Social Contracts and World Bank Country Engagements Lessons from Emerging Practices IEG MESO EVALUATION © 2019 International Bank for Reconstruction and This work is a product of the staff of The World RIGHTS AND PERMISSIONS Development / Bank with external contributions. The findings, The material in this work is subject to copyright. The World Bank interpretations, and conclusions expressed in this Because The World Bank encourages dissemination 1818 H Street NW work do not necessarily reflect the views of The of its knowledge, this work may be reproduced, in Washington, DC 20433 World Bank, its Board of Executive Directors, or the whole or in part, for noncommercial purposes as Telephone: 202-473-1000 governments they represent. long as full attribution to this work is given. Internet: www.worldbank.org The World Bank does not guarantee the accuracy of Any queries on rights and licenses, including the data included in this work. subsidiary rights, should be addressed to World The boundaries, colors, denominations, and other Bank Publications, The World Bank Group, 1818 H Attribution—Please cite the report as: World Bank. information shown on any map in this work do not Street NW, Washington, DC 20433, USA; fax: 202- 2019. Social Contracts and World Bank Country imply any judgment on the part of The World Bank 522-2625; e-mail: Engagements: Lessons from Emerging Practices. concerning the legal status of any territory or the pubrights@worldbank.org. IEG Meso Evaluation. Independent Evaluation endorsement or acceptance of such boundaries. Group. Washington, DC: World Bank. Cover Photo: Victor Ekpuk, I Be You You Be Me, 2007, acrylic on wood. Image courtesy of the World Bank Group Art Program. Social Contracts and World Bank Country Engagements Lessons from Emerging Practices IEG Meso Evaluation October 2, 2019 Contents Abbreviations ..................................................................................................................................................... v Acknowledgments ........................................................................................................................................... vi Overview ............................................................................................................................................................vii 1. Introduction ...................................................................................................................................................1 Objectives and Scope .................................................................................................................................2 Methodology .................................................................................................................................................5 2. Social Contract Diagnostics .....................................................................................................................7 The Added Value of Social Contracts Diagnostics ............................................................................8 Limitations .................................................................................................................................................... 12 3. Operational Implications ........................................................................................................................ 16 Social Contracts in Transition................................................................................................................. 16 Status Quo Social Contracts .................................................................................................................. 22 4. Discordances with the World Bank’s Business Model................................................................. 27 5. Conclusions and Lessons ....................................................................................................................... 32 Social Contract Diagnostics ................................................................................................................... 32 Operational Implications and Challenges ......................................................................................... 33 Overcoming Operational Challenges ................................................................................................. 34 Bibliography..................................................................................................................................................... 37 Boxes Box 1.1. Social Contract Definition ................................................................................................................3 Box 2.1. A Promising Example of Social Contract Research in the Africa Region ....................... 14 Box 3.1. Helping Tunisians Renew Their Social Contract ................................................................... 20 Box 4.1. Lessons from the World Bank’s Use of Governance and Social Frameworks ............. 30 Figures Figure 1.1. Social Contracts Conceptual Framework ...............................................................................4 iii Contents Figure 2.1. Development Challenges Explained by Social Contracts in Systematic Country Diagnostics ................................................................................................................................... 10 Figure 3.1. Coherence Analysis in the Middle East and North Africa Region............................... 18 Figure 3.2. Coherence Analysis .................................................................................................................. 24 Tables Table 1.1. Case Study Selection......................................................................................................................6 Table 2.1. Systematic Country Diagnostics Using Social Contracts Framing ..................................9 Table 3.1. Middle East and North Africa Strategy Pillar 1: Renewing the Social Contract ........ 17 Appendixes Appendix A. Methodology ......................................................................................................................... 45 Appendix B. Conceptual Framework ...................................................................................................... 56 Appendix C. Case Study Summaries....................................................................................................... 62 Appendix D. Coherence Analysis ............................................................................................................. 71 Appendix E. Stocktaking of Governance and Social Development Analytical Frameworks ........................................................................................................................................................................ 84 iv Abbreviations ASA Advisory Services and Analytics CMU Country Management Unit CPF Country Partnership Framework CSO civil society organizations DPF development policy financing FY fiscal year GP Global Practice IEG Independent Evaluation Group MENA Middle East and North Africa Region OECD Organisation for Economic Co-operation and Development PEA political economy analysis SCDs Systematic Country Diagnostics UN United Nations WDR World Development Report v Acknowledgments This evaluation was prepared by an Independent Evaluation Group (IEG) team led by Estelle Raimondo and comprising Anis Dani (senior consultant, social development specialist), Ann Flanagan (senior economist), Gaby Loibl (program assistant), Joy Butscher (junior professional officer), Kenza Haouche (consultant, data scientist), Maurizia Tovo (senior consultant, social protection specialist), Peter Knaack (consultant, political economist), Qihui Chen (consultant, data scientist), Rasmus Heltberg (lead evaluation officer), Rebecca Riso (consultant, analyst), and Vanessa Corlazzoli (consultant, evaluation specialist). Maximillian Ashwill was the report’s primary editor. The evaluation was conducted under the overall direction of Alison Evans, Director- General, Evaluation, and the guidance and supervision of Emanuela Di Gropello, manager, Corporate and Human Development, Auguste Tano Kouame, director, Human Development and Economic Management, and Sophie Sirtaine, director, Strategy and Operations. Peer reviewers for this evaluation were Dr. Barbara Nunberg (visiting scholar at the Robert F. Wagner Graduate School of Public Service at New York University and former World Bank Governance Expert); Prof. Indermit Singh Gill (professor of practice of public policy in the Sanford School of Public Policy and formerly director for development policy in the Office of the Chief Economist at the World Bank); and Professor Mouhoub El Mouhoud (professor of economics, University of Paris Dauphine). The team is grateful for the very useful feedback provided by the Council of Chief Economists of the World Bank, the technical counterpart for this evaluation. The team also thanks the World Bank team working on the flagship report on social contracts in Africa, notably, Bernard Harborne (lead social development specialist), Debbie Isser (lead governance specialist), Indhira Santos (senior economist), and Mathieu Cloutier (young professional) for the fruitful knowledge sharing throughout the evaluation cycle. Finally, the team is thankful to the many World Bank managers and staff who engaged with us and gave generously of their time for interviews. Thanks are also due to the country offices of Guatemala, Niger, Tunisia, and Uzbekistan for their support during our country visits. vi Overview Highlights Social contract diagnostics are useful analytical innovations with relevant operational implications, especially in countries where social contracts are in transition. However, the World Bank’s involvement in social contract renewal carries efficacy and reputational risks and poses some operational challenges because it is at odds with some aspects of the World Bank’s business model. To derive more value from social contracts diagnostics within the constraints of the World Bank’s operating environment, the World Bank must think critically about the conditions under which a social contract diagnostic can effectively guide its country engagements. This also requires more focus on citizen engagement, building partnerships, and building staff’s social contract–related capabilities. The World Bank emphasizes the the risks, challenges, value added, and importance of social contracts to eliminate opportunities of using social contracts to poverty and boost shared prosperity. In the shape the institution’s country 2014 World Bank Group Goals, the World engagements. This learning-oriented Bank calls for social contracts that prioritize evaluation, requested by members of World the poor while creating the conditions for Bank senior management, generates lessons equitable growth (World Bank Group from the World Bank’s experience using 2014c, 20). Social contracts can be social contract diagnostics to help countries understood as the implicit, mutual reshape their social contracts. It does this by bargaining over what citizens expect from (i) evaluating the quality and value added the state, and what the state can legitimately of social contract diagnostics; (ii) assessing demand of citizens in return. The World how social contract diagnostics are Bank has identified broken social contracts translated into operations; (iii) identifying as causing development challenges in client the risks and challenges of integrating social countries and proposed solutions to help contract diagnostics into operations; and societies reshape their social contracts. Yet (iv) drawing lessons on how to overcome while some countries with dysfunctional these challenges. Several issues are outside social contracts are undergoing social the evaluation’s scope. The evaluation does contract transitions, other countries’ social not assess social contracts or their contracts remain unmovable. transitions, the relevance or appropriateness of the social contract concept in general, or The evaluation’s objective is to inform the the effectiveness of the World Bank’s World Bank’s management and Board on vii Overview operations in solving social contract issues. nontraditional data sources, including Further, the evaluation does not question barometer surveys and perception-based the importance of understanding and data; they analyzed historical social contract influencing social contracts to the World change, and identified social contract Bank’s mission, an idea already endorsed stakeholders involved in the bargaining and by the Board. the power imbalances that lead to dysfunctional social contracts. Findings on Social Contract Diagnostics However, the World Bank misses The World Bank increasingly uses social opportunities to engage key counterparts contract diagnostics to understand regional when it does not disseminate social contract and country-specific development diagnostics. In all country cases studied for challenges. Simply defined, social contract this evaluation, the World Bank’s diagnostics help understand what citizens dissemination of its social contract expect from the state, what the state can diagnostics was limited. The World Bank, legitimately expect from citizens, and any therefore, overlooks opportunities to disequilibrium between the two. At the demonstrate its deep country knowledge, country level, this evaluation identified 21 build reform coalitions, and improve its Systematic Country Diagnostics (SCDs) that image among some citizens and key societal use a social contract framing to diagnose actors. When disseminated, these and explain complex development diagnostics have been well received by challenges such as entrenched inequalities, stakeholders. poor service delivery, weak institutions, and why decades of policy and institutional Currently, the World Bank has no formal reforms promoted by external development conceptual framework or shared actors could not fundamentally alter understanding of social contracts, leading to countries’ development paths. invalid uses. As a result, the World Bank sometimes advocates for “new social Social contract diagnostics add significant contracts” without precisely understanding value to the World Bank’s work, provided how to accomplish that goal. Within the 21 they are carefully selected, executed, and SCDs examined by the evaluation team, disseminated. Such diagnostics help teams there is uneven rigor in how the term social understand policy failures, local political contract is used, leading to ambiguity over dynamics, and intractable development how social contract diagnostics can guide challenges and contribute to building actions. Moreover, some World Bank partnerships. This evaluation finds that documents use social contracts as an well-executed social contract diagnostics economic concept but ignore their share similar features: they were conducted sociopolitical dimensions. For external by multidisciplinary teams with experts interviewed, stripping social governance, social development, and contracts of their sociopolitical dimensions political economy experts; they relied on results in insufficient reflection on how viii Overview social contracts change and the roles that control by the Development Effectiveness external actors, such as the World Bank, can team. legitimately and effectively play in their The World Bank has shown agility in renewal. Evidence from this evaluation’s responding to emerging social contract five case studies shows that the World opportunities outside MENA. For instance, Bank’s misunderstanding and lack of after the election of a new president in consensus on its role leaves it overly Uzbekistan, the World Bank confident in its ability to design and shape opportunistically proposed framing a new social contract transitions. The forthcoming implicit social contract in the country’s 2018 Africa Region flagship report on social Performance and Learning Review, a contracts seeks to overcome some of these drastic departure from its earlier strategy. limitations. The World Bank’s proposed National Findings on Operational Implications Citizen Engagement Framework goes beyond embedding citizen engagement The World Bank’s ability to translate social mechanisms into World Bank projects. By contract diagnostics into operations contrast, in Nepal, the World Bank opted depends on whether social contracts are in for a more gradual approach to the transition or static. The 2015 Middle East country’s social contract transition. Only and North Africa (MENA) strategy after the country’s debate on federalism was represents the World Bank’s most settled did the World Bank adjust its ambitious attempt to translate social country program to Nepal’s social contract contract diagnostics into country operations transition. during a social contract transition. This strategy assumes the Arab Spring Yet, helping countries transition to new unraveled the “old” social contract and social contracts poses considerable risks to describes what the World Bank Group the World Bank’s efficacy and reputation. considers to be a necessary “new social The MENA strategy acknowledges this, contract.” Anchoring the MENA strategy saying there are risks that stakeholders around social contract renewal created new could perceive the World Bank as taking policy reform opportunities and improved sides in a protracted political transition; the World Bank portfolio’s coherence and sensitive analytical work damages the targeting of lagging regions. Maintaining World Bank’s privileged relationship with this coherence when the Region’s lending governments; and the intervention or social commitments rose significantly required transition could fail despite the World strong leadership from the Country Bank’s best efforts. These risks materialized Management Unit (CMU) and regional Vice in several MENA countries. Presidency Unit and thorough quality In countries where social contracts are static, the World Bank’s ability to align its lending portfolio with social contract ix Overview diagnostics is limited. In these “status quo” contract bargaining power in specific countries, citizens tend to be disengaged, sectors and localities. However, the 2018 have low capacity to mobilize, and have Independent Evaluation Group evaluation low expectations of the state’s ability to Engaging Citizens for Better Development deliver services and enhance citizens’ well- Results found that the majority of citizen being. In these situations, the state has few engagement activities embedded in World incentives to adjust its role in the social Bank projects are limited in what they can contract and deliver better services. The accomplish. The evaluation also found that World Bank’s interventions show the World Bank has effective, yet continuity with the past and tend to focus underexploited, channels to renegotiate key on strengthening service delivery. By features of the social contract by facilitating contrast, social contract issues that reinforce citizen dialogues with various levels of state society’s capacity to honor its side of the actors. The SCDs’ social contract diagnostics contract, such as citizen engagement or reinforce the point that incremental government accountability mechanisms, are change—through noncore, nonlending less frequently translated into operations, instruments—is unlikely to significantly despite SCDs calling for this. The World shift state-citizen relationships in countries Bank has no concerted approach to ensure with static social contracts. its lending does not prolong or exacerbate broken social contracts. Findings on Discordance with the Business Model In status quo countries, the World Bank uses nonlending activities to address The World Bank faces multiple challenges specific social contract issues, but these and constraints when helping countries activities are not commensurate with the renew their social contracts—challenges large and transformative efforts needed for shared by other donors who use the World Bank to meet its stated governance diagnostics to guide their ambitions. Instruments include trust- operations. First, social contract funded activities, Advisory Services and interventions conflict with the World Bank’s Analytics (ASA), and citizen engagement current client-driven engagement approach, mechanisms embedded in projects. The especially in status quo countries. This World Bank uses ASA to dig deeper into model prioritizes interventions requested social contract issues, and in turn, ASA can by borrowing central governments’ be leveraged to create social contract executive. However, social contract changes. However, ASA are still too rarely thinking must consider other parties to the used, and they have limited impact when contract besides the central government, they are not disseminated to citizens and such as other state and nonstate actors, civil society. The World Bank also embeds citizens and other societal forces. Second, citizen feedback mechanisms into its social contract transitions are long-term traditional projects, which can shift social processes that do not fit into shorter-term x Overview country programs and lending instruments. Offices or the Development Economics Vice Third, engaging in social contract Presidency should further invest in transitions poses risks to the World Bank’s understanding social contracts: how they reputation as a trustworthy government change, how they vary within countries, partner. Some staff interviewed said the what actors and institutional mechanisms social contract agenda could damage the contribute to social contract transitions, how World Bank’s carefully maintained image of institutions affect social contract bargaining, a neutral technocracy. Fourth, the World and how social contracts relate to