K E Y C O N C E P T S A N D O P E R AT I O N A L I M P L I C AT I O N S I N T W O F R A G I L E S T AT E S THE CASE OF SIERRA LEONE AND LIBERIA A JOINT INITIATIVE BY THE WORLD BANK’S FRAGILE AND CONFLICT-AFFECTED STATES GROUP (OPCFC) AND THE UNITED NATIONS DEVELOPMENT PROGRAMME’S BUREAU FOR CRISIS PREVENTION AND RECOVERY (BCPR) STATE-BUILDING: KEY CONCEPTS AND OPERATIONAL IMPLICATIONS IN TWO FRAGILE STATES CONTENTS BY SUE INGRAM page 3 1. INTRODUCTION 4 2. KEY CONCEPTS IN THE DISCOURSE ON STATE-BUILDING AND FRAGILITY 4 2.1 State-building as a response to fragility 6 2.2 What do we mean by state-building? 7 2.3 Building blocks for state functioning 8 2.4 What makes state-building different in fragile states? 10 3. USING A STATE-BUILDING LENS TO INFORM PRACTICE 10 3.1 Assessing programming approaches using a state-building lens: Sierra Leone and Liberia 12 3.2 Operational considerations highlighted by the country missions 12 3.2.1 Invest in developing a broadbased understanding of the political economy and the drivers of state-building 14 3.2.2 Consider how programming may impact on the political settlement and the political processes underpinning it, and possible downstream consequences 20 3.2.3 Consider the impact of aid modalities on state-building 24 3.2.4 Think and work system-wide, and for the long term 26 3.2.5 Match the development approach to the context 29 3.2.6 Take ethical responsibility for champions in contested environments 31 4. WHAT CAN AN OVERWORKED COUNTRY OFFICE DO? BOXES page 9 BOX 1: Sources of legitimacy 15 BOX 2: Constitutional revision 17 BOX 3: Electoral jockeying in Sierra Leone and Liberia 21 BOX 4: UN Joint Vision and Multi-Donor Trust Fund for Sierra Leone 22 BOX 5: Elements of the parallel public sector in Sierra Leone 23 BOX 6: Governance and Economic Management Assistance Program (GEMAP) 27 BOX 7: The politics of exclusion 28 BOX 8: Anti-Corruption Commissions T H E W O R L D BA N K — U N D P 1 1. INTRODUCTION T his paper aims to provide a very distilled summary flict-affected settings. It is not meant as the definitive word of the concepts shaping the discourse around on state-building—this work is being undertaken through state-building in fragile, conflict-affected situa- other processes.2 Nor should it be seen as an assessment of tions, and to explore some of the operational implications for the programs of UNDP and the World Bank in Sierra Leone international development practitioners working in these set- and Liberia: neither the terms of reference for the country tings, drawing on experience from two post-conflict countries. missions nor the methodology used would support such an The paper arises out of a collaboration between UNDP’s assessment. Rather, the ideas set out here are an ex-post Bureau for Crisis Prevention and Recovery and the World facto distillation of insights into state-building gained from Bank’s Fragile and Conflict-Affected Countries Group to the two missions—an opportunity to reflect more broadly on strengthen their analytical work and guidance to country the complex and often unpredictable interplay at work where offices in the area of state building, and to extend intera- local and international actors engage in this domain. gency cooperation at headquarters and field level. As part of The material in this paper is organised around four this collaboration, the two groups agreed to jointly undertake themes: two country studies to look at specific aspects of state-build- I Current concepts and theory on state-building; ing and the wider lessons this work might suggest for future I Our practical experience with applying a state-building engagement on state-building in other fragile settings. The lens to specific aspects of programming in Sierra Leone selection of the case studies sought two different contexts and Liberia; where state fragility shaped the development approach, I Some operational considerations on approaching state- including one country with an integrated peacekeeping mis- building in fragile, conflict-affected settings; and sion, where both UNDP and World Bank Country Offices were I Proposals for what an overworked country office can do to ready to engage with the initiative, and where there was the support state-building. potential to build on ongoing and planned activities in the This paper sits alongside a detailed report on “Donor areas of state and institution building. Applying these crite- Support for Capacity Development in Post-Conflict States: ria, Sierra Leone and Liberia were identified. Reflections from Two Case Studies in West Africa” which was This paper, and the operational guidance it proposes, is a also developed as part of the UNDP-World Bank collaboration product of the missions to Sierra Leone and Liberia, and its and field missions. principal audience is country office staff in fragile and con- T H E W O R L D BA N K — U N D P 3 2. KEY CONCEPTS 2.1 IN THE DISCOURSE STATE-BUILDING ON STATE-BUILDING AS A RESPONSE AND FRAGILITY TO FRAGILITY F or the last decade, the international community has While there is no firm consensus on precisely what defines been preoccupied with the consequences of state a fragile state or situation, there is broad agreement on the fragility, which directly threatens the security and essential attributes, including weak institutions and gover- wellbeing of populations within the territory of the state and nance systems, and a fundamental lack of leadership, politi- wider regional and global security, and seriously retards cal will and/or capacity to deliver on key public goods, espe- progress towards achieving the Millennium Development cially in terms of protecting the poor (Rocha Menocal 2009, Goals. p3; OECD 2010, p146;). The state-building agenda is a FIGURE 1: COMMON TASKS IN PEACEBUILDING AND STATE-BUILDING IN THE POST-CONFLICT PERIOD Peacebuilding State-building • Restoring basic administrative capacity and a functioning civil service • Consolidating (and sometimes • Political settlements & • Strengthening public financial renegotiating) peace agreements agreement on the rules of management and economic policymaking • Establishment of an interim or transitional the game • Support to political governance: building government • Security sector reform leadership capacities of key state • Early recovery, critical infrastructure, (including justice, rule decision-makers and executives, employment generation, and of law, and policing) parliamentary performance, and civil livelihoods restoration • Constitution-making society participation • Refugee and IDP repatriation processes and • Decentralization management at the and civilian protection strengthening of core central and intergovernmental level • Transitional justice, amnesty, governance institutions and support to local governance and prosecution for war crimes • Electoral processes • Supporting national and local • Rebel-to-political party transformation • Delivery of basic social “democratic dialogue” and • Disarmament, demobilization and services multistakeholder processes reintegration of ex-combatants • Developing conflict and governance crisis-response capacities (NB: this is a representative, not an exhaustive, list, depicting common tasks in peacebuilding and state-building as currently practiced) Source: Wyeth and Sisk, 2009, p16 direct policy response to these conditions. utes, they remain two very distinct processes and may, at In 2007, the OECD published its Principles for Good times, pull in quite different directions (Rocha Menocal International Engagement in Fragile States and Situations. 2009, p14; Wyeth and Sisk 2009, p1). Their complementa- These positioned state-building as the central objective in ry domains are summarised in Figure 1 (taken from Wyeth addressing fragility. The Principles emphasise that interna- and Sisk 2009, p16). tional engagement should aim to build the relationship While peacebuilding aims to create the conditions for sta- between state and society, concentrating on two main areas: bility, this is not in itself sufficient to overcome fragility. first, supporting the legitimacy and accountability of states Lasting stability can only come through resilience. Resilience and, second, strengthening the capacity of states to fulfil —that is, the ability to cope with internal and external shocks their core functions. —is characterised as the opposite of fragility (OECD 2008b, Fragility is often—although not necessarily—associated p12) and it is the condition of resilience that will stop a with conflict, which is both a cause and an effect of fragility. country from spinning back into conflict when under pres- As a result, the discourse on state-building has been signifi- sure. Positive state-building, predicated on inclusive political cantly developed against the backdrop of countries emerging processes, is seen as an important component of the process from conflict, and the analysis of state-building and peace- by which states move from fragility to resilience (OECD building often go hand in hand. The two should not, howev- 2008a). er, be conflated. While they share some fundamental attrib- T H E W O R L D BA N K — U N D P 5 2.2 society, state-building is quintessentially political in char- acter: it is about how power and authority is used, and in whose interests; WHAT I State-building is not a technical process but a transaction- al one: it is essentially concerned with how the state inter- DO WE acts with society i.e. its legitimacy, responsiveness and MEAN BY accountability; and I State-building is principally an endogenous process, STATE-BUILDING? shaped by national actors. These elements—which are discussed further below—in turn inform the objectives and approaches of state-building interventions. STATE-BUILDING IS STILL QUITE A NEW AREA of development The social contract is central to the discourse on state- theory, with extensive analytical work currently underway to building. At its most elemental, this goes back to the tease out the concepts, processes and operational responses. Enlightenment concept of the individual surrendering person- As for the term “fragility,” there is no settled definition of the al sovereignty to the collective state in exchange for the term “state-building.” At its most value-neutral, state-build- maintenance of social order through rule of law. Over several ing “is the process through which states enhance their abili- centuries of European state-building, society’s expectations ty to function” (Whaites 2008, p4). Most definitions are of the state and the state’s expectations of its citizens have more normative, and converge around several common ele- expanded from the most basic bargain of taxation in return ments. These are captured in a recent OECD definition which for territorial security to a much broader suite of benefits and characterises state-building as: protections (OECD 2008b, Annex A). Locked into this is the notion of mutual responsibility and accountability: the an endogenous process to enhance American revolutionary slogan “no taxation without represen- capacity, institutions and legitimacy tation” captures the spirit. Inclusive social bargaining around of the state driven by state-society the construction of the social contract deepens its legitima- cy, and platforms and processes that foster participation in relations. (OECD 2008a) the ongoing negotiation of the common weal help to maintain its vigour. A subsequent DFID policy paper adds an important gloss State-building is “founded on political processes to nego- to this definition: tiate state-society relations and power relationships among elites and social groups” (OECD 2008a, emphasis added). In all contexts, state-building is These processes determine the character of the engagement principally about strengthening the between citizens and the state, and the extent to which states relationship between the state and are able to effectively negotiate and respond to societal society, and developing effective ways expectations without recourse to violence. Where there is a to mediate this relationship. (DFID 2009, p4) mismatch between expectations and performance, it can result in political tensions that may play out in instability or This gloss is the key to understanding the fundamental lead to a renegotiation of the political settlement (OECD paradigm shift that the state-building discourse embodies. 