51429 Effective Poverty Reduction Strategies in Fragile and Conflict -Affected Countries: Lessons and Suggestions Vincent Fruchart, Per Egil Wam and William Webster September 2009 | Conflict, Crime & Violence Issue Note | Social Development Deparment | The World Bank Group Effective Poverty Reduction Strategies in Fragile and ConflictAffected Countries: Lessons and Suggestions Vincent Fruchart, Per Egil Wam and William Webster Social Development Department Sustainable Development Network The World Bank Group September 2009 This publication was developed and produced by the Social Development and Urban Development Families of the World Bank, which are both part of its Sustainable Development Network (SDN). Documents in the Social Development Conflict, Crime and Violence Issue Note Series are published informally and circulated to encourage discussion and comment within the development community. The findings, interpretations, judgments, and conclusions expressed in this paper are those of the author(s) and should not be attributed to the World Bank, its affiliated organizations, members of the Board of Executive Directors, or the governments they represent. Cover Photo: Eric Miller For additional copies of this paper, please contact: Social Development Department The World Bank 1818 H Street, NW Washington, DC 20433 Fax: 202 522 3247 Email: socialdevelopment@worldbank.org Printed on Recycled Paper Table of Contents Foreword ................................................................................................................................................. ii 1. Introduction .......................................................................................................................................... 1 2. Facing the Challenges: Adapting PRSs to the History and Threat of Violence ...................................... 2 Destructiveness of Violent Conflict ...................................................................................................... 2 Credibility of the PRS ........................................................................................................................... 4 Different Poverty in ConflictAffected Countries ................................................................................. 5 Different Institutions in ConflictAffected Countries ............................................................................ 6 3. Responding to the Challenges: PRS Effectiveness................................................................................. 8 Realism ............................................................................................................................................... 8 Prioritization ....................................................................................................................................... 9 Context Specificity ............................................................................................................................ 10 Flexibility ............................................................................................................................................ 11 Concreteness ...................................................................................................................................... 11 4. Reducing the Challenges: World Bank Assistance to PRSs ......................................................................... 13 Conflict Perspective ............................................................................................................................ 13 Relevant Analysis ............................................................................................................................... 15 Policy Support ................................................................................................................................... 16 Strengthening National Capacities ..................................................................................................... 17 Examples of Success .......................................................................................................................... 18 Conclusion: Meeting the Challenges....................................................................................................... 20 Bibliography........................................................................................................................................... 21 Boxes: Box 1: Mahaweli Ganga Project, Sri Lanka ................................................................................................ 5 Box 2: PRS Participation Lessons ............................................................................................................. 9 Box 3: Context Specificity through Conflict Analysis ................................................................................ 11 Box 4: Facilitation of Concrete Policy Action .......................................................................................... 12 Box 5: Screening Indicators of Conflict ................................................................................................... 14 Box 6: Example of Policy Matrix (Burundi) .............................................................................................. 18 i Foreword This Note was prepared by Per Egil Wam (TTL, SDV), William Webster (consultant, main author of the note) and Vincent Fruchart (consultant). The team would like to thank reviewers Nora Dudwick (PRMPR) and Gabriel Demombynes (LCSPP) for their helpful comments and suggestions, and counterparts in Afghanistan, Burundi, Central African Republic, Cote d'Ivoire, Haiti, and Sierra Leone for important guidance and feedback. The team would also like to express its gratitude to all others who provided information and inputs during the drafting of this Note, including Shonali Sardesai and Benjamin Petrini (SDV), and Donata Garrasi, Stephanie Kuttner and Tim Carrington (consultants). All errors and omissions are the responsibility of the authors alone. The note was prepared under a farranging analytical and technicalassistance program jointly undertaken by the World Bank's Social Development Department and the United Kingdom's Department for International Development (DFID). The program, which concluded in 2008, focused on the particular challenges to effective poverty reduction in fragile and conflictedaffected countries and has provided support in countries such as Afghanistan, Burundi, Central African Republic, Haiti, Liberia and Sierra Leone. The program has prepared the following reports and issue notes: Toward a ConflictSensitive Poverty Reduction Strategy: A Retrospective Analysis (World Bank 2005, 2007) "The Security Sector and Poverty Reduction Strategies" (Dissemination Note, World Bank, July 2008) The Security Sector and Poverty Reduction Strategies (Issue Note, World Bank, February 2009) Poverty Assessments in Fragile and ConflictAffected Countries: Strengthening the Conflict Perspective (Issue Note, World Bank, June 2009) "The Security Sector and Poverty Reduction Strategies" (Dissemination Note, World Bank, July 2008) ii 1. Introduction Poverty Reduction Strategies (PRSs) face impressive challenges in the poorest countries, the countries that most need skillful planning. Most commonly, PRSs are hindered by weak organizational capacity in government and the private sector, and by dependence on donors for finance and expertise, which undermines country ownership. In fragile states and situations1, these challenges are augmented by weak public institutions, the predominance of elitedriven or neopatrimonial governments, questionable political will and legitimacy to address deepseated problems underpinning poverty, and sometimes lack of full government control of national territory. Conflictaffected countries, which are usually poor and fragile, face perhaps the most difficult challenges of all, adding unresolved conflict issues, militarization of society, and conflictbased and crimebased violence to the problems of other countries. Finally, largescale violence makes previously poor states almost destitute. Violence exacerbates the challenges facing governance and development in poor countries across the board. In addition, many of the social and economic factors that drive violence in these countries are closely related to poverty. For PRSs, the aspects most crucially affected are country ownership, ability to forecast, and resources even to plan, much less implement, development strategies. Despite these difficulties, sixtyfive lowincome countries have developed or are preparing Poverty Reduction Strategy Papers (PRSP), and about twenty of these experienced violent conflict in the ten years before they published their PRSPs. Other PRS countries are considered fragile and at risk of violent conflict. For these countries, success or failure will depend on addressing and overcoming the challenges characteristic of conflict. 