various Bank staff’s limited social contract–related development outcomes, such as shared skills are an obstacle to implementing social prosperity, economic performance, and contract changes. Country managers and institutional stability. For example, the directors say it is rare to find staff who have World Bank could build on its research on in-depth sector knowledge and analytical the importance of social contracts for skills combined with a good understanding development in Africa. In addition, this of societal and political dynamics. It was research and refined understanding should clear in all case study countries the World inform a systematic process to help CMUs Bank deployed few, if any, staff with decide when to undertake social contract expertise in governance and social diagnostics. development to translate social contract Defining the World Bank’s role in social diagnostics into country program contract renewal, specific to each country’s implementation, and to advise and engage social contract dynamics, can improve the on relevant interventions. World Bank’s coherence. This evaluation Lessons identifies three sets of conditions that the World Bank should take into account to To overcome these incongruities and derive gain traction on social contract issues. These more value from its social contract conditions include the following: diagnostics, the World Bank may want to better account for client countries’ social • The presence or absence of a social contract dynamics, define its role according contract renegotiation within the to these dynamics, and reflect on how it country. engages with clients, citizens, and partners • The nature of the country’s social to catalyze change. The following lessons contract bargaining. This includes emerged from the analysis. whether the social contract benefits the The World Bank may want to refine its few or the many; whether it relies on understanding of social contracts and be corruption and clientelism; whether more strategic in when and how to citizens and civil society can participate undertake social contract diagnostics. To in the bargaining; whether the state is this end, World Bank teams in charge of responsive to citizens’ social contract research, such as Regional Chief Economist demands; and whether the state has the xi Overview capacity and legitimacy to fulfill these exploiting its other nonlending channels of demands. influence. • The World Bank’s comparative In both types of countries, the World Bank advantages in specific social contract should focus on how it delivers its country situations. This includes its positioning programs. compared with other development First, building social capital with citizens partners and its reputation and social beyond the citizen engagement mechanisms capital among citizens and civil society. embedded in projects can help mobilize In countries undergoing social contract support for reforms. Such an effort is transitions, the World Bank could consistent with previous Independent contribute more effectively to social contract Evaluation Group recommendations in renewal in four ways: (i) ensuring that Engaging Citizens for Better Development social contract diagnostics are well executed Results and would entail broadening citizen and strategically disseminated; (ii) ensuring engagement beyond projects to the national that country engagements are coherent with level. the social contract diagnostics by following Second, relying more on partnerships can diagnostics with sector- or reform-specific increase civic mobilization and political political economy analyses; (iii) transformation. CMUs should convene encouraging strong country director broad country-level coalitions beyond leadership on social contract issues; and (iv) government actors and across election building country teams’ capacity on cycles. This is necessary because the World sectoral, governance, social development, Bank does not have a mandate, or a and citizen engagement expertise. comparative advantage, to manage the In countries where the social contract is political dimensions of social contract static, CMUs in partnership with Global renewals. Practices could use social contract Third, better mobilization of staff diagnostics more effectively to guide its specializing in social contract issues by country engagement in three ways: (i) Global Practice management and strong adjusting its stated goal of helping countries leadership by country directors can help renew their social contracts to something improve the World Bank’s capacity to assist more precise and achievable; (ii) developing countries to renew their social contracts. an approach to safeguard against lending Country directors play a pivotal role in that prolongs or exacerbates broken social overcoming institutional fragmentation and contracts; and (iii) more deliberately negotiating sensitive social contract reforms with governments. xii 1. Introduction 1.1 The World Bank emphasizes the importance of social contracts to eliminate poverty and boost shared prosperity. In the exposition of its twin goal strategy, the World Bank calls for “social contracts within each country demanding that the poor be a priority in the policy environment that supports the growth process” (World Bank 2014c, 20). Through the voice of its president, its high-visibility publications, and its Systematic Country Diagnostics (SCDs), the World Bank has identified broken state- societal agreements—embedded in social contracts—as causing development challenges in client countries. The World Bank has proposed avenues for change and committed to helping societies reshape their social contracts. Such social contract renewals are at the core of the 2015 regional strategy in the Middle East and North Africa (MENA), were proposed to address growing distributional tensions in Europe and Central Asia and promoted to sustain inclusive growth in East Asia and Pacific (Bussolo et al. 2018; World Bank 2018c). Also, according to the 2019 World Development Report (WDR), creating new social contracts will be fundamental to overcome the challenges linked to technological advances in the labor market (World Bank 2019). 1.2 Social contracts help explain development outcomes. Social contracts can be understood as the implicit, mutual bargaining over what citizens expect from the state, and what the state can legitimately demand of citizens in return. The concept of “social contracts” was popularized in 17th-century political philosophy but has evolved and is increasingly integrated in development research and practice (McCandless 2018). There is increasing awareness that development outcomes—such as inequality, human capital, quality of services, and intergenerational mobility—cannot be explained solely by differences in incomes, policy, or demography. Rather, these different outcomes are also linked to social norms, a country’s history, societal expectations of the state, and the institutional channels that define state-society relationships. This state-society relationship is at the heart of the social contract (Alston et al. 2013). In turn, development processes also shape the trajectory of the social contract. The diversity of social contracts, even in countries with similar histories, geographies, and income levels, has prompted research on the determinants of social contracts and their influence on development outcomes. For example, scholars have studied the relationship between failed social contracts and the onset of war (Azam and Mesnard 2003; Addison and Murshed 2001). Other research uses social contracts to explain taxation, public services, conflict and fragility, welfare regimes, and social protection (Hickey 2011; Ho 2018; Isakjee 2017; Nugent 2010) and the policies behind them. 1.3 Several factors are destabilizing social contracts across regions. Research on social contracts also shows they maintain “punctuated equilibriums,” that is, long 1 Chapter 1 Introduction periods of relative stability, followed by shorter periods of abrupt transitions (Alesina et al. 2012; Alston et al. 2013; Acemoglu and Robinson 2013). World Bank research shows that social contracts are under stress from global trends—in migration, climate change, and disruptive technologies—or unique country circumstances, such as conflict, perceived inequalities, and public service dissatisfaction (World Bank 2011b, 2016b, 2018c). Nevertheless, although these stress factors are causing some countries’ social contracts to change, or undergo social contract transitions, other social contracts are static. Objectives and Scope 1.4 The evaluation’s objective is to inform the World Bank’s management and Board on the risks, challenges, value added, and opportunities of using social contracts to shape the World Bank’s country engagements. This learning-oriented evaluation, requested by the World Bank’s senior management, generates lessons from the World Bank’s experience using social contract diagnostics to help countries reshape their social contracts. The evaluation will inform World Bank teams that have integrated social contracts into their strategy, such as the MENA Region, or are considering it, such as the Africa Region and the Fragility, Conflict, and Violence Global Theme, among others. It will also inform the ongoing operationalization of the 2017 WDR and the next generation of SCDs. This evaluation builds on the recent evaluation by the Independent Evaluation Group (IEG) on citizen engagement (World Bank 2018b). 1.5 The evaluation examines how various parts of the World Bank understand social contracts and how they diagnose them. It also examines how the World Bank uses social contract diagnostics to shape country engagements. It does this by (i) evaluating the quality and value added of social contract diagnostics; (ii) assessing how social contract diagnostics are translated into operations; (iii) identifying the risks and challenges of integrating social contract diagnostics into operations and drawing lessons on how to overcome these challenges. The evaluation’s examination period was delimited to fiscal year (FY)13 to FY19. This period’s start date coincides with when the World Bank established its twin goals, demonstrating the importance the World Bank places on social contracts to achieve these goals. MENA’s 2015 regional strategy, which focuses the Region’s work on helping renew social contract, was also formulated during this time and, thus, constitutes a special focus of the evaluation. 1.6 Several issues are outside the evaluation’s scope. The evaluation does not assess social contracts or their transitions, the relevance or appropriateness of the social contract concept in general, or the effectiveness of the World Bank’s operations in solving social contract issues. Further, the evaluation does not question the importance 2 Chapter 1 Introduction of understanding and influencing social contracts to the World Bank’s mission, an idea already endorsed by the Board. Rather, it takes stock of social contract knowledge to assess the World Bank’s understanding of social contract risks and opportunities and its role in helping countries reshape them, especially through the integration of social contract diagnostics into country engagements. 1.7 The evaluation is based on an original conceptual framework. The Organisation for Economic Co-operation and Development (OECD) offers a minimalist definition of social contracts as a “dynamic agreement between state and society on their mutual roles and responsibilities” (OECD 2009, 17). Underpinning this agreement is a set of political economy dynamics described by the United Nations Development Programme (UNDP) in box 1.1, which this evaluation adopts as its working definition. Box 1.1. Social Contract Definition As defined by the United Nations Development Programme, “the social contract refers to processes by which everyone in a political community, either explicitly or tacitly, consents to state authority, thereby limiting some of her or his freedoms, in exchange for the state’s protection of their rights and security and for the adequate provision of public goods and services. This agreement calls for individuals to comply with the state’s laws, rules, and practices in pursuit of broader common goals, such as security or protection, and basic services. The validity and legitimacy of a social contract may be gauged by the extent to which it creates and maintains an equilibrium between society’s expectations and obligations and those of state authorities and institutions, all amidst a context of constant flux” (UNDP 2016, 9). Social contracts are, first and foremost, shaped by the state-society interface. This is the intersection between society’s expectations about what it should receive from the state, and what the state can legitimately demand from citizens in return. Social contracts define the boundaries of what society perceives as acceptable policy. This evaluation focuses on policies highlighted in SCDs or other Region-specific research. These policies are typically for social and governance issues related to jobs, taxes, security, public services, and subsidies and social protection. Social contract stability and robustness over time depends on society’s levels of cohesion, which requires tacit and common agreements on what citizens expect from the state and the state’s capacity to deliver on these expectations. Source: UNDP 2016. 1.8 To understand how social contracts are shaped, how they change, and how external development actors, such as the World Bank, contribute to this change, a more elaborate conceptual framework is necessary. Hence, the evaluation team constructed the conceptual framework based on expert consultations, a structured review of the World Bank’s social contract research, and an academic literature review on social contracts. Figure 1.1 shows that the conceptual framework consists of four elements: (i) the main players that shape social contracts (corners of the rectangle); (ii) the types of influence or power these players exert on each other in the bargaining process (arrows); 3 Chapter 1 Introduction (iii) the nature of the institutional space where social contract bargaining takes place (center of the rectangle); and (iv) the ways each actor engages with existing social contracts at a given point in time. In many countries, there are multiple social contracts, with subnational variations in state-society relationships; however, this evaluation focuses on national-level social contracts, which is consistent with the World Bank’s scope for social contract diagnostics. A more detailed explanation of the framework is in appendix B. Figure 1.1. Social Contracts Conceptual Framework Source: Independent Evaluation Group. Note: Depending on the country, state actors that are influential on the social contract can comprise actors at multiple levels of the state structure and within the various branches of government. Non-state actors may include organized groups of elites, labor unions, private companies, nongovernmental organizations, and civil society organizations, as well as traditional and religious groups. External actors that can influence the bargaining among domestic actors include bilateral and multilateral agencies, foreign governments, and international nongovernmental organizations, as well as international organized terrorist or crime groups, and so on. 4 Chapter 1 Introduction Methodology 1.9 The evaluation’s main question was, “When are social contract frameworks useful in shaping World Bank country engagements?” The team assessed how well the World Bank understands and diagnoses social contracts in its research and SCDs and how coherently it links its Country Partnership Frameworks (CPFs) and lending and knowledge portfolio to social contract diagnostics. When the team found a mismatch between the diagnostics and country engagements, they identified the reasons for it. 1.10 The methodology relies on five data sources: (i) a structured literature review, (ii) a coherence analysis, (iii) case studies, (iv) a stocktaking of World Bank and other development partners’ adoption of governance diagnostic frameworks, and (v) key informant interviews. The evaluation design matrix in appendix A describes the full methodology. • Structured literature review: The evaluation team undertook a structured literature review of World Bank strategy documents, global and regional research on social contracts, and SCDs that explicitly used a social contract diagnostic. The team compared this internal literature to external research on social contracts from the Middle East, North Africa, and Latin America. • Coherence analysis: The team analyzed the coherence among social contract diagnostics, the CPFs or the Country Engagement Notes, and the composition of the lending and knowledge portfolio for countries where an explicit social contract diagnostic was adopted in the SCD (n = 21) and for all countries in the MENA Region (n = 12). Moreover, the team reviewed the portfolio’s evolution using the Completion and Learning Review as a reference document to identify portfolio changes that took place after the social contract diagnostic. This review was complemented by interviews. For countries in the MENA Region, the coherence analysis examined the portfolio’s overarching trajectory to identify how the type, nature, and location of operations were influenced by pillar 1 of MENA’s Regional strategy, which was to renew social contracts. • Case studies: The evaluation team conducted five case studies: Guatemala, Nepal, Niger, Tunisia, and Uzbekistan. These case studies assessed the coherence and usefulness of the World Bank’s social contract diagnostics. The team chose the cases based on discussions with technical counterparts and other World Bank experts. The team looked for variability in (i) whether the World Bank adopted an explicit social contract to guide its operation, (ii) whether the country’s social contract is in transition, and (iii) whether the case’s social contract issues differed from the other cases. This variance allowed the team to assess coherence across 5 Chapter 1 Introduction contexts and generate context-specific lessons. Table 1.2 summarizes the case study selection. The team adopted a detailed case protocol to ensure data collection and analytical consistency across cases. • Stocktaking: To draw lessons from World Bank and other development partners’ experience with applying governance and social diagnostics to their operations, the evaluation team conducted a stocktaking of six analytical frameworks and diagnostic tools used by the World Bank and four frameworks used by other development partners. The goal was to identify the strengths and weaknesses of these approaches, the extent to which they led to changes in country engagement, programming, or operations as well as challenges to operationalization. The stocktaking included a review of evaluative evidence on the application of these frameworks as well as interviews with key informants who were involved in the design and implementation of these frameworks. Table 1.1. Case Study Selection World Bank Social Contract Social Contract in Country Diagnostic Transition (Stage) Other Characteristics Guatemala Yes No Middle-class exit and ethnic fragmentation Nepal No Yes (advanced) Transitioning out of fragility, ethnic fragmentation Niger Yes No Fragile state Tunisia Yes Yes (middle) Considered “success” of the Arab Spring Uzbekistan No Yes (early) Resource-rich and planning economy Source: Independent Evaluation Group. • Key informant interviews: The evaluation team conducted 163 interviews. Specifically, the team consulted three groups of key informants. (i) World Bank managers and senior staff in Regions, Country Management Units (CMUs)— including vice presidents, chief economists, and country directors and managers—and the Global Practices (GPs) and Global Themes that play a direct role in social contract–related issues—including the Poverty; Fragility, Conflict, and Violence; Governance; Social Protection; and Social Development GPs; (ii) staff and managers who worked for the MENA Region at some point between FY15 and today; and (iii) external academics, think tanks, civil society organizations (CSOs), nongovernmental organizations, labor unions, citizens, and government representatives. 6 2. Social Contract Diagnostics Highlights • Social contract diagnostics are useful analytical innovations; they explain seemingly disparate and intractable development challenges and policy failures in client countries through a single approach. • When well executed, social contract diagnostics add value to the World Bank’s work: they enhance teams’ understanding of the key sociopolitical issues undermining countries’ development, they help teams mitigate the political and governance risks in operations, and they build partnerships and policy reform alliances. • Yet, the World Bank could better use these diagnostics to demonstrate its improved understanding of the societies in which it operates, to improve its reputation among nongovernment actors, and to help citizens hold their governments accountable. • The World Bank’s social contract diagnostics have some limitations: they do not have a common conceptual framework, nor do they articulate a clear role for the World Bank, leading to casual or invalid uses that undermine their analytical utility and the legitimacy of the World Bank’s recommendations on preferred types of social contracts. 