2008a, OECD 2010, p151). When we unpack the elements of the definition, they all turn A state-building perspective emphasises that functioning on this essential dynamic: institutions depend not only on their technical design, but on I State-building positions state-society relations – the social the social context within which they operate. “Formal institu- contract—at centre stage; tions need to be rooted in society otherwise they risk becom- I Because it is operating at the interface between state and ing mere shells or being captured by private or patrimonial 6 S TAT E - B U I L D I N G — K E Y C O N C E P T S A N D O P E R AT I O N A L I M P L I C AT I O N S I N T W O F R A G I L E S TAT E S interests” (OECD 2008a, para 10). This reflects directly on how the state performs and how it engages with society: its 2.3 responsiveness to the interests and expectations of citizens; its accountability to them for the way it exercises the powers BUILDING conferred upon it; and the legitimacy with which it acts i.e. the level of popular acceptance of its actions. While these BLOCKS elements have also formed part of the discourse on gover- FOR nance, the concept of legitimacy in particular has assumed greater prominence in the analysis of state-building. Lack of STATE legitimacy is seen as a major contributor to state fragility FUNCTIONING because it undermines state authority, and therefore capaci- ty (OECD 2010b, p7). As an endogenous process, state-building is something WHILE THE CONCEPTUAL MODELS FOR STATE-BUILDING are still that is done from within. Outsiders such as international evolving, one commonly applied framework is structured development partners can at best try to promote the process, around the basic building blocks for state functioning (see, but they cannot drive or control it. This presents a dilemma for example, Whaites, Fritz and Rocha Menocal and DFID). for international partners: how and how far can they seek to These are: use development assistance to guide what are essentially internal processes (Wyeth and Sisk 2009, p5). Another com- the political settlement mon dilemma is how best to engage where the political process is dominated by an elite which patently governs in its essential capabilities which the state must own interests. What, then, are the entry points for interna- have to survive; and tional partners to guide the state-building process in a benign direction without inadvertently shoring up the inter- expected capabilities which citizens look ests of elites at the expense of the wider society? to the state to provide, and which shore up its legitimacy. The political settlement is “the forging of a common under- standing, usually among elites, that their interests or beliefs are served by a particular way of organising political power” (Whaites 2008, p4). Settlements are represented as span- ning the continuum from negotiated peace agreements to long term accommodations, usually enshrined in a constitu- tion (Brown and Grävingholt 2009, p5). In essence, the set- tlement spells out the rules of the game, providing the insti- tutional underpinning for state functioning. There is a convergence of thinking around the essential capabilities of the state as: I the maintenance of security across the territory; I establishment and maintenance of the rule of law; and I collection of revenue to finance state functions. Without these capabilities, the state cannot establish effective dominion over its territory and even minimally sat- isfy its side of the social contract. T H E W O R L D BA N K — U N D P 7 The expected capabilities of the state, and the extent of their scope and performance, will vary widely from society to 2.4 society. Two areas of capability which consistently recur in the literature are: I developing and managing the conditions for economic WHAT growth; and MAKES I basic service delivery and livelihood security. STATE-BUILDING The characterisation of capabilities as essential or expect- DIFFERENT ed does not imply that the former are paramount or the first IN FRAGILE priority in state-building. Delivery of the expected capabilities helps to create the conditions of social and economic wellbe- STATES? ing that contribute to internal stability and resilience in the longer term. While there is broad agreement around what constitutes THE NUMBER OF STATES HAS TRIPLED over the last sixty years3 basic state functioning, there is considerably less confidence as a result of the decolonisation of the European empires in about how to put it into practice beyond the traditional, Africa, Asia and the Pacific, and more recently the breakup “technical” response of working with state actors to build of the Soviet Union and the former Yugoslavia. Almost all the capacity. “Evidence-based knowledge about ‘what works’ in countries included in the rollcall of fragile states have building and reforming states is surprisingly limited, despite emerged in this period.4 In many of these fragile states, there the numerous ‘public administration reform’ and ‘capacity was little or no sense of shared national identity at the time building’ projects that donors have supported” (Fritz and of their formation, the underpinning social contract for the Rocha Menocal 2008, p6). state was weak or non-existent, and shifting elite alliances have been left largely unchecked by the formal institutions to govern in their own interests. These factors contributed to the conditions for fragility. In fragile states, the gap between the model of the ration- al-legal bureaucratic state of academic literature and devel- opment practice and institutional forms on the ground is often very wide. These differences emerge along several axes: In fragile states, several systems operate alongside each other: the formal, the informal and the customary. While in the Western state the formal system is preeminent and there is a clear separation between public and private spheres, state-society relations elsewhere are more likely to be influenced by informal and customary rules, and personal relations based on kinship and community provide the basis of trust and the channel for accessing political and economic benefits (OECD 2010b pp8-9). In many areas, day to day activities may be framed and arbitrated within customary rules rather than within the rules of the formal system. Donors, on the other hand, largely engage with the formal system. In fragile states, the sources of legitimacy BOX 1 play out differently to the pattern seen in western states. The four sources of legitimacy widely discussed in the lit- erature—input or process legitimacy; output or performance legitimacy; shared beliefs; and international legitimacy—are summarised in Box 1. While none of these sources of legitima- cy exists in isolation, and no state relies solely on one of them, their interaction is critical to how state-society relations play out in a particular context, and impacts on fragility (OECD 2010b p12). Donors tend to focus on process and perform- ance legitimacy; this, however, can be hard to achieve given the characteristics of fragile states (ODI 2009, p2). In fragile states, the state is generally unable to establish itself as the highest SOURCES OF LEGITIMACY political authority and to penetrate and shape society. INPUT OR PROCESS LEGITIMACY This can manifest as a very limited territorial reach beyond When the legitimacy of the state is tied to agreed rules the national capital and main urban centres, as very limited of procedure through which the state takes binding decisions and organises people’s participation. In capacity to take and execute decisions that bind the society Western states these rules will be mainly formal (usual- as a whole, or as a very limited range of public goods. This ly enshrined in the constitution). In traditional political translates into heightened conflict and social contestation, orders, process legitimacy will be based on customary regions of lawlessness, incapacity to raise revenue and pro- law or practice. vide even a minimum level of public goods, and extreme lev- OUTPUT OR PERFORMANCE LEGITIMACY els of human insecurity. Defined in relation to the performance, effectiveness and quality of services and goods that the state delivers. SHARED BELIEFS Including a sense of political community, and beliefs shaped by social practices and structures, political ide- ologies, religion and tradition that allow people to see the state or other form of public authority as the over- arching, rightful authority. INTERNATIONAL LEGITIMACY Recognition of the state’s sovereignty and legitimacy by external actors, which in turn has an impact on its internal legitimacy. From OECD 2010b, pp11-12. T H E W O R L D BA N K — U N D P 9 3. 3.1 USING A ASSESSING PROGRAMMING STATE-BUILDING APPROACHES LENS TO USING A STATE-BUILDING LENS: INFORM PRACTICE SIERRA LEONE AND LIBERIA S ierra Leone and Liberia are small states on the coast converged during a decade of violent civil conflict in the of West Africa which share a common border and cer- 1990s, fuelled by a cross-border trade in guns and diamonds. tain parallels in their colonial and post-colonial histo- Poor governance and fourteen years of civil war trans- ry. Territory in both was acquired around two centuries ago to formed Liberia from a middle income country—albeit one resettle former slaves, and in both countries a profound politi- where growth did not translate into development5—to one of cal, economic and cultural gulf developed between the occu- the poorest countries in the world. GDP fell 90 percent pied coastal settlements and the indigenous interior. While between 1987 and 1995, one of the largest economic col- Sierra Leone became and remained a British colony until lapses ever recorded, and by the end of the conflict external 1961, in 1847 Liberia became the first independent black debt stood at 800 percent of GDP, making it proportionately African state, albeit one ruled by a small Americo-Liberian oli- one of the most indebted countries in the world. One third of garchy to the exclusion of the interests of the indigenous pop- the population was displaced over the course of the conflict, ulation. The histories of the Sierra Leone and Liberia again and around ten percent of the population died. A comprehen- sive peace agreement was signed in 2003 which provided for Entry points nominated by the Liberian Country Offices were: the establishment of a National Transitional Government and I Rule of law; the deployment of a UN peacekeeping force. A large UN I Sub-national state-building; and Peacekeeping Mission remains in place. I Support to the legislature. Sierra Leone was similarly impoverished by decades of unstable autocratic governments serving the interests of a A small team made up of World Bank (Fragile and Conflict- narrow political elite and by the predations of the civil war. In Affected States Group) and UNDP (Bureau for Crisis the course of the nineties, the economy contracted by an Prevention and Recovery) headquarters staff and two consult- average 7.1 percent annually and the government ran up ants (state-building and capacity development) visited Sierra massive budget deficits. Half the country’s population was Leone and Liberia for two weeks each8 to study the entry points displaced during the conflict, and around two percent lost nominated by the Country Offices, using a state-building lens. their lives. A peace accord was signed in 1999 establishing This is perhaps a rather heroic description for what was still an a transitional government, and a UN peacekeeping mission inchoate frame of analysis. Broadly and in part intuitively, we was deployed. This was replaced by a UN Integrated Office in approached the task using several points of reference: 2005 and subsequently by the UN Peacebuilding Office— I Backgrounding on the political history and political econ- the first of its kind—in 2008. omy to inform the context in which programming is As countries emerging from conflict,6 both Liberia and anchored; Sierra Leone confront magnified challenges associated with I A consideration of how programming was positioned with- the volatile transitions from war to democracy, the appalling in and contributed to the political settlement; extent of social dislocation and brutalisation that accompa- I A consideration of how programming contributed to the nied the conflicts, and the overshadowing role of UN peace- development of state capacity; keeping operations as an initial point of engagement by the I A consideration of how programming responded to social international community. Both countries sit close to the bot- expectations of the state, both in terms of what the state tom of the Human Development Index, and both perform does, and how it does it (using accountability as a default poorly on the Country Policy and Institutional Assessment measure of the integrity of the social compact); and that provides the basis of the World Bank’s classification of I A consideration of how programming aligned with and 7 states as fragile (World Bank 2006, p4). supported perceptions of legitimacy (e.g. the formal jus- Both the World Bank and UNDP Country Offices in Sierra tice system may be seen as inaccessible, inappropriate, Leone and Liberia agreed to collaborate with the fragile states corrupt and “unjust” by local communities who find much teams in their respective headquarters to hold their program- greater legitimacy in customary law). ming up to a state-building lens. While none of the four Country Offices had a state-building strategy as such, a num- In the limited window of time, the exercise was more an ber of their programs were seen to be contributing broadly to opportunity to test the usefulness of the parameters than to state-building processes and in each country the two offices apply them in any systematic way to build up an overall picture together identified specific entry points to be considered by of state functioning and explore entry points for future work. the joint field missions. A further consideration in the nomi- Nonetheless, the use of a state-building lens provided a useful nation of entry points was that they should engage comple- way of understanding the impact both of current areas of work mentary programming interests of the World Bank and UNDP. and prospective priorities, and the findings are set out in the country reports prepared immediately following the missions. The Country Offices in Sierra Leone nominated the follow- The exercise argues for the merits of looking at specific ing entry points: areas of programming through a state-building lens as a I Public sector reform; straightforward means of engaging with this discourse, I Support to Parliament; improving program impact and avoiding “doing harm.” It also I Decentralisation and local government; and argues for the value of commissioning a broader assessment I The effectiveness of program implementation units. of state-building and associated entry points – something this exercise was not able to do. T H E W O R L D BA N K — U N D P 11 3.2 OPERATIONAL CONSIDERATIONS HIGHLIGHTED BY THE COUNTRY MISSIONS THIS SECTION DISCUSSES SIX POINTS of operational guidance highlighted by the missions to Sierra Leone and Liberia: I Building a broadbased understanding of the political economy and the drivers of state-building; I Understanding how programming may impact on the political settlement and the political processes underpin- ning it, and possible downstream consequences; I Understanding the impact of aid modalities on state- building; I Working system-wide and for the long term; I Matching the development approach to the context; and contexts. This call is consistently and forcefully repeated I Taking ethical responsibility for champions in contested across the contemporary literature on state-building. environments. Because state-building is quintessentially political and These six areas