1 Although different organizations define `fragile states' differently, they seem to agree that a key characteristic is the quality of governance and state institutions: "...when state structures lack political will and/or capacity to provide the basic functions needed for poverty reduction, development and to safeguard the security and human rights of their populations." (OECD, 2007) "...those where the government cannot or will not deliver core functions to the majority of its people, including the poor." (DFID, 2005) "...countries facing particularly severe development challenges such as weak institutional capacity, poor governance, political instability, and frequently ongoing violence or the legacy effects of past severe conflict." (World Bank, 2007) 1 2. Facing the Challenges: Adapting PRSs to the History and Threat of Violence The World Bank's 2005 PRS Review reiterated four core principles, that PRSs be: 1) countrydriven; 2) medium to longterm; 3) comprehensive and resultoriented; and 4) partnershiporiented. The Review also recognized the challenges that all poor countries face, that "actions will need to be sequenced and tailored to country conditions," and that "expectations, therefore, need to be realistic." The Review pointed out that attention to countryspecific circumstances was more important in fragile and conflictaffected states than in other poor countries, and should extend to specific consideration of security issues, governance reform, and causes and consequences of violent conflict. The Review suggested that PRSs in these countries take advantage of work in parallel peace processes, to integrate poverty reduction and conflict prevention. The Review was largely correct, but too mild. A welldesigned PRS offers the best hope for many countries to move from poverty and conflict to development and stability, but the promise of poverty reduction can evaporate in a conflictaffected country, because: Violence creates new poverty, faster than most PRSs can reduce it; The credibility of the PRS suffers if conflict issues are not taken into account; Violenceinduced poverty is different, for individuals and in the aggregate, and will require interventions that are different; and Violence transforms economic, political, and social structures, invalidates previous data, refutes most previous analysis, and therefore forces a careful, focused, fresh planning framework on any PRS that hopes to be effective. In view of these points, it is not just expectations that need to be realistic, but strategies, plans, and actions. In addition to the call for sequencing and countryspecificity, we must focus on clear articulation of priorities, and concrete goals reached in small, judicious, manageable, flexible steps as we work to help poor societies emerge from violence and the threat of violence. Destructiveness of Violent Conflict A recent study estimated that civil wars reduced GDP growth by two percentage points per year of duration of the conflict. This result is bad enough, but it is an average, and includes large economies with relatively small conflicts. In small or poor countries, the damage is often considerably greater. World Development Indicators lists five countries that have had singleyear losses of more than 40 percent of GDP since 1980: Liberia 1990, Rwanda 1994, Georgia 1992, Armenia 1992, and Iraq 2003. All (except Armenia, devastated by an economic blockade that was violenceinspired) accompanied violent internal conflict. Violence is not just destructive of wealth: violence is the cause of the worst shortterm local destruction in the past thirty years. In many cases, the destruction is longterm. An econometric analysis of costs of conflict found that seven years after the most intense conflict, GDP per capita was more than 40 percent lower in Guinea Bissau, and more than 25 percent lower in Rwanda, than it would have been if the violence had not occurred. The paper also found strong evidence that at least onefifth of poor people in each country would not have been poor if the conflicts had not taken place2. 2 BoubacarSid Barry, Humberto Lopez and Quentin Wodon. "The Impact of Conflict on GDP Per Capita and Poverty: Results for GuineaBissau and Rwanda", in Fruchart, Wam, and Wodon, eds. (2009, forthcoming). 2 There is increasing reason to believe that simple analysis of composite variables like GDP undercounts the costs of war. For example, the analysis mentioned above, of the impact of violence on the Rwandan economy, noted the considerable recovery in the year after the genocide, and reduced its assessment of economic impact accordingly. Yet the recovery was hugely influenced by an outpouring of relief assistance to Rwanda; it was not a reduction in the cost of violence, it was a partial assumption of the cost by the world community. Measures of economic output do not begin to capture other human and social losses. The destruction of human life, health, homes, and security alone mandate paramount attention to violent conflict. These losses are magnified by the general depletion of the entire society's resources for recovery. Human capital is lost to violence, not just through the death of educated, acculturated adults, but through the flight of threatened populations into refugee camps, and the longterm migration of suddenly insecure middle classes overseas. The knowledge base for maintaining, let alone rebuilding, societal welfare shrinks with the loss of each mind. Violence that affects local, regional, or national government will also dissipate administrative capital. Any shifts in political power, including electoral changes or simply the death of a leader, lead to re ordering of priorities, and to organizational and staffing changes in government. These changes often send projects and policy agreements back to the drawing board, erasing hardwon consensus and delaying official action. These changes are more likely when, as a result of violence, resource allocations shift and a crisis atmosphere hijacks government operations. The changes reach their destructive peak if control of the state is changed by armed force, which injects ideologies and programs (or the lack of them) that were often outside the range of disputes in the previous institutional arrangements. In addition, revenue losses from reduced economic activity, and loss of territorial control for the government, weaken the government's ability, at all levels, to deliver services and to plan for medium term peace, recovery, and development. Economic institutions weaken as traditional trade routes are blocked, credit dries up, property rights are undermined, contracting systems lapse, markets fail, and banking systems evaporate. In many places, the cash economy thins out, or is replaced by foreign currencies. Relative prices change in reflection of new shortages, and can no longer serve as reliable guides for mediumterm planning or intertemporal welfare comparisons. The loss of economic institutions increases the cost of doing business across the country, further shrinking the economy, and reducing policy options for growth. Finally, violence distorts and dissolves social capital, as individuals lose trust in authority to keep them safe, lose trust in peers, especially strangers, to be wellintentioned, lose contact with spatial or communicative communities, and find social norms of behavior fading and failing. There may be a tendency to undervalue social capital, because it is difficult to define precisely, much less measure. However, the bridging aspects of social capital, that open opportunities for the poor, beyond the narrow support group of village or kin, are precisely the components that seem most damaged by extended violence. As one study of postconflict societies observed, "Even if other forms of capital are replenished, economic and social development will be hindered unless social capital stocks are restored." (Colletta and Cullen, 2000: p. 4) The sheer ability of violence, or the pervasive fear of violence, to comprehensively reduce all these aspects of the welfare and capacity of a society, qualifies it as one of the greatest, if not the greatest, threat to development. It eminently qualifies for the title of "development in reverse." (Collier et al., 2003: p. ix) The eruption of civil war may be relatively rare, but the threat of violence is far more common, and the list of tools to prevent violence is growing. Given the harm that may be averted, there are powerful incentives to use and augment those tools. In the meantime, there are equally powerful 3 reasons to know the risks of violence and the extent of fear in a community, and to plan development policy accordingly. Credibility of the PRS The scale of destruction in civil wars or other major violence is usually massive, obvious, and dramatically important. Nevertheless, planners and governments are often reluctant to raise sensitive issues, and therefore appear to neglect consideration of the causes and consequences of violent conflict. Nobody forgets the violence; it simply becomes "resolved," or "irrelevant," or "behind us." However, even if the violence deescalates, the social and economic factors that caused the violent conflict are likely to remain. There are many good reasons not to dwell on conflict issues, including avoidance of exacerbating recent tensions. If the conflict is perceived to have been ended by a victory and defeat, the victors are often anxious to consolidate their position by establishing that the conflict no longer threatens stability, that there are no residual issues to be resolved; conversely, the losers may see no benefit in reminding the world of their status. Many violent conflicts end this way; one analyst estimates that 90 percent of civil wars fought for control of government since 1955 ended with a victory for either government or rebels. (Fearon, 2007). Even if the conflict ended with negotiated compromise or stalemate, all parties may be wary of upsetting the tenuous balance. However, the consequent denial of reality damages internal and external communication and relations. Some donors are understandably frustrated, often incredulous, at the apparent lack of realism in poverty reduction planning. If governments devise strategies that cover only nonaffected regions, they are not literally planning to reduce poverty in the whole country. If they prepare strategies that treat poverty