2.1 The World Bank increasingly uses social contract diagnostics to understand regional and country-specific development challenges. Simply defined, social contract diagnostics help us understand what citizens expect from the state, what the state can legitimately expect from citizens, and any disequilibrium between the two. The World Bank uses several channels to conduct social contract diagnostics. At the regional level, all Regions, except South Asia, have published flagship research reports that diagnose social contract issues and identify avenues for reshaping them, starting with MENA in 2013. At the country level, this evaluation identified 21 SCDs that use a social contract framing to diagnose and explain complex development challenges such as entrenched inequalities, poor service delivery, low domestic resource mobilization, high level of informal labor; and intractable governance issues, such as weak institutions and corruption and cronyism. The first country study that used social contract diagnostics was the 2006 study entitled A New Social Contract for Peru: An Agenda for Improving Education, Health Care, and the Social Safety Net (Cotlear 2006). The first SCD using a social contract diagnostic was Bulgaria in 2015. Alternatively, social contract diagnostics can also be used with other World Bank diagnostic tools, such as risk and resilience assessments. This chapter uses the evaluation’s structured review of the World Bank’s research and diagnostics, compared with the broader literature on social contracts, to assess the utility and quality of social contract diagnostics. 7 Chapter 2 Social Contract Diagnostics The Added Value of Social Contracts Diagnostics 2.2 The World Bank has produced influential diagnostics on social contract issues. The structured literature review conducted by the evaluation team showed that the World Bank is among five leading development policy actors engaging with the concept of social contracts, along with the UNDP, the OECD, the U.S. Agency for International Development, and the U.K. Department for International Development. While the UNDP and the OECD have principally researched social contracts in the context of fragility and peace building, the World Bank’s research has used social contracts more broadly to shed new light on intractable development challenges, such as weak governance, inequality, and poor-quality services. A focused look at social contract research in North Africa, the Middle East, and Latin America confirmed the World Bank’s research on social contracts has been very influential in these regions. Social contract research uses a wide range of data, especially perception-based data, and qualitative case studies. For instance, the World Bank researched Latin American social contracts, particularly its fragmentation and how the middle-class opts out from a social contract from which they see little benefit by avoiding taxes and using the private sector for basic services such as health, education, and electricity (Ferreira et al. 2013). The research showed that this social contract fragmentation contributed to persistent inequalities in the region. Other organizations operating in Latin America also relied on the World Bank’s social contract research to guide their own research. For example, in 2014, the Economic Commission for Latin America and the Caribbean’s Compacts for Equality: Towards a Sustainable Future and, later, OECD’s Latin American Outlook 2018: Rethinking Institutions for Development both relied on the World Bank’s social contracts research. 2.3 Social contract diagnostics demonstrate the importance of perception-based data in explaining complex phenomena. Perception-based data are data that capture people’s perceptions of the state’s efficacy, legitimacy, and corruptibility and their satisfaction with state services and their standard of living. Interviews with chief economists and high-level researchers show that the successful use of perception-based data in social contract diagnostics justifies relying on this type of data, a notion long contested within the World Bank. After the Arab Spring, several flagship reports from MENA’s Chief Economist’s office successfully used perception-based data to explain the brewing discontent, which standard development indicators could not explain. The research contrasted conventional data showing economic growth, poverty reduction, and improved equality with perception-based data, which relied on an eclectic mix of economic, political, and sociological data, showing grievances and high levels of dissatisfaction (Ianchovichina 2018). As a result, the MENA Region institutionalized perception-based data in its regional strategy results framework as a measure of key 8 Chapter 2 Social Contract Diagnostics performance indicators. Similarly, a Europe and Central Asia flagship report shows that perception-based data better explains feelings of inequality and discontent with the Region’s social and political situation than do traditional measures of income and resource distribution (Bussolo et al. 2018). 2.4 At the country level, social contract diagnostics are adaptable to many contexts. All interviewees felt social contract diagnostics were pertinent in most contexts, from fragile states where citizens expect little from the state; to middle-income countries where the emerging middle class tends to opt out of publicly provided services; to higher-income countries where globalization and technological advancement challenge social contracts. The adaptability of social contract diagnostics was shown in the 21 diverse SCDs in which it was applied. As shown in table 2.1, these countries span four regions and include all income groups, fragile and nonfragile states, and resource-rich and resource-poor countries. Table 2.1. Systematic Country Diagnostics Using Social Contracts Framing Middle East and Latin America and the North Africa Region Africa Europe and Central Asia Caribbean Egypt, Arab Rep. Guinea-Bissau Belarus Chile Iraq Mali Bulgaria Costa Rica Jordan Mauritania Russian Federation Guatemala Lebanon Mauritius Tajikistan Paraguay Morocco Sierra Leone Haiti Tunisia South Africa Source: Independent Evaluation Group. 2.5 Social contract diagnostics can explain disparate and seemingly intractable development challenges within a single analytical framework. All interviewees thought social contract diagnostics could explain phenomena that resist conventional thinking on market or government failures, or other paradigms. Social contract diagnostics explain development challenges and the persistence of distortionary economic policies because of unequal bargaining power among citizens, the state, and nonstate players, including the private sector. For example, in MENA, it was the reciprocal obligations between the state and citizens that maintained distorting food and energy subsidies on the one hand and repression on citizen voices on the other. This evaluation’s structured review of the 21 SCDs shows that, echoing the emergent academic literature on the topic, the World Bank has used social contracts to explain multiple policy failures related to social exclusion, unemployment, governments’ lack of accountability, and inadequate social protection and service delivery, among others (figure 2.2). The SCDs leverage social contract diagnostics to explain why decades of policy and institutional reforms from external actors could not fundamentally alter countries’ development paths. For 9 Chapter 2 Social Contract Diagnostics instance, Guatemala’s 2015 SCD identifies a fragmented social contract between “two Guatemalas.” What the SCD calls the “first Guatemala” is urban, wealthier, and of Hispanic decent and benefits from the current social contract; the “second Guatemala” is rural, poorer, and more indigenous and is largely excluded from the current social contract. The second Guatemala expects and receives little from the state. The SCD shows the legacy of this fragmented social contract is rooted in the economic and social dynamics of colonial times, particularly discriminatory property rights and massive land expropriations, which created imbalanced political power and the inability of the “second Guatemala” to renegotiate the social contract. Figure 2.1. Development Challenges Explained by Social Contracts in Systematic Country Diagnostics Social exclusion & low citizen voice 19 Lacking jobs & private sector 13 Low transparency & accountability 12 Poor service delivery 10 Insufficient social protection 8 Fragility, crime & violence 6 Weak resource mobilization & management 6 Overcentralized state 2 0 2 4 6 8 10 12 14 16 18 20 Number of diagnostics (n = 22) Source: Independent Evaluation Group. Note: n = 22; all 21 SCDs with social contract framing are included as well as the MENA regional strategy. 2.6 When well executed, social contract diagnostics help the World Bank mitigate governance and political risks. Social contract diagnostics build on a two-decade-long effort by World Bank governance and social development specialists to ensure that World Bank diagnostics and operations systematically consider the complex social, political, and institutional settings through which policies are designed and implemented. As such, SCDs with a well-executed social contract diagnostic apply specific analytical tools, such as fragility assessments, political economy analyses (PEAs), and institution and governance reviews. These diagnostics are typically conducted by interdisciplinary teams composed of governance, social development, social protection, and poverty or fragility specialists. Recent SCDs that used a social contract diagnostic benefited from analytical guidelines that emerged from the 2017 WDR on Governance and the Law. Social contract diagnostics have led some CMUs to revive the use of PEAs for specific sectors or operations, which can make the CMU’s operational portfolio more relevant and feasible. In the Maghreb CMU, a three-year 10 Chapter 2 Social Contract Diagnostics programmatic Advisory Services and Analytics (ASA) initiative applies PEAs to the social contract renewal process. This evaluation’s Tunisia case study confirms the usefulness of these PEAs on four flagship operations, including a youth employment project, an education reform project, a development policy financing (DPF) project, and a state-owned enterprises governance reform ASA. The PEAs precisely identified the causes of past implementation failures and the risks to implementing new operations. It helped the World Bank team mitigate high political risks and trade carefully between entrenched interests that were opposing reforms. It provided practical solutions, such as building a reform coalition that targets specific groups within unions and the government, compelling specific groups who resist change to adjust their behavior, and adopting gradual reforms rather than the immediate, or “big bang,” reforms that were planned. 2.7 When disseminated, these diagnostics have been well received by stakeholders. Social contract diagnostics can build bridges between the World Bank and the societies it studies. For example, in South Africa, the country team deliberately decided to include South Africans in conversations on the SCD’s design and widely disseminated the SCD. The goal was to overcome the World Bank’s reputation in South Africa, which was negatively tainted by the structural adjustments’ era. As a result, the World Bank partnered with local researchers to analyze South Africa’s social contract situation to identify constraints to achieving the World Bank’s twin goals. The country team used the SCD to genuinely learn about the country and apply this knowledge to operations, which in turn made operations less unilateral and more responsive to South Africans’ expectations. Interviewed SCD authors emphasized that the SCD was widely distributed and considered by government officials, academics, CSOs, and development partners in South Africa to be of high analytical value. 2.8 The World Bank misses opportunities to engage key counterparts when social contract diagnostics are not disseminated. In all country cases studied for this evaluation, the World Bank showed limited dissemination of its social contract diagnostics. The reasons for not disseminating varied: the SCD dissemination was subsumed in the CPF dissemination with little attention to the diagnostic, research teams moved on to other tasks, or the diagnostic was considered too controversial by the government. Engaging with policy makers and civil society in the country at an early stage over the concept and implications of the diagnostics can also prove useful. Failing to engage counterparts on the diagnostics, the World Bank misses opportunities to demonstrate the depth of country knowledge it acquired by doing the diagnostic, to improve its poor image among some citizens and key societal actors, and to build reform coalitions. In Tunisia, where the World Bank developed a high-quality social contract diagnostic in its SCD that built on social contract research taking place at the regional level and on in-depth analytical work 11 Chapter 2 Social Contract Diagnostics on state capture in several sectors of the Tunisian economy, the World Bank did not disseminate its diagnostic. CSOs, development partners, and members of the Tunisian governments interviewed concurred that the World Bank missed an opportunity to show an intimate familiarity with the social contract transition for which the country received the 2015 Nobel Peace Prize. In sharing its diagnostic more widely, the World Bank could have improved its image among citizens and skeptic CSOs. It could also have enhanced its ability to convene partners to act on social contract issues, especially in areas where the World Bank does not have a comparative advantage and needs partners to step in, such as on political institutions strengthening. Limitations 2.9 Currently, the World Bank has no formal conceptual framework nor a shared understanding of social contracts. A systematic look at the World Bank’s knowledge work that uses a social contract diagnostic shows that social contracts are understood and conceptualized differently across World Bank documents. Some, such as the 2012 and 2017 WDRs take a decidedly historical and institutional perspective on social contract formation. Others, such as the 2019 WDR and the East Asia and Pacific Region flagship report Riding the Wave, focus on emerging and future threats that might destabilize current social contracts. Moreover, some documents focus on how formal and informal institutions shape social contract bargaining, whereas others, such as the 2019 WDR and the 2018 Europe and Central Asia Region flagship report Toward a New Social Contract: Taking on Distributional Tensions in Europe and Central Asia, only focus on the policies that emerge from this bargaining and ignore the negotiation process altogether (Bussolo et al. 2018). World Bank staff and chief economists interviewed for this evaluation highlight that there is no shared understanding of “social contracts” across the World Bank despite the popular use of the term in documents. They decry the term’s casual, undefined use or misuse as a euphemism for authoritarianism or predatory states. 2.10 Social contract diagnostics have uneven quality. In addition to the 21 SCDs that have social contract diagnostics, the evaluation identified 13 others that mentioned “social contract” casually. Within the 21 SCDs, there is uneven rigor in how “social contract” is used, leading to ambiguity over how social contract diagnostics can guide actions. Almost half of the SCDs (10 of 21) diagnose a broken social contract without defining social contracts. Another quarter (5 of 21) employ a state-focused definition that emphasizes the public sector’s role in delivering public goods and services. Only a third (7 of 21) emphasize citizens’ and other stakeholders’ participation in social contract bargaining. Moreover, in some SCDs, social contract diagnostics use superficial insights into the historical legacies that influence social contracts. However, when social contract diagnostics are well executed they mobilize local research and expertise, including an 12 Chapter 2 Social Contract Diagnostics array of longitudinal perception-based data, to capture the social contract’s institutional roots and current dynamics, providing a detailed picture of the social contract’s multiple stakeholder and bargaining groups. 2.11 As a result, the World Bank sometimes advocates for “new social contracts” without precisely understanding how to accomplish that. Many World Bank documents use “social contracts” as an economic concept but ignore their sociopolitical dimensions. For example, the 2018 Europe and Central Asia Region flagship report Toward a New Social Contract: Taking on Distributional Tensions in Europe and Central Asia explicitly states, “the report put an economic interpretation on the concept [of social contract]” and defines it as the “individuals’ agreement for the broad outline of economic policies if the outcomes of these policies coincide with their preferences” (Bussolo et al. 2018, 3). Redefining social contracts as an economic equilibrium helps justify the World Bank’s foray into social contract renewal and its recommendations for what social contracts should look like. However, for interviewed external experts, stripping social contracts of sociopolitical dimensions is problematic because it does not properly reflect how social contracts change and the roles that external actors, like the World Bank, can legitimately and effectively play in their renewal. The ongoing regional research on social contracts in Africa seeks to overcome some of these limitations by investigating how different forms of social contract bargaining relate to development outcomes. The research explores how the region’s social contracts are produced and how they change over time. It brings into focus the nexus between politics, citizen-state power relations, and development outcomes (see box 2.1). 2.12 The World Bank has not yet articulated the specific conditions under which it can effectively help countries transition from one social contract to another. As discussed earlier, the World Bank has used social contract diagnostics in a diverse set of countries, irrespective of the country’s social contracts and its trajectory of change. Currently, there is no systematic consideration for when, where, and how to conduct social contract diagnostics. Instead, diagnostics are undertaken on an ad hoc basis, mostly when SCD task team leaders feel that social contracts are relevant for the country’s diagnostic. Paradoxically, outside of MENA, most of the countries where the World Bank has undertaken a social contract diagnostic are not renegotiating their social contracts, whereas countries that have recently experienced social contract transitions, such as Burkina Faso, The Gambia, Kazakhstan, and Serbia, do not have a social contract diagnostic. Even in the case of MENA, where there has been high-quality research on the shared characteristics of broken social contracts, staff working in the region still do not generally understand the social contract idiosyncrasies in each country, a necessity for any country’s social contract renewal. The structured document review and interviews with external experts and World Bank staff involved in knowledge work on 13 Chapter 2 Social Contract Diagnostics social contracts all coalesce to show that the World Bank has no shared understanding of how to apply social contract diagnostics to different country contexts and political settlements. 2.13 The World Bank has not sufficiently reflected on the role it can legitimately and effectively play in renewing social contracts. Except for the 2017 WDR and forthcoming research on social contracts in Africa (see box 2.1), none of the documents using a social contract diagnostic and proposing policy solutions for social contract renewal clearly identify the role of external actors and their comparative advantages and disadvantages in helping countries transition from one social contract to another. Yet, as the next chapter shows, the extent to which the World Bank can effectively help countries change their social contracts crucially depends on whether these countries are in transition or not. Moreover, the World Bank’s capacity to help countries renew their social contracts also depends on the relative influence and social capital the institution has at any point in time within a country. The World Bank’s economic redefinition of social contracts places faith in the capacity of external actors to bring “solutions” to weak social contracts by changing formal rules, building the capacity of state actors, and changing policies. According to many interviews with staff and scholars and evidence from this evaluation’s five case studies, these misunderstandings and the lack of consensus on its role leaves the World Bank overly confident in its ability to design and shape the pace of social contract transitions. Box 2.1. A Promising Example of Social Contract Research in the Africa Region At the initiative of the Chief Economist for the Africa Region, the World Bank is currently undertaking regional research (to be finalized in FY20) on the importance of the social contract for development in Africa. The research aims to inform policy dialogue and development practice by focusing on the sociopolitical settings that enable or constrain effective policies from being successfully implemented. This study has the potential to enhance the World Bank Group’s understanding of social contracts through four main contributions. First, the study team is proposing a conceptual framework to capture how different social contracts are produced in Africa, how they change over time, and how they relate to various development outcomes in terms of stability, equity, and prosperity. Second, through longitudinal case studies the study team is generating in-depth evidence on the social contract’s dynamics, its multiscalar nature, and the various bargaining mechanisms at play in various contexts. Third, building on the case study evidence and a review of robust indicators, the study team is developing a measurement framework to characterize social contracts and track their change trajectories. The research is a multidisciplinary endeavor. It is co-led by the Social Protection and Job, Social Development, and Governance Global Practices. It relies on the expertise of academics who worked extensively on Africa and sought collaborations with civil society. Source: Independent Evaluation Group. 