of guidance are discussed further below. endogenous, it has to be approached as such. To engage effectively in support of state-building, it is essential to understand the colonial and post-colonial history of the coun- try, the institutional legacies, the economic and societal pro- file of the country, the character and nature of the interaction 3.2.1 between the formal, informal, traditional and shadow sys- Invest in developing tems, who the players are and what they stand to gain or lose from change, where the fault lines are in the society and the a broad-based understanding root causes of conflict past and present. Without this knowl- of the political economy edge, we are flying blind, and may inadvertently do real harm: and the drivers “In almost all cases, the biggest risks of doing harm in state-building emerge from a lack of deep and detailed histor- of state-building ical and local knowledge of the political processes, political settlements, patterns of state-society relations and sources of legitimacy in the countries where donors are operating. While in many countries where donors are operating the costs of THE FIRST OF THE OECD’S Principles for Good International gaining this type of knowledge may be too high to be justified, Engagement in Fragile States and Situations (OECD 2007) is in fragile states the costs of not having such in-depth knowl- “take context as the starting point.” The accompanying text edge are far too high.” (OECD 2010a, pp120-121) emphasises the need for international actors to understand One approach to building understanding of context is to the specific context in each country and to use political analy- commission a political (or political economy) analysis. DFID sis to adapt international responses to country and regional led the pack with its Drivers of Change studies which consid- 12 S TAT E - B U I L D I N G — K E Y C O N C E P T S A N D O P E R AT I O N A L I M P L I C AT I O N S I N T W O F R A G I L E S TAT E S er the dynamic interaction between three sets of factors: structures, institutions and agents (DFID 2009b, p9). Many other donors have developed broader governance assessment tools which, in some instances, include significant elements of political analysis.9 In other circumstances, one-off studies are commissioned to contextualise a particular area of devel- opment, such as sub-national governance. While the tools and studies have proliferated, a question mark hovers over their impact. As yet there appears to be lit- tle evidence that political analysis is prompting donors to question underlying development assumptions (Unsworth 2008), and the challenges of linking analysis to action and establishing the right incentives to ensure that findings are put into practice operationally are frankly acknowledged (DFID 2009b, p24). In Sierra Leone, the mission had access to two political economy analyses—a Drivers of Change study undertaken by BUILDING A DFID, and a World Bank Report on Governance and Political BROAD-BASED Economy Constraints to World Bank CAS Priorities in Sierra UNDERSTANDING Leone. Although no similar studies were available for Liberia, OF THE POLITICAL for both countries the team had access to a rich stream of ECONOMY information and analysis embedded within PRSPs, country assessments, program documents, Truth and Reconciliation Access/commission political economy KEY MESSAGES Commission Reports, Security Council reports, academic pub- analysis and act upon the intelligence that lications,10 and reports from authoritative international organi- the analysis provides; sations and think tanks such as the International Crisis Group, the Centre for Humanitarian Dialogue and the Carter Center. Engage with others in-country who are Increasingly, such material is publicly available or shared analysing the country context and institu- within the development community. The Accra Agenda for tional developments, and share information; Action commits donors to conducting joint assessments of governance and the causes of conflict and fragility, engaging Consider establishing a network of actors with developing country authorities and other relevant stake- who are authoritative on local context and engage with them regularly to ground truth holders to the maximum extent possible (Article 21), and political and social developments with a DFID has committed to conducting its political economy view to identifying implications for current analyses wherever possible with the wider development com- and future programming. munity to encourage shared understanding and joint action (DFID 2009b, p1). The most recent guidance on conducting political assess- ments emphasises that they should not be a one-off exercise, but a dynamic process where knowledge is continuously analysis is not simply the province of the visiting expert com- updated over time and fed back into programming (DFID missioned to produce a report. Particularly in fragile contexts 2009b, p20). Consistent with this, donors have been encour- which are in a state of flux and where analysis can quickly aged to invest more heavily in “the generation and dissemi- become out of date, programme staff should be thinking nation of good quality, accessible local data and related pol- about issues around incentive structures, becoming familiar icy analysis” (Unsworth 2008). Importantly also, political with the major players, looking for emerging change agents T H E W O R L D BA N K — U N D P 13 3.2.2 and identifying who stands to gain or lose from particular Consider how programming developments. This in turn suggests that staff should spend significantly more time in-country in order to gain an under- may impact on the standing of the political and social dynamics of the environ- political settlement and ment within which they are working (Unsworth 2009). In Sierra Leone, the team was impressed by the approach the political processes of the integrated UN Peacebuilding Mission to sharing ongo- underpinning it, and ing political analysis and reporting across the UN system, possible downstream with the heads of the political and civil affairs units of the Mission included in the regular meetings of UN agency consequences heads. Political affairs units within UN missions are a useful source of analysis and day to day political reporting, and it is important that development personnel involved in state- THE SECOND OF OECD’S Principles for Good International building ensure they have continuing access to these and Engagement in Fragile States and Situations (OECD 2007) is other sources of analysis and interpretation in order to “do no harm.” A recently published report Do No Harm: “ground truth” developments as they unfold and assess the International Engagement for State-building (OECD 2010a) impact on current and planned programming decisions. documents how donor actions can harm state-building process- While political analysis is of its very nature potentially es, and urges donors to look for both the intended and unin- sensitive, there are also sound arguments for involving local tended consequences that may flow from their interventions. actors in its development and sharing assessments with gov- “Donors can inadvertently do harm when the resources ernment and national partners where feasible. This can only they deliver or the policy reforms they advocate exacerbate help to make for a more productive exchange in arriving at rather than mitigate the conditions for violent conflict, or common understandings. There is rarely a single version of weaken rather than strengthen the state as a site of decision reality, so it is important to understand all major perspectives making and policy formation over the deployment of public and to bring them to the discussion. DFID, one of the strong resources.” (OECD 2010a, p29) advocates of analysis, suggests that the scope for engaging The potential for harm can arise either from what we do i.e. government and national partners be judged on a case by the substance of the intervention, or how we do it i.e. the case basis: “In difficult political environments, full disclo- modality employed. This section of the paper addresses the sure of findings may serve to undermine relationships and substance of our interventions, and the next section looks at fuel tensions. However, in more permissive contexts, the ben- the modalities we employ, drawing on the experience in Sierra efits of working with national governments and other partners Leone and Liberia. Both sections signpost activities which can often outweigh the costs.” (DFID 2009b, p22) impinge directly on the authority and capacity of the state. Post-conflict state-building engages significantly around the institutions that determine the distribution and exercise of power and authority, and this section explores the attendant risks by considering four areas that generally feature in peace settlements and associated state-building agendas:11 constitu- tional development; conduct of elections; transitional justice; and decentralisation. Each had ongoing currency and elements of controversy during our missions in Sierra Leone and Liberia. CONSTITUTIONAL DEVELOPMENTS BOX 2 CONSTITUTIONS EMBODY THE CHARACTER of the political settlement (see Box 2), and constitutional reform is often a priority in post-conflict peacebuilding and state-building. In Sierra Leone and Liberia, constitutional reform is on the agenda of both the international community and the govern- ment, although the sense of urgency and commitment seems rather greater on the international side. In Sierra Leone, a Constitutional Review Committee was set up in early 2007 and reported to the President in early 2008.12 Despite references as far back as 2007 in UN doc- uments to the prospects of early constitutional change,13 lit- tle seems to have happened. At a meeting of the Security Council on 8 June 2009, the Foreign Minister for Sierra Leone reported that the Constitutional review had been referred to the relevant Cabinet Committee and the Security CONSTITUTIONAL REVISION Council, in its 15 September 2009 resolution on Sierra Leone, emphasised the engagement of the UN mission with Political transitions are often marked—for good or government to support the constitutional reform process. ill—by the overhaul of the national constitution. The In Liberia, although constitutional reform is positioned constitution of Sierra Leone, for example, has served prominently on the donor policy agenda, domestic commit- as a barometer of the country’s changing political for- ment to it appears equivocal at best. During our mission, tunes over the last half century. At independence, the constitution was framed around a Westminster system international interlocutors spoke critically of the country’s of government; in 1971, in the wake of an unsuccess- history of “imperial presidency” and domination by the exec- ful military coup, Parliament declared the state to be utive, and argued forcefully for a reform of the constitution as a republic, the Prime Minister morphed into the part of the process of institutional realignment. In February President and the constitution was amended accord- 2009, the UN Secretary-General’s report to the Security ingly; in 1978, the constitution was overhauled to Council on Liberia noted that the President had constituted a establish a one-party state; and in 1991 the constitu- tion was once again amended to reinstate a multi- Constitutional Reform Task Force, in accordance with a rec- party system, although its implementation stalled ommendation from the Governance Commission. A less posi- after a military coup in 1992. tive picture of progress emerged from the Secretary-General’s August 2009 report to the Security Council, which advised that the Task Force had yet to be constituted. A fair interpre- tation of the delays and the prominence accorded the issue since they directly determine who will hold power and who in Security Council reporting would be that there is little local loses out. The very language—”winning” or “losing” an elec- enthusiasm for the constitutional reform agenda. tion—captures the essence of competition. Pursuing elections as a means to resolve unresolved armed ELECTIONS conflict or in a country where political allegiances are along ethnic lines can either exacerbate conflict or, more certainly, TYPICALLY PEACE AGREEMENTS and the mandates of leave its roots and causes unaddressed (OECD 2010a, p41). peacekeeping missions established in their wake include the The 1997 election in Liberia, held under the terms of the conduct of elections as important milestones along the path Abuja II peace agreement, saw the notorious warlord Charles towards the consolidation of legitimate political authority.14 Taylor voted in as President in a landslide. The choice for vot- However elections are also a lightning rod for contestation ers was between voting Taylor in or a return to war.15 T H E W O R L D BA N K — U N D P 15 “…fear of what Charles Taylor might do if he were to lose 2007 and the rule change may well have worked against them, the election apparently played a great role in consternating as previously their election was assured if they were assigned a many to vote for him. Perhaps the best expression of the grim high enough slot on their party’s district ticket. paradoxes that catapulted Taylor into power was indicated by The fairness of elections is determined by much more the common electoral rendition by the teeming Liberian than what happens on and around polling day. At least as youths who supported Taylor: “He killed my ma, He Killed My important as the electoral event is the electoral machinery— pa, I’ll vote for him.”16 the system of rules and procedures governing voter registra- The form of the electoral process that is brokered as part tion, the accuracy and completeness of the roll, the relative of a political settlement or peace agreement can have a pro- size of constituencies, the criteria for assigning voters to par- found effect on the legitimacy of electoral outcomes. In ticular constituencies, boundary demarcation, and voter Liberia, the Comprehensive Peace Agreement signed