as though violence played no role in shaping it, that is reason to suspect that they have missed some of the extent or depth of poverty, or that they are misunderstanding the perceptions of their own populations, or that they are depending on previolence understandings and models beyond their useful applicability. The combination of donor frustration and government denial undermines the very cooperation that the PRS process is designed to promote. By the same token, very few populations in conflictaffected countries are unaware of the history of violence. If the government proposes strategies to reduce poverty without reference to the violent causes of poverty, the strategies may be accepted as development choices of the government, and may even be welcomed, but they will clearly not seem to be based on a shared understanding of the country's situation. The PRS process, considered to depend crucially on national, not just governmental, ownership of purposes, priorities, and policies, will be invalidated from the start. Deception will become another tool of persuasion and negotiation, and later appeals to truth and collective purpose will ring hollow. Policy may be implemented with unspoken understandings, but the weakness of trust in conflict affected countries means that effective PRSs cannot. The effort to hide causes and consequences of conflict may be perceived, especially by the country's population, as a method of concealing unresolved or new grievances, or even as a way to favor particular interest groups. Even in nonviolent settings, it is not hard to find examples of development efforts that were sabotaged, that provoked riotous opposition, or that even led to violence, because of conflict factors that were ignored or unseen in the planning stage. Massive displacements of populations to make way for infrastructure projects are only the most dramatic cases; strengthening one bureaucratic office at the expense of another, for technical reasons, in apparent ignorance of their nontechnical roles as bastions of political factions, may be more common. In many cases, it was the action that favored or harmed an identifiable interest group that exacerbated ­ or created ­ a conflict factor (as example, see Box 1 on the Mahaweli Ganga in Sri Lanka). However, 4 when mention of known conflict factors is neglected in subsequent development strategies, it facilitates suspicion, distrust, and a renewed sense of grievance and coercion. Box 1: Mahaweli Ganga Project, Sri Lanka Design: Location: Country's largest river flowing northeast through sparsely populated dry zone in the East Investments: Development program beginning in 1970s with irrigation schemes and rural infrastructure Beneficiaries: Resettlement to newly irrigated areas of about 30,000 families from densely populated South and West Financing: Many international donors, including the World Bank with six credits totaling $450 million (in 2001 dollars) between 1970 and 1998 Consequences: Demographic change: resettlement of 30,000 Sinhalese into areas considered part of the `Tamil homeland' Development priorities: large investments into Mahaweli project, while investments in old irrigation facilities serving Tamil and Muslim farmers further east were diminishing Context: a sense among Tamils of increasing discrimination, e.g. in the education system and recruitment to the civil service Political perception: the Mahaweli project was widely seen by Tamils as a scheme by the Sinhalese government to take control of traditional Tamil areas Outcome: Project played into growing political conflict in Sri Lanka For details see "Project Performance Reassessment Report, Sri Lanka, Third Mahaweli Ganga Development Project", Operations Evaluation Department, The World Bank, June 28, 2004 and http://www.conflictsensitivity.org/?q=node/121 International actors can mitigate the risk that conflictrelated issues will be missed in the PRSP, by offering methodological support and facilitation in the preparation of the PRS paper, as the World Bank did in Burundi in 2006. The Bank helped the government organize a workshop to assess the conflict sensitivity of its PRS (then under preparation), open to all stakeholders. As a result, thanks mostly to the mediating presence of international actors, it become possible for the national PRSP team to address the land issue, which until then had been carefully avoided by Burundian planners aware of the very high political sensitivity of the issue. This workshop, and the organization of a subsequent technical workgroup with international facilitation, allowed the Burundian stakeholders to start looking at the issue in a more dispassionate manner and include some initial steps towards its resolution. Different Poverty in ConflictAffected Countries A society that has undergone violent conflict will be exceptional if it has not seen changes in the characteristics of its poor population. First, there will be poor people who have been killed, driven away from the country, or, occasionally, enriched. To replace these departures from the ranks of the poor, some formerly welloff people will have lost assets and livelihoods, and now be among the poor. These newly poor may have higher educational achievements and stronger health histories than other poor people, but may have fewer coping skills for their new standards of living. 5 A more visible demographic shift is likely to be within the ranks of the poor. Violence and insecurity exert uneven influences on a population. Combatants (often young men) are disproportionately at risk of death or injury, and are disproportionately likely to leave home. As livelihoods decline, vulnerable populations ­ the elderly, the poorest, the unhealthy, children ­ are more likely to die or be driven from their home areas. As danger increases, the slightly better off, and young adults, are more likely to seek safety and economic opportunity elsewhere. In theory, these influences may reinforce each other, or may cancel each other out; in practice, violence often shifts family structures to smaller units, headed by females. Just as the threat of violence reaches more people than acts of violence do, so displacement usually affect many more people than physical injury does. Mostly as result of violent conflict some 26 million people worldwide are internally displaced while an additional 16 million are refugees outside their countries' borders.3 In their temporary housing, these people are highly visible and often muchstudied, but most such populations are temporary. When they return to their place of origin, these populations will find changed circumstances. Ageold land holdings may be contested, relations with neighbors who remained throughout the conflict may be strained, status relations are likely to have shifted, infrastructure is damaged or gone, the landscape is transformed, and the future is considerably more risky. In almost all cases, social support networks built over generations will have been weakened or crippled. Returnees and "stayees" alike will have lost ground in their efforts to establish community and build livelihoods. In numbers, populations that are shifted by largescale violence may be dwarfed by the steady rural urban flow that has strained the resources of so many cities in developing countries, but their local impact may be just as high. In addition, these populations will seem similar to the populations ­ often, themselves ­ that occupied the same spaces before violent conflict, but the similarity will be deceptive. Development interventions based on preconflict information will be at greater risk of failure, and the danger may be well hidden by traditional diagnostics. Some differences will show up in traditional measures, and these can be used as potential indicators of the need to search for more hidden differences. For example, a study of health in rural Burundi found a strong correlation between exposure to violent conflict and reductions in the heights of children (Bundervoet et al., 2008). Another study found that civil violence reduced the primaryschool attendance rate of girls in Tajikistan, for reasons apparently not connected to the destruction of school facilities (Shemyakina, 2006, p. 33). Measurable social outcomes may be a clue that other aspects of physiological and psychological change, not so easily quantified, should be investigated as well. Some of the effects of violence on individuals may be less easily observed, including changed attitudes towards personal or financial risk, different ambitions and expectations, and different interpretations of social, political, or economic information. These changes may not directly contribute to poverty, but they may change responses to economic incentives, and at least temporarily reduce the efficiency of all but the most rudimentary economic production. Different Institutions in ConflictAffected Countries In addition to changing the abilities and perceptions of individuals, violence changes their institutional setting. Economic systems are warped, beyond sheer loss of assets, by changes in relative prices that reflect uneven shortages, by the loss of trading and supply relationships, by the weakening of the rule of law for enforcing contracts and property rights, and by increased discounting of the future as a result of heightened awareness of risk. 3 Figures according to UNHCR and Internal Displacement Monitoring Centre (IDMC) websites, February 2009. 