14 Chapter 2 Social Contract Diagnostics 2.14 To sum up, social contract diagnostics add significant value to the World Bank’s work, provided they are carefully selected, executed, and disseminated. Well-executed social contract diagnostics help teams understand policy failures, local political dynamics, and intractable development challenges and contribute to building partnerships. This evaluation finds that well-executed social contract diagnostics share similar features: they were conducted by multidisciplinary teams with governance, social development, and political economy experts; they relied on nontraditional data sources, including barometer surveys and perception-based data; they analyzed historical social contract change; identified social contract stakeholders involved in the bargaining; and identified the power imbalances that lead to dysfunctional social contracts. To derive more value from social contract diagnostics, the World Bank could encourage broader and more strategic dissemination, better define its role in social contract transitions, and develop a criterion for when and how to conduct these diagnostics. 15 3. Operational Implications Highlights • In countries experiencing social contract transitions, the World Bank has coherently aligned its country strategies and operations with its social contract diagnostics. • However, supporting social contract transitions also poses considerable risks to the World Bank’s efficacy and reputation. • In “status quo” countries where social contracts are static, the World Bank has difficulties aligning operations with its stated goal to help countries renew their social contracts. • In “status quo” countries, the World Bank uses nonlending activities to address specific social contract issues, but these activities are not commensurate with the large and transformative efforts needed for the World Bank to meet its stated ambitions. 3.1 The World Bank’s ability to translate its social contract diagnostics into operations depends on whether social contracts are static or in transition. In the MENA Region, for example, the World Bank took advantage of the Arab Spring to realign its regional strategy toward helping countries renew their social contracts. In countries where the World Bank has diagnosed broken social contracts but where there is no internal movement to renegotiate them, or “status quo” social contracts, the World Bank uses nonlending instruments, such as ASAs, trust-funded activities, and project- embedded citizen engagement mechanisms to overcome resistance to change, but to limited effect. This chapter uses in-depth case studies, a coherence analysis, and interviews with World Bank staff and stakeholders to delve deeper into these issues. Social Contracts in Transition 3.2 The 2015 MENA strategy represents the World Bank’s most ambitious attempt to translate social contract diagnostics into country operations. In 2015, the World Bank made social contract renewal one of four pillars in its new regional strategy for MENA. This strategy was premised on the World Bank’s assumption that the “Arab Spring” unraveled the “old” social contract, which was characterized by citizens’ expectations that the state provides food and fuel subsidies, free health and education services, and public sector jobs for those educated enough to attain them, but citizens had limited ability to hold the government accountable for these services and little voice to renegotiate the contract (Devarajan 2016). The MENA strategy delineates what the World Bank Group considers to be a necessary “new social contract” and the three priority areas to achieve social contract renewal: boosting private sector jobs, improving 16 Chapter 3 Operational Implications the quality of services, and enhancing citizen engagement (see table 3.1). The MENA strategy prioritized the social contract pillar in four countries transitioning from one social contract to another, including the Arab Republic of Egypt, Jordan, Morocco, and Tunisia (Devarajan 2016). The MENA strategy achieve its vision in three ways: (i) relying on financial, advocacy, and knowledge sharing partnerships; (ii) using the World Bank Group’s own lending exclusively for projects that will renew the social contract while emphasizing public over private goods, improving the policy environment for private sector investments, and strengthening citizen inclusion and government accountability; and (iii) leveraging the Bank Group’s knowledge work to drive, rather than follow, lending and spur public debates and reform coalitions. Table 3.1. Middle East and North Africa Strategy Pillar 1: Renewing the Social Contract Opportunities Quality Services Citizen Engagement Promote broad-based Strengthen public institutions for more Support enabling reforms and sector development efficient and effective service delivery legislation (for example, access to information) Strengthen skills that Modernize information and Develop performance-based match market demands communication technology systems and aspects of sectoral governance accountability institutions Work on labor markets to Create environment for greater private Enhance mechanisms for promote formal job sector investment in infrastructure and accountability creation services via the International Finance Corporation and the Multilateral Investment Guarantee Agency Design “capture-proof” Modernize the social protection system Facilitate consensus building business policies to foster among stakeholders for sensitive entrepreneurship policy reforms Build inclusion by Build on local success stories (including Mainstream beneficiary feedback supporting lagging regions via third sector providers) Source: World Bank Group (2015a), MENA Strategy Board Presentation. 3.3 Anchoring the MENA strategy around social contract renewal improved the World Bank portfolio’s coherence with its diagnostic. The coherence analysis and interviews with Regional staff and country directors confirm that MENA’s strategy added discipline to the World Bank’s Regional program, creating coherence, even at a time when new lending commitments in the Region rose significantly from $1.5 billion in FY12 to $6.4 billion in FY18. Figure 3.1 shows that SCDs, for most countries of the region, identified social contract issues and integrated them into the CPFs. Once in the CPF, social contract issues translated into knowledge and operational work. Notably, however, citizen engagement operations were missing, which is discussed later. Maintaining this coherence required strong leadership from the CMU and Regional Vice Presidency Unit and thorough quality control by the Development Effectiveness team. 17 Chapter 3 Operational Implications This countered GPs’ incentives to implement their own priorities, which may or may not be aligned with the strategy. Each operation put forward by GPs needed to demonstrate that it contributed to social contract renewal and did not reinforce the “old broken” social contract. As a consequence of this filter, several projects and CPFs were redesigned because they were inconsistent with the vision of social contract renewal. MENA’s strategy became a powerful tool in external policy dialogues to explain to clients and partners why the World Bank would support certain project proposals and not others. 3.4 The MENA Region strategy’s social contract framing also improved the World Bank’s targeting of lagging regions. The social contract renewal process increased the social contract participation of people living in lagging regions. This, as a result, allowed the World Bank to better target operations to these regions. In Tunisia, for example, five recently approved projects target youth and women in lagging regions. Government and civil society counterparts interviewed in Tunisia recognized that deliberately targeting lagging regions made the World Bank’s operations more relevant than other development partners’ operations, whose targeting is less intentional. This is because Tunisia’s transition required greater participation from lagging regions in the social contract. Figure 3.1. Coherence Analysis in the Middle East and North Africa Region Jobs and Private Decentralization Service Delivery Social Inclusion Accountability Transparency Management Engagement and Citizen Protection Financial Tax and Sector Social and Egypt, Arab Rep. (2015) ◼◼◼◼ ◼◼◼◼ ◼◼◼◼ ◼◼ ◼◼◼◼ ◼◼◼◼ ◼◼◼ Iraq (2017) ◼◼◼ ◼◼ ◼ ◼ Jordan (2016) ◼◼◼◼ ◼◼◼ ◼ Lebanon (2015) ◼◼◼ ◼◼◼◼ ◼◼ ◼ ◼◼ ◼ Morocco (2018) ◼ ◼◼ ◼ ◼◼ Tunisia (2015) ◼◼◼ ◼◼◼◼ ◼◼◼ ◼◼◼ ◼◼◼◼ West Bank and Gazaa ◼◼◼◼ ◼◼◼◼ ◼◼◼◼ ◼◼ Yemen, Rep.a ◼◼◼◼ ◼ ◼◼ ◼ ◼ ◼ Given as ◼ Given as program ◼ Included in lending ◼ Included in ASA recommendation in recommendation in portfolio approved portfolio approved SCD CPF pillars since SCD since SCD Source: Independent Evaluation Group. Note: ASA = Advisory Services and Analytics; CPF = Country Program Framework; SCD = Systematic Country Diagnostic. a. Middle East and North Africa strategy 2015. 18 Chapter 3 Operational Implications 3.5 The MENA Region’s strategy’s emphasis on social contracts created new policy reform opportunities. Renewing social contracts provided a useful rationale for countries to embark on necessary but unpopular reforms. For example, the need to reduce oil and gas subsidies became urgent with the ups and downs of the oil prices. Yet, subsidies are very popular, and governments maintain them to avoid possible civil unrest. Country Directors and task team leaders who advocated for replacing subsidies with targeting cash transfers found social contract framing useful in policy dialogues with reluctant governments. These reform advocates emphasized that cash transfers empower citizens to decide on what to spend, as opposed to the government deciding for them through subsidies. This made the reforms more palatable. The World Bank, equipped with modeling tools and technical capacity, took advantage of oscillating oil prices to motivate governments to reform politically sensitive subsidies in several countries in the region (Verme and Araar 2017). In Tunisia, the World Bank is helping revamp the country’s social protection system from a subsidy-heavy model to a targeted cash transfer model. Box 3.1 illustrates the reform’s complex and inherent challenges. 3.6 Social contract framing is obvious in certain projects within MENA’s Regional portfolio. Several MENA operations include all the main principles of social contract renewal as laid out in the MENA strategy. These principles include making private sector jobs resistant to elite capture, bringing informal jobs into the formal economy, targeting lagging regions, and empowering youth and women. According to government and civil society stakeholder interviews, Tunisia’s youth economic inclusion project, or Moubadiroun, is a good example of how these principles successfully infiltrated MENA’s project designs. First, the project targets disadvantaged men and women between the ages of 18 and 35 who live in lagging regions with employment opportunities. This stands in contrast to the employment programs of the “old social contract” that were prone to elite capture. Second, the project eliminates barriers under the “old social contract” that prevented small enterprises from entering the formal economy and reaching national and international markets. This was accomplished by developing value chains and better linking businesses in the country’s lagging regions to established markets. Third, the project includes a scorecard mechanism for program participants to rate the quality of service providers. 19 Chapter 3 Operational Implications Box 3.1. Helping Tunisians Renew Their Social Contract Tunisia is the only case of uninterrupted peaceful transition toward a new social contract after the Arab Spring. Since the revolution, Tunisia has achieved major institutional changes enshrined in the new constitution that lays the foundations for a nascent democracy, a more decentralized state, and a more open economy. A dynamic civil society formed in the aftermath of the revolution and is actively taking part in the redefinition of the Tunisian social contract and in holding governments accountable for following through on their promises. Yet, this peaceful transition toward a new social contract remains at risk. Risks that threaten the achievement of a new social contract include an alarming macro-fiscal situation with rapid inflation, growing public debt, and structural fiscal imbalances; weak implementation capacity within the central and decentralized governments, which jeopardizes the concretization of constitutional reforms; and major blockages on reforms that are crucial to overcome the critical situation of the Tunisian economy. Bilateral and multilateral donors have mobilized a lot of support to help Tunisia in this transition. The World Bank has been a key partner in helping the country undertake major reforms toward a new social contract. Through a mix of budget support, investment projects, and technical assistance, the World Bank has very coherently organized an ambitious portfolio to respond to its social contract diagnostics. Although progress is undeniably taking place on several fronts, the World Bank is facing hurdles in delivering on its portfolio. Despite the political astuteness of the World Bank’s coun try team, key reforms that it supports are currently at a standstill or have had to be aborted, including reforms of the pension system, education sector, subsidy regime, and state-owned enterprises. Challenges include (i) navigating a very complex political economy situation and reaching consensus on reforms that remain unpopular with the powerful Tunisian Labor Union or some parts of the civil service; (ii) overcoming powerful vested interests linked to the “old social contract” that resist opening the economy; (iii) undoing the moral hazard created by the series of budget support that occurred despite a wave of new recruitments and salary increase in the public sector that went against the World Bank and International Monetary Fund recommendations; (iv) and continuing to gain traction in policy dialogues despite having reached its prudential limit on lending. Source: Independent Evaluation Group. 3.7 The World Bank has shown agility in responding to emerging social contract opportunities outside MENA. For instance, after the election of a new president in Uzbekistan, the World Bank opportunistically proposed a new implicit social contract framing in the country’s 2018 Performance and Learning Review, a drastic departure from earlier World Bank programing. The Performance and Learning Review identified new reforms that the World Bank was willing to support financially, but that would require fundamental changes in how the government interacts with citizens. As a result of the Performance and Learning Review, the World Bank now elevates citizen engagement and state institutional reforms to a focus area of the updated World Bank country strategy. The World Bank now proposes a National Citizen Engagement Framework that goes well beyond World Bank projects. This push for national citizen engagement required the World Bank to be opportunistic because of Uzbekistan’s 20 Chapter 3 Operational Implications scarcity of CSOs, state-controlled media, and limited spaces for citizens’ voices. It is also a departure from the approaches of Uzbekistan’s other development partners. For instance, the Asian Development Bank and bilateral donors have not embraced citizen engagement in Uzbekistan, while the European Union prefers financing World Bank programs rather than taking a position on citizen engagement. 3.8 In Nepal, the World Bank opted for a more gradual approach to the country’s social contract transition. After the end of the civil war, Nepal embarked on a lengthy political transition. During the 10 years leading up to a new constitution, the country experienced acute political instability, including 20 different ruling coalitions. During the country’s divisive debate on federalism, the World Bank opted for a wait-and-see approach and abstained from promoting decentralization, despite having a clear comparative advantage on this topic in the South Asia Region. Only after this debate was settled did the World Bank adjust its country program to Nepal’s social contract transition. This gradual approach was deemed crucial by the World Bank team in Nepal because, as the 2017 Risk and Resilience Assessment recognizes, the incomplete social contract’s inequality and power imbalances continue to represent a source of potential conflict. 3.9 Helping countries transition to new social contracts poses considerable risks to the World Bank’s efficacy and reputation. The MENA strategy acknowledges this, saying there are risks that (i) stakeholders could perceive the World Bank as taking sides in a protracted political transition, (ii) sensitive analytical work damages the World Bank’s privileged relationship with governments, (iii) the World Bank’s funds land in “the wrong hands,” and (iv) the intervention or social transition could fail despite the World Bank’s best efforts. These risks materialized in several MENA countries and are largely unavoidable. This is because social contract bargaining is inherently a political process, with a multiplicity of actors entering the policy arena, where many voices must be heard. In Nepal, like in Tunisia, the social contract transition redefines the political settlement, restructuring the state from a centralized to a decentralized power structure and transforming society in the process. Interviewed staff in both countries said these profound changes and unstable political alliances made it difficult to maintain the World Bank’s technical neutrality and make progress on key reforms. For example, as illustrated in box 3.1, several World Bank–supported reforms to state-owned enterprises or pensions in Tunisia failed because the reforms were challenged by the most powerful workers’ union or because they ran counter to vested interests within the government and powerful economic elites with ties to the old regime. 3.10 The World Bank can draw three lessons from its experience operationalizing social contract diagnostics in countries undergoing social contract transitions. First, the 21 Chapter 3 Operational Implications World Bank can use social contract diagnostics to shape country engagements when the diagnostic is well executed and strategically disseminated. This builds social capital with stakeholders that are party to the social contract, including CSOs, trade unions, political parties, private sector actors, farmers’ organizations, and business associations. Currently teams have little incentive to disseminate their analytical work. Second, the World Bank is more effective at operationalizing these diagnostics when they are followed by sector- or reform-specific PEAs and there is strong country director leadership. This leadership is crucial when there is resistance from GPs or governments to this operationalization. However, the rapid changes in leadership mean that capabilities, interests, and commitments to such deep transformation vary considerably over time. Third, country teams must possess the right sectoral, governance, social development, and citizen engagement expertise. Yet, the composition of country teams does not always reflect consideration of skill mix. Teams also benefit from political savviness, agility to respond to opportunities, and programmatic flexibility, as the next chapter explains. Status Quo Social Contracts 3.11 Many countries maintain their social contract status quo, making social contract transitions difficult. Countries that are not experiencing a social contract transition, nor an internal movement to renegotiate core features of the contract, are maintaining a social contract status quo—we call these “status quo” countries. In these countries, the social contract can be fraught, exclude a significant part of the population, or fail to meet citizens’ expectations, yet it is static. For instance, citizens can be disengaged, have low expectations about the ability of state actors to deliver services that would enhance their well-being, or have low capacity to mobilize. As a result, state actors have no incentives to deliver better services. In these situations, using a social contract diagnostic can provide valuable insight to anticipate the possibility of eruption of violence or the breakdown in trust. 3.12 For example, in Niger, the social contract is lopsided, but it is unlikely to change soon. This is because the social contract is well established, with a powerful governing elite, very few nongovernment players, and a largely powerless and disconnected citizenry. Niger’s SCD offers a refreshingly frank diagnostic of Niger’s political settlement and the state-citizen relationship. It also calls for a paradigm shift toward a “citizen-centered approach” to service delivery that reaches citizens directly, empowers them to make their own decisions, and depends less on public intermediaries. However, Niger’s status quo social contract seems unlikely to change because there is no clear dynamic to dramatically modify the current equilibrium. This, despite the country facing multiple poverty traps, enduring serious security and humanitarian risks, and 22 Chapter 3 Operational Implications having low-quality public service delivery. Moreover, governments, often the guardians of the status quo, are unlikely to assent to World Bank efforts to fundamentally alter the social contract status quo, even when it is broken. 