in 2003 information and education—which can work to include or expressly provided for the temporary suspension of some pro- exclude, to enfranchise or disenfranchise. visions of the constitution relating to elections—a deal In Sierra Leone, the National Electoral Commission is reportedly done to keep the parties at the negotiating table. keen to see technical support provided much earlier in the One of the suspended provisions was the requirement that electoral cycle and to continue after an election to support members of the Legislature be elected with an absolute ongoing development of capacity. Although the Commission majority (involving where necessary a run-off ballot between has done some good work on the national electoral machin- the two top ranked candidates). Instead, a first-past-the-post ery, the machinery for local elections appears less robust: the system was adopted and, with a plethora of political parties demarcation of local boundaries has been criticised as using and independents contesting the election, winning candi- outdated census data that does not capture subsequent pop- dates were elected on very small percentages of the total ulation movements and provisional ward rolls were not dis- vote. This change is blamed for delivering a Legislature of played for verification, leaving many people disenfranchised. dubious composition and ability. Criticisms of the performance of the electoral machinery In Sierra Leone the electoral rules were also changed to in Liberia are more extensive: poorly demarcated constituen- facilitate elections in the wake of conflict. The 1996 elec- cy boundaries; one-off mobile registration which failed to tion, which marked a brief intermission in the sequence of reach all constituents; disenfranchisement of voters who were military governments, adopted proportionally elected multi- not given adequate documentation and poor location of member district constituencies in the place of single member polling stations. The National Electoral Commission is itself local constituencies. The change was explained in terms of worried about the enormous logistical challenges of the ter- the difficulty in conducting a reliable voter registration due to rain and access and the shortcomings of a rural network that the high rate of population mobility caused by the conflict. has not been serviced for two decades. Major effort is need- The change was retained for the first election following the ed before the next election to delineate constituency bound- peace agreement, held in 2002, but reversed in the lead-up aries (one of the constitutional requirements suspended by to the 2007 election. the Peace Settlement which is now operative again) and to The reversion to single member constituencies in 2007 prepare for run-off elections now that the constitutional pro- meant that individual candidates had to engage with their con- vision for an absolute majority has been reinstated. The elec- stituents and declare how they would represent them—a real toral roll also needs updating to incorporate many internally test of democracy. On the downside, the change is portrayed as and externally displaced Liberians who have moved since the intensifying political contestation by putting candidates head last election. to head in a very personal competition for voter support. The In both Sierra Leone and Liberia, the elections scheduled 2007 national elections saw attempts to block candidates from over the next two years are already casting a long shadow (see campaigning in some localities, and the 2008 local elections Box 3). This highlights the significance of careful analysis, were marred by the intimidation of independent and women planning and implementation through the entire electoral candidates in the lead-up and by a very low voter turnout on cycle, given that electoral-related violence can manifest itself polling day. The number of women MPs elected dropped in in different forms and intensities in each of its phases.17 16 S TAT E - B U I L D I N G — K E Y C O N C E P T S A N D O P E R AT I O N A L I M P L I C AT I O N S I N T W O F R A G I L E S TAT E S BOX 3 TRANSITIONAL JUSTICE ELECTORAL JOCKEYING TRANSITIONAL JUSTICE MECHANISMS such as Serious IN SIERRA LEONE AND LIBERIA Crimes Commissions and Truth and Reconciliation Commissions are a common element of peace agreements, In Sierra Leone, the electoral cycles for the President, becoming “a priority state function in the aftermath of con- the Parliament and local councils converge in 2012, and the political space is already becoming more con- flict” (World Bank/UNDP 2005, p7). They respond to the tested in what is basically a two party system. A ache in society to see wrongs admitted and pursue reconcili- national election in 2007 handed government to the ation. At the same time, they are inherently political: deci- party which had long been out of office and carried the taint of its 1980s introduction of a one-party state. sions about what and whom to focus on can become highly While the transition occurred peacefully, there is dis- controversial, and their proceedings can accentuate griev- quiet that the government is growing increasingly par- ances and, on occasions, produce tectonic shifts in the rela- tisan, and party support is increasingly dividing along tionships of power and authority within a country. Perceived regional and ethnic lines. The new Cabinet appointed in March 2009 is dominated by northerners, and more government delays in responding to their findings may also than half of them are from the President’s own dis- become a lightning rod for popular resentment and political trict. The colours of the ruling party are increasingly mobilisation. in evidence at government sponsored events and on In Sierra Leone, the work of the Truth and Reconciliation government buildings. In March 2009 serious violence broke out between supporters of the two parties, Commission proceeded swiftly and relatively uncontroversial- egged on by partisan radio stations, and the opposi- ly and its 5000 page report was released in October 2004, tion party headquarters was destroyed and female four years after the Commission’s establishment. However staff raped. On the same weekend, a rural by-election was disrupted by violence and postponed, with a very NGOs were subsequently critical of the long delays in imple- low voter turnout on the rescheduled polling date menting the recommendations of the Report, including a probably skewing the outcome. Fears of ongoing polit- reparations program for war victims as foreshadowed in the ically motivated violence are growing as the electoral 1999 Peace Agreement. In late 2007 the United Nations and stakes build, as is unease about the consequences should electoral results be seriously discredited. Sierra Leone’s Human Rights Commission urged the Government to produce an implementation strategy for the In Liberia there is already nervousness about the like- report, and a year later the Government, with UN assistance, ly political jockeying around the Presidential and strengthened the organisational arrangements for repara- Legislature elections due in 2011, with a proliferation of parties, weak party organisation and a wide open tions; initial payments to victims were made in early 2010. field increasing the potential for instability. Although The Report of Liberia’s Truth and Reconciliation the 2005 Presidential election delivered a candidate Commission, which was released during the team mission to who is highly regarded by the international communi- ty, her party holds only a handful of seats in a the country, dropped like a grenade into the leadership circle, Legislature which controversially includes a number naming many prominent figures. Over 100 factional leaders of former warlords and their backers. Over the four and associates were recommended for prosecution, and a years since the election, the Legislature is seen to “non-exhaustive” list of 50 people were named as financiers have played a spoiling game, stalling the passage of annual budget bills and other key legislation underpin- or supporters of the warring factions and recommended for ning the President’s reform program. When the mis- public sanctions, including barring from public office for 30 sion visited in mid 2009, the President’s own political years. The list included the internationally feted President future was under a cloud, with the release of the Truth and a number of other politicians and serving members of the and Reconciliation Commission report which recom- mended that she and others be barred from political Government. office for 30 years. A number of Legislators had also The message here is that, once set in train, transitional been identified by the Truth and Reconciliation justice can create its own force field which must be under- Commission as implicated in the violence of the past. stood and sensitively handled. As one senior Liberian observed to us, “While discourse about impunity and recon- ciliation can be unsettling and carry risk to state-building, the T H E W O R L D BA N K — U N D P 17 process of state-building can never be successful if these actors generally prefer some variant of deconcentration. challenges are ignored or left unattended.” However national and international actors alike will all be speaking in terms of “decentralisation,” leaving considerable DECENTRALISATION room for misunderstanding and frustration, and it is critical to ensure clarity on all sides about what form is intended in THE INTERNATIONAL DEVELOPMENT COMMUNITY has a particular development setting. often placed its policy weight behind decentralisation as a The jury is still out on how best to approach decentralisa- means of enabling citizens to participate more directly in gov- tion. One wide-ranging review found that many of the promis- ernance processes and empowering those previously exclud- es of decentralisation had not been met, or the results had ed from decision-making,18 and as a way of improving access been mixed. No consistent evidence was found to document 19 to public goods and services at the periphery. The term that decentralisation had improved efficiency, equity or service “decentralisation” comprehends a broad sweep of institution- delivery, although the picture was more promising in terms of al arrangements. Within a unitary state, decentralisation may participation. Importantly, the review found that the success of simply mean the delegation of administrative decision-mak- decentralisation efforts may depend as much on contextual ing from the centre to offices located closer to the people factors—such as the character of the regime, the degree of (deconcentration); alternately, it can involve the transfer of power-sharing at the centre, the ethnic constellation or whether functions or authority from central levels of government to the policy was adopted in response to local or international sub-national institutions governed by locally elected repre- advocacy—as on the design of the particular model of decen- sentatives (devolution). In a federal system, decentralisation tralisation.20 Another overview of the impact of decentralisa- involves a constitutional division of sovereignty between the tion in Africa concluded that the creation of stronger local gov- constituent states and the federation as a whole. Typically, ernments with control over revenue has typically served to sub-national actors are seeking the transfer of elements of decentralise corruption rather than eradicate it, and found no political decision-making and fiscal authority, where national evidence of gains in terms of fiscal efficiency.21 Both Sierra Leone and Liberia were characterised by high- patchier across Cabinet and within the bureaucracy, and its ly centralised and non-inclusive government and very weak implementation has been driven by a Project Implementation penetration of public goods and services into the rural areas. Unit established outside the relevant government department In the period of post-conflict reconstruction, decentralisation and operating quite independently of it. By the time of our was accordingly high on the international agenda. Its process mission in mid 2009, the majority of functions earmarked for of adoption in the two countries, however, has been rather transfer to the districts had yet to be devolved, and funds and different. staff had yet to follow functions. With a change in local gov- In Sierra Leone, decentralisation was a major plank of the ernment minister earlier in 2009, the process at last post-conflict recovery strategy, strongly advocated by donors appeared to be gaining some momentum. and embraced by the then President who reportedly had him- In Liberia the move towards decentralisation has been self once been a district administrator. The machinery for rather more sedate and locally orchestrated. Current policy22 local government was set up soon after the 1999 peace set- commits the Government to an improved system of gover- tlement, and was backed up by a substantial World Bank pro- nance that is more localised and more responsive to the gram which built on initial work by UNDP. While decentrali- needs and aspirations of citizens across the country, and sation had high level political commitment, support was decentralisation is foreshadowed although it is also empha- sised that “this process will take years, possibly decades, to complete.” During the mission’s visit in mid 2009, the Governance Commission released the draft of a National Policy on Decentralization and Local Governance which pro- posed a system of elected county governments within a uni- tary state. Meanwhile the Liberian Government has moved ahead with some deconcentration of public administration, appointing superintendents at provincial level to coordinate and manage government functions. Even without political decentralisation, much can be done —with political will—to improve the availability and respon- siveness of government services at the local level and to broaden the opportunities for local communities to contribute TAKING CARE to planning and allocative decisions