6 Almost inevitably, power relationships change, between individuals, among groups, between groups and the formal instruments of administration, and between the country and the outside world. The conflict will have winners and losers, and the winners will likely have new priorities. This will be most obvious at the national or regional level, but the changed pattern of power and goals often ripples down to the most intimate relationships, between landlord and tenant, shopkeeper and assistant, wife and husband. In addition, the basis for establishing power may change. It is highly likely that armed might will have gained in leverage, but other sources of power, such as lineage, age, gender, group affiliation, technical ability, wealth, traditional symbolic status, access to channels of communication, and foreign connections, may increase or decline in importance. Along with power changes, the basis for and pattern of networks of trust and cooperation may change, often to favor familiar, close, similar, or related individuals. The accumulated loss of familiarity will to some degree act as negative education: many individuals will find themselves not just earning subsistence livings, but earning unfamiliar livings, in unfamiliar geographic or social settings. The speed with which they restore their ability to function in cooperative groups will depend on the history of violence, the persistence of the conflicts that prompted the violence, and the quality and level of peacemaking efforts and development interventions ­ and on luck. Violent conflicts come in all sizes, intensities, durations, and spatial distributions. The degree of destruction, and the extent and pervasiveness of the transformation of society, will vary widely, as will the ability of the society to recover. However, the direction of change due to violence will almost always be toward the removal of welfare and the slowing of progress. Given the consequences of errors, all interventions whether project, plan, or strategy will be improved by reexamination after a violent conflict to determine whether any predictable harm can be avoided. 7 3. Responding to the Challenges: PRS Effectiveness In the face of considerable uncertainty, loss of historical continuity, and the potentially disastrous consequences of mistakes, it might seem prudent for a country to cease all efforts at poverty reduction, at least until extensive further planning and analysis can provide more confident guidance. However, no country has the luxury of waiting for confidence. Moral and political pressures will preclude it. In addition, welldesigned poverty reduction efforts will contribute to the understanding necessary to support accelerated development. Warnings about the dangers of overly ambitious PRSs in conflictaffected countries are always wise, but they do not imply postponement of all poverty reduction planning. Rather, they highlight the need for PRSs, or equivalent strategies that may carry different names, to be initially modest, focused, and manageable. Effective strategy, in a fragile or conflictaffected state, will include, besides constant attention to conflict factors, adherence to five guiding principles: Realism; Transparent prioritization; Context specificity; Flexibility; and Concreteness. These principles will not be sufficient. Each country will need to determine aspects of a strategy that captures its own political vision. However, they may be necessary. The aspects of a strategy that do not follow the five principles may be irrelevant or harmful. Realism Realism will involve matching goals to resources: natural resources, financial resources, knowledge, organizational capacity, and political tolerability. An overstretched development program will not only fall short, it will take credibility and trust down with it. It will be more uncertain than a smaller program, because it will have more ways to fail. While it fails, it will open time for violence to reemerge. Large comprehensive strategies may be tempting, because they can embrace a wide variety of demands from all sides of the conflict, and because they can be presented as forceful responses to urgent societal needs. However, shortterm expediency can build expectations that lead to longterm disillusionment and resentment, and more conflict. This is true of overambitious strategies that are accurately publicized, and, even more commonly, of oversold strategies. The communication of unrealistic promises to the public leads to loss of confidence. Smaller, more careful, periodically updated strategies can demonstrate success sooner, can respond to the frequent changes in circumstances that characterize conflictaffected states, can avoid wasting scarce development resources, and can provide knowledge in more palatable doses for planners and populace alike. Realism should be a conscious goal from the outset of project preparation, especially in the dialogue with stakeholders through the participation process. In conflict settings PRS participation processes will face challenges such as weak capacity of the state and other stakeholders, fragmentation of society, divisions and exclusionary policies, ongoing violence and insecurity, and limited access to parts of the country. In conflict contexts participatory strategies may need to focus less on technical aspects and policy issues than on creating space for dialogue, inclusion, and reconciliation, and on building social capital. They should enlist key stakeholders, including informal actors like traditional authorities, to help represent the priorities of those who have been directly affected by the conflict. The objectives for participation will be 8 shaped by a realistic assessment of the ability of the government to manage expectations in the short and long term, and to act upon the outcomes of participation. Overambitious plans for participation may put excessive strain on institutions and systems that are just recovering from conflict, and on a civil society that is still coping with survival and reconstruction. If inclusive participation is not possible at the early stages of a PRS, then objectives for strengthening participation can be integrated in the strategy itself, as part of a governance reform program. Box 2: PRS Participation Lessons In the Democratic Republic of Congo (DRC) and in Burundi participation strategies focused on reaching out to populations that were most affected by the conflict, including in remote areas in the country. Civil society organizations that were active in conflict affected zones received training in participatory methodologies and support in carrying out participatory appraisals. The effort widened the spectrum of views included in the final PRSs, and contributed to building local capacities for future engagement in policymaking and development programs. In Bosnia Herzegovina the strong involvement of civil society organizations and the inclusion of all ethnic groups in the PRS participation process contributed to reaching consensus around the national priorities for reconstruction and development and to national reconciliation. The PRS was the first domestic policy document approved by the local multiethnic structures after the end of the conflict. In Afghanistan the government's clear objective for PRS participation was to reach out to the provinces, to enlist the support of local stakeholders, including local political, administrative and traditional leaders, and to inform and reassure local populations about the ongoing efforts of the central government to deliver development. Conversely, the more technical aspects of PRS elaboration, such as defining objectives, strategies and policy actions, which require strictly technical expertise can provide a setting for productive initial expansion of participation. In Haiti, at the request of the Government, the World Bank provided facilitation support for the complex task of elaborating the PRS policy matrices (a task often farmed out to consultants, in spite of its crucial importance for Government's policy planning). The workgroups, which gathered sectoral ministries staff, along with technical staff from the civil society, private sector and international organizations, led to very fruitful discussions. In the case of the education sector, it was the first time that all the participants had sat at the same table, which led to first lengthy debates to clear more than a decade of misunderstanding and then to the beginning of a real dialogue. Prioritization Transparent prioritization sounds like a recipe for intensified conflict and renewed violence, because it invites clarification of disagreements. However, prioritization can highlight common goals. If there aren't many common goals, even a controversial prioritization of povertyreduction goals can, if it is explicit, clarify issues, improve the chances of successful implementation, and reduce the risk of dashed expectations. Conversely, failure to be clear about priorities and tradeoffs simply pushes difficult conflict issues to the future, raises fears that the issues will be permanently ignored, and leaves implementation decisions vulnerable to hijack by selfserving political constituencies. When the choices are not transparently shown at the start, there will be ample room for suspicion about the legitimacy of the process by which they are made later. 