3.13 The World Bank has not been able to radically change its lending portfolio in status quo countries. Only two of the 21 SCDs that use a social contract diagnostic judge the country’s social contract as strong, the rest identify weaknesses, broken features, or even the absence of a social contract and call for radical changes to the World Bank’s development approach. The coherence analysis shows the World Bank has not been able to adopt radical changes to its lending portfolio in status quo countries. For instance, the 2015 Guatemala SCD recognizes that a “big push” is needed to break the country out of its negative “self-reinforcing dynamic” (Sanchez, Scott, and Lopez 2016). Despite this, the World Bank’s current list of proposed loans is similar to the list agreed to prior to the SCD calling for a “big push” and limited to what Guatemala’s Congress would approve. For example, the proposed development policy loan (P160667) and the Transparency and Efficiency in Tax Administration (P153366) project maintain the World Bank’s focus on combating corruption, increasing the tax base, and improving the central government's public finance management. The current portfolio’s Investment in Urban Infrastructure project (P143495) focuses on low-income urban dwellers much like the World Bank–supported Barrio Seguro (Safe Neighborhood) program of the previous CPF. And the current Crecer Sano project operates similarly to the Hambre Cero (Zero Hunger) Pact of 2012. 3.14 In status quo countries, the World Bank focuses on strengthening service delivery. In countries where social contracts are static, the World Bank’s ability to align its lending portfolio with social contract diagnostics is limited. This evaluation’s coherence analysis assesses the World Bank’s consistency in identifying social contract issues, proposing reforms to address these issues in SCDs and CPFs, and, subsequently, including these reforms in its lending and nonlending portfolios. As shown in figure 3.2, the World Bank’s coherence varies significantly depending on the type of reform. Social contract issues that pertain to the state’s side of the contract, such as improving service delivery and enhancing the enabling environment for jobs, are systematically part of the World Bank’s CPF and translated into operations. By contrast, social contract issues that reinforce society’s capacity to honor its side of the contract, such as citizen engagement or government accountability mechanisms, are less frequently translated into operations, despite SCDs calling for this. For example, several SCDs highlight issues with eroding media independence, judicial systems that fail to protect the vulnerable, and perception of widespread corruption as damaging social contracts. Yet interventions to remedy these issues are scarce in World Bank portfolios. 23 Chapter 3 Operational Implications Figure 3.2. Coherence Analysis Given as Given as program Included in lending Included in ASA recommendation recommendation portfolio since SCD portfolio since SCD in SCD in CPF Given as recommendation in SCD Given as program recommendation in CPF Included in lending portfolio since SCD Included in ASA portfolio since SCD 16 14 14 14 12 12 12 Number of Countries 12 11 10 10 10 10 9 8 8 7 7 6 6 6 6 6 5 4 4 4 3 3 3 2 2 1 1 1 1 1 0 0 0 0 Source: Independent Evaluation Group. Note: n = 19: only countries with both SCDs and CPFs. ASA = Advisory Services and Analytics; CPF = Country Partnership Framework; SCD = Systematic Country Diagnostic. 3.15 In status quo countries, the World Bank has no concerted approach to ensure its lending does not exacerbate broken social contracts. Moreover, World Bank staff and managers do not agree on how this can be done. Some interviewees believe the World Bank’s emphasis on creating jobs and strengthening service delivery effectively avoided exacerbating inequalities and undermining state-society relationships. Others warned that World Bank loans can inadvertently cause harm when they are not well aligned with social contract diagnostics. The evaluation’s case studies and OECD’s “Do No Harm” paper (OECD 2010) observe other ways that donors risk reinforcing broken social contracts. World Bank loans can do harm when they reinforce local patronage, weaken incentives for elites to “buy into” social contract renewals, raise citizen expectations of the state beyond what the state can reasonably meet, or create a parallel delivery system that competes with those of state agencies. 3.16 In status quo countries, the World Bank uses nonlending instruments to influence social contracts. These instruments include ASAs, trust-funded activities, and 24 Chapter 3 Operational Implications citizen engagement mechanisms embedded in projects. ASAs that dig deeper into social contract issues can be leveraged to create change, especially when the ASA is disseminated to citizens and civil society (Devarajan and Khemani 2018). Figure 3.2 shows the World Bank can use ASAs more often to address social contract issues when lending is constrained by uncooperative governments. In Niger, for example, ASAs promote inclusiveness, transparency, and responsible public sector management, all of which contribute to the quality of social contracts. In fact, eight out of 20 ASAs in Niger are focused on governance; most of the others are focused on service delivery. Two of Niger’s cross-sectoral ASAs reinforce the CPF’s emphasis on women’s empowerment and the SCD’s citizen-centric approach. A few ASAs have destabilized social contracts and led to catalytic effects in status quo countries. For example, a three-year Gender and Social Exclusion Assessment, co-led by the World Bank and Nepali government at a time when Nepal’s social contract transition had not really started, resulted in a landmark publication in 2006 (World Bank 2006b). The publication demonstrated that any effort to renew the country’s social contract would have to address the historical exclusion of Nepali women and ethnic and cast discrimination. As a result of this noncore ASA, gender and social inclusion became important parts of Nepal’s development discourse and social contract discussion for the year to come. Nepal’s National Planning Commission committed to implementing the report’s recommendations. The government subsequently adopted sectoral strategies, implemented various forms of affirmative action, hired gender and social inclusion specialists in several agencies, and monitored outcomes through a series of surveys. 3.17 The World Bank embeds citizen feedback mechanisms into its projects, which can shift social contract bargaining power. The World Bank increasingly uses citizen engagement mechanisms to increase citizens’ trust and shift service providers’ incentives. In Niger, all stakeholders interviewed for this evaluation were satisfied with the World Bank’s increased attention to citizen engagement and feedback. They highlighted that social safeguards forced reluctant governments to pay attention to citizens. In Uzbekistan, nongovernmental organizations filed a 2013 claim to the World Bank’s Inspection Panel that the World Bank’s Rural Enterprise Support Project endorsed child and forced labor. The panel eventually dismissed the claim, but the World Bank established a third-party monitoring mechanism for Uzbekistan’s cotton harvest and used lending to pressure the government to change national labor practices. Through the third-party monitoring, the World Bank trained local CSOs to legally and independently monitor the harvest season for labor abuses and provided these monitors with communication channels to the government, including an anonymous hotline to voice grievances. 25 Chapter 3 Operational Implications 3.18 The World Bank’s nonlending instruments create useful incremental changes, but these will not be enough to fundamentally reform social contracts. Social contract diagnostics in SCDs are clear on this point: incremental change is insufficient to shift state-citizen relationships in countries with static social contracts. An IEG evaluation, Engaging Citizens for Better Development Results, found that most citizen engagement activities embedded in World Bank projects are limited in what they can accomplish (World Bank 2018b). These mechanisms are helpful but not deep enough to trigger institutional changes. In that sense, the World Bank may want to rein in its rhetoric on helping countries renew social contracts at the national level and be more realistic in what it can achieve with its existing tool kit. The World Bank can be more strategic in affecting change if it pays closer attention to the multiplicity of social contracts within countries. The IEG evaluation also found the World Bank has effective, yet underexploited, channels to broker citizen dialogues with the state. One such example was identified in the Guatemala case study, which shows the World Bank helped give voice to indigenous communities through a trust fund activity that organized dialogues between business leaders and indigenous groups under the International Labour Organization Convention 169. The World Bank managed to broker these consultations, despite a long history of mistrust between these two groups, and in the process, rebalanced negotiations on land and hydropower disagreements. 3.19 Three key lessons emerge from the World Bank’s experience operationalizing social contract diagnostics in countries where the social contract is static. First, the World Bank’s ability to change social contract dynamics and its capacity to align its lending portfolio with its diagnostic are constrained in countries where the social contract is not being internally renegotiated. In these situations, the World Bank may want to adjust its stated goal of helping countries renew their social contracts to something more precise and achievable. Second, the World Bank could develop an approach to ensure that its lending does not prolong or exacerbate broken social contracts. Third, outside of lending, the World Bank can more proactively and deliberately exploit its other channels of influence—such as convening power, citizen engagement, and research and analytics—to address social contract issues. Doing so requires having engaged regional and CMU leadership. 26 4. Discordances with the World Bank’s Business Model Highlights • The World Bank’s ability to support countries in renewing their social contracts is hindered by several aspects of its business model and staff’s limited skills. • The challenges that the World Bank faces in operationalizing social contract diagnostics echo similar challenges it faced with other political and governance frameworks and are shared by other development partners. 4.1 The World Bank faces multiple challenges and constraints in operationalizing its ambition to help countries renew their social contracts. This chapter identifies the challenges that the World Bank faces when attempting to coherently translate its social contract diagnostics into effective operations both in countries where the social contract is in transition and where it is static. The chapter specifically identifies some mismatch between the World Bank’s ambition to help countries renew social contracts and some key features of the World Bank’s business model. 4.2 The nature of social contract interventions conflicts with the World Bank’s client- driven engagement approach, especially when social contracts are static. Since the World Bank’s good governance reforms in the 1990s, which sought more client ownership over the World Bank’s development interventions, the institution has strengthened its client-driven model, prioritizing interventions on the formal request of borrowing governments. The World Bank tends to define its “client” narrowly, as the central government, and primarily the executive branch. However, this operating logic is not entirely conducive with social contract thinking, which must consider other parties to the contract, such as citizens and other societal forces, as well as other government branches and tiers, such as subnational authorities that play a critical role in service delivery. World Bank staff are quick to observe that the World Bank, as an external actor, has a limited scope for engagement and authority to engender lasting change in social contracts, especially when the government is the guardian of the status quo. One staff member put it poignantly: “If the government is captured by a self- serving elite that wants to maintain the social contract status quo can the World Bank really push for more inclusive institutions as long as it works with that government?” 4.3 Engaging in social contracts poses risks to the World Bank’s reputation as a trustworthy government partner. Some staff interviewed expressed concern that the social contract agenda could damage the World Bank’s carefully maintained image of neutral technocracy, which is essential for its external legitimacy and client country 27 Chapter 4 Discordances with the Business Model relationships. They emphasized the complexity of engaging on social contract themes and the feeling of “walking on tightropes.” They highlighted several examples where it was necessary for the World Bank to refrain from engaging in sociopolitical issues. For example, in the wake of Uzbekistan’s 2005 violent repression of the Andijan protest, several bilateral donors and representatives from the European Bank for Reconstruction and Development voiced concerns over human rights violations. In retaliation, the Uzbek government ousted the European Bank for Reconstruction and Development and reduced interactions with bilateral donors. Opportunistically, both the World Bank and the Asian Development Bank stepped into this void and increased their lending portfolios. Similarly, in Guatemala, UN-sponsored anticorruption investigations and the International Commission against Impunity in Guatemala fueled the government’s suspicions and resistance to foreign oversight. As a result, the World Bank’s country office in Guatemala takes great care to meet with lobbyists, parliamentarians, and government officials to convey impartiality. Most interview sources said the World Bank has been successful in maintaining its image as an apolitical, technocratic organization in Guatemala. That said, any association with the UN or International Commission against Impunity in Guatemala in the eyes of Guatemala’s elite could jeopardize the World Bank’s operations in the country. 4.4 Social contract renewal is a long-term process that does not fit into shorter-term country programs and lending instruments. Renewing social contracts requires complex institutional and territorial reforms and new legislation in sensitive areas such as pensions, taxation, and public sector organization, which can take a decade to pass and another decade to fully implement. A closer look at CPFs casts doubt on the feasibility of aligning three- to five-year programs with long and encompassing social contract transitions. For instance, in Guatemala, Niger, and Tunisia, the SCDs explicitly argue that policy makers need to address many development challenges at once, and advocate for development policies that create synergies and multiply the effects of single actions rather than merely aiming for incremental change. Yet, the CPFs exclude SCD recommendations that are not government priorities or work to the Bank Group’s comparative advantages. World Bank operations also tend to focus on delivering measurable outputs within relatively short five-year time frames, which is probably not long enough to renew social contracts. This approach may be a pragmatic necessity to ensure effectiveness, but it also raises questions of how a social contract diagnostic can be translated into effective client country engagement. 4.5 Traditional financing and technical assistance constrain the efficacy of interventions to renew social contracts. The World Bank’s influence and bargaining power remains relatively limited because of its limited role as a lender, and the increasing competition of other financial institutions. Even so, using budget support to 28 Chapter 4 Discordances with the Business Model address the social contract distortions of a political equilibrium is problematic (Devarajan and Khemani 2018). The World Bank used DPF to relieve fiscal pressures from reform-minded governments and to nudge reform-skeptic governments toward social contract renewals. In both cases, the World Bank’s model of quickly disbursing money is at odds with building demand for reforms beyond a few government “political champions.” In all case study countries, there were instances of governments accepting DPF prior actions to release loan tranches, and then lapsing into inaction or even reversing reforms. Staff involved in operations that seek to address core social contract issues emphasized that design and implementation schedules are frustrated by the long and sensitive political bargaining that requires World Bank technical staff to engage in contentious negotiations among political factions and divided civil society groups. Government counterparts interviewed expressed frustration with the disbursement metrics by which the World Bank judges the performance of these interventions and argued in favor of longer-term approaches to programming. 4.6 Staff’s limited social contract–related skills are an obstacle to implementing social contract renewal. Examples of this were evident in all country case studies. CMUs report it is rare that staff combine in-depth sector knowledge and analytical skills with a good understanding of societal and political dynamics. It was clear in all case studies that there are not enough social development specialists, with expertise in citizen engagement issues, and governance experts, with expertise in political economy thinking, to staff all relevant interventions. For example, the World Bank’s Nepal and Uzbekistan offices only recently obtained a social development expert on staff. In Tunisia, the World Bank has only one governance specialist for the country’s entire portfolio. CMU leadership laments their lack of control over team composition. 4.7 The World Bank’s internal competition and incentive-to-lend structure limit its motivation to engage in social contract transitions. Interviewees point out that GPs are reluctant to dedicate staff time and supervision budgets to elements outside their core practices, such as integrating citizen engagement mechanisms into projects. This might explain why, despite the Uzbek government providing a window of opportunity to prioritize citizen engagement in the country program, there are currently no citizen engagement projects among the 17 loans and 10 ASAs in Uzbekistan. That said, the World Bank’s citizen engagement portfolio in Uzbekistan is expected to rise from zero to four projects by 2021, a modest increase. Moreover, internal competition among GPs to lend in core business limits their incentive to engage in noncore areas required for social contract transformation. 4.8 Other donors who adopt governance framework diagnostics, which are similar in some ways to social contract diagnostics, share these challenges. A review of four 29 Chapter 4 Discordances with the Business Model governance frameworks developed by other development partners showed similar challenges to those faced by the World Bank in social contract diagnostics. Some of these challenges for donors are related to the nature of these diagnostics: they are of uneven quality, are overly descriptive with few clear practical suggestions on levers of change, and they have limited participatory processes and dissemination mechanisms. Other challenges may relate to the tool’s incongruity with the agency’s operating model. For example, institutionalizing PEA and governance and anticorruption framing within the U.K. Department for International Development was challenging because of the agency’s “Value for Money” agenda, which seeks to always demonstrate what results are bought with the U.K. taxpayers’ money and the lack of staff and director incentives to act on the diagnostics’ recommendations (DFID 2011; Duncan and Williams 2012). The U.S. Agency for International Development’s various unfruitful attempts at institutionalizing PEAs reveal the difficulty in changing operating cultures to think strategically about political issues. The agency’s latest reincarnation of PEAs attempts to create that cultural shift with its “Thinking and Working Politically” approach (Menocal et al. 2018). Lessons from the World Bank’s use of similar frameworks are summed up in box 4.1. Box 4.1. Lessons from the World Bank’s Use of Governance and Social Frameworks Since the 1990s, the World Bank has had experience using various analytical frameworks to provide decision makers with a comprehensive understanding of the country context. These include political economy analysis (PEA); poverty and social impact analysis (PSIA); governance and anti-corruption analysis; institutional and governance review; fragility assessments, and citizen engagement roadmaps. A review of the World Bank’s experience with these frameworks offers relevant insights. Usefulness. Given the diversity of client countries, any framework must be flexible. These frameworks were flexible enough in terms of methods and data sources to allow the World Bank to tailor them to different countries and sectors. Many could also be used to study a specific problem or reform, which made them easier to operationalize. However, flexibility led to ambiguity on how to use the frameworks and affected the real and perceived quality of the analysis. There was sometimes insufficient guidance to country offices and insufficient effort to build in-country capacity. Uptake. Some frameworks, specifically PEA, PSIA, and citizen engagement gained longevity when the World Bank formally mandated their use in operational processes. Moreover, multidonor trust funds ensured long-term funding and technical expertise for PEA and PSIA. Once trust funding ceased, the World Bank’s use of these frameworks declined markedly, and communities of practice became less active. Independent Evaluation Group evaluations found partial use of political, governance, and fragility analyses in Systematic Country Diagnostics, Country Partnership Frameworks, Poverty Reduction Strategy Papers, and development policy financing (DPF) and policy dialogue. For example, PSIA was included in some DPFs but not 30 Chapter 4 Discordances with the Business Model consistently integrated into all. Strong PEA contributed to the success of DPF implementation, and lack of PEA led to weak implementation. Challenges faced by earlier frameworks are similar to those faced by social contracts. Staff need adequate time, resources, and expertise. Optimal use of studies to support reform may require making clients the change agents, involving local actors in conducting the research, and integrating findings upstream in clients’ processes rather than treating studies as research outputs. All of this is hard for a task-oriented organization such as the World Bank. Moreover, even as high-quality analytical studies unpacked the political and governance factors that inhibited development, the World Bank sometimes had limited ability to act on these factors. Political elites were resistant to change, and the World Bank sought to maintain good relationships with government clients, to implement its programs. Real or perceived political sensitivities sometimes limited the distribution of reports that client governments perceived as sensitive. Some studies were not published because their findings were controversial. Source: Independent Evaluation Group. 