.Forthemostpart,however,thisanalysisrevealslittleimpactofBRAKDP on social cohesion as capturedforexamplebymeasuresof as conflict resolution, community publicgoodsprovision,andinvolvementinassociationallife. OneexplanationforthelackofgainsmaybetheshorttimeperiodBRAKDPranfor.Previous workonKDPinIndonesiahasshownthatincreasesincollectiveactionandparticipationtendto occurafterthreetofourprogramcycles(Barron,DiproseandWoolcock2006).BRAKDPranfor only one year. If so, this analysis suggests that sustained involvement may be a necessary conditionforprogrameffectiveness. 55 5. IMPACTSONTRUSTINLOCALGOVERNMENTANDSTATESOCIETYRELATIONS In a postconflict context, it is crucial for the government to win people's trust to regain legitimacy.ThisisparticularlytrueinAcehwherethegovernmentisassociatedwithprevious oppression and where people are generally skeptical of the government's intentions and capacity(MSR2009). It was hypothesized that, if properly implemented, BRAKDP might provide a channel for the government(atalllevels)todemonstratetangiblyitsabilitytoaddressthepopulation'sneeds. This might in turn generate greater faith and trust in village and governmental institutions, reinforcingthetransitionfromwartopeace.Conversely,iftheprogramwasnotimplemented fairly, transparently and in a timely manner, it risked further alienating the population, potentiallycreatingspaceforantigovernmentelementstowinthemover.Doesparticipation inBRAKDPincreasetrustinlocalgovernmentandstrengthenstatesocietyrelations? 5.1 TrustinCommunityDecisionMaking Asacommunitydrivendevelopmentproject,BRAKDPprovidedvillagerswiththeopportunity to participate directly in decisions over how funds would be spent. We explore whether exposure to this participatory methodology has changed how decisions are made within communitiesandhowindividualsbelievedecisionsshouldbemade. Byexposingcommunitiestoparticipatorydecisionmaking,BRAKDPishypothesizedtoleadto higherlevelsofoverallsatisfactionwithhowdecisionsaremadeatthevillagelevel.Weseein Table5.1thatlevelsofsatisfactionwithvillagedecisionmakingarehighacrossallkeygroupsof interest.Thereisnosignificantdifferenceinsatisfactionwithdecisionmakingacrosstreatment andcontrolareas. TABLE5.1:SATISFIEDWITHDECISIONS Individualsin Individualsin Simple Difference control treatment difference accountingfor communities communities (se) selection (N) (N) (se) All 0.91 0.91 0.00 0.02 (1,217) (1,079) (0.02) (0.04) Conflictvictims 0.89 0.92 0.03 0.02 (453) (525) (0.03) (0.05) Mostconflictaffected 0.88 0.94 0.07* 0.03 (280) (368) (0.03) (0.07) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspendingcapacity, theirquadraticandcubedterms,andtheirinteraction.Question:Overall,howsatisfiedareyouwiththewaythatdecisionsthat affectallcommunitymembersaremadeinyourvillage?Source:ARLS 56 Table 5.2 describes the groups that individuals believe play the biggest role in community decisionmaking. The largest share feels that villagers play the greatest decisionmaking role, with elders/traditional leaders and village government playing the second and third largest roles, respectively. However, there is evidence that, accounting for selection, individuals feel villagersplaylessofanimportantroleinBRAKDPcommunities(aresultthatisrobustinboth alternativespecifications),despitetheprogram'semphasisonparticipatorymethodologies. TABLE5.2:VILLAGERS'ROLEINDECISIONMAKING Individualsin Individualsin Simple Difference control treatment Difference Accountingfor communities communities (se) Selection (N) (N) (se) Sharebelievingthatvillagersdoplaythemostimportantrole All 0.44 0.42 0.02 0.16* (1,224) (1,082) (0.04) (0.08) Conflictvictims 0.46 0.46 0.00 0.25** (455) (526) (0.05) (0.12) Mostconflictaffected 0.43 0.42 0.01 0.17 (282) (367) (0.06) (0.13) Sharebelievingthatvillagehead/governmentdoesplaythemostimportantrole All 0.15 0.16 0.01 0.04 (1,224) (1,082) (0.02) (0.05) Conflictvictims 0.16 0.15 0.02 0.02 (455) (526) (0.04) (0.08) Mostconflictaffected 0.16 0.16 0.00 0.10 (282) (367) (0.05) (0.09) Sharebelievingthatelders/traditionalleadersdoplaythemostimportantrole All 0.40 0.41 0.01 0.10 (1,224) (1,082) (0.04) (0.09) Conflictvictims 0.36 0.38 0.02 0.18 (455) (526) (0.06) (0.13) Mostconflictaffected 0.38 0.40 0.02 0.06 (282) (367) (0.07) (0.14) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspendingcapacity, theirquadraticandcubedterms,andtheirinteraction.Tableonlyreportsresultsforresponsesreportedbyatleast2percentin thepopulation. Question:Imaginethatthevillagereceivesfundstoinvestinimprovinginfrastructureinthevillage.Adecisionneedstobemade abouthowthefundsshouldbespent.Whowouldlikelyplaythebiggestroleinmakingthedecision?Source:ARLS Table 5.3 examines whether individuals believe their