through formal consulta- THAT PROGRAMMING tive processes. It will be informative to see how far Sierra DOES NOT UNDERMINE Leone and Liberia have moved towards this goal in five years STATE-BUILDING time, and the relative impact of devolved and deconcentrat- Identify programming that directly impacts ed models. KEY MESSAGES on the distribution of power and the stand- As a mechanism for the redistribution of power and influ- ing of political elites—notably in areas ence, decentralisation produces clear winners and losers. such as constitutional reform, electoral One group of collateral losers in both Sierra Leone and planning, transitional justice and decen- Liberia are Members of Parliament, who feel they have lost tralisation. authority through the loss of obvious influence over the dis- When programming in these areas, consid- tribution of public goods and services. One MP in Sierra er the effects of institutional design on the Leone measured the loss of influence in the new seating distribution of power and influence, identi- arrangements at local functions, where MPs are put at the fy potential winners and losers, and assess low tables on plastic chairs while vice-Ministers sit at the the downstream consequences of program high table. MPs’ support for decentralisation will be impor- design. tant if the policy is to succeed, and appropriate ways to engage them can and should be found. T H E W O R L D BA N K — U N D P 19 3.2.3 Consider the ects, supporting day-to-day project execution, and developing project specific monitoring data. impact of aid modalities USE OF GOVERNMENT BUDGET AND POLICY PROCESSES on state-building IN SIERRA LEONE, the four leading donors—DFID, EC, the World Bank and the African Development Bank—entered into THE PROVISIONS OF PEACE SETTLEMENTS generally include a an MoU with the Government in 2006 establishing a mechanism for the formation and installation of a transition- Framework for Multi-Donor Budget Support,24 and 13 percent al government—this was the case in both Sierra Leone and of total ODA is provided through the budget, with a further 14 Liberia. Settlements crystallise the power relations at the percent being provided through other forms of programme- time of negotiations (Brown and Grävingholt 2009, p.11), based approaches.25 Over the last few years, a number of ini- and this is reflected in the composition of transitional govern- tiatives have been designed as Sector Wide Approaches 23 ments, which may be highly compromised. This poses a (SWAPs) or implemented through Multi-Donor Trust Funds fundamental dilemma for the international community, (MDTFs). The World Bank’s two new activities in the area of whose efforts to build capacity and support basic government governance (the Basic Services Programme and the Integrated functions may serve to shore up the standing of unsavoury Public Financial Management Reform Project), for example, political figures and provide a rich stream of development are designed respectively as a SWAP and a MDTF. The dollars for predatory elites. Ministry of Health is also supported through a SWAP. Even where the composition of the government is uncon- In Liberia, almost all ODA is dispersed outside the Budget troversial, its organisational capacity has generally been hol- and country systems.26 In their joint Country Assistance lowed out by years of conflict, the exodus of the most educat- Strategy for Liberia for 2009-2011, the World Bank and the ed and internationally mobile citizens, the disintegration of African Development Bank have indicated that they will use administrative systems and processes and the widespread Budget support as a modality, in part as a signal to other destruction of the physical infrastructure for public adminis- development partners to channel more funds through the tration and service delivery. Budget.27 These dilemmas directly shape the selection of aid modal- Efforts are being made in both Sierra Leone and Liberia to ities. Despite the injunctions of the Paris Declaration on Aid align aid strategies with government policies, to harmonise Effectiveness and the Accra Agenda for Action, the feasibili- donor approaches and to strengthen government coordination ty of providing direct budget support or even of working of donor effort. In both countries, the World Bank and UNDP directly with paralysed government systems often appears have structured their country assistance strategies around the remote. Yet, to the extent that aid bypasses government sys- national Poverty Reduction Strategy. In Sierra Leone, the tems, it can undermine the authority and legitimacy of the work of all UN agencies, funds and programmes is coordinat- state by neutering the policy-making role of government and ed through a Joint Vision which integrates the UN’s contribu- the coherence of the national Budget as the principal vehicle tion to the PRSP, its mandate from the Security Council and for public policy, and undercutting the authority of its contribution to the goals of the UN Peacebuilding Parliament in the appropriation of public funds and oversight Commission. In a major innovation, a singe Multi-Donor Trust of public expenditure. Fund for the UN system in Sierra Leone is proposed to under- Projectised aid not only retards the development of regu- write the implementation of the Joint Vision (see Box 4). lar government processes for policy development and imple- In Liberia, development assistance is coordinated through mentation, but also creates huge transaction costs for weak a high level Reconstruction Development Committee, estab- governments. Government officials, rather than using their lished in 2006 as a platform for dialogue between the gov- working time to plan and manage government programs, are ernment and development partners on implementation of the absorbed by the demands of donor liaison: negotiating proj- Government’s PRSP. The Committee is chaired by the 20 S TAT E - B U I L D I N G — K E Y C O N C E P T S A N D O P E R AT I O N A L I M P L I C AT I O N S I N T W O F R A G I L E S TAT E S President and includes key Ministers and major development THE EMERGENCE—AND WIND-BACK—OF A partners. In Sierra Leone, aid coordination and management PARALLEL PUBLIC SECTOR is being strengthened to support implementation of the Government’s Agenda for Change (PRSP) and the Ministry of ONE OF THE MOST CHALLENGING DILEMMAS for donors in Finance and Economic Development is being restructured to post-conflict settings is how to simultaneously build capacity more effectively attract development assistance into the and deliver results: the temptation is to trade investment in Budget. The head of the UN Peacebuilding Mission and government capacity off against urgency through short to UNDP are supporting these changes. medium term skill substitution, while espousing capacity development over the medium to long term. The risk is that, in the interim, a parallel public sector is constructed that BOX 4 becomes extremely difficult to dismantle, and local capacity is choked off. In both Liberia and Sierra Leone, parallel public sectors have emerged. The symptoms are more acute in Sierra Leone where post-conflict state-building kicked off three years ear- lier. Sierra Leone’s public service is characterised by an age- ing senior management cadre close to retirement, a “missing middle” and a concentration of 80-85% of staff at junior UN JOINT VISION AND MULTI-DONOR clerical grades, and an attendant inability to translate govern- TRUST FUND FOR SIERRA LEONE ment decisions into actionable and actioned administrative processes in any meaningful way. Instead, policy implemen- The UN Joint Vision is an integrated strategy across tation and budget execution is driven by a constellation of UNIPSIL and the 17 UN organisations represented in parallel arrangements, from Project Implementation Units to Sierra Leone, replacing the more traditional UN special units and positions heavily subsidised by donors (see Development Assistance Framework (UNDAF). Unlike Box 5, next page). the costly and time consuming process for develop- Sharp salary disparities between mainstream civil servants ment of an UNDAF, it was developed and negotiated and those in parallel arrangements have siphoned critical skills locally over a few months, led by the Executive Representative of the UN Secretary-General. At its out of government, and are corrosive to the morale and motiva- core is a seven page outline of five strategic priorities tion of those left behind. The Government and donors are now and their rationale. This is a simple document in lan- caught in a Catch 22: the parallel system has developed to the guage that the government understands. It cuts point where it is indispensible, but the costs of sustaining it through the plethora of plans and strategies of individ- make it unaffordable in the longer term. At the same time, ual organisations (a stocktake one year back identi- efforts at civil service reform over the last few years have been fied 35 strategies for Sierra Leone across donors and agencies.) largely stillborn. The situation has reached a tipping point, and the solution will not be easy to find or implement. Following the launch of the Joint Vision, the UN coun- try team moved to develop integrated programmes LEGITIMACY: A TWO-WAY STREET and to establish a single Multi-Donor Trust Fund to support the contribution of all UN organisations con- IT WAS PUT FORCEFULLY TO THE MISSION that high donor tributing to the implementation of the Joint Vision, which has been costed at $345million over four years. visibility can undermine the legitimacy of the state. In the This is a significant innovation to strengthen coordina- words of one Minister: tion and efficiency and streamline delivery. “Every donor wants to hang their flag on what they do. If the perception of the people is that donors do everything, then voters won’t care who they vote for because the interna- tional community meets their needs.” T H E W O R L D BA N K — U N D P 21 We were ourselves struck by the high visi- BOX 5 ELEMENTS OF THE PARALLEL PUBLIC SECTOR IN SIERRA LEONE bility of the international presence as a default government: down to government SEVERAL CATEGORIES OF PERSONNEL are operating offices in the districts still being painted in alongside mainstream public servants in Sierra Leone, the UN colours of blue and white. At a subtler and together these categories of personnel are perform- level the proliferation of parallel systems in ing most higher level policy development, planning, government, as well as creating major down- budgeting and policy implementation functions, leaving stream challenges for the coherence and sus- routine administration of procedures to the mainstream tainability of public administration, can also public service. The parallel public sector is made up of the following categories: compromise the authority of government. But, at times, working through government OCCUPANTS OF KEY PUBLIC SERVICE POSTS in receipt of becomes a bridge too far. In Liberia, during salary supplements: donors and government have agreed the life of the Transitional Government, cor- to salary supplementation of certain officials, mainly at ruption and mismanagement of public funds the senior level, in order to either recruit or retain them; had grown so serious that, in the view of inter- LOCAL TECHNICAL ADVISERS (LTAS), mainly recruited national donors, it threatened Liberia’s transi- from the private sector and civil society in Sierra Leone tion and the prospects for stable peace. or from the diaspora to work under contract to donors Responding to the threat, lead donors jointly within the public sector, but not occupying establishment conceived a highly interventionist program— positions. In some departments and agencies, they have the Governance and Economic Management become the core of the workforce and dominate the Assistance Program—which inserted interna- technical/professional ranks. Most heads of divisions and technical units in the Ministry of Finance, for exam- tional personnel with signing authority into ple, are headed by LTAs, and the salary costs for the 45 several critical areas of fiscal management LTAs in the Ministry is five times the total wages bill of the and oversight (see Box 6). The Program was Ministry; reluctantly agreed to by the Transitional Government and such was its scope and NON-NATIONAL TECHNICAL ADVISERS (TA): linked to bilater- potential duration that it was determined that al/multilateral development projects or supplied by groups such as the Overseas Development Institute and the Tony the plan should be submitted to the Security Blair Foundation; Council for endorsement and adjustment to the mandate of the Peacekeeping Mission. PROJECT IMPLEMENTATION UNITS (PIU)S: PIUs are GEMAP has been described as: responsible for the execution of significant development projects/programmes supporting government functions, and take a variety of forms. They may be strictly separate “…a possible compromise from the structures of Government as in the case of the between two unpalatable policy Decentralisation Secretariat, or completely integrated as options for international actors in the Ministry of Health. in post conflict situations: imposition of a temporary COMMISSIONS AND OTHER AUTONOMOUS TRANSITIONAL international trusteeship, or the STRUCTURES: these are organisations outside and paral- lel to the main structures of government. Their mandate exercise of a full range of action is to manage the delivery of a particular task or function. by a transition regime drawn They differ from PIUs in that they are not linked with the from leaders of former warring functions of a particular department or agency. parties, in which considerations of popular legitimacy, compe- tence and commitment may be secondary.”28 22 S TAT E - B U I L D I N G — K E Y C O N C E P T S A N D O P E R AT I O N A L I M P L I C AT I O N S I N T W O F R A G I L E S TAT E S While GEMAP was forced on the Transitional Government, BOX 6 the Government elected in 2006 is closely engaged in its management: implementation is supervised by an Economic Governance