31 5. Conclusions and Lessons 5.1 This evaluation finds that social contract diagnostics are useful analytical innovations with relevant operational implications. This is especially true in countries where social contracts are in transition. However, the World Bank’s involvement in social contract renewal carries efficacy and reputational risks and poses some operational challenges because it is at odds with some aspects of the World Bank’s business model. To overcome these operational challenges, the World Bank must think critically about the conditions under which a social contract diagnostic can effectively guide its country engagements. This would also require more focus on citizen engagement, building partnerships, and building staff’s social contract capabilities. Social Contract Diagnostics 5.2 The World Bank’s use of social contract diagnostics has provided insight into country clients’ development challenges. A social contract diagnostic is a single approach that can identify the causes of seemingly disparate development challenges. This approach goes beyond isolated diagnostics on fragility, citizen engagement, and institutions and governance. Well-executed social contract diagnostics have explained past World Bank implementation failures and explained complex phenomena, such as the Arab Spring, using perception-based data. 5.3 The World Bank is missing opportunities by not sufficiently disseminating social contract diagnostics to stakeholders. When disseminated, social contract diagnostics are well received by a range of counterparts and other stakeholders and can improve the World Bank’s reputation. Diagnostics also help the World Bank convene policy makers, civil society, and country-level development partners by sharing a common understanding of the root causes of client countries’ challenges and shedding light on innovative solutions. However, the World Bank rarely disseminates the social contract diagnostics carried out for SCDs nor ASAs that focus on social contract issues. The World Bank therefore misses opportunities to demonstrate its deep understanding of the societies in which it operates and build its credibility and reputation among potential reform partners. 5.4 The World Bank often lacks a deep understanding of social contract dynamics and its context-dependent role in their transformation. There are instances of the World Bank mischaracterizing social contracts in SCDs and strategic documents. Governance, fragility, and social development experts notably point out that the World Bank at times strips social contracts of their sociopolitical dimensions and reinterprets them as an economic concept. The World Bank has not sufficiently reflected on its legitimate role or the contextual conditions necessary to effectively help countries renew social contracts 32 Chapter 5 Conclusions and Lessons and has in some documents conveyed a certain hubris by assuming that it can renew countries’ social contracts. Operational Implications and Challenges 5.5 The World Bank’s use of social contract diagnostics to inform development carries the potential for transformative change, but also risks. In several countries already renegotiating their social contracts, the World Bank uses social contract diagnostics to inform its portfolio composition, subnational targeting, and advocate for essential yet unpopular reforms. However, social contract transitions pose significant risks to the World Bank’s operations. Specifically, it is difficult for the World Bank to maintain its perceived technical neutrality while progressing on key social contract reforms among unstable political coalitions. Additionally, the social transition could fail despite the World Bank’s best efforts. 5.6 The World Bank’s ability to operationalize its social contract diagnostics is limited in countries where there is no social contract transition under way. Despite high- quality diagnostics, the World Bank’s ability to coherently translate its social contract diagnostics into lending operations in status quo countries is limited. In these countries, the World Bank often focuses on improving service delivery to help the government meet citizens’ expectations, without a clear strategy to safeguard against reinforcing or prolonging flawed social contracts through its lending. If the World Bank wishes to influence social contracts in countries not undergoing social contract transitions, it tends to rely on nonlending activities, trust-funded work, analytical work, or citizen engagement commitments to influence social contracts at the margins. However, this approach is often not strategically orchestrated and does not constitute the radical change from “business as usual” called for in social contract diagnostics. 5.7 To conclude, this evaluation exposes incongruities between the World Bank’s social contract ambitions and key features of the World Bank’s business model that constrain the implementation of its approach to social contracts. The World Bank is both a “mission-driven” and a “client-driven” institution and, as such, must deal with several built-in contradictions that a social contract framing lays bare. These incongruities can become salient when there are contradictory demands from different parts of the World Bank’s authorizing environment, or when the World Bank’s research and analytics reveal reform pathways that client governments refuse to take. Thus, it is worth reflecting on how the World Bank can overcome these incongruities in relation to social contract diagnostics, and beyond. 33 Chapter 5 Conclusions and Lessons Overcoming Operational Challenges 5.8 To overcome these incongruities and derive more value from its social contract diagnostics, the World Bank may want to better account for client countries’ social contract dynamics, define its role according to these dynamics, and reflect on “how” it engages with clients, citizens, and partners to catalyze change. This evaluation draws lessons from the World Bank’s experience with translating social contract diagnostics into its country engagements, and from past and other donors’ experience with similar governance frameworks. The following lessons emerged from the analysis. 5.9 The World Bank may want to refine its understanding of social contracts and be more selective in when and how to undertake social contract diagnostics. To this end, World Bank teams in charge of research, such as Regional Chief Economists Offices and the Development Economics Vice Presidency, may want to further invest in understanding social contracts: how they change, how they vary within countries, what actors and institutional mechanisms contribute to social contract transitions, how institutions impact social contract bargaining, and how social contracts relate to various development outcomes, such as shared prosperity, economic performance, and institutional stability. For example, the World Bank could build on its research on the importance of social contracts for development in Africa. Currently, decisions to undertake social contract diagnostics are ad hoc. Therefore, the Regional Vice Presidency Units and CMUs could be more purposeful in deciding when and how to undertake social contract diagnostics. This evaluation’s evidence suggests that social contract diagnostics are particularly desirable in countries undergoing social contract transitions, or where broken social contracts undermine World Bank operations. 5.10 Defining the World Bank’s role in social contract renewal, specific to each country’s social contract dynamics, can improve the World Bank’s coherence. Clearly articulating the factors that contribute to legitimate and transformative social contract interventions should guide World Bank’s actions. This evaluation identifies three factors that contribute to World Bank success in applying a social contract framing to its development approach. These factors include the following: • The presence or absence within the country of a social contract renegotiation. • The nature of the country’s social contract bargaining. This includes whether the social contract benefits the few or the many, whether it relies on corruption and clientelism, whether citizens and civil society can participate in the bargaining, whether the state is responsive to citizens’ social contract demands, and whether the state has the capacity and legitimacy to fulfill these demands. 34 Chapter 5 Conclusions and Lessons • The World Bank’s comparative advantages depend on the local contextual conditions, such as the World Bank’s positioning compared with other development partners on social contract issues and its reputation and social capital among citizens and civil society. 5.11 In countries undergoing social contract transitions, the World Bank could contribute more effectively to social contract renewal in four ways. First, CMUs should ensure that social contract diagnostics are well executed, and that key state and nonstate actors are engaged on the diagnostics. This builds social capital with stakeholders that are party to the social contract, including CSOs, trade unions, political parties, private sector actors, farmers’ organizations, and business associations. Engaging state actors in the legislative and judiciary branches, as well as subnational and local authorities, would also be beneficial. Second, the World Bank CMUs could follow diagnostics with sector- or reform-specific PEAs. Third, World Bank senior management and Board should ensure coherence between its portfolio of operations and its social contract diagnostics, including introducing some flexibility with some aspects of the business model that preclude effectively delivering on social contract transformation. This may include flexibility on the length of loans and CPF periods, performance targets, and engagement processes. The Board and senior management have already applied these types of exceptions to conventional supervision and the traditional business model when engaging in fragile states. For example, in the Republic of Yemen the International Development Association portfolio was innovatively reprogramed to pivot from a conventional peace-time portfolio to an adaptive crisis response. Fourth, GPs should build staff capacity on sectoral, governance, social development, and citizen engagement expertise. 5.12 In countries where the social contract is static, the World Bank could use its social contract diagnostic more effectively to guide its country engagement in three ways. First, teams in charge of drafting CPFs may want to adjust their stated goal of helping countries renew their social contracts to something more precise and achievable. Second, the World Bank management could develop an approach to safeguard against lending that prolongs or exacerbates broken social contracts. Third, outside of lending, World Bank GPs jointly with CMUs can more proactively and deliberately exploit its other channels of influence— such as convening power, citizen engagement, and research and analytics—to address social contract issues. 5.13 In both types of countries, the World Bank should focus on “how” it delivers its country programs—more specifically, how it engages citizens, how it forges partnerships, and how it mobilizes the right skill sets. 35 Chapter 5 Conclusions and Lessons 5.14 Building social capital with citizens beyond the citizen engagement mechanisms embedded in projects can help mobilize support for reforms. This is consistent with IEG’s recommendations from its citizen engagement evaluation (World Bank 2018b). Such an effort would entail broadening citizen engagement beyond projects to the country level. For all stakeholders interviewed, the World Bank has potential to engage more extensively with society through innovative and continuous communication. For example, the World Bank could build on existing good practice of setting up country-level World Bank–CSO committees to engage regularly throughout the World Bank country engagement cycle, monitor specific projects or commitments, share results of specific analysis, and keep abreast of specific reform efforts. The World Bank could also consider more systematic engagements with state actors outside the central government and executive branch. In many countries service delivery is undertaken at the subnational level, and the judiciary and legislature have outsized roles to play. This could help mobilize a critical mass of support for reforms. The World Bank could also consider using its data, research, and analytics more systematically to empower citizens to bargain for social contracts; this would start by broadly disseminating its social contract diagnostics. 5.15 Better relying on partnerships can help achieve civic mobilization and political transformation. To do so, World Bank CMUs could convene broad country-level coalitions beyond government actors and across election cycles. This is necessary because the World Bank does not have a mandate nor a comparative advantage to manage the political dimensions of social contract renewals. 5.16 Better mobilizing staff specialized in social contract issues and strong leadership by country directors can help improve the World Bank’s capacity to renew social contracts. All interviewees, including at the vice president level, observed that social contract renewal requires specialized skills in social development and governance and strong leadership from Country Managers and Directors. They play a pivotal role both in overcoming institutional fragmentation and in negotiating with governments on sensitive reforms that typify social contract renewal. The evaluation also found that when World Bank staff in charge of carrying out social contract diagnostics are also involved in shaping country programs and portfolios, this helps improve coherence. 36 Bibliography Acemoglu, D., and J. Robinson. 2008. The Role of Institutions in Growth and Development. Washington, DC: World Bank. ———. 2013. Economics versus Politics: Pitfalls of Policy Advice. Journal of Economic Perspectives 27 (2): 173–92. Addison, Tony, and Syed Mansoob Murshed. 2001. From Conflict to Reconstruction: Reviving the Social Contract. Discussion Paper 2001/048, UNU-WIDER, Helsinki. Alesina, Alberto F., Guido Cozzi, and Noemi Mantovan. 2012. The Evolution of Ideology, Fairness and Redistribution. NBER Working Paper No. 15587 (December), National Bureau of Economic Research, Cambridge, MA. https://onlinelibrary.wiley.com/doi/full/10.1111/roiw.12388 Alston, Lee J., Marcus Andre Melo, Bernardo Mueller, and Carlos Pereira. 2013. Changing Social Contracts: Beliefs and Dissipative Inclusion in Brazil.” Journal of Comparative Economics 41 (1): 48–65. Azam, Jean-Paul, and Alice Mesnard. 2003. Civil War and the Social Contract. Public Choice 115 (3): 455–75. Andrews, M., L. Pritchett, and M. Woolcock. 2017. Building State Capability. Evidence, Analysis, Action. Oxford: Oxford University Press. Atkinson, A. B., T. Piketty, and E. Saez. 2009. Top Incomes in the Long Run of History in Top Incomes: A Global Perspective. Oxford: Oxford University Press. Benabou, R. 2000. Unequal Societies: Income Distribution and the Social Contract. American Economic Review 90 (1): 96–129. Birdsall, N. 2008. The Development Agenda as a Global Social Contract; or, We Are All in This Development Boat Together. Center for Global Development. Lecture presented at the Dutch Scientific Council (WRR), The Hague, December 8. Bussolo, Maurizio; Davalos, Maria E.; Peragine, Vito; Sundaram, Ramya. 2018. Toward a New Social Contract: Taking on Distributional Tensions in Europe and Central Asia . Europe and Central Asia Studies; Washington, DC: World Bank. https://openknowledge.worldbank.org/handle/10986/30393 Capoccia, Giovanni, and Daniel R. Kelemen. 2007. The Study of Critical Junctures: Theory, Narrative and Counterfactuals in Institutional Analysis. World Politics 59 (3): 341–36. Carothers, T. 2002. The end of the transition paradigm. Journal of Democracy 13 (1): 5–21. Cherrier, C. 2015. Examining the Catalytic Effect of Aid on Domestic Resource Mobilization for Social Transfers in Low-Income Countries. UNRISD Working Paper 2015-3, United Nations Research Institute for Social Development, Geneva. 37 Bibliography Clingendael Institute for the Netherlands Ministry of Foreign Affairs. 2008. Framework for Strategic Governance and Corruption Analysis (SGAGA). The Hague: Clingendael Institute. Cogneau, D. 2012. The Political Dimensions of Inequality During Economic Development. Region et Developpement (35): 12–36. Cogneau, D., and J. D. Naudet. 2007. Who Deserves Aid? Equality of Opportunity, International Aid, and Poverty Reduction. World Development 35 (1): 104–20. Cotlear, D. 2006. A New Social Contract for Peru: An Agenda for Improving Education, Health Care, and the Social Safety Net. World Bank Country Study. Washington, DC: World Bank. De Waal, A. 1996. Social Contract and Deterring Famine: First Thoughts. Disasters 20 (3): 194–205. https://doi.org/10.1111/j.1467-7717.1996.tb01033.x. Devarajan, S. 2016. An Exposition of the New Strategy “Promoting Peace and Stability in the Middle East and North Africa.” Washington, DC: World Bank. Devarajan, S., and S. Khemani. 2018. If Politics Is the Problem, How Can External Actors Be Part of the Solution? In Institutions, Governance and the Control of Corruption, edited by K. Basu and T. Cordella, 209–51. International Economic Association Series. Cham, Switzerland: Palgrave Macmillan. DfID (Department for International Development). 2005. Briefing Using Drivers of Change to Improve Aid Effectiveness. A DFID Practice Paper, London, DfID. ———. 2005. How to Note: Lessons learned – Planning and Undertaking a Drivers of Change Study. A DFID Practice Paper, London, DfID. ———. 2011. The Politics of Poverty: Elites, Citizens and States: Findings from Ten Years of DfID Funded Research on Governance and Fragile States 2001–2010. London: DfID. http://www.dfid.gov.uk/Documents/publications1/evaluation/plcy-pltcs-dfid-rsch-synth- ppr.pdf. Duncan, A., and G. Williams. 2012. Making Development Assistance More Effective Through Using Political-Economy Analysis: What Has Been Done and What Have We Learned ? Development Policy Review 30 (2): 133–48. Economic Commission for Latin America and the Caribbean. 2014. Compacts for Equality: Towards a Sustainable Future. 35th session of ECLAC: https://repositorio.cepal.org/bitstream/handle/11362/36693/LCG2586SES353e_en.pdf Epstein, G. S., and I. N. Gang. 2016. Taxation, Social Protection, and Governance Decentralization. WIDER Working Paper 2016/101, UNU-WIDER, Helsinki. Ferreira, Francisco H. G., Julian Messina, Jamele Rigolini, Luis-Felipe López-Calva, Maria Ana Lugo, and Renos Vakis. 2013. Economic Mobility and the Rise of the Latin American Middle Class. Washington, DC: World Bank. doi: 10.1596/978-0-8213-9634-6 38 Bibliography Fisher, Jonathan, and Heather Marquette. 2013. Donors Doing Political Economy Analysis: From Process to Product (and Back Again?). ISA Annual Convention panel on “Politicising or Depoliticising Aid? The Political Economy of Political Economy Analysis,” San Francisco, 3–6 April. Fleury, S. 2017. The Welfare State in Latin America: Reform, Innovation and Fatigue. Reports in Public Health 33 (July). Franzoni, J. M. 2008. Welfare Regimes in Latin America: Capturing Constellations of Markets, Families, and Policies. Latin American Politics and Society 50 (2): 67–100. Freund, Caroline, Antonio Nucifora, and Bob Rijkers. 2014. All in the Family: State Capture in Tunisia (English). Policy Research Working Paper WPS 6810, World Bank Group, Washington, DC. Handoussa, H. 2008. The Egypt Human Development Report 2008. Egypt’s Social Contract: The Role of Civil Society. Cairo: United Nations Development Programme. Herrera, J., M. Razafindrakoto, and F. Roubaud. 