participation in village meetings is valuable. Specifically, we ask whether respondents believe they are influential in shaping the outcomesofcommunitydecisionmakingprocesses.Overall,thedatasuggestthatbeliefsabout 57 efficacy are relatively low. Moreover, the results suggest that individuals (especially conflict victims) in treatment communities believe they are less influential than individuals in control communities,althoughtheseresultsdonotsurviveaccountingforselection. TABLE5.3:POLITICALEFFICACY Sharethatbelievetheyplayan Individualsin Individualsin Simple Difference influentialroleindecisionsatleast control treatment difference accounting someofthetime communities communities (se) forselection (N) (N) (se) All 0.38 0.35 0.03 0.01 (1,193) (1,063) (0.03) (0.06) Conflictvictims 0.43 0.33 0.10** 0.02 (441) (522) (0.05) (0.10) Mostconflictaffected 0.39 0.30 0.09 0.04 (273) (367) (0.05) (0.12) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:Whendecisionsaremadeonissuesthataffectall villagers,doyoufeelthatyoupersonallyplayaninfluentialroleinaffectingtheoutcome,forinstance,whenyouspeakat villagemeetingsortrytopersuadeothers?Source:ARLS 5.2 TrustinGovernment One possible outcome of the BRAKDP program is higher levels of awareness and trust in government, as the program was an attempt by government to delivery on key needs in the immediateaftermathoftheconflict. Behavioralmeasures Table5.4reportstheresultsofanattempttomeasurehowwillingindividualsaretosupport theactivitiesofthedistrictgovernmentusingabehavioralmeasure.Tomeasuretrustinlocal governmentdirectly,weimplementedaprocedureinwhichrespondentsweregivenRp.10,000 (around US$ 1) and asked to decide (in private) how much they wished to contribute to developmentactivitiesadministeredbythedistrictgovernmentandhowmuchtheywishedto keepforthemselves.Themoneytheywantedtosendtothedistrictgovernmentwastobeput inanenvelope;alltheseenvelopeswerelaterdeliveredtothedistrictoffice. The share of funds sent to local government serves as a measure for trust in the ability or willingnessoflocalgovernmenttousefundswell;differencesinwillingnesstocontributecan serve as a measure of program effects. In effect the measure captures the willingness to pay localtaxes.Wefoundthatrespondentstendedtosendabout2025percentoftheendowment, keeping the rest for their personal use. There is, however, no difference in level of trust exhibitedbyindividualsintreatmentandcontrolcommunities,bothinthesimplecomparison andinthecomparisonthattakesaccountofselectioneffects. 58 TABLE5.4:TRUSTINDISTRICTGOVERNMENT ShareofRp.10,000contributedto Individualsin Individualsin Simple Difference districtgovernmentforinvestment control treatment difference accounting inlocaldevelopment communities communities (se) forselection (N) (N) (se) All 0.25 0.24 0.00 0.00 (1,225) (1,090) (0.03) (0.05) Conflictvictims 0.22 0.21 0.01 0.02 (455) (528) (0.03) (0.06) Mostconflictaffected 0.24 0.20 0.04 0.08 (282) (369) (0.03) (0.06) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Exercise:RespondentsweregivenanenvelopewithRp. 10,000andaskedtodecidehowmuchtheywishedtokeepforthemselvesandhowmuchtheywantedtocontributetothe localgovernmentfordevelopmentinthearea.Themoneycontributedwasthentransferredtothedistrictgovernmentoffice. Source:ARLS Attitudinalmeasures Confidenceingovernmentcanalsoberecordedthroughresponsestoattitudinalquestions.We nowturntotrustinthelocalvillageapparatus.Table5.5describeshowindividualsresponded toahypotheticalsituation:ifagrantofRp.100million(aroundUS$10,000)wasgiventothe village,whatsharedotheythinkshouldbemanagedbythevillagegovernmentasopposedto distributed directly to villagers? By this measure, respondents exhibit relatively low levels of trustinthevillageapparatus,generallybelievingthatonlyaboutonethirdofthefundsshould be given to village government. There is no different in responses between treatment and controlcommunities. TABLE5.5:TRUSTINVILLAGEAPPARATUS ShareofRp.100milliongrant Individualsin Individualsin Simple Difference respondentbelievesshouldbe control treatment difference accounting managedbyvillageapparatusas communities communities (se) forselection opposedtodistributeddirectlyto (N) (N) (se) villagers All 0.37 0.36 0.01 0.03 (1,221) (1,084) (0.02) (0.04) Conflictvictims 0.33 0.35 0.02 0.03 (454) (528) (0.03) (0.06) Mostconflictaffected 0.33 0.35 0.02 0.02 (281) (369) (0.03) (0.07) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Question:IfagrantofRp.100millionweremadeavailable tosupportthisvillage,howmuchshouldbemanagedbythevillageapparatustobeusedfordevelopmentprojectstohelpthe villageandhowmuchshouldbedividedupandgivendirectlytoindividualvillagerstouseastheyseefit?Source:ARLS 59 We also asked