Steering Committee chaired by the President with an international development partner as deputy, and Committee members include senior government personnel, donors and civil society. This supervisory arrangement was described as one of the most innovative aspects of the pro- gramme when it was reviewed in 2006.29 GEMAP is now winding down and, given its profile and the wider interest it has attracted as a development modality, it may be timely to evaluate it in these terms. GOVERNANCE AND ECONOMIC MANAGEMENT ASSISTANCE PROGRAM (GEMAP) GEMAP ADDRESSES SIX AREAS: improving revenue capture across SOEs and agencies, including through the deploy- ment of international experts with co-sign- ing authority; TAKING CARE THAT AID MODALITIES improving budgeting and expenditure man- DO NOT UNDERMINE agement; STATE-BUILDING improving procurement practices and the In programming, determine how and to granting of concessions; KEY MESSAGES what extent activity can be aligned with government policy and systems (including establishing effective processes to control budget systems) corruption, including the establishment of an Anti-Corruption Commission and sup- Maximise coherence with government poli- port for the investigation of serious fraud, cy and systems within the constraints of corruption and economic crimes; the operating environment supporting key institutions such as the Identify and develop opportunities for General Auditing Office, the General donor harmonisation and the use of multi- Services Agency, the Governance Reform donor programs and trust funds Commission and the Contracts and Monopolies Commission; and Before establishing a PIU or installing num- bers of technical advisors and/or salary capacity-building. assisted personnel, develop a transition plan and timetable for reintegrating func- The implementation of GEMAP is guided by an Economic tions sustainably into government struc- Governance Steering Committee, chaired by the President tures and budget and with a development partner as deputy chair. Selected Ministers, chairs of key institutions including the Be sensitive about the optics of donor Governance Reform Commission and local heads of badging. multilateral organisations and governments are members. T H E W O R L D BA N K — U N D P 23 3.2.4 Think and work gent of advisers, and it is pumping out audit reports which are exposing high levels of malfeasance and administrative inca- system-wide, pacity across government organisations, and publishing them and for the on its website. However that is where the matter ends: the Legislature is not following up on audit reports, and bottle- long term necks in the courts and suborned juries are leaving suspects unprosecuted and defendants unconvicted. The Auditor- General has decried the demoralising effect of this on his STATE-BUILDING, unlike institution building, entails concen- staff: “they are taking risks and making enemies, but to no trating on how power and authority are distributed and exer- effect.” In justice to donors, efforts are being made to resus- cised, and this in turn necessitates thinking and working citate the Public Accounts Committees in both houses of the across all elements of the state and at the intersection Legislature, but there is a prevailing pessimism about the between state and society. It means being alert to the ripple capacity of the current Legislature to pursue audit reports with effects of actions in one area on other areas, and following any vigour and even a revitalised Legislature is not sufficient processes transactionally through the system rather than try- in itself to ensure the implementation of audit findings. ing for a technical fix through one organisation, no matter In Sierra Leone the links in the audit chain are rather how significant that organisation may be. It also entails work- stronger although the audit process seems rather tamer over- ing with a wider spectrum of organisations and interests— all. The Office of the Auditor-General has been strengthened, including societal groupings—to achieve improvements in and the Public Accounts Committee is active and engaged: it the way the state functions. is holding hearings and taking evidence from officials, press- ing for a change in the standing orders to hold public hear- FOCUS ON THE END GAME, AND WHAT’S ings, and has recently held public meetings in the districts in NEEDED TO GET THERE concert with the Anti-Corruption Commission and the Office of the Auditor-General. While these moves are in the right STATE-BUILDING DEMANDS FLEXIBILITY, and some nifty direction, the audit circle still needs to be closed with action footwork at times. Importantly, it is necessary to keep your on audit findings by government departments and agencies, objectives to the forefront, and find creative ways to get there. and action as necessary in the courts—something which There is no fixed path: it is not a case of fixing this organisation apparently has not yet arisen. The audit program in Sierra or that, but of working with all the bits in the system that can Leone is also rather less activist than in Liberia, at this stage contribute—and are willing to contribute—to the end game. concentrating on the backlog of audited financial accounts. Accountability, for example, is a defining quality in the The acid test will come if it moves further into performance character of the relationship between state and society, and audits that put the spotlight on the current administration. strengthening accountability is a high priority for donors. However concentrating on one organisation within the web of STRENGTHEN THE HAND OF SOCIETY IN THE accountability, such as supreme audit, is unlikely to effect STATE-SOCIETY EQUATION significant change in isolation. To impact on accountability, action is needed at other points along the chain such the PARLIAMENTS SPAN STATE AND SOCIETY in the exercise of Public Accounts Committee of the Parliament, the Public their preeminent functions of popular representation, law- Service Commission, the Ministry of Finance, the finance and making and executive oversight. Typically in fragile states the programme management areas of line agencies, the police, preeminent authority of the Parliament has been eclipsed by the courts, the media and civil society groups. the executive, and part of the process of recalibrating the In Liberia, donors have invested very substantially in relationship between state and society is the reinstatement of strengthening the General Auditing Commission. Its staff Parliament’s authority, capacity and legitimacy. Yet in the numbers have been increased to around 360, including an donor world, Parliaments usually receive patchy, highly pro- activist Auditor-General from the diaspora and a large contin- jectised assistance. 24 S TAT E - B U I L D I N G — K E Y C O N C E P T S A N D O P E R AT I O N A L I M P L I C AT I O N S I N T W O F R A G I L E S TAT E S This is the case in both Sierra Leone and Liberia, where ical institutions acquired at independence. Liberia does, the Parliaments are receiving minor, piecemeal assistance however, offer some hopeful examples of how the perform- from various donors, but there is little systematic effort to ance of the representational function can be improved. An ad invest in their revitalisation. In the case of Liberia, there is hoc committee of the Legislature has prepared a comprehen- also some suspicion of the Legislature (as it is termed) itself, sive Modernisation Plan which includes a component which has a chequered membership and a recent track record focussed on representation, some legislators have already of stalling the legislative program of a reforming executive. opened constituency offices, and the National Democracy The understandable concerns over the motives of the legis- Institute has supported legislators to undertake constituency lators obscure an important fact—that it is the proper function consultations. of a legislature to act as an independent counterweight to exec- A vibrant civil society is an essential ingredient in the utive power under a Constitution that calls for equality and state-building recipe: to articulate the interests of citizens, to coordination among the branches of government. Although the monitor the activity of the state and to draw attention to gov- relations between Legislature and Ministers have been charac- ernment performance. In both Sierra Leone and Liberia, the terised by mutual suspicion, this now seems to be changing. team met with civil society organisations that were playing a Discussion inside the Executive has shifted to improving coor- strong role in taking the arguments up to government and out dination with the Legislature and taking the lead through early to the people. Very often, however, donor engagement with and constructive engagement with relevant legislators and civil society is quite narrowly focused and the organisations committees on emerging policy which will flow through into assisted tend to be those led by educated elites and far legislation. This is the machinery beginning to work. removed from the citizens whose interests they represent, The Legislature’s activism has produced some positive however ably. There is also a danger of asymmetric empower- outcomes, bringing a new transparency and public attention ment: where citizens are empowered to demand services to the budget process. For the first time, the draft Budget is without a parallel empowerment of the government to provide receiving broad exposure through public committee hearings these services (OECD 2010a, p18). and their live broadcast on radio, and Ministers are reported- Too often we fail to engage, or to engage adequately, with ly responding by appearing before the committee much bet- traditional systems and structures although these constitute ter prepared. This new activism, however, will only be as con- the fundamental basis of local social and political organisa- structive as the capacity of the Legislature itself, and here tion in states where the majority of the population is rural and the concerns are legion. lives by subsistence agriculture, and where the footprint of Where donors do engage with Parliaments, they generally the state is shallow at best. Traditional systems pre-date the focus on the oversight function and, to a lesser extent, on state and continue to operate alongside it. Importantly, they lawmaking. The strengthening of Parliamentarians’ represen- often have a high level of legitimacy with their members and tative role receives scant attention, although it is—in theory generally meet their members’ needs to a far greater degree at least—the means by which citizens shape the direction of than does the state itself. public policy and expenditure priorities and the day to day In Liberia, the team was struck by the disinterest—and decisions of government. In both Sierra Leone and Liberia, as underlying embarrassment and hostility—shown for the tradi- in many other countries, constituents complain that the elec- tional system by various government officials with whom we tion is the last they see of their representatives who prompt- met. The Traditional Council of Liberia spoke forcefully to us ly retreat to the capital to re-emerge only when the next elec- about their exclusion from the formal system of government tion is called. For their part, Members of Parliament com- and their very limited contact with the international commu- plain about the unrealistic demands of their constituents for nity. We sense that there is considerable room to engage and direct and personal material assistance, and their frustration work more closely with the traditional system at both nation- at their inability to deliver. al and local level.30 In part, this reflects the mismatch of political paradigms There are some positive examples of this happening. The between customary and modern political systems and poor work of the Carter Center in the area of local justice is popular understanding of the attributes of the national polit- impressive in its bridging of traditional and formal systems. T H E W O R L D BA N K — U N D P 25 3.2.5 We also commend the inclusion of a representative of the Match the Paramount Chiefs on the governing body for the Extractive Industries Development Initiative in Liberia. This offers a development approach model for other consultative and advisory processes. More to the context broadly, there are many potential mechanisms for supporting dialogue between formal and traditional systems. IF THE FIRST PRINCIPLE for good international engagement in PLAN FOR LONG TIME HORIZONS, fragile states and situations is to take the context as the start- AND BE VERY REALISTIC ABOUT WHAT ing point, the corollary is that the insights this analysis pro- IS ACHIEVABLE vides must then inform the specifics of the development approach. There is no blueprint for state-building that can be NONE OF THE PROGRAMMING OBJECTIVES for state-build- applied from country to country: each situation is unique, ing are going to happen fast: Rome wasn’t built in a day, and and the context will determine where to start, how to neither is the average fragile state. In an environment where sequence and what to emphasise. institutions are weak and dysfunctional and there is a need to One of the Ministers with whom we spoke in Liberia deplored work on many fronts simultaneously, overloading the avail- the contingents of international consultants who step off the able capacity with comprehensive support in all sectors plane with their prefabricated solutions from Guatemala, remains a challenge. It is important to design interventions Cambodia or wherever. Another interlocutor, in similar terms, with this in mind, to be very realistic about what is achiev- observed that donors and international development practition- able and the timeframe needed, and to plan to work incre- ers are often too preoccupied with best practices. He empha- mentally and flexibly over the long haul. sised that, while it is important to learn from experience else- where, international partners can be more helpful when they serve as resource people informing local people about what is done elsewhere while working with them to shape their own institutions and approaches based in the local reality. A key theme in Liberia was that the conflict was, at its core, about exclusion (see Box 7) and this translated into a strong call for improved communication, voice and participa- THINKING AND WORKING SYSTEM-WIDE, tion in decision-making. A further issue that stood out in Liberia was the pervasive legacy of endemic and savage vio- AND FOR THE LONG TERM lence over more than a generation, which has touched every concentrate on the end game, and what’s village and every person’s life in some way. A recent study put KEY MESSAGES needed to get there: be prepared to work the levels of major depressive disorders and post-traumatic flexibly across a range of organisations in stress disorders at 40 percent and 44 percent respectively of the process the adult household-based population.31 Interlocutors in Liberia referred to the debilitating effects of such psycholog- consider how to strengthen engagement at ical burdens on a range of behaviours ranging from dealing the state-society interface e.g. through pro- grammatic support to Parliaments and with risks, short versus long term thinking and the willingness through working with a wider spectrum of and ability to collaborate, especially outside the boundaries civil society, including the traditional system of the family. In this environment, capacity development ini- tiatives that have the effect of making people feel less plan for long time horizons, and be very secure—personally, financially, socially and physically—will realistic about what is achievable have little chance of succeeding. Lofty capacity development interventions are likely to bump up against people’s own sur- vival strategies at all levels. 