2007. Governance, Democracy and Poverty Reduction: Lessons Drawn from Household Surveys in Sub-Saharan Africa and Latin America. International Statistical Review 75 (1): 70–95. https://www.jstor.org/stable/41508450. Hickey, S. 2011. The Politics of Social Protection in Africa: What Do We Get from a ‘Social Contract’ Approach? Working Paper, European Report on Development, European University Institute Report, Paris. https://ec.europa.eu/europeaid/sites/devco/files/erd-socpro- poverty-hickey-20100617_en.pdf. Ho, S. 2018. Social Contracts and Public Goods Provision. Foreign Affairs, May 23. https://www.foreignaffairs.com/sponsored/social-contracts-and-public-goods-provision. Ianchovichina, Elena. 2018. Eruptions of Popular Anger: The Economics of the Arab Spring and Its Aftermath. MENA Development Report. Washington, DC: World Bank. Isakjee, A. 2017. Welfare State Regimes: A Literature Review. IRIS Working Paper Series 18, University of Birmingham, Birmingham. Jones, S., and F. Tarp. 2016. Does Foreign Aid Harm Political Institutions? Journal of Development Economics 118: 266–81. Keefer, Philip. 2011. Collective Action, Political Parties, and Pro-Development Public Policy. Asian Development Review 28: 94–118. Kouakou Kouame, W. A. 2018. Contrat social et mobilisation des ressources domestiques dans les pays en developpment. (PhD dissertation, Université de Sherbrooke). https://savoirs.usherbrooke.ca/bitstream/handle/11143/12748/Kouame_Wilfried_Anicet_ Kouakou_PhD_2018.pdf?sequence=1&isAllowed=y. Lavers, T., and S. Hickey. 2015. Investigating the Political Economy of Social Protection Expansion in Africa: At the Intersection of Transnational Ideas and Domestic Politics. Effective States and 39 Bibliography Inclusive Development Working Paper 47, Effective States and Inclusive Development Research Centre, Manchester. Mahoney, James. 2000. Path Dependence in Historical Sociology. Theory and Society 29 (4): 507–48. McCandless, Erin. 2018. Reconceptualizing the Social Contract in Contexts of Conflict, Fragility and Fraught Transition. Wits School of Governance Working Paper Series, University of the Witwatersrand, Johannesburg. Menocal, Alina Rocha, Marc Cassidy, Sarah Swift, David Jacobstein, Corinne Rothblum, and Ilona Tservil. 2018. Thinking and Working Politically Through Applied Political Economy Analysis: A Guide for Practitioners. Center of Excellence on Democracy, Human Rights and Governance. Washington, DC: USAID. https://usaidlearninglab.org/sites/default/files/resource/files/pea_guide_final.pdf Mkandawire, T. 2016. Colonial Legacies and Social Welfare Regimes in Africa. An Empirical Exercise. Working Paper 2016-4, United Nations Research Institute for Social Development, Geneva. Mouhoud, El M. 2012. Political Economy of Arab Revolutions: Analysis and Prospects for North-African Countries. Mondes en développement 2: 35–50. https://doi.org/10.1111/j.1467- 7717.1996.tb01033.x. North, Douglass C., John Joseph Wallis, Steven B. Webb, and Barry R. Weingast. 2007. Limited Access Orders in the Developing World: A New Approach to the Problems of Development. Policy Research Working Paper 4359, World Bank, Washington, DC. North, D., J. Wallis, and B. Weingast. 2009. Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History. New York: Cambridge University Press. Nugent, P. 2010. States and Social Contracts in Africa. New Left Review 63: 35–68. OECD (Organisation for Economic Co-operation and Development). 2009. Concepts and Dilemmas of State Building in Fragile Situations: From Fragility to Resilience. OECD Journal on Development 9 (3): 61–148. https://doi.org/10.1787/journal_dev-v9-art27-en. ———. 2010. Do No Harm: International Support for Statebuilding. OECD Conflict and Fragility Paper. Paris: OECD. OECD/CAF/UN ECLAC. 2018. Latin American Economic Outlook 2018: Rethinking Institutions for Development, OECD Publishing, Paris, https://doi.org/10.1787/leo-2018-en. Ofreneo, R. E. 2017. World Bank Bats for ‘New Social Contract’ for East Asia. Business Mirror, December 13. https://businessmirror.com.ph/2017/12/13/wb-bats-for-new-social-contract- for-east-asia/ Rodrik, D. 2006. Goodbye Washington Consensus, Hello Washington Confusion? A Review of the World Bank’s Economic Growth in the 1990s: Learning from a Decade of Reform. Journal of Economic Literature 44 (December): 973–87. 40 Bibliography Rousseau, Jean-Jacques. 1762. Du Contrat Social, Ou Principes du droit politique. SIDA (Swedish International Development Cooperation Agency). 2005. Methods of Analyzing Power: A Workshop Report. Stockholm: SIDA. ———. 2006. Power Analysis: Experiences and Challenges. Stockholm: SIDA. ———. 2013. Power Analysis: A Practical Guide. Stockholm: SIDA. Sanchez, Susana M., Kinnon Scott, and J. Humberto Lopez. 2016. Guatemala: Closing Gaps to Generate More Inclusive Growth. Systematic Country Diagnostic. Washington, DC: World Bank. UNDP (United Nations Development Programme). 2012. Governance Measurements for Conflict and Fragility: A Comparative Inventory. Oslo: UNDP. ———. 2016. Social Contract in Situations of Conflict and Fragility. Oslo: UNDP. ———. 2018. Forging Resilient Social Contracts. Oslo: UNDP. USAID. 2018. Thinking and Working Politically through Applied Political Economy Analysis: A Guide for Practitioners. Washington, DC: USAID https://usaidlearninglab.org/sites/default/files/resource/files/pea_guide_final.pdf Verme, P., and A. Araar, eds. 2017. The Quest for Subsidy Reforms in the Middle East and North Africa Region: A Microsimulation Approach to Policy Making. Natural Resource Management and Policy Series. Cham, Switzerland: Springer International. Weaver, Catherine. 2008. Hypocrisy Trap: The World Bank and Poverty Reform. Princeton, NJ: Princeton University Press. World Bank. 2000. Reforming Public Institutions and Strengthening Governance. A World Bank Strategy. Public Sector Group. World Bank, Washington, DC. ———. 2002. Institutional and Governance Reviews: A New Type of Economic and Sector Work. PREM Notes Number 75, World Bank, Washington, DC. ———. 2003. A User’s Guide to Poverty and Social Impact Analysis. Washington, DC: World Bank, Poverty Reduction Group (PRMPR) and Social Development Department (SDV). ———. 2004. The Poverty Reduction Strategy Initiative. An Independent Evaluation of the World Bank’s Support through 2003. Washington, DC: World Bank, World Bank Operations Evaluation Department. ———. 2005. The Poverty Reduction Strategy Initiative. Findings from 10 Country Case Studies of the World Bank and IMF Support. Washington, DC: World Bank, World Bank Operations Evaluation Department and IMF Independent Evaluation Office. ———. 2006a. Poverty and Social Impact Analysis of Reforms: Lessons and Examples from Implementation. Washington, DC: World Bank. 41 Bibliography ———. 2006b. Unequal citizens: gender, caste and ethnic exclusion in Nepal: Summary (English). Washington, DC: World Bank. http://documents.worldbank.org/curated/en/201971468061735968/Summary ———. 2008a. Good Practice Note: Using Poverty and Social Impact Analysis to Support Development Policy Operations. Washington, DC: World Bank. ———. 2008b. The Political Economy of Policy Reform: Issues and Implications for Policy Dialogue and Development Operations. Washington, DC: World Bank. ———. 2009a. Assessing the Use of Poverty and Social Impact Analysis in World Bank Development Policy Loans. Washington, DC: World Bank. http://www.worldbank.org/en/topic/poverty/brief/poverty-and-social-impact-analysis- psia ———. 2009b. Development Policy Operations. Washington, DC: World Bank. ———. 2009c. Poverty and Social Impact Analysis (PSIA): Reviewing the Link with In-Country Policy and Planning Process Synthesis Report. Washington, DC: World Bank. ———. 2010a. Analyzing the Effects of Policy Reform on the Poor: An Evaluation of the Effectiveness of the World Bank Support to Poverty and Social Impact Analysis. Independent Evaluation Group. Washington, DC: World Bank. ———. 2010b. An Evaluation of the Effectiveness of World Bank Support to Poverty and Social Impact Analyses. Independent Evaluation Group. Washington, DC: World Bank. ———. 2010c. Governance and Anti-Corruption Diagnostics: Guidance and Tools for Implementation, Monitoring, and Assessment in the Field. Washington, DC: World Bank. ———. 2010d. The Political Economy of Reform: Moving from Analysis to Action, A Global Learning Event. Final Report. Washington, DC: World Bank. ———. 2011a. An Evaluation of Political-Economic Analysis in Support of the World Bank’s Governance and Anticorruption Strategy. IEG Working Paper 2011/4, World Bank, Washington, DC. ———. 2011b. Political Economy Impact Stories. Washington, DC: World Bank. ———. 2011c. World Development Report. Washington, DC: World Bank. https://siteresources.worldbank.org/INTWDRS/Resources/WDR2011_Full_Text.pdf ———. 2014a. Problem-Driven Political Economy Analysis: The World Bank’s Experience. Washington, DC: World Bank. ———. 2014b. Strategic Framework for Mainstreaming Citizen Engagement in World Bank Group Operations: Washington, DC: World Bank. http://documents.worldbank.org/curated/en/266371468124780089/pdf/929570WP0Box380 ategicFrameworkforCE.pdf. 42 Bibliography ———. 2015a. MENA Economic Monitor: Inequality, Uprising and Conflict in the Arab World . Washington, DC: World Bank. ———. 2015b. The Poverty Focus of Country Programs. Washington, DC: World Bank. https://ieg.worldbankgroup.org/sites/default/files/Data/Evaluation/files/poverty_focus_c p.pdf ———. 2015c. Poverty and Social Impact (PSIA). Washington, DC: World Bank. http://www.worldbank.org/en/topic/poverty/brief/poverty-and-social-impact-analysis- psia ———. 2016a. Making Politics Work for Development: Harnessing Transparency and Citizen Engagement. Policy Research Report. Washington, DC: World Bank. ———. 2016b. Poverty and Social Impact Analysis (PSIA) Final Report . Multi-Donor Trust Fund Annual Report. Washington, DC: World Bank. ———. 2016c. The Role of Political Economy Analysis in Development Policy Operations . Independent Evaluation Group Learning Product. Washington, DC: World Bank. ———. 2017a. An Early Stage Assessment of the Systematic Country Diagnostics and Country Partnership Framework Process and Implementation. Independent Evaluation Group. Washington, DC: World Bank. ———. 2017b. Fragility Assessments and Risk and Resilience Assessments: Review of Progress and Lessons Learned (2012–2016). Washington, DC: World Bank. ———. 2017c. World Bank Group Country Engagement. Washington, DC: World Bank. https://ieg.worldbankgroup.org/sites/default/files/Data/Evaluation/files/scd-cpf.pdf. ———. 2018a. Citizen Engagement. http://www.worldbank.org/en/about/what-we- do/brief/citizen-engagement. ———. 2018b. Engaging Citizens for Better Development Results. Independent Evaluation Group. Washington, DC: World Bank. ———. 2018c. Riding the Wave: An East Asian Miracle for the 21st Century. World Bank East Asia and Pacific Regional Report. Washington, DC: World Bank. ———. 2019b. World Development Report: The Changing Nature of Work. Washington, DC: World Bank. World Bank Group. 2014. The World Bank Group Goals: End Extreme Poverty and Promote Shared Prosperity. Washington, DC: World Bank. https://openknowledge.worldbank.org/handle/10986/20138 Yanguas, Pablo, and David Hulme. 2014. “ Can Aid Bureaucracies Think Politically? The Administrative Challenges of Political Economy Analysis (PEA) in DFID and the World Bank.” 43 Bibliography ESID Working Paper 33, Effective States and Inclusive Development Research Centre, Manchester. 44 Appendix A. Methodology Evaluation Questions The main question that motivated this evaluation is, “Under what circumstances is a social contract lens useful in shaping World Bank’s country engagements to achieve better development outcomes?” Underlying this overarching question were two main lines of inquiry: (i) How does the World Bank understand, conceptualize, and diagnose disequilibria in social contracts, and how has it articulated its role in helping countries reshape them? (ii) How coherent is the link between World Bank country diagnostics of social contracts and its country engagement? The first question dealt with the relevance of the conceptualization and articulation of the World Bank’s role, across countries, based on relevant endogenous and exogenous factors that help define the nature and dynamic of social contracts. The second question addressed the consistency of applying a social contract lens within countries, between diagnostic and practice. In answering this question, the evaluation sought to identify the operational risks and challenges that translating social contract diagnostics represent and to draw lessons for overcoming these challenges. These questions are articulated into subquestions—see box A.1. Box A.1. Evaluation Questions Overarching evaluation question: Under what circumstances is a social contract lens useful in shaping World Bank country engagements to achieve better development outcomes? i. How does the World Bank understand, conceptualize and diagnose social contracts, and how has it articulated its role in helping countries reshape them? a) How does the World Bank understand and diagnose disequilibria in social contracts and how does it explain social contracts’ influence on development outcomes? b) How consistent are the World Bank’s understanding and diagnostics of social contracts with the literature on the topic? c) How has the World Bank articulated its role in helping countries reshape social contracts and to what extent has a social contract framework facilitated the flexibility and adaptability of the World Bank’s approach in cases of shifting social bargains? ii. How coherent is the link between World Bank country diagnostics of social contracts and its country engagement? a) In countries where the World Bank has identified imbalances in social contracts through country diagnostics, to what extent and how has this diagnostic shaped its country engagement? What is being done differently and why? b) To what extent and how has the first pillar of the Middle East and North Africa strategy been translated into country programs, lending, advisory work, and policy dialogues in the Region? 45 Appendix A Methodology iii. For the World Bank, what are the advantages, risks, and limitations of using social contracts to frame countries’ development challenges and opportunities for better development outcomes? Source: Independent Evaluation Group. Evaluation Components The evaluation methodology was made up of five components. These are presented in table A.1 and described in detail in this section. The evaluation matrix at the end of this appendix summarizes the structure of the evaluation design (see table A.3). Table A.1. Evaluation Components Evaluation Components Description Literature reviews A review of the World Bank’s research, analytical, and diagnostic work on social contracts was conducted. It was then put in dialogue with external academic and policy literature on social contracts, with a specific emphasis on Latin America and the Caribbean and the Middle East and North Africa. Coherence Analysis A structured review of Systematic Country Diagnostics, Country Program Frameworks, or Country Engagement Notes, lending, and Advisory Services and Analytics was conducted for 33 countries that had an explicit social contract diagnostic or were covered by the Middle East and North Africa strategy. Case studies Five in-depth country case studies were conducted on Guatemala, Nepal, Niger, Tunisia, and Uzbekistan. Stocktaking of governance and social A review of the World Bank’s use of six governance and social diagnostics analytical frameworks and of four other donors’ frameworks was conducted to draw lessons from past and comparative experiences. Key informant interviews Interviews with 163 key informants were conducted, including World Bank staff and management, external experts on social contracts, client governments, civil society organizations, and citizens. Source: Independent Evaluation Group. Literature Reviews The evaluation team undertook two types of literature review to answer questions 1a, 1b, and 1c: • The first review focused on the World Bank’s knowledge work on social contracts, including regional and global research, Systematic Country Diagnostics (SCDs), and Advisory Services and Analytics (ASA). Keyword searches were conducted in ImageBank and operation portals. A screening of 46 Appendix A Methodology abstract with inclusion and exclusion criteria was applied, leading to a core set of research and diagnostic documents. This core set was enriched through interviews with World Bank researchers and a limited snowballing from the reference list. A full-text review of 42 papers was then conducted to identify how the World Bank defines, measures, and understands social contracts, and whether and how it articulates its roles in shaping them (questions 1a and 1c). • The second review was meant to put the World Bank’s literature in dialogue with the outside literature to answer question 1b specifically. The team partnered with the World Bank Library services to conduct an efficient search. The review focused on development policy–relevant literature published in the past 15 years that used social contracts to understand country development challenges. Google Scholar, EconLit, and Scopus databases were used for the review. Twenty-two papers were identified as highly relevant and were read fully. In addition, Region-specific searches of literature on Latin America and the Caribbean and the Middle East and North Africa (MENA) were conducted to trace the World Bank’s research on social contract influence. The Web search was completed with interviews with 10 experts on social contracts in either Region. Given the ongoing work on social contracts in Africa, the evaluation team partnered with the World Bank team currently working on a flagship regional report in Africa and kept abreast of their conceptual development. The evaluation team participated in two seminars organized with academic experts of social contracts in Africa. Coherence Analysis A detailed description of the methodology for this analysis is included in appendix D. Sampling and Selection Countries included in this review were those with an SCD that mentioned the keywords “social contract” and/or “social compact” at least five times. An exception was countries in the MENA Region, which were included even if they did not have a unique SCD (or if the SCD did not mention the keywords). For these 12 countries, the 2015 MENA regional strategy was used as the diagnostic document because of the regional strategy pillar 1, which aimed to renew social contracts. 47 Appendix A Methodology Using this methodology, 33 countries1 were identified as in scope, and SCDs (or MENA regional strategy) were reviewed as a “starting point” for relevant social contract information. This included how the SCDs diagnose key elements of the current social contract, and how they articulate those areas in need of reform. SCDs served as the baseline because they are the instrument used by the World Bank Group to identify, through in-depth analysis and stakeholder consultations, the most important challenges and opportunities faced by a country. An SCD informs each new Country Partnership Framework (CPF). Data Analysis The team subsequently reviewed the corresponding CPF or Country Engagement Note (CEN)—the key strategy document guiding the Bank Group’s support to a member country. The CPF/CEN pillars were reviewed and extracted to facilitate the coherence analysis between the challenges diagnosed (in SCDs) and the reform areas formally agreed on (in CPFs/CENs). The lending and nonlending (ASA) portfolios approved since the SCD were downloaded from Business Intelligence to further track coherence—namely, the consistency between the CPF/CEN pillars and financed activities. Finally, to assess the previous programmatic engagement within these 33 countries, the Independent Evaluation Group (IEG) looked at IEG-validated Completion and Learning Reviews for prior engagements relevant to the social contract diagnostic elements. The Completion and Learning Review would have to have covered the period immediately preceding the SCD. When a Completion and Learning Review was not available, a World Bank Performance and Learning Review for the preceding period was reviewed instead. A coding template was developed and administered in Microsoft Access to consistently extract relevant data from the documents reviewed. Data analysis was carried out in Microsoft Excel by thematically coding the data exported from the Access database, and graphically presenting various parts of the analysis. 1Algeria; Bahrain; Belarus; Bulgaria; Chile; Costa Rica; Djibouti; the Arab Republic of Egypt; Guatemala; Guinea-Bissau; Haiti; Iran; Iraq; Jordan; Kuwait; Lebanon; Libya; Mali; Mauritania; Mauritius; Morocco; Oman; Paraguay; Qatar; the Russian Federation; Saudi Arabia; Sierra Leone; South Africa; Tajikistan; Tunisia; United Arab Emirates; West Bank and Gaza; the Republic of Yemen. 