9 Prioritization need not be exhaustive among all possible choices. A large, ambitious, allencompassing strategy can be prioritized, or a small, lean one. Nor does prioritization constrain timing; resources and absorptive capacity will dictate the pace of implementation. However, where choosing priorities proves difficult, either because of political sensitivities or because of conceptual ambiguities and complexity, it may be useful to agree on a small number of the highest priorities for an initial strategy, postponing clarification of secondary priorities for later stages in an iterated, evolving strategy. Conflictaffected countries are not alone in requiring prioritized strategies, but they face more severe pressures and constraints than other poor countries, and the harm from unfocused development programs would be concomitantly greater. Where other poor countries might concentrate on six or eight primary goals, a conflictaffected country might be better advised to select two or three. Context Specificity Each country knows its own context and history. It should hardly be necessary to admonish planners to be countryspecific when they develop PRSs. However, there will be persistent pressure on governments to conform to international standards for PRSs, and to produce strategies that can be "objectively" evaluated and compared with strategies in other countries. These pressures are understandable, and they should be acknowledged, but they must not preclude careful attention to each country's unique environment and condition. The country's environment includes neighboring countries, and allies and enemies and trading partners and competitors worldwide. It also includes countries where its citizens have migrated (the "diaspora"). The country's environment includes nonstate actors, such as the members of the diaspora themselves, corporations that do business in the country, NGOs that work in the country, and consumers of its exports. All of these actors can act as forces for conflict resolution or exacerbation, for poverty or prosperity. Each country's environmental constellation contains a unique list of actors, with unique patterns of interaction with the country. The country's condition includes not just the characterization of present economic activity, political institutions and social relations, but also the historical path to the present. History informs the memories of the populace, and their expectations and habits. It identifies what has worked, what has failed, what has not been tried, and, sometimes, why. For a conflictaffected country, the history includes not just the effects of violence, but the societal factors that expressed themselves in the violence. What were the stresses that violence revealed? What were the understandings and misunderstandings? What were the structural flaws that allowed violence to emerge? All strategies depend on a shared model of the country and its poverty, but strategies that take careful account of the country's attributes will extend and deepen the shared perceptions, and achieve a chance at plausibility and success. 10 Box 3: Context Specificity through Conflict Analysis Conflict analysis can contribute to insights into how poverty is created, exacerbated, and structured in conflictaffected countries. For example, both conflict analysis and traditional poverty assessments are likely to look at social cleavages. However, conflict analysis goes beyond describing the cleavages to study how policies play a role in shaping these cleavages. In the period leading up to the war in Burundi, there were four main cleavages in society with the ethnic divide (HutuTutsi) being the strongest. The conflict analysis for Burundi (Brachet and Wolpe, 2004) reveals that as economic depression and resource scarcity worsened, the concentration of power among the Tutsis led the divisions between the "haves" and "havenots" to coincide with ethnic and regional distinctions. While poverty increased in the country as a whole, the Hutus faced greater hardship and poverty. Conflict analysis helped reveal how policy choices exacerbated political inequality, which in turn deepened poverty. Conflict analysis can show why perceptions of relative poverty sometimes differ from statistical measures. The Niger Delta is a case in point (World Bank, 2008). Poverty statistics show that the communities in the Niger Delta are better off than those in other parts of the country, and that poverty has decreased in the Delta. However, levels of selfassessed poverty and perceptions of inequality indicate strong feelings of economic marginalization, because the communities feel that despite residing in the resourcerich region of the country, they do not enjoy commensurate benefits and that they therefore are "poor" citizens (or poorer than their counterparts in other regions). Flexibility Violence brings uncertainty and the likelihood of unexpected change. Strategies that depend on current conditions will be at least partially invalidated when those conditions change, unless they can shift with the tides. Strategies that are built of small, clearly prioritized components, without explicit provision for flexible change built in from the start, will still be able to adjust future components without the loss of initial gains, but with a potential loss of time. Strategies that have flexibility built into their design, the assumption of midcourse corrections, and automated procedures for reexamination and redirection, will be poised to expect changes, and will be finding less resistance and delay when they occur. Flexibility includes openness to changes in the details of programs, such as deadlines and field activities, in the choice of programs to support leading poverty reduction priorities, or even in the alignment of the priorities themselves. Flexibility can be aided by listing options in the PRSP, including the priorities, programs, actions, and monitoring measures chosen, and others considered. Concreteness All strategies will in the end be implemented through a set of policy actions, in budget allocations, procurement of goods and services, and execution of reforms and projects. The justifications and evaluations may be expressed in abstract principles and standards, but the eventual results on the ground, measured or not, will be concrete. Therefore, the question of how concrete to be in PRSs is a question of timing and agency. Will it be made concrete at the design stage, by planners? At the implementation stage, by untrained local officials acting on their own initiative? Something concrete will emerge over time; if it is nobody's decision, the concrete result will appear almost accidental, surely unplanned, and likely unhelpful. The more specific a PRS can be in its initial statement of policy actions, the more it is able to shift from defining objectives to designing actions, the more likely it is to be effective. There is a possibility of delay in the specification of details of policy actions, because the loss of ambiguity reduces the ability of conflicting groups to see their own goals reflected in vague language, 11 and to agree to that language. The potential gain, however, is considerable. Finding specific steps to agree on will solidify the cohesiveness of adversaries, will reassure them about the process for agreement, and will allay fears about the insertion of unwelcome actions in the implementation stage. The limitation of actions to shared priorities will help keep PRSs within realistic bounds. In addition, the concreteness of the stated policy actions will facilitate their implementation and comparison of outcomes with the initial PRS. Box 4: Facilitation of Concrete Policy Action A facilitation method to encourage concreteness in planning has been developed and tested in three fragile states, Burundi, Central African Republic and Haiti, in the context of the program on effective PRS in conflictaffected countries. It includes two main steps, each facilitated by a neutral, outside facilitator: A/ Scoping Workshop The workshop is organized around a specific theme, such as security sector reform or postconflict issues (the more limited the scope, the better), and includes the key stakeholders from three categories: i/ Government officials and expert civil servants; ii/ civil society and private sector representatives; iii/ international experts. The purpose is to arrive at a general consensus on the most important issues pertaining to the topic, as well as on the main possible strategies to be pursued. With targeted research questions provided by the facilitator, it is possible to identify very specifically the issues to be addressed in the context of the PRS as well as the general strategies to be pursued. B/ Policy Matrix Workgroups A policy matrix workgroup then focuses on a specific sector or topic (e.g., justice reform, agriculture, government capacity building). The group members are drawn from the participants in the workshop (although it is possible to add newcomers). It is important that government officials be part of the group, along with other experts (national and international). The ideal group size is about 12 people, and it is important to have as much variety as possible in the profile of the participants. The objective of the workgroup is to build policy matrices structured in terms of global objectives (one or several), intermediary objectives (or strategies, to reach the global objectives), and specific actions that are feasible within the PRSP time horizon along with the necessary results and performance indicators. These actions being specific can be easily costed, as well as developed with precise output and outcome indicators The five principles ­ realism, prioritization, concreteness, flexibility, and countryspecificity ­ reinforce each other in almost trivially obvious ways. Countryspecificity, builtin flexibility, and concreteness will tend to reinforce realism, as far as budgets and capacities go. Prioritization and realism will ease the path to designed flexibility, by reducing the choices necessary at any stage. Realism, prioritization, and concreteness will hardly be possible without thorough attention to the specific circumstances of the country. The demonstration effect for the population and outside observers, of the country's ability to find common goals, express them in clear actions, and match the results to the stated intentions, without selfserving afterthefact interpretation, will go a long way to restoring credibility lost to the conflict. 