26 S TAT E - B U I L D I N G — K E Y C O N C E P T S A N D O P E R AT I O N A L I M P L I C AT I O N S I N T W O F R A G I L E S TAT E S AVOID IMPORTED “OFF-THE-SHELF” BOX 7 SOLUTIONS THERE IS A STRONG TEMPTATION, when confronted with the challenges of state-building, to use the familiar and apparently effective template of the Western state: a tempta- tion acted out many times over by individual advisers brought in from positions in donor governments who tend to default to the tried and true systems and processes from back home. This is not a sound move: “…the formal institutions of the Western state derive their capacity and legitimacy from a long history of interaction between state and society, and cannot just be reproduced by transferring those same institutional models into different THE POLITICS OF EXCLUSION social, cultural, historical and political contexts.” (OECD 2010b, p8) Systemic capacity is a further contextual consideration. Liberia’s modern history has been marked by exclu- Organisations, institutions and processes do not operate in a sion, and that history is still playing out. For much of the life of the state, most of the indigenous population vacuum: they are sustained or subverted by their environ- was excluded from participation in the political ment, from the basic corporate infrastructure they need for process and access to public goods. As the formal their day to day operations to the wider institutional arrange- barriers were gradually dismantled from the middle of ments required to complement and give effect to their activ- the 20th century, systemic discrimination continued to ities. Development experience is littered with failed initia- exclude large segments of the population. Liberia is tives, be it the sophisticated business processes of new pub- unique in that, unlike a decolonised state where the exclusionary colonial power has departed, the lic management that are too complex to graft onto simple descendents of the original oligarchy are still present, budgetary and human resource systems, or new organisations still powerful and still deeply resented. such as Anti-Corruption Commissions which depend on strong political support and an effective justice system to From the outset of the modern state, land was focus function effectively (see Box 8, next page). for injustice and exclusion, from the dubious acquisi- Both Liberia and Sierra Leone have established Anti- tion of land by the American Colonisation Society to the tying of franchise to freehold land ownership in the Corruption Commissions as part of the donor-led strategy for original Constitution, to the Government’s assertion of recovery, in environments where poor governance lay at the public ownership of traditional lands. The Constitution heart of the conflict and ongoing corruption was threatening also maintained the interior “provinces” in a legally the political settlement. In both countries, the Commissions subordinate relationship to the coastal “counties” have been slow to get off the ground and get runs on the until the mid 1960s. Communications between the board. In the case of Liberia, the establishment of the centre and the periphery were weak throughout the twentieth century, with an inadequate road system Commission was specified as part of the package of measures that became impassable in the wet season and mini- required under the Governance and Economic Management mal broadcast radio and telecommunications cover- Assistance Program, although it was not set up until 2008, age. The UN Peacekeeping Mission (UNMIL) has estab- with Liberia’s controversial legislature delaying the passage lished the first radio station broadcasting across the of enabling legislation.32 A year on, the UN Secretary-General country, and the mission also largely maintains the was reporting to the Security Council that the operational country’s roads. capacity of the Commission was improving and it had begun investigations into two major cases. However the absence of supporting legislation, including a public service code of con- T H E W O R L D BA N K — U N D P 27 duct, continued to hamper its effectiveness.33 Sierra Leone’s Commission has been in place for rather longer—since 2000—but its results are lacklustre. It started out badly, being described as too hamstrung by politics to be either independent or effective.34 With the change of government in 2007, the then Commissioner—a relative of the outgoing President—resigned. Although a new Anti-Corruption Act was passed in 2008, the Commission continues to attract criti- cism for being heavy on public pronouncements and light on prosecutions. WORK WITH WHAT’S ALREADY THERE WHERE POSSIBLE WHEN DONORS SURVEY THE WRECKAGE of the post-con- flict state, there is a strong urge to start from scratch and re- engineer systems and processes. But current thinking coun- sels strongly against this. However fragile the institutions, it is not an institutional tabula rasa and institutional legacies constitute important reference points (Fritz and Rocha Menocal 2007, p20). Donors are now enjoined to build on what is already there, accompanying and facilitating domes- tic processes, leveraging local capacities and complement- ing, rather than crowding out, local initiatives (Rocha Menocal 2009 p18, citing Cliffe and Manning). As part of MATCHING THE DEVELOPMENT this approach, it is important to look for what capacity APPROACH TO THE CONTEXT exists—human and systemic—and build on what is there as this will help to restore confidence and competence. There is no blueprint for state-building: The work of the Carter Center with the justice sector in KEY MESSAGES programming priorities, sequencing and Liberia is a powerful example of this process at work. The programming elements must be informed Carter Center was invited by the President to put a transition- by the local context al justice program in place until the formal justice sector could be revitalised. It quickly became apparent that it was Avoid off-the-shelf approaches based on western models—tailor the approach to not a case of rebuilding the justice sector in rural areas—it complement local institutions and local had never really functioned there. Across the country, there is capacity also a massive level of distrust in the formal system.35 The approach of the Carter Center is to strengthen dispute reso- Wherever feasible, build on existing institu- lution at the local level by bridging the formal and tradition- tions and capacity, and avoid crowding out al systems, working to the principle of accessible communi- local initiative and local people ty-based justice for all.36 28 S TAT E - B U I L D I N G — K E Y C O N C E P T S A N D O P E R AT I O N A L I M P L I C AT I O N S I N T W O F R A G I L E S TAT E S 3.2.6 Take ethical responsibility BOX 8 for champions in contested environments A CENTRAL PLANK of post-conflict state-building is arresting the vicious cycle of poor governance, corruption and violence, and early investment is made in the revitalisation of the accountability machinery—supreme audit, anti-corruption commissions, ombudsmen and the like—and the formation of transitional justice bodies to investigate past misconduct and deal with the perpetrators. When, as donors, we promote such organisations, we can expose their members to serious risk – both now and into the future. And these are countries where life is cheap and retribution can be bought with a very modest payment. In our meeting with the Liberian Auditor-General, he spoke passionately about the risks that his own staff were taking in their jobs, and his concerns are well founded as elsewhere officials who have upheld public integrity and the empowerment of citizens have been ruthlessly intimidated and on occasions killed. In Solomon Islands, for example, the ANTI-CORRUPTION COMMISSIONS families of senior government officials were menaced and a The evidence is now in on Anti-Corruption Commissions in fragile settings, and it is not encouraging. In 2008, the World Bank’s Independent Evaluation Group released its report Public Sector Reform: What Works and Why? Its sobering assessment is that direct meas- ures to reduce corruption, such as anti-corruption laws and commissions, rarely succeed. The report recom- mends that, where strong political will and an effective judicial system are absent, anti-corruption efforts should focus on indirect measures such as strengthen- ing public financial management and personnel man- agement and information systems, making better infor- mation available to the public in ways that stimulate public demand for better processes, and building the capacity of demand-side institutions like the legisla- ture, the audit office and the media. This comes down to repairing what is there: more laborious perhaps, but ultimately more effective. T H E W O R L D BA N K — U N D P 29 highly respected member of the National Peace Council was murdered presumably in retribution for his role in purging the special constables from the police force. And in Timor-Leste, UN locally engaged electoral staff were targeted and killed by the militias in the explosion of violence following the 1999 popular consultation. As international actors, what are our obligations towards those we put on the firing line? We are very keen to identify and promote champions, but are we still there for them if things go sour? And are there situations where we should step KEY MESSAGES back from the preferred position of working with and through TAKING ETHICAL RESPONSIBILITY national personnel? For example, when aggressive audit is FOR CHAMPIONS required, is it preferable to use international personnel who can move in and out quickly, before the audit results are Recognise the duty of care towards nation- al personnel whose functions and authority released? In Liberia, for example, the massive level of corrup- threaten powerful elites tion under the Transitional Government was exposed through a series of audits undertaken by a team of EC auditors. Assess and act in a way that minimises risk There is a real duty of care that donors must reflect on to local actors when engaging in state-building, to protect those brave indi- viduals who are prepared to challenge the power and the plunder of long-standing elites, and act in a way that min- imises risk to local actors. 4. WHAT CAN AN OVERWORKED COUNTRY OFFICE DO? T o work effectively in the business of state-building, we need to recognise it for the political process which it is, to work to understand as best as possi- ble the context in which we are operating, and to always analyse the potential consequences of our actions. This is an area that demands a high degree of honesty and integrity by international actors working with fragile and conflict-affected states if we are to avoid doing harm. This paper has attempted to show some of the ways in which we can focus and calibrate our approach as develop- ment practitioners to positively support state-building processes, and minimise the inadvertent harm we might do along the way. In considering what country offices can do in this area, we are very aware of the many competing demands on them, and of the day to day realities of development pro- gramming and implementation in fragile states. For these use a state-building lens in program design reasons, we have sought to keep our recommendations sim- and evaluation ple and achievable. Our four basic recommendations are out- In fragile and conflict-affected states, it is high risk to lined below. approach programming as a purely technical exercise. In all areas, programming needs to be held up to a state-building develop and apply political analysis lens in order to understand how the program objectives and action areas and the program modalities will shape and be The case for obtaining and using political and political shaped by the institutions of the state and the political economy analysis is set out in section 3.2.1. The message is process. Section 3.1 describes some of the criteria that we simple: first, find out what analysis already exists and how used when holding programming up to a state-building lens, current it is, and where it is not available or not well-devel- and Section 3.2 outlines some of the programming implica- oped, commission it. Second, keep the analysis current and tions. In many development organisations, a gender assess- use it to “ground-truth” developments on the ground and ment is built into both program design and program evalua- assess their likely impact on current and planned program- tion. In fragile and conflict-affected states, where state- ming decisions. Third, promote a culture (and incentive building is an overarching development goal, we would structure) within the Country Office which encourages staff to encourage program managers to include a state-building build a strong understanding of the country context and to impact assessment as a standard requirement in the design anchor their work in that understanding. and evaluation of programs.37 T H E W O R L D BA N K — U N D P 31 consider engaging state-building expertise ENDNOTES 1 The author wishes to acknowledge the valuable suggestions and comments To strengthen the focus on state-building, country offices on this paper received from reviewers—Javier Fabra, Verena Fritz, Patrick could consider engaging a small number of state-building Keuleers, Claudia Melim-McLeod, Nadia Piffaretti, Amos Sawyer and Timothy Sisk – and members of the UNDP BCPR and World Bank FCAS Group—Greg and country experts on a standing basis to update political Ellis, Ana Paula Fialho Lopes, Eugenia Piza-Lopez and Lucy Turner. 