48 Appendix A Methodology Case Studies The methodological design relied heavily on case study evidence to generate lessons from experience. Case studies were necessary evaluation components for three reasons. First, the complexity and systemic nature of the evaluand required a granular understanding of the context in which the World Bank’s country engagement is embedded and a good understanding of the social contract dynamic. Case studies are the preferred method to gain this level of empirical depth. Second, the formative nature of the evaluation lent itself particularly well to a case-based design, where experience in various contexts could be understood in depth and contrasted across contexts. Third, the evaluation team needed to gather the perspectives of a wide range of stakeholders, on all sides of the social contract bargaining (see appendix B for conceptual framework) to appropriately study the World Bank’s work on social contract. Case studies are the preferred method to gather and synthesize a wide range of perspectives. Sampling and Selection The team chose the cases based on discussions with technical counterparts and other World Bank experts. In selecting the cases, the team looked for variability on: (i) whether the World Bank adopted an explicit social contract to guide its operation; (ii) whether the country’s social contract is in transition; and (iii) whether the case’s social contract issues differed from the other cases. This variance allowed the team to assess coherence across contexts and generate context-specific lessons. Table A.2 summarizes the case study selection. A few other cases were considered for the analysis but had to be abandoned because the evaluation mission’s timing and topic were deemed inconvenient by CMUs. These cases are Bulgaria, Myanmar, Nepal, and South Africa, which had to be conducted on a desk-based basis. Table A.2. Case Study Selection World Bank Social Contract Social Contract in Country Diagnostic Transition (Stage) Other Characteristics Guatemala Yes No Fragmented social contract Nepal No Yes (advanced) Transitioning out of fragility, ethnic fragmentation Niger Yes No Fragile state Tunisia Yes Yes (middle) Considered “success” of the Arab Spring Uzbekistan No Yes (early) Resource-rich and planning economy 49 Appendix A Methodology The selection of key informants to be interviewed during the mission was prepared carefully. The objective was to balance the number of interviews with the three groups of stakeholders that are party to the social contract bargain: (i) governments (at the central and local levels), (ii) citizens and civil society organizations, and (iii) nonstate actors with bargaining power (for example, donors, firms). To identify the relevant key informants the team consulted the country office but also relied on external sources, specifically researchers who had worked on each country’s social contract as well as members of civil society who are active on social contract issues. These were identified using platforms such as LinkedIn, or case leads’ own networks. This selection process was necessary to identify key informants that would provide an independent yet informed perspective on the World Bank’s work in the country. Data Collection and Analysis The team adopted a detailed case protocol to ensure data collection and analytical consistency across cases. Case authors were all experienced researchers with political economy, social development, or governance expertise and in-depth country knowledge. For each case study a minimum of 25 interviews (up to 40) were conducted. Nepal was an exception, as it was conducted as a desk-based review with five complementary interviews. The case narratives generated by each researcher were then discussed as a group to share and contrast the experiences and drawing within- and across-case lessons. Appendix C presents summaries of each case. Stocktaking of Governance and Social Diagnostic Frameworks The evaluation team conducted a stocktaking of the World Bank’s experience with six governance or social analytical and diagnostic frameworks and approaches and four frameworks used by other donors. The objective of this stocktaking was to capture key lessons from previous experience with operationalizing these types of frameworks into World Bank’s country engagement. The review was guided by the following questions: • What are the strengths and weaknesses of past approaches? • To what extent did the approaches lead to significant changes in World Bank’s country engagement, policy dialogue, and programming? • What challenges did they face? • Why have they fallen out of practice? • How could a social contract framework eschew or move beyond these challenges? 50 Appendix A Methodology The following frameworks, analytical lenses, or diagnostic tools adopted by the World Bank were analyzed as part of the review: political economy analysis (PEA), problem- driven economic analysis, governance and accountability diagnostic (GAC), poverty and social impact analysis, institutional and governance review, citizen engagement, and risk and resilience assessment (also known as fragility assessments). The following donor’s frameworks were also part of the review: the U.K. Department for International Development’s Drivers of Change, the Netherlands’ Strategic Corruption and Governance Analysis, the Swedish International Development Cooperation Agency’s Power of Analysis, and the United States Agency for International Development’s Political Economy Analysis. For a comprehensive overview of each of the models, refer to the United Nations Development Programme’s Governance Measurements for Conflict and Fragility (UNDP 2012). To conduct the review, the evaluation team gathered all the evaluative evidence available on each framework, including IEG reports, self-assessment reports, and reports conducted by third parties. In addition to the review of evaluative evidence, the evaluation team interviewed 10 experts who had been involved in the operationalization of the frameworks as well as members of the World Bank’s Political Economy Analysis Community of Practice. A detailed report was written to synthesize all the information. Appendix E provides an overview of this background paper. Key Informant Interviews Interviews were conducted to answer all evaluation questions. Specifically, the following groups of key informants were consulted: • Authors of Research, SCDs, and ASAs that are focused on social contracts • World Bank management and senior staff in Region’s Vice Presidential Units and in Country Management Units (CMUs): vice presidents, chief economists, country directors, and managers and directors, program leaders, country program coordinators, country economists, and development effectiveness offices • World Bank task team leaders and staff who worked on projects that are addressing contracts-related issues (Social Development; Governance; Social Protection; Poverty; Gender; and Fragility, Conflict, and Violence) 51 Appendix A Methodology • In the context of field-based case studies: academics, members of think tanks, civil society organizations, labor unions, citizens, and government representatives The total number of interviews conducted for this evaluation amounts to 163. Limitations The limitations of this evaluation revolve around two main sets of challenges related to the construct and external validity of the findings. • First, the evaluation’s most pressing challenge was to adequately capture a latent concept, whose definition is heterogeneous and contested, whose explicit use within the institution is emergent and whose past use is mostly implicit, and for which staff’s knowledge and experience is primarily tacit. These characteristics can threaten the construct validity of the findings. To mitigate this risk, the evaluation used a careful search and identification strategy applied to both its external literature review and to its review of internal research and analytics, relied on a large range of interviews with academics and internal key informants from various disciplines, and constructed an original conceptual framework that was discussed and validated by the technical counterparts and the team currently working on social contracts in Africa. • The second set of challenges stems from the idiosyncratic nature of the social contract, which poses threats to the external validity of the findings of the evaluation. Notwithstanding the diversity of social contracts, the World Bank’s regional studies and other literature have identified common traits that characterize the dynamics of social contracts, which is the primary contextual factor of interest for this evaluation. In addition, the objective of this evaluation is not to define social contracts at the country level, but rather to identify the World Bank’s framework to understand and take social contracts into account in its country engagement. To address the external validity challenge, the evaluation used two strategies: First, it reviewed the universe of SCDs that had an explicit social contract diagnostic to generate information across a wide range of contexts and systematically analyzed patterns emerging from these countries. Second, it selected case studies that could corroborate and refine these patterns and included cases where there was no formal social contract diagnostic, as an “out- of-sample” test to the findings. 52 Appendix A Methodology Table A.3. Evaluation Design Matrix Key Questions Data Collection and Analysis Methods Sampling and Selection Overarching evaluation question: Under what circumstances is a social contract lens useful in shaping World Bank country engagements to achieve better development outcomes? 1. How does the World Bank understand, conceptualize, and diagnose social contracts, and how has it articulated its role in helping countries reshape them? 1a. How does the World Bank understand and Structured literature review Keyword searches with inclusion/exclusion criteria diagnose social contracts and how does it for Advisory Services and Analytics, Systematic explain social contracts’ influence on Country Diagnostics (SCDs), Country Partnership development outcomes? Frameworks, project appraisal documents, World Bank global and regional research and strategies for FY13 to today 1b. How consistent are the World Bank’s Structured literature review Keyword search with inclusion/exclusion criteria for definitions and measurement of social Expert interviews public policy and international development– contracts with the literature? related literature 2008–18, both peer-reviewed journal and gray literature Purposive selection of experts based on literature review, with diverse regional and policy area expertise 1c. How has the World Bank articulated its role Structured document review See above: focus on strategic documents in helping countries reshape social contracts Key informant interviews Purposive selection of key informant within and to what extent has a social contract Regions, Global Practice; and Country Management framework facilitated the flexibility and Unit adaptability of the World Bank’s approach in cases of shifting social bargain? 53 Appendix A Methodology Key Questions Data Collection and Analysis Methods Sampling and Selection 2. How coherent is the link between World Bank country diagnostics of social contracts and its country engagement? 2a. In countries where the World Bank has (1) Coherence analysis (1) See above: focus on countries where the SCD diagnosed issues with social contracts through has diagnosed issues with social contracts country diagnostics, to what extent and how (2) Five case studies (interviews with World Bank has this diagnostic shaped its country staff, academics, governments, civil society (2) Purposive sample of cases where there has engagement? organizations, trade unions) been a clear diagnosis of issues with the social contract in the SCD. Maximizing variation by social contract issues diagnosed and country context 2b. To what extent and how has the first pillar (1) Coherence analysis with separate analysis for (1) FY15 to today: population of activities of the Middle East and North Africa (MENA) MENA strategy been translated into country (2) Purposive selection of key informants who have programs, portfolios, and policy dialogues in (2) Key informant interviews of past and current worked in MENA at some point since 2014 the region? MENA management and senior staff (year before strategy was made public) 2c. For the World Bank, what are the strengths (1) Key informant interviews (including within the (1) See above and limitations of using social contracts as a context of case studies—see above) framework to diagnose countries challenges (2) See above and opportunities for better development (2) Expert interviews outcomes? (3) See above (3) Structured literature review s 54 Appendix A Methodology Reference UNDP (United Nations Development Programme). 2012. Governance Measurements for Conflict and Fragility: A Comparative Inventory. Oslo: UNDP. 55 Appendix B. Conceptual Framework The objective of this conceptual framework was to guide the evaluation and capture the relevant elements at play in shaping social contracts in various types of countries to serve as a backbone for the evaluation design. This framework has been refined throughout the evaluation process. Definition Box B.1. Social Contract Definition The social contract refers to processes by which everyone in a political community, either explicitly or tacitly, consents to state authority, thereby limiting some of their freedoms, in exchange for the state’s protection of their rights and security and for the adequate provision of public goods and services. This agreement calls for individuals to comply with the state’s laws, rules, and practices in pursuit of broader common goals, such as security or protection, and basic services. The validity and legitimacy of a social contract may be gauged by the extent to which it creates and maintains an equilibrium between society’s expectations and obligations and those of state authorities and institutions, all amidst a context of constant flux. Source: UNDP 2016, 9. A minimalist definition of the social contract is offered by the OECD as a “dynamic agreement between state and society on their mutual roles and responsibilities” (OECD 2009, 17). Underpinning this agreement is a set of political economy dynamics that are well captured in a more specific definition proposed by the United Nations Development Programme (box B.1), which this evaluation adopts as its working definition. To understand how social contracts are shaped, how they change over time and how external development actors, such as the World Bank, can contribute to this change, a more elaborate conceptual framework is necessary. The evaluation team reviewed a rich literature, consulted with experts, and received feedback from several stakeholders within the World Bank to arrive at a conceptual framework represented in figure B.1. The framework is made up of four building blocks: (i) the main players that intervene in shaping social contracts (corners of the rectangle); (ii) the various types of influence or power that the players exert on each other in the bargaining process (sides of the rectangle); (iii) the nature of the institutional space where social contract bargaining takes place (center of the rectangle); (iv) and the ways in which each set of actors engages with existing social contracts at a given point in time. In many countries, social contracts are multiscalar—with subnational variations in the nature of state-society 56 Appendix B Conceptual Framework relationships. However, in line with the World Bank’s use of social contracts framing, this evaluation focuses only on national-level social contracts. Key Players and Their Relationships Social contracts are first and foremost shaped by a state-society interface, and the intersection between society’s expectations about what it should receive from the state, and what the state can legitimately demand from citizens in return. Underpinning the idea of social contracts is the norm of universalism. The extent to which social contracts remain in equilibrium and robust over time depends on the level of social cohesion. Within society, a tacit common agreement and norms must exist, for instance on taxes, redistribution, provision of public sector services, tolerance for informality and corruption, or trust in a given government. On the other hand, the stability of social contracts depends on the state’s capacity to hold its side of the bargain and deliver the level of protection and services that society expects. Social contracts define the boundaries of what are perceived as acceptable policy alternatives and are expressed in policy choices for specific policy systems—on security, social protection, labor markets, public goods and services, or transfer and taxes. A fundamental element of social contracts is the license given to the state by citizens to exert power and authority. In some countries, this authority is predatory, or exercised through coercion or self-interest and material rewards alone. In the long run, however, social contracts that are maintained only through predatory authority tend not to survive, as society’s loyalty to a predatory or extractive state is short lived, and citizens may instead decide to exit the social contract or voice demands for change, either peacefully or violently. Depending on the situation, exit strategies can include migration or giving up on public services for private ones, accompanied with a refusal to pay taxes or abide by the “rules of the game.” Thus, states’ legitimacy is key to the endurance of social contracts (DfiD 2011; UNDP 2016; McCandless 2018). While a state-society interface is at the core of social contracts, a conceptual framework that simplifies this bargaining as bilateral would be misconstrued. In all countries, other players—such as private firms or other market players, political and economic elites, nongovernmental organizations, civil society or religious organizations, as well as organized members of the diaspora—contribute to shaping social contracts. In some countries military and organized crime groups can be influential actors as well. The conceptual framework identifies four categories of actors with influence on the social contract. There are inherent limitations to this group, including that groups are not mutually exclusive, and that citizens are not a monolithic group. There can be different social contracts with various subgroups within the same country. 57 Appendix B Conceptual Framework Figure B.1. Conceptual Framework Source: Independent Evaluation Group. 58 Appendix B Conceptual Framework These nonstate actors can play a multiplicity of roles in shaping social contracts. In fragile countries, where the state is absent from specific regions or sectors, these actors can de facto replace the state and provide protection or basic services to the populace. Conversely, in some countries, external actors circumvent society’s role in domestic resource mobilization and its associated accountability function. These actors can also play a mediating role in social contracts bargaining, by representing the interests of groups of citizens or by determining the distribution of resources within society, including via market mechanisms. Some of these players (including firms) take advantage of their bargaining power to shape policies to their own narrow advantage— the so-called state capture. Firms or elites connected to the political system are able to obtain preferential treatments or barriers at entry for newcomers, at a significant cost for the economy and the well-being of the broader population (World Bank 2017, 33). This phenomenon was very common in the Middle East and North Africa and one of the causes of the collapse of the “old social contract” (for example, Ianchovichina 2018; Rijkers et al. 2015). Put differently, the ruling coalition creates and exploits rents so that powerful groups and individuals find it in their interest to refrain from destabilizing the social order, the so-called limited access order. In addition to internal players, external actors and supranational forces also influence social contracts (from international agencies, to peacekeeping forces and foreign powers, and global market institutions and shocks). For example, the renegotiation of social contracts in the Middle East cannot be understood without capturing the influence of the “external pact” made up of foreign actors (Mouhoud 2012; Ianchovichina 2018). Bargaining The extent to which social contracts can be dynamically reshaped and renegotiated depends in large part on the nature of the institutional and political space within which bargaining takes place (World Bank 2017). The literature highlights two defining features of this institutional space: (i) the extent to which bargaining takes place through formal and impersonal institutions (for example, election, court system, grievance redress mechanisms, market mechanisms) or informal and personal channels; and (ii) the level of contestability of the bargain (North, Wallis, and Weingast 2009; Acemoglu and Robinson 2008; Keefer 2011; World Bank 2016). In most countries where the World Bank operates, informal rules, networks, and personalized relationships still influence how power, resources, and public goods are allocated among different groups. Given the high levels of informality in fragile and transitional contexts, the social contract can be at risk if locally legitimate institutions are set aside—or made instruments of control—by formal authorities (UNDP 2016, 12). Likewise, if the social contract is the product of a narrow agreement among elites with little contestability in the policy arena 59 Appendix B Conceptual Framework or few channels for citizens to voice demands and concerns, imbalances between citizens’ expectations and state provision cannot be resolved through dynamic channels and can lead to violent uprising (World Bank 2017). Trajectory of Change Ultimately, various social contracts may explain different development trajectories and outcomes in terms of equity, stability, and prosperity. Social contracts can be deeply ingrained, with long periods of apparent stasis or equilibrium of “social forces.” Yet even within these periods of stability, social contracts are dynamic, with gradually evolving expectations from society, changing bargaining power in ruling coalitions, or developing policy systems. During these periods of equilibrium, policies tend to change only incrementally or have limited influence on the core of the social contract, due to the “stickiness” and path dependence of institutions, and entrenched interests in ruling coalitions