12 4. Reducing the Challenges: World Bank Assistance to PRSs What can the World Bank and other international bodies do to help countries produce Poverty Reduction Strategies, and take ownership of them? Outsiders can help best by lending clear support to five guiding principles of PRSs in fragile and conflictaffected states. They can: Promote a conflict perspective; Perform relevant analysis; Provide policy support; Strengthen national capacities; and Share examples of effective PRSs. Each of these areas is important for increasing the effectiveness of a conflictsensitive PRS. The following include suggestions for World Bank engagement in each area. Conflict Perspective Adopting a conflict perspective will help donors reinforce the awareness that conflictaffected countries have unique problems that may require unique strategic approaches, not just the intensification of approaches taken by all poor countries. Although the conflict perspective should pervade all support from donors, it seems to require, at a minimum: encouragement for initial caution in poverty reduction strategies, conflict screening for all countries at risk of new or renewed violent conflict, and treatment of violence and coercion as components of poverty. In many cases, countries cannot wait even for the beginning of rigorous conflict analysis before they make decisions about poverty reduction in the short and medium term. Before analysis, prudence will dictate that countries proceed very carefully, with manageable, shortterm, simple policy actions. Prudence does not require inaction; it does not even favor inaction in a rapidlychanging environment of continued conflict and enormous need. It does require, when the future is uncertain, the present is deceptively unfamiliar, budgets are tight, and the consequences of failure are horrendous, that countries not overreach, and not make irreversible mistakes. Despite strong political pressures to please disparate constituencies, shortterm, costeffective, toppriority, simple interventions form the only justifiable strategy in most conflictaffected countries. Donors should encourage such caution and manageability. The world is littered with the wreckages of illconsidered development programs. The first step in analysis and planning is to determine whether a country is in fact conflictaffected or prone to violent conflict. This is trivial in the case of countries that are emerging from largescale violence, but at some point even these countries may be considered to have put their conflicts behind them, and to be no longer at risk. However, countries that have not suffered a recent violent conflict may show signs of danger. Poor economic performance, weakened governments and social institutions, population or environmental pressures, outside political pressure, or political power shifts, may all be harbingers of war. The World Bank's Conflict Analysis Framework (CAF) offers one suggestion for a screening process to determine how essential it will be to plan deliberately for the danger of violent conflict. There are plausible criteria other than those proposed in the CAF, but some form of simple, plausible, transparent test should be routinely required to determine the risk of violence in a country's mediumterm future. 13 Box 5: Screening Indicators of Conflict The first step in the CAF is a risk screening process which consists of nine main indicators, which aim to capture the deteriorating environment in a country. In general, as the number of indicators checking positive for a country increases, the importance of conducting a full conflict analysis in the country also increases. Violent conflict in the past 10 years Low per capita GNI High dependence on primary commodity exports Political instability * transformation of the state structure * breakdown of law and order Restricted civil and political rights Militarization * high defense spending as a ratio of GNI and large army as proportion of population * the availability of arms among nonstate actors Ethnic dominance of state institutions or the economy Neighboring countries affected by conflict High youth unemployment For details, see World Bank. 2005. "The Conflict Analysis Framework (CAF)." Social Development Department, The World Bank, Washington, DC. The first version of the CAF was developed in 2002, updated in 2005, and it has been used in a variety of contexts since. Violence and the threat of violence are as much manifestations and components of poverty as malnutrition, ignorance, homelessness, or powerlessness. If you can't go out at night, if you are fearful about sending your children to school, if your husband is missing, you are poorer, at any level of other material welfare, than if these conditions did not hold. If insecurity keeps you from planning and investing for the long term, the value of your capital has declined: again, you are poorer. Many current discussions of violence and poverty treat violence as a cause or consequence of poverty. It is both, but it is more. Unless violence is recognized as an aspect of poverty, measures to reduce either will fail to serve the needs of the poor in conflictaffected countries. International institutions can ensure that this point is remembered. As an example, the 2005 Kosovo Poverty Assessment saw violence and the risk of violence as an aspect of the insecurity encountered by the population. It noted that the people themselves articulated the lack of security as being an essential part of poverty, i.e. a sense of vulnerability to a reduction in different dimensions of their wellbeing from risks at national, community and household levels. Among the physical insecurities assessed in the report were interethnic tension and violent crimes, including trafficking of human beings and domestic violence.4 Significantly, the Kosovo report included support of conflict prevention as one of its recommended policy directions. A PRS may not be the most effective tool to mitigate the risks related to ethnic divisions, but it would nevertheless be important to account for such security risks in the planning of policy actions. 4 World Bank, 2005d. 14 Relevant Analysis Donors today provide a wide range of analyses to support and strengthen development in poor countries, but conflictaffected countries could benefit from intensification of analysis of specific topics, including: research into the differences between structural poverty and conflictinduced poverty, regional analysis of both poverty and conflict, refinement of conflict analysis tools, and conflictsensitive poverty assessments. A corollary to the notion of violence as a component of poverty is the claim made above that violence induced poverty is qualitatively different, even in nonthreat, noninjury manifestations, from chronic poverty or poverty from other causes. Research may confirm or refute assertions that, for example, war torn populations will exhibit different responses to disease, returned refugees will entertain different attitudes toward the use of land and the reliability of community, or, the threat of violence will change poor people's willingness to plan for the long term. These and many other plausible hypotheses will benefit from testing, for the strength of their applicability, for the ways in which they interact with each other and vary across countries, and for their implications for poverty reduction strategies. The more countries know about the ways violence and threat change their populations' responses, the better they can design programs to meet their needs. In addition to the lessknown consequences of violence, there is still much to understand about the antecedents of violence. Despite the importance of violent conflict for all countries, threatened directly or not, far less is known about the causes of violence, or how to foresee its outbreak, than is known about poverty reduction, wealth creation, sustainable development, and governance in developing countries. This knowledge gap is mirrored by a consensus gap, with diverse academic, policy, and advocacy voices emphasizing a confusing array of alternative violence factors. Few understandings could be more important for a poor country in a fragile situation than understanding of the provenance and course of violent conflict. Any substantial advance in understanding the interactions of violent conflict, social structure, political power, and economic development will pay immediate dividends in the potential for poverty reduction. As a preliminary contribution to this understanding, the World Bank is continuing to refine the Conflict Analysis Framework, to help identify factors that may affect the likelihood of renewed conflict. In any given country, many of the factors will likely be secondary or even unimportant, and other factors may dominate; the large number of detailed factors for consideration (more than 30 in five thematic categories) is a testament to the great variety of violent conflicts around the world, and to the multiplicity of perspectives on the causes of those conflicts. Despite the fact that the CAF was developed for analysis of conflict, the factors it considers may be influential contributors to poverty even in the absence of violent conflict. In addition, the World Bank is applying the evolving CAF. In 2006 the Bank commissioned a conflict analysis in the Central African Republic, which was performed by national consultants with World Bank supervision. In 2007, the report was edited and distributed in the context of a workshop organized to help the Government better take into account conflict factors in its PRS analysis and recommendations. The document was well received by all parties and served as a basis for the discussions held on security issues and their impact on poverty. Conflict analysis tools are useful, and will be improved steadily over time, but they will not suffice. There are many ways to approach the problem of conflict analysis, and many questions left to be asked, let alone answered. The World Bank and other international organizations can continue to devote resources to better understanding of the effects of conflict on poverty, and viceversa. 