individuals to report their confidence in a set of actions they could take that mightbeeffectiveinimprovingthesituationintheirvillage.WefocusinTable5.6ontheshare of respondents reporting that subdistrict, district, and provincial authorities could make a differenceinaddressinglocalchallenges. Onlyaboutonethirdofrespondentsbelievethatgovernmentalauthoritiesarebestplacedto improve the situation in the village. However, the survey results suggest that treatment communities, and conflict victims in particular, were more likely to believe this than control communities(aresultthatisnotconsistentwiththeevidenceinthebehavioralgame).These resultsaresignificantlyweakenedintherobustnesstestshowever. TABLE5.6:CONFIDENCEINEFFECTIVENESSOFSUBDISTRICT,DISTRICT,ANDPROVINCIALGOVERNMENT Shareofrespondentsreportingthatsubdistrict, Individualsin Individualsin Simple Difference district,orprovincialauthoritieswouldbeamong control treatment difference accounting the"mosteffective"or"nextmosteffective" communities communities (se) forselection groupstoimprovethesituationinthevillage (N) (N) (se) All 0.31 0.32 0.01 0.17* (1,225) (1,090) (0.04) (0.09) Conflictvictims 0.32 0.32 0.00 0.28** (455) (528) (0.05) (0.14) Mostconflictaffected 0.32 0.34 0.01 0.32** (282) (369) (0.06) (0.15) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Exercise:Respondentswereaskedtorankalistofactions theycouldtaketoimprovethesituationintheirvillage.Tenoptionswerepresentedincludingcomplainingtoavillagehead, subdistrictofficial,districtofficial,provincialofficial,religiousleaders/elders,KPA/GAM;appealingtolocalNGOsor internationalorganizations;expressingopinionduringelections;takingpartinprotests;orresortingtoviolence.Source:ARLS Knowledgeofgovernment Finallyweconsideraseriesofknowledgequestions.Table5.7providesinformationabouthow awarerespondentsareoftheleadersatvariouslevelsofgovernment.Individualswereasked whethertheycouldnametheheadofthesubdistrict,thedistrict,andthegovernor,aswellas inwhichyearthenextelectionwouldbeheld.Theadvantageofthiskindofquestionisthatit can capture effects of increased levels of contact or interaction with government and is not easilysusceptibletomisrepresentation. Overallwefindthatknowledgeofmorecentralpolitics(thedateofthepresidentialelection, thegovernorofAceh)isgreaterthanknowledgeoflocalpolitics(thenamesofdistrictandsub district leaders). Only about onethird of respondents can provide the names of subdistrict heads.Thereisnoevidencethatindividualsintreatmentcommunitiesexhibitgreaterorlesser awarenessofgovernment. 60 TABLE5.7:AWARENESSOFGOVERNMENT Individualsin Individualsin Simple Difference control treatment difference accounting Shareofallindividualswhocan communities communities (se) forselection correctlyname... (N) (N) (se) Theheadofthesubdistrict 0.28 0.33 0.05 0.02 (1,225) (1,090) (0.04) (0.06) Theheadofthedistrictorregency 0.62 0.53 0.09** 0.08 (1,225) (1,090) (0.04) (0.08) ThegovernorofAceh 0.63 0.68 0.05 0.03 (1,225) (1,090) (0.03) (0.06) The year in which the next presidentialelectionwillbeheld 0.61 0.64 0.02 0.04 (1,225) (1,090) (0.04) (0.06) ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Source:ARLS 5.3 AttitudesaboutGovernance BRAKDP also aimed to expose individuals to transparent and participatory decisionmaking. Doesthisexposureleadtogreatersupportfordemocracyintreatmentcommunities?Tofind out,weaskedrespondentsaseriesofquestionsonpotentiallydivisiveissuesofgovernancein theircommunities.TheresultsarereportedinTable5.8. TABLE5.8:SUPPORTFORDEMOCRACY Individualsin Individualsin Simple Difference control treatment difference accounting communities communities (se) forselection (N) (N) (se) Agreethatweshouldbemoreactivein 0.33 0.33 0.00 0.01 questioningtheactionsofourleaders. (1,224) (1,087) (0.04) (0.07) Agreethatleadersshouldnotfavor 0.98 0.98 0.00 0.01 theirownfamilyorethnicgroup. (1,223) (1,086) (0.01) (0.02) Agreethatallshouldbepermittedto 0.22 0.26 0.04 0.08 takepartinimportantdecisions (1,221) (1,084) (0.04) (0.07) Agreethatwomenshouldhavethe 0.32 0.25 0.07* 0.02 samerolesasmeninpositionsof (1,221) (1,088) (0.04) (0.08) authorityinvillagegovernment ***Significantat99%;**Significantat95%;*Significantat90%.Thetablereportsestimatedpopulationmeans,standard errorsandsampleN's(wheretotalsamplesizeis2,315),aswellasthedifferenceforpopulationsintreatmentandcontrol communitiesusingleastsquaresandinstrumentalvariableregressions.IVregressionscontrolforconflictandspending capacity,theirquadraticandcubedterms,andtheirinteraction.Source:ARLS 61 Ingeneral,citizensexhibitdeferencetotheirleaders;onlyabout33percentfeeltheyshouldbe moreactiveinquestioningtheirleaders.Citizensalsoshowlowlevelsofsupportforpermitting everyone to participate in important decisions. Citizens overwhelmingly agree, however, that leadersshouldrepresenteveryone,ratherthanfavortheirownfamilyorethnicgroup.Overall, therearenodifferencesinattitudesforindividualsintreatmentandcontrolcommunities. 