2 The International Network on Conflict and Fragility, hosted by OECD, of analysis, test program plans and undertake state-building which both UNDP and the World Bank are members, is currently developing impact assessments. a detailed policy guidance note on state-building in fragile situations, which should be released in 2010. The UK Department for International Development released a practice paper Building Peaceful States and Societies in March 2010, and the World Bank is currently developing an work beyond executive government, analytical framework and guidance note on governance and public sector and protect change agents reform in fragile and conflict-affected states. The first European Report on Development, which was launched in October 2009, also took as its theme Overcoming Fragility in Africa. 3 In 1945, 51 states became founding members of the UN, although not all Section 3.2.4 discusses broadening engagement to extant states joined at the time. There are now 192 member states of the include a wider spectrum of state institutions and non-state United Nations, the most recent addition being Timor-Leste which became an independent state in 2002. actors including customary authorities in programming, and 4 There is no universal list of fragile states, and the basis for their identifica- section 3.2.6 highlights our duty of care towards the cham- tion varies somewhat across organisations of the international community. An OECD Factsheet (Ensuring Fragile States Are Not Left Behind, December pions we embrace to challenge the abuses of elites. State- 2007) identified 38 fragile states. Of these 38, only four have been inde- pendent states for over sixty years: Afghanistan since 1919; Haiti since building is about society and about people, and it is impor- 1804; Liberia since 1847; and Tonga arguably since 1875 when it adopted tant that we work not just with the state but at the interface its Constitution, although it accepted British Protectorate status from 1900 - 1970. Liberia was never colonised, having been settled by former slaves on between state and society, and respect and protect the inter- land acquired by the American Colonisation Society in 1822. 5 An economic survey of Liberia published in 1966 which became a classic ests of those who are leading positive change. characterised the economy as one of growth without development. Although wealth from primary industries such as rubber and mining was considerable, it was not invested in the physical and social infrastructure of roads, agricul- ture, health and education necessary for development. (Robert W. Clower et. al. Growth Without Development: an economic survey of Liberia, North- Western University Press, 1966). 6 The rollcall of fragile states comprehends a spectrum of political and eco- nomic circumstances and they are far from homogenous. These differences were recognised in specific business models which the World Bank devel- oped to work with countries in crisis: deterioration; prolonged crisis or impasse; post-conflict or political transition, and gradual improvement (World Bank 2006, pix) 7 The World Bank defines a country as a fragile state if it is a low- income country or territory, IDA-eligible, with a CPIA score of 3.2 or below. Countries are considered “core” fragile states if their CPIA is below 3.0 or “marginal” fragile states if their CPIA score is between 3.0 and 3.2. The CPIA scores are described as providing guidance on the “spectrum” of fragility and not as constituting hard and fast rules. 8 The team visiting Sierra Leone comprised: Sue Ingram (consultant), Marcus Lenzen (UNDP), Ana Paula Fialho Lopes (World Bank) and Peter RECOMMENDATIONS Morgan (consultant). The team visiting Liberia comprised Greg Ellis (World Bank), Sue Ingram, Peter Morgan, Eugenia Piza-Lopez (UNDP) and Lucy Turner (UNDP). 9 In 2008 the OECD provided members of the DAC Network on Governance Develop and apply political analysis with the final results of a survey of Governance Assessment tools which iden- tified 46 that were in use or under development in member states (OECD use a state-building lens in program 2008c). 10 For example Amos Sawyer, a former Liberian President, current chair of design and evaluation the Governance Commission and academic political scientist, has published extensively on the politics of Liberia, e.g. Beyond Plunder: Towards Consider engaging state-building expertise Democratic Governance in Liberia and The Emergence of Autocracy in Liberia: Tragedy and Challenge. 11 For a breakdown of the constituent elements of 27 peace agreements that Work beyond executive government, ended civil wars made in the period 1990-2006, see Suhrke et. al. (2007). and protect change agents 12 Agence de Presse Africaine, 14 January 2008. 13 For example, the 14 May 2007 Report of the Peacebuilding Commission to the Security Council refers to the prospect of a referendum on constitu- tional amendments in tandem with the 2007 elections (A/61/901- S/2007/269 at para 13). 32 S TAT E - B U I L D I N G — K E Y C O N C E P T S A N D O P E R AT I O N A L I M P L I C AT I O N S I N T W O F R A G I L E S TAT E S 14 For example, a mapping of 27 peace accords signed after 1989 identified building, for example, includes a state-building impact assessment tool provisions for elections and political parties in 85 percent of cases, with an (OECD 2010a, p141). average high level of specificity: see Suhrke, Astri, Torunn Wimpelmann and Marcia Dawes (2007) Peace Processes and State-building: Economic and Institutional Provisions of Peace Agreements, Chr Michelsen Institute, p25. 15 The Carter Centre, Observing the 1997 Special Elections Process in BIBLIOGRAPHY Liberia, p43. Brown, Stephen and Jörn Grävingholt (2009) “Framing Paper on Political 16 Liberian Truth and Reconciliation Commission Report, pp130-131. Settlements in Peacebuilding and State-building,” Note for the OECD-DAC International Network on Conflict and Fragility (draft of 23 August 2009). 17 UNDP (2009), Elections and Conflict Prevention: A Guide to Analysis, Planning and Programming, New York and Oslo. DFID (2009a) “Building the State and Securing the Peace,” Emerging Policy Paper, DFID, London. 18 UNDP, Governance for Sustainable Human Development, 1997, pxiii. 19 World Bank, Decentralization in Client Countries: An Evaluation of World DFID (2009b) “Political Economy Analysis—How To Note,” DFID practice Bank Support 1990-2007, Washington, 2008, pxix. paper, DFID, London. 20 Arild Schou and Marit Haug (2005) Decentralisation in Conflict and Post- DFID (2010) “Building Peaceful States and Societies,” DFID Practice Paper, Conflict Situations Norwegian Institute for Urban and Regional Research, DFID, London. Oslo. 21 Oxford Analytica, “The outlook for decentralisation in Africa,” Daily Brief, Fritz, Verena and Alina Rocha Menocal (2007) “Understanding State- Building from a Political Economy Perspective”, Report for DFID’s Effective 2009. and Fragile States Teams, Overseas Development Institute, London. 22 Government of Liberia, Poverty Reduction Strategy 2008-11, pp85-86. 23 That this was the case in Sierra Leone and Liberia is captured in the fol- OECD (2007) Principles for Good International Engagement in Fragile States, OECD, Paris. lowing observation: “The wars in Liberia and Sierra Leone confronted policy makers and mediators with a fundamental challenge: how to deal with armed OECD (2008a) State Building in Situations of Fragility: Initial Findings— non-state actors who appeared less interested in politics than plunder, and August 2008, OECD, Paris. who, therefore, were extremely difficult to accommodate in any rational polit- ical settlement.” (Lansana Gberie “Bringing Peace to West Africa: Liberia OECD (2008b) Concepts and Dilemmas of State Building in Fragile and Sierra Leone”, paper for the Africa Mediators’ Retreat, Centre for Situations: From Fragility to Resilience, OECD, Paris. Humanitarian Dialogue p65). 24 World Bank (2008) Second Governance Reform and Growth Credit, OECD (2008c) Survey of Donor Approaches to Governance Assessment, Program Document. February 2008, http://www.oecd.org/dataoecd/58/32/42258487.pdf 25 OECD 2010a, p65. OECD (2010a) Do No Harm: International Support for State-building, OECD, 26 IDA, IFC and African Development Fund, Joint Country Assistance Paris. Strategy for the Republic of Liberia for the period FY09-FY11, p3. 27 supra p1. OECD (2010b) The State’s Legitimacy in Fragile Situations—Unpacking Complexity, (draft) OECD, Paris. 28 Renata Dwan and Laura Bailey, Liberia’s Governance and Economic Management Assistance Programme: A Joint Review by the Department of Overseas Development Institute (2009) “‘State-building for Peace’: navigat- Peacekeeping Operations’ Best Practices Section and the World Bank’s ing an arena of contradictions,” Briefing Paper, http://www.odi.org.uk Fragile States Group 2006, p6. Rocha Menocal, Alina (2009) “State-building for Peace: A new paradigm for 29 supra p22. international engagement in post-conflict fragile states?” Workshop paper for 30 Sensitivity to context is—as ever—critical in determining how and where the European Report on Development to engage. We were advised, for example, that in Liberia the traditional chiefs http://erd.eui.eu/BackgroundPapers/ERD-Background_ have been losing standing as “lords of the soil” through the manipulation of Paper-Rocha-Menocal.pdf (accessed 25 January 2010) their selection under President Tubman, the subsequent introduction of chiefly elections in the 1970s and their association with armed factions dur- Suhrke, Astri, Torunn Wimpelmann and Marcia Dawes (2007), Peace ing the conflict. It was suggested that they cannot be compared in stature or Processes and State-building: Economic and Institutional Provisions of Peace authority to chiefs in some other parts of West Africa such as Ghana, Nigeria Agreements, Chr. Michelsen Institute. or Côte D’Ivoire; within Liberia, their standing remains relatively higher among http://www.cmi.no/publications/publication/?2689=peace-processes-and- communities of northern and northwestern Liberia than in the southeast. state-building 31 Kirsten Johnson et al “Association of Combatant Status and Sexual Unsworth, Sue (2008), “Is Political Analysis Changing Donor Behaviour?” Violence with Health and Mental Health Outcomes in Postconflict Liberia” Paper presented at the Development Studies Association annual conference, Journal of the American Medical Association, vol 300, no. 6, 13 August 2008, http://www.devents.org.uk/Change-unsworth.doc 2008. 32 This was just one of several important pieces of legislation reported as Whaites, Alan (2008) “States in Development: Understanding State-build- delayed in the UN Secretary-General’s 15 August 2008 report to the Security ing,” DFID Working Paper, DFID, London. Council on Liberia. Other major bills included the Code of Conduct Act and the Defence Act. World Bank/UNDP (2005), Rebuilding Post-Conflict Societies: Lesson from a Decade of Global Experience, Workshop Report, New York, 19-21st 33 United Nations Security Council, Nineteenth Progress Report of the September 2005. Secretary-General on the United Nations Mission in Liberia, S/2009/411, 10 August 2009. World Bank Independent Evaluation Group (2006), Engaging with Fragile 34 International Crisis Group, Sierra Leone: the State of Security and States: An IEG Review of World Bank Support to Low-Income Countries Under Stress, Washington. Governance, ICG Africa Report No. 67, Sept 2003. 35 Deborah Isser, S. Lubkemann and S. N’Tow, Looking for Justice: Liberian Wyeth, Vanessa and Timothy Sisk (2009), “Rethinking Peacebuilding and Experiences and Perceptions of Local Justice Options (draft), United States State-building in Fragile and Conflict-Affected Countries: Conceptual Clarity, Institute of Peace, 2009. Policy Guidance and Practical Implications,” Discussion Note for the OECD- 36 Jimmy Carter, “Reconstructing the Rule of Law: Post Conflict Liberia,” DAC International Network on Conflict and Fragility (draft of 15 May 2009). Harvard International Review, Fall 2008. 37 The recent OECD report, Do No Harm: International Engagement in State- T H E W O R L D BA N K — U N D P 33 The World Bank 1818 H Street NW Washington, DC 20433 USA www.worldbank.org/fragilityandconflict UNDP New York Office Bureau for Crisis Prevention & Recovery, UNDP One United Nations Plaza, DC1 20th floor New York, NY 10017 www.undp.org/cpr UNDP Geneva Office Bureau for Crisis Prevention & Recovery, UNDP 11-13 Chemin des Anemones Chatelaine, CH-1219 Geneva Switzerland www.undp.org/cpr Photo Credits © 2010, The World Bank and UNDP Arne Hoel: cover (both inside and out), and pp. 9-10, 12-15, 18-19, 21, 23, 27-32. p. 2: Lianqin Wang/The World Bank p. 3-4, 7: Curt Carnemark/The World Bank Radio Nederland Wereldomroep (Flickr Creative Commons): p. 8