15 One potential extension of existing conflict analysis is beyond the national borders. No country is any longer a master of its own fate, especially when it comes to violent conflict. Wars spill across borders, as refugees arrive from the next country, fighters seek safe bases, or natural resources and narcotics attract national or private armed interests. Most countries are far less qualified to analyze their neighbors' conflict situations than their own, so international organizations may be the best source of such regional analysis. The World Bank can encourage and perhaps even sponsor the studies in unstable regions. The multicountry benefits should be obvious and noncontroversial. Finally, the quality of the PRS will depend crucially on the quality of the poverty information it relies on. The primary source of information for building PRSPs is the Poverty Assessment (PA). Many PAs, like many PRSs, turn a blind eye to conflict issues, or omit conflict regions. When they do, they provide distorted information that makes effective national poverty reduction largely a matter of luck. If PAs incorporate conflict analysis to reveal how the identity, situation, perceptions, and prospects of the poor have been affected by conflict, and how their responses to development interventions will in turn be changed, the PAs can contribute constructively to development and poverty reduction strategies. The ways the conflict perspective in PAs can be strengthened are discussed in a separate issue note: Poverty Assessments in Fragile and ConflictAffected Countries: Strengthening the Conflict Perspective (World Bank, 2009). Policy Support Many conflictaffected countries can benefit from donors' experience in policy formulation. Policy support may take the form of policy design for a specific sector that is especially relevant to conflict affected countries, such as the security sector. Alternatively, policy support may come in the form of encouragement for policy directions that are highly advisable for conflictaffected countries, such as an emphasis on contextspecificity in poverty reduction strategies, or a shift in emphasis from poverty reduction goals to clear articulation of policy actions. One of the highest priorities will be the security sector. People and governments in conflictaffected countries regularly identify insecurity as a major obstacle to poverty reduction. Weak, ineffective, and unaccountable security sectors are often at the heart of state fragility and violent conflict that deepens poverty. The provision of security is a core function of the state and a necessary condition for the delivery of essential services for poverty reduction. In order to devise effective poverty reduction strategies, countries need to integrate efforts to improve security, including thorough reform and strengthening of the security sector. While the World Bank does not usually provide direct assistance to the security sector, it does have a mandate to ensure security considerations are integrated into national development strategies and to provide assistance in its core areas of competence, including economic governance and public finance management for the security sector. The World Bank has prepared an issue note that discusses entry points for engagement in the security sector in countries where insecurity is a major contributor to poverty and vulnerability, and that presents several country cases that show diverse approaches to incorporating security sector issues in PRSs.5 Continued research into and promotion of security sector development and integration will contribute crucially to the effectiveness of PRSs in conflictaffected countries. The security sector is a concrete example of attention to the particular context of a conflictaffected country, because security in many cases has been part of the problem. However, poverty reduction strategies will gain through their identifiable relevance to each country's individual circumstances. Despite considerable lip service to the unique nature of each country, and despite the genuine belief of country offices and Bank staff that they adhere to this principle and focus on their country, there are 5 Garrasi, Donata, Stephanie Kuttner and Per Egil Wam, 2009. 16 influences that militate against it, including the strong desire for intercountry comparisons, the need to be able to transfer lessons from one country for enlightenment of another, and the desire to treat countries equitably. The preference for apolitical analysis and planning may also feed the desire to remain at an apparently neutral technical level in country analysis and advice. From this point of view, the more generalizable a country's PRS principles and priorities, the less likely they are to be politically motivated and the more likely they are to be transferable from one country to another. In addition, the urgency of conflict resolution and poverty reduction may increase the appeal of proven "offtheshelf" planning approaches. The preference for following proven paths in the initial stages of poverty reduction is mistaken. National ownership, public credibility, and program success all depend on poverty reduction strategies that clearly respond to the poverty and conflict factors that distinguish each country. One way to ensure that a country's characteristic features are the background for poverty reduction strategies is to keep the strategy as concrete as possible. The most concrete part of the PRSP is in the statement of policy actions. Concentration on the quality of policy actions, whether reforms, regulations, or investments, is the surest way to keep PRSs concrete, realistic, and countryspecific. The World Bank and other donors can emphasize that including highquality policy actions is a key to the overall effectiveness of the PRS. Bank staff can remind governments that there is no single road to success in PRSs and that initial PRSs, though contextspecific, can be relatively simple, pending clearer understandings of the changes brought by violence and the efficacy of various poverty reduction approaches. Subsequent PRSs will undoubtedly include more complete prioritization, more thoroughly inclusive participation, and a wider coverage of all sectors. Strengthening National Capacities Many countries lack the midlevel and senior governmental expertise to articulate priorities, mediate complex conflicts, and structure detailed development plans, especially in ways that are readily translated into language the international organizations are accustomed to. To some degree, technical assistance can bridge this capacity gap. However, the goal is to build capacity, not simply supplement it. Bank staff have had success in Burundi, Central African Republic, and Haiti, with an approach that leads officials and civil society representatives through the process of translating the policy objectives into concrete and feasible actions (see box 4 for details on this approach). This facilitation has the triple advantage of conferring a sense of country ownership of the resulting strategy, drawing out the attention and participation that is often necessary for the delicate resolution of conflict, and thoroughly training local planning and administrative officials, so that they can independently draft papers for future iterations of PRSs. Box 6 shows a small sample of policy actions to support the global policy objective of strengthening peace and security in Burundi. The output indicators reinforce the concreteness of each action. 