5.4 ConclusionsonTrustinLocalGovernmentandStateSocietyRelations WefindlittleevidencethatBRAKDPresultedinchangesinkeyindicatorsofattitudestowards government,suchasawarenessoflocalgovernmentandsupportfordemocracy.Onthewhole villagers feel they play a role in local decisionmaking, although individual perceptions of politicalefficacyarelow.Thereissomeevidencethatvillagersfeellessinfluentialincommunity decisionmaking in BRAKDP areas, despite the program's empowerment goals. Additionally, thereislittleevidenceofenhancedtrustinvillagegovernmentorindistrictgovernmentusinga behavioral measure. BRAKDP does appear to have resulted in higher levels of confidence in subdistrict, district or provincial level government as expressed in survey responses, which could indicate that credit for the program is being attributed to governments at these levels. This might be consistent with the fact that subdistrict level facilitators play a key role in the programandthattheprogramwasmanagedbyBRA'sprovincialoffice.Incontrast,districtlevel governmentplayedamuchsmallerroleinprogramsothelackofanimpactofBRAKDPonfaith indistrictgovernment,asrecordedthroughthebehavioralgame,isnotsurprising.Again,the lackofimprovementsintrustinlocalgovernmentmaybearesultoftheshorttimeperiodof theprogram. 62 6. COMMUNITYDEVELOPMENTANDEXCOMBATANTREINTEGRATION A surprising result of this study is that on a number (although not on all) measures BRAKDP appears to be associated with lower rates of acceptance of excombatants. In particular, we find that BRAKDP contributed to reported lower levels of acceptance of excombatants for bothconflictvictimsandvillageleaders.Onothermeasures,suchasreportsofsocialtensions and on selfreporting of acceptance by excombatants, we do not find evidence of adverse effects,althoughwedonotfindpositiveeffectseither.33 Collectively, these results run against the belief held by many practitioners of demobilization and reintegration programs that social investments in postconflict areas are an especially effectivewayofensuringtheacceptanceofreturningexcombatants.Whilethereisnowbroad consensusaroundthisview,thereisinfactlittleevidencethatacommunityfocusedapproach isequallyorevenmoreeffectivethananapproachfocusedonexcombatants(ofcourse,often programsfocusonbothgroups).Thusitisstrikingthatinthisanalysisofaprogramtargetedat civilian victims we find no evidence of positive effects and some evidence of adverse effects. Whataccountsfortheseadverseimpacts?Anumberofdifferentexplanationsarepossible. 6.1 ExCombatantCaptureofBRAKDPFunds? One plausible explanation is that BRAKDP increased tensions between excombatants and civilians in project areas because of concerns that exTNA, the combatant wing of GAM, appropriated funds or because conflict victims felt that the allocation to excombatants was unfair. The evidence on this hypothesis is mixed. In Table 2.12 we saw that the most common complaintamongconflictvictimswasthatexcombatants(orPETAorIDPs)benefitedtoomuch from the BRAKDP program (13 percent of victims agreed with this statement). Was this the case? Table6.1showsthatexcombatantsreceivedBRAKDPbenefitsatratesapproximatelyequalto those of noncombatants. Individuals in households with excombatants received goods from BRAKDP in about the same proportion as civilian victims and nonvictims with no ex combatants in their households. Most households with excombatants also report as conflict victims. These households fare marginally worse than other conflict victims (the difference is notsignificant),whileexcombatantnonvictimsfaremarginallybetterthanothernonvictims (althoughagainthisdifferenceisnotsignificant). 33 Ingeneralselfreportingofacceptanceuniversallysuggestsfewproblemsofexclusion.Allofthefew(7)casesof reportedproblemsareamongexcombatants,buttheestimatedrateofproblemsisthesameinprojectand comparisonareas. 63 TABLE6.1:SHAREOFEXCOMBATANTSANDCIVILIANSRECEIVINGBENEFITS(PROJECTAREASONLY) Shareof...whoreported Nonconflictvictims Conflictvictims Total receivedbenefits Householdswithnoex 40 44 42 combatants (640,000) (602,000) (1,243,000) Householdswithexcombatants 43 42 42 (3,900) (19,500) (23,500) Total 40 44 42 (645,000) (622,000) (1,267,000) Thetablereportspopulationaverages(withestimatedpopulationN'sbelow).Source:ARLS Thesefindingssuggestthat`capture'didnottakeplaceonalargescale. Whatthenexplainsapparentdissatisfaction?Recallthatbydesign,excombatantsshouldnot have been direct beneficiaries of the BRAKDP process. (Former combatants had their own programsandsoweremeanttohavebeenexcludedfromtheprogram).Forthisreason,any benefits they received could be seen to be too much even if excombatants fared no better thancivilianpopulationsinthedistributionofprogramfunds.Thus,whileitseemsclearthatex combatantsdidnot`capture'BRAKDPfunds,theydidsucceedinobtainingfundsdestinedfor noncombatants at rates approximately equal to noncombatants. This could account for the adverseeffectsoftheprogramonthesocialacceptanceofexcombatants.Ingeneral,thedata supportsthisinterpretation. 6.2 AlternativeExplanations Asecondhypothesisisthatexcombatantsmayhaveresentedthefactthattheywereunable todisproportionatelybenefit.WhileBRAranseparateprogramsforexcombatants,fundswere oftenlateinbeingdisbursed,wereoflowlevelspercapita(WorldBank2006),andmanyex combatantsmissedout(MSR2009).Thismayhaveincreaseddemandsfromexcombatantsfor alargeshareoftheBRAKDPpie.Ifthisledtomisbehavioronthepartofformercombatants (as on occasions it did ­ Morel et. al. 2009) this may in turn have increased community resentmentandhadanegativeimpactonacceptance. AthirdhypothesisisthatBRAKDPhelpedempowercommunitiestostanduptodemandsfrom excombatants. This may have happened through a number of different mechanisms. One is that,byfocusingattentiononanindividual'sstatusasa`conflictvictim',theBRAKDPprogram mayhavefacilitatedaprocessofblameassignmentinwhichformerfighterscametobeseenas responsiblefortheinjuriessufferedbythepopulationduetothewar.Thiscouldhavereduced acceptance. Another mechanism might be that the participatory process gave conflict victims moreconfidencetostanduptothedemandsofformercombatants.Itispossible,forexample, thatbyprovidingindependentrevenuesources,BRAKDPlessenedthedependenceofvillagers onexcombatantstructures,whichcouldleadtoadeclineinacceptanceofexcombatants Wedonothavedatatotestsuchaclaimbutthesupervisionmissionssuggestthat,atleastin someareas,conflictvictimsweremorewillingtostanduptoexcombatantdemands(Morelet. al.2009).Ifthisisthecase,itisdifficultexantetosaywhetherthisispositiveornegativefor sustainablepostconflictstability.InAceh,therehavebeengrowingcomplaintsfrommanythat 64 the influence of former combatants is too great. 34 If programs like BRAKDP increase the counterveiling power of civilian community members, it may create an important check and balance to GAM power, for untrammeled control over resources and decisionmaking could leadtofreshresentment(andpotentiallyuprisings)inthefuture(Barron2009).Conceptualized as such, an increase in minor tensions between former combatants and civilians may be positive for postconflict stability, as long as such tensions do not escalate into more serious conflictastheyrarelyhavetodate.35 ***** Ourdatadonotpermitustoassessthevalidityofthesedifferenthypothesesadequately.We believe however that this would be a fruitful area for further research. More broadly, future work on postconflict reintegration and development programs should take into account the possibilityofadverseinitialimpactsoncommunitycombatanttiesandtoseekfurtherevidence thatmightshedlightonthemechanismsatworkandtheultimateimpactsonpeace. 34 The20062007localexecutiveelections(attheprovincialanddistrictlevels)andthe2009localparliament electionssawmanymembersoftheformerrebelgroupaccedetopositionsofpower(ClarkandPalmer2008;MSR 2009).Thishasgivenmanyformercommandersaccesstoandcontroloverstateresources,forexamplethrough theawardingofconstructioncontracts(Aspinall2009). 35 Wenote,however,thatmeasuresinthedatasuggestthattheinfluenceofexcombatantsandKPAstructuresin villagesismodestatbest.Almostnorespondentsclaimedthatexcombatantsplayedadominantroleinvillage decisionmakingeitherintreatmentorcomparisonareasandonlyabout1percentofrespondentsfeltthat complainingtoexcombatantorKPA(thecommitteeputinplacebyGAMtohelpformercombatantstransitionto civilianlife)structureswouldbeaneffectivewaytoresolvealocalproblem.Onemightalsoexpecthigherlevelsof individualandcollectiveefficacyasaresultofBRAKDP,especiallyamongconflictvictims,ifthishypothesiswere true.However,asreportedearlier,thereislittleevidencethatBRAKDPpositivelyimpactedefficacy. 65 7. CONCLUSIONS Established in the immediate aftermath of the conflict in Aceh, BRAKDP was designed to deliver communitybased reintegration assistance through a participatory mechanism that emphasized local ownership and transparency in decisionmaking. Its particular focus was on conflictvictims,giventhemanyotherinitiativestargetedatexcombatantsattheconflict'send. BRAKDPhadamultiplicityofobjectives,butthreewereofparticularimportance.Theprogram sought:(a)toimprovethematerialwellbeingofitsbeneficiaries,particularlyconflictvictims;(b) torebuildsocialcohesioninAcehnesecommunities,manyofwhichfacedthedifficulttaskof reintegratingexcombatantsandIDPs;and(c)tohelpbuildfaithandtrustthatgovernmental institutionscoulddeliverinmeetingtheneedsofindividualsandcommunities. ThispaperpresentsevidenceontheimpactofBRAKDPonmaterialwellbeing,socialcohesion, andtrustingovernment.Inassessingimpact,weexamineoutcomesagainsta`nullhypothesis' ofnoeffect.Foreachoutcome,weaskwhatwewouldhaveexpectedtoseeifBRAKDPhadno impact and assess how different what we actually observe is from this null outcome. Establishingthecausalimpactoftheprogramonoutcomesinthiswayischallenginghowever becausetheprogramwasimplementedincommunitiesthatdifferedsystematicallyfromthose thatdidnotreceiveBRAKDP.Wedrawonavarietyofstatisticaltechniquestodealwiththese biases that result from selection into the program in order to produce valid estimates of the program'simpact. This complexity, though unfortunate, appears to have been necessary. In many instances, we foundthatourestimatedprogrameffectsdiffersubstantiallyfromwhatwouldbeinferredfrom simplecomparisonsofmeans:asimplecomparisonofoutcomesin`treated'and`control'areas wouldhaveledtoerroneousinferences.Butitalsocomeswithcosts.Inthiscase,wefindthat evenafteraccounting for selection, many of the messages emanating from this research lack the clarity that one seeks in an evaluation of this form. In many cases where impacts are apparent,theyarenotrobusttoalternativespecifications.Thecoremethodologicallessonwe draw, one that is increasingly being appreciated in the evaluation community, is that evaluations are most effective and give clearest answers when evaluation considerations are built into the design of a project. The first best approach is to use some form of randomized intervention,whenpossible.Whenthisisnotpossibleasecondbestapproachistoapplyafully replicable selection rule that determines treatment status from a continuous underlying prioritization variable(s). It is also clear that indepth qualitative work is necessary to satisfactorily understand outcomes and (most importantly) the processes through which they eventuate. Nevertheless,theresultsaresupportiveofthefollowinggeneralconclusions: BRAKDPhadmixedsuccessintargetingconflictvictimsasbeneficiaries.TheBRAKDPprogram soughttotargetconflictvictims.Insomerespects,itwassuccessful.Onaverageconflictvictims fared better than nonconflict victims in large part because the geographic prioritization rule used ensured that subdistricts with a disproportionately large share of conflict victims were 66 more likely to receive the program in the first round and were more likely to receive larger grants in that round. This resulted in significantly higher levels of per capita assistance being provided to conflict victims than to those who were less affected by the conflict. However, targeting within communities was less successful, with conflict victims generally faring no better than nonconflict victims. Indeed, while conflict victims and nonvictims were equally likely to benefit from the program within subdistricts, conflict victims were more likely to reportthattheirpreferredprojectswerenotselectedforimplementation.Sincetheprogram was terminated before completion, about half of the conflict victims in Aceh were never reachedatall. BRAKDP is associated with welfare gains and improvements in perceptions of wellbeing. AlthoughcommunitiesweregivendiscretiontoallocateBRAKDPfundstoprivatetransfersor public goods, the vast majority elected to distribute the cash directly to households. We find evidencethatthesecashtransfersareassociatedwithanincreasedownershipofassetsamong households in general and conflict victims in particular. Moreover, we find evidence that the program contributed to a substantial increase in the farming of productive land (a near doubling for conflict victims). There is also evidence of a substantial program effect on the incidenceofpovertyasreportedbyvillageheads.Thesegains,however,donot(yet)translate into broader welfare improvements, as reflected in health, schooling, and community infrastructure. There is little evidence that BRAKDP generated improvements in social cohesion or improved awareness of or faith in governmental institutions at the village or at higher levels. Levels of social acceptance of excombatants and IDPs, reported social tensions and conflict among groups,andobservedlevelsofcommunityefficacyarebroadlysimilarbetweenthosevillages that received BRAKDP and those that did not, even after accounting for selection. There is some evidence that BRAKDP is associated with less acceptance of excombatants by conflict victimsinprojectareas,althoughthereisnoevidencethatthesetensionsescalatetoviolence. Someofthesefindingsarenotsurprising.BRAKDPranforjustoneyear,perhapslimitingthe gainsinsocialcohesionthatmayhavebeenpossibleiftheprogramhadbeenimplementedfor multiple cycles. 36 The program functioned mainly as a mechanism for transferring private benefits to households. While participatory and transparent communityprocesses were a requiredelementofprojectselectionineachvillage,thevastmajorityofcommunitiesmoved quicklytodistributetheblockgrantsascashpaymentstocommunitymembers.Thesetransfers areassociatedwithwelfaregains(intermsofassetownershipandthecultivationofland)and with reported reductions in the incidence of poverty, but there is not yet evidence that they haveeffectsonbroaderwelfareimprovementsintermsofaccesstohealthoreducation. It is striking that we find evidence for a welfare effect on individuals in BRAKDP areas but, contrarytothepresuppositionsofcommunitybasedprogramsofthisform,verylittleevidence that BRAKDP resulted in higher levels of social cohesion. This finding differs from results 36 Thisleadstotwofurtherconclusions.First,plannerswhoseektouseCDDtypeprogramsinpostconflict contextsshouldthinkaboutdesigningprogramsthatrunovermultiplecycles.Second,methodologically,itmaybe prematuretoassessthesocialandstatesocietyimpactsofprogramsafteroneyearofoperation. 67 reported in other research on the effects of communitydriven programs. In one randomized evaluation of a communitydriven reconstruction program in Liberia, for example, there was evidencethattheCDDprocesscontributedtogainsinsocialcohesion,eventhoughtherewere fewpositiveimpactsonmaterialwelfare(Fearon,HumphreysandWeinstein2009). There are of course many differences between the Liberia case and the case examined here. OnedifferenceofimportanceisthatinLiberia,allprogramfundswereusedforpublicgoods projects whereas in Aceh, village allocations were primarily distributed privately. While the decisionmakingprocesscomponentswerethesameacrossthetwocases,thepredominance ofprivategoodsinAcehmighthavelimitedimprovementsincohesionthatcouldcomefrom the joint production of collective goods. This raises the question: Does the impact of communitydrivenreconstructiondependonwhetherdevelopmentmoneyisspentonprivate orpublicgoods?37 The predominance of cash transfers may also account for the evidence of increased tensions withexcombatantsinBRAKDPcommunities.Totheextentthatexcombatantsbenefitedfrom cash disbursements that were, at least in principle, intended for conflict victims, increasing social tension may have been the result. Alternatively, excombatants might have resented theirinabilitytocontroltheprogramanddisproportionatelybenefit(inparticular,becauseof problems with other reintegration programs targeting excombatants), or the programs may haveempowercivilianconflictvictimstostanduptoexcombatantdemands. These patterns point to two fundamental tensions within the CDD model. First, many of the goalsofCDDmaydependuponprocessesthatarebroughtintoplayconditionalonparticular types of activities being implemented (joint selection of projects, community oversight of implementation,etc.)ButinsofarasaCDDmodelallowscommunitiesfullcontrolovertheuse of finances, these processes can be bypassed which may eliminate the gains in social cooperation and faith in government which CDD is intended to generate. Second, and as observed elsewhere, CDD programs, including those with peacebuilding aims, can lead to tensionsbetweengroupsthroughtheirpromotionofcompetitionoverfiniteresources.Inthe longruntheymayleadtoastrongerbasisforpeace,throughempoweringgroupsandbuilding local institutions. But in the short run they can lead to social divisions. Weighing these (potential)shortandlongrunimpactsisimportantinpostconflictenvironments. 37 IninitialexplorationofthishypothesisusingtheAcehdata(comparingoutcomesbetweenareasinwhich villagesoptedforpublicgoodstooutcomesinareaswherevillagesoptedforprivategoods),wedonotfind supportforthehypothesisthatadverseoutcomesareduetothefocusonprivategoods.Placesthatselected groupgoodsweresignificantlylesslikelytobeacceptingofexcombatants(theeffectisalsonegativebutweaker inareasthatselectedprivategoodsonly).Ononlyonemeasure(participationinassociations)areeffectsmore positiveforgroupsthatengagedinsomepublicgoodsproduction.Theconclusionswecandrawfromthisdataare howeverlimitedsincecommunitiesselfselectintopublicorprivategoodsprojects. 68 REFERENCES Alatas, Vivi (2005). "An Evaluation of the Kecamatan Development Program". Jakarta: World Bank. Aspinall,Edward(2009)."CombatantstoContractors:ThePoliticalEconomyofPeaceinAceh." Indonesia87:April. Barron, Patrick (2009). "The Limits of DDR: Reintegration Lessons from Aceh" in Small Arms Survey,SmallArmsSurveyYearbook2009.Cambridge:CambridgeUniversityPress. Barron,PatrickandAdamBurke(2008)."SupportingPeaceinAceh:DevelopmentAgenciesand InternationalInvolvement".PolicyStudiesNo.47.Washington,D.C.:EastWestCenter. Barron, Patrick, Rachael Diprose and Michael Woolcock (2006). "Local Conflict and DevelopmentProjectsinIndonesia:PartoftheProblemoraPartofaSolution."PolicyResearch WorkingPaperNo.4212.Washington,D.C.:WorldBank. Chopra,JaratandTanyeHohe(2004)."ParticipatoryIntervention."GlobalGovernance10:289 305. Clark, Samuel and Blair Palmer (2008). "Peaceful Pilkada, Dubious Democracy." Indonesian SocialDevelopmentPaperNo.11.Jakarta:WorldBank. Cliffe, Sarah, Scott Guggenheim and Markus Kostner (2003). "CommunityDriven Reconstruction as an Instrument in WartoPeace Transitions." Conflict Prevention and ReconstructionWorkingPaperNo.7.Washington,D.C.:WorldBank. Fearon, James, Macartan Humphreys, and Jeremy Weinstein (2009). "Can Development Aid ContributetoSocialCohesionAfterCivilWar?EvidencefromaFieldExperimentinPostConflict Liberia."AmericanEconomicReviewPapers&Proceedings99(2):287­291. Humphreys, Macartan and Jeremy Weinstein (2007). "Demobilization and Reintegration." JournalofConflictResolution51(4):531567. Internal Crisis Group (ICG) (2007). "Aceh: PostConflict Complications." Asia Report No. 139. Jakarta/Brussels:ICG. InternationalOrganizationforMigration(IOM)andHarvardUniversity(2007).APsychological NeedsAssessmentofCommunitiesin14ConflictAffectedDistrictsinAceh.Jakarta:IOM. Mansuri,GhazalaandVijayendraRao(2004)."CommunityBased(andDriven)Development:A CriticalReview."WorldBankResearchObserver19(1):139. 69 McLaughlin,Karrie,AdamSatuandMichaelHoppe(2007)."KecamatanDevelopmentProgram QualitativeImpactEvaluation".Jakarta:WorldBank. Morgan, Stephen Lawrence and Christopher Winship (2007). Counterfactuals and Causal Inference:MethodsandPrinciplesforSocialResearch.Cambridge:CambridgeUniversityPress. Morel, Adrian, Makiko Watanabe and Rob Wrobel (2009). "Delivering Assistance to Conflict Affected Communities: the BRAKDP Program in Aceh." Indonesian Social Development Paper No.13.Jakarta:WorldBank. MSR (2009). MultiStakeholder Review of PostConflict Programming in Aceh: Identifying the FoundationsforSustainablePeaceandDevelopmentinAceh.BandaAceh/Jakarta:MSR. Muggah,Robert(ed.)(2008).SecurityandPostConflictReconstruction:DealingwithFightersin theAftermathofWar.LondonandNewYork:Routledge. Rosenbaum, P. R., and Rubin, D. B. (1983). "The Central Role of the Propensity Score in ObservationalStudiesforCausalEffects,"Biometrika70:41­55. Rosenbaum, P., and Rubin, D. B. (1984). "Reducing Bias in Observational Studies Using SubclassicationonthePropensityScore."JournaloftheAmericanStatisticalAssociationVol79: 516524. SwedishMinistryofForeignAffairs(2006).StockholmInitiativeonDisarmamentDemobilisation Reintegration:FinalReport.Stockholm:MinistryofForeignAffairs. Voss, John (2008). "Impact Evaluation of the Second Phase of the Kecamatan Development PrograminIndonesia."Jakarta:WorldBank. World Bank (2006). GAM Reintegration Needs Assessment: Enhancing Peace through CommunitylevelDevelopmentPlanning.Jakarta:WorldBank. 70 INDONESIANSOCIALDEVELOPMENTPAPERS No. Title Authors(s) Date 1 The Dynamics of District Governance: Forums, Budget Processes and LuthfiAshari May04 Transparency Dynamika Pemerintahan Kabupaten: Forum, Perencanaan Anggaran danTransparenci 2 ViolenceandConflictResolutioninNonConflictRegions:TheCaseof PatrickBarron Aug04 Lampung,Indonesia DavidMadden 3 Mobilizing for Violence: The Escalation and Limitation of Identity YuhkiTajima Aug04 Conflicts 4 More Than Just Ownership: Ten Land and Natural Resource Conflict SamuelClark(ed.) Dec04 CaseStudiesfromEastJavaandFlores Bukan Sekedar Persoalan Kepemilikan: Sepuluh Studi Kasus Konflik TanahdanSunberDayaAlamdariJawaTimurdanFlores 5 Crisis, Social Ties, and Household Welfare: Testing Social Capital AnnaWetterberg Apr05 TheorywithEvidenceFromIndonesia 6 Village Corruption in Indonesia: Fighting Corruption in Indonesia's AndreaWoodhouse Apr05 KecamatanDevelopmentProgram 7 CountingConflicts:UsingNewspaperReportstoUnderstandViolence PatrickBarron May05 inIndonesia JoanneSharpe 8 Aceh:ReconstructioninaConflictEnvironment AdamBurke Oct05 Afnan 9 Media Mapping: Understanding Communications Environments in JoanneSharpe Apr07 Aceh ImogenWall 10 Conflict and Community Development in Indonesia: Assessing the PatrickBarron Jul06 ImpactoftheKecamatanDevelopmentProgram RachaelDiprose MichaelWoolcock 11 Peaceful Pilkada, Dubious Democracy: Understanding Aceh's Post SamuelClark Aug08 ConflictElections BlairPalmer 12 CommunityBased Reintegration in Aceh: Assessing the Impacts of PatrickBarron Dec09 BRAKDP MacartanHumphreys LauraPaler JeremyWeinstein Papersareavailableonlineatwww.conflictanddevelopment.org 71