17 Box 6: Example of Policy Matrix (Burundi) OBJECTIVES ACTIONS INSTITUTION IN OUTPUT INDICATOR OUTCOME INDICATOR DATA END DATE CHARGE SOURCE Security Sector Global objective: Strengthening peace and security 1/ Remove residual factors of insecurity Signature and Gvt Burndi Peace accord signed Absence of armed End 2006 implementation of peace violence in the most accord with the FNL affected provinces Demobilisation or MDNAC Number of FNL FNL operates only as End 2006 integration of FNL MISP combattants: a political party. combattants into the FDN SE/CNDRR integrated into FDN Absence de armed (National Defense Forces) or integrated into PNB violence in the most PNB (Burundi National demobilized affected provinces Police) Implementation of socio SE/NCDRR Number of demobilized Numbe of demobilized End 2007 economic reintegration combattants benefiting combattants who had support programs for from the reintegration received reintegration demoblized combattants programs support and have a paid activity Audit of all development MinPlan Number of programs Number of demoblized End 2007 programs to ensure that they CNCA audited combattants include measures designed SE/CNDRR benefiting from these to help the reintegration of measures demoblized combattants Continuation of training / SE/CNDRR Number of training End 2007 education programs for progams already demobilized children Number of demobilized child soldiers having gone back to school Number of demobilized child soldiers having received training A further way to increase the effectiveness of the PRS concurrently with strengthening national capacity in design and implementation is to encourage integration of strategy preparation into established planning and budgeting processes. This will increase the country ownership of the strategy, and the likelihood that it will keep within financially realistic bounds. Similar approaches, combined with sharing of data, extended training, and dissemination of information about the pitfalls of conflictblind development programming, can bring substantial benefits to poverty reduction strategies. Examples of Success With the advantage of a worldwide perspective, the World Bank can publicize effective PRSPs. Although the details and challenges of each PRS will vary from country to country, the overall principles are fully generalizable. Documents that successfully incorporate conflict sensitivity, and that follow the five guiding principles for PRSs in conflictaffected countries, can illustrate and reinforce the desired approach; promotion of the strategies will emphasize the Bank's commitment to the principles. Later, 18 the Bank is ideally suited to following the results of implementation, distilling lessons, from the design of PRSs to results in poverty alleviation without violence, and transmitting the lessons learned to countries with less experience. One example that may merit examination is the recently published PRSP of Liberia. The Liberian PRS has, in addition to four pillars (Consolidating Peace and Security, Revitalizing the Economy, Strengthening Governance and the Rule of Law, Rehabilitating Infrastructure and Delivering Services), five crosscutting issues: Gender Equity, Environment, HIV and AIDS, Children and Youth, and Peacebuilding. This twodimensional approach, which allows each crosscutting issue to cast each pillar in a different perspective, also creatively distinguishes between restorative responses to violence, in "Consolidating Peace," and constructive responses, in "Peacebuilding." The entire structure of the PRSP shows awareness of the need for conflict sensitivity in poverty reduction strategies. 19 Conclusion: Meeting the Challenges The difficulties of preparing a comprehensive poverty reduction strategy in a conflictaffected country have seemed insurmountable to some observers. One option they have suggested is to postpone PRSP until the conflicts have subsided, and considered strategies can be prepared with more assurance of success. This seems an unnecessary prudence, and a lost opportunity. Even the monumental challenges facing poverty reduction strategies in conflictaffected countries can be ameliorated. Countries can keep their strategies lean, flexible, and focused. The focus should be on the new country circumstances, especially the challenges offered by the changed forms of poverty and its interrelationship with violent conflict. The World Bank and other outside organizations can provide guidance, information, standards, and encouragement. Countries have often made progress in poverty alleviation despite the damage of violence, but this has all too often been with the help of outside donors, with the loss of country ownership of the process, and of the country's ability to direct development in line with domestic priorities. The political, social, and economic sustainability of such development is questionable, and the pace is usually far too slow. Where outsiders have carried out the technical work and made, even dictated, the difficult political choices, even though the guiding principles may have otherwise been more or less followed, the country has not gained enough. Countries will need to be weaned from excessive donor assistance if they are to pursue their own goals and help their own people sufficiently. Effective PRSs in conflictaffected states will have the additional benefit of being "conflict reduction strategies". They will have to be, because violent conflict makes poverty worse and poverty reduction more difficult. When conflict understanding permeates the Poverty Reduction Strategy, the country will be on the road to building prosperity with peace. 20 Bibliography Brachet, Juanna and Howard Wolpe. 2004. "Peace and Development in Burundi." Working Paper No. 15, Conflict Prevention and Reconstruction Unit, The World Bank. Brainard, Lael, and Derek Chollet, eds. 2007. Too Poor for Peace? Global Poverty, Conflict, and Security in the 21st Century. Washington, DC: Brookings Institution Press. Bundervoet, Tom, Philip Verwimp, and Richard Akresh. 2008. "Health and Civil War in Rural Burundi." MICROCON Research Working Paper 5, MICROCON, Brighton, UK. Bundervoet, Tom. 2006. "Estimating Poverty in Burundi." HiCN Working Paper 20, Households in Conflict Network (HiCN), The Institute of Development Studies, University of Sussex, Brighton, UK. Chen, Siyan, Norman V. Loayza, and Marta ReynalQuerol. 2007. "The Aftermath of Civil War." Policy Research Working Paper 4190, The World Bank, Washington, DC. Colletta, Nat J. and Michelle L. Cullen. 2000. Violent Conflict and the Transformation of Social Capital. The World Bank, Washington DC. Collier, Paul. 1999. "On the Economic Consequences of Civil War." Oxford Economic Papers 51: 168183. ----. 2004. "Development and Conflict." Centre for the Study of African Economies, Department of Economics, Oxford University, Oxford, UK. ----. 2007. The Bottom Billion. Why the Poorest Countries are Failing and What Can Be Done About It. New York, NY: Oxford University Press. Collier, Paul, V.L. Elliot, Havard Hegre, Anke Hoeffler, Marta ReynalQuerol, and Nicholas Sambanis. 2003. Breaking the Conflict Trap: Civil War and Development Policy. Policy Research Report, The World Bank and Oxford University Press, Washington, DC. DFID. 2005. "Why We Need to Work More Effectively in Fragile States." Department for International Development, London, UK. EnbergPedersen, Lars, Louise Andersen, and Finn Stepputat. 2008. "Fragile Situations: Current Debates and Central Dilemmas." DIIS Report 2008:9, Danish Institute for International Studies, Copenhagen. Fearon, J. 2007. "Iraq's Civil War." Foreign Affairs, 86 (2), March/April. Fruchart, Vincent, Shonali Sardesai and Per Egil Wam. 2009. "Poverty Assessments in Fragile and ConflictAffected Countries: Strengthening the Conflict Perspective." Social Development Department, The World Bank, Washington, DC. Fruchart, Vincent, Per Egil Wam, and Quentin Wodon, eds. 2009. Social and Poverty Analysis in Conflict Affected Countries Tools and Case Studies from SubSaharan Africa. Washington, DC: The World Bank. Forthcoming. 21 Garrasi, Donata, Stephanie Kuttner and Per Egil Wam. 2009. "The Security Sector and Poverty Reduction Strategies." The World Bank, Washington, DC. Goodhand, Jonathan. 2001. "Violent Conflict, Poverty and Chronic Poverty." CPRC Working Paper 6, Chronic Poverty Research Centre (CPRC), University of Manchester, Manchester, UK. Justino, Patricia. 2006. "On the Links between Violent Conflict and Chronic Poverty: How Much Do We Really Know?" CPRC Working Paper 61, Chronic Poverty Research Centre (CPRC), University of Manchester, Manchester, UK. Justino, Patricia. 2007. "On the Links between Violent Conflict and Household Poverty: How Much Do We Really Know?" MICROCON Research Working Paper 1, MICROCON, Brighton, UK. Lopez, Humberto, and Quentin Wodon. 2005. "The Economic Impact of Armed Conflict in Rwanda." Journal of African Economies, Centre for the Study of African Economies, Oxford University Press. Luckham, Robin, Ismail Ahmed, Robert Muggah, and Sarah White. 2001. "Conflict and Poverty in Sub Saharan Africa: An Assessment of the Issues and Evidence." IDS Working Paper 128, The Institute of Development Studies, University of Sussex, Brighton, UK. OECD. 2007. "Principles for Good International Engagement in Fragile States and Situations." Development Assistance Committee, OECD, Paris. Rice, Susan E., Corinne Graff, and Janet Lewis. 2006. "Poverty and Civil War: What Policymakers Need to Know." Global Economy and Development Working Paper 2, The Brookings Institution, Washington, DC. Sambanis, Nicholas. 2004. "Poverty and the Organization of Political Violence: A Review and Some Conjectures." Yale University, New Haven, CT. Shemyakina, Olga. 2006. "The Effect of Armed Conflict on Accumulation of Schooling: Results from Tajikistan." HiCN Working Paper 12, Households in Conflict Network (HiCN), The Institute of Development Studies, University of Sussex, Brighton, UK. Stewart, Frances, Valpy FitzGerald, and Associates, eds. 2000. War and Underdevelopment. Volume 1. The Economic and Social Consequences of Conflict. New York, NY: Oxford University Press. World Bank. 2005a. "2005 PRS Review. Balancing Accountabilities and Scaling up Results." The World Bank and International Monetary Fund, Washington, DC. World Bank. 2005b. "The Conflict Analysis Framework (CAF)." Social Development Department, The World Bank, Washington, DC. World Bank. 2005c. "Toward a ConflictSensitive Poverty Reduction Strategy: Lessons from a Retrospective Analysis, Vol. 1 of 1." Report N. 32587, Social Development Department, The World Bank, Washington, DC. World Bank. 2005d. "Kosovo Poverty Assessment: Promoting Opportunity, Security, and Participation for All." Report No. 32378XX, The World Bank, Washington, DC. World Bank. 2007. "IDA 15: Operational Approaches and Financing in Fragile States." The World Bank, Washington, DC. 22 World Bank. 2008. "Republic of Nigeria: Niger Delta Social and Conflict Analysis." Sustainable Development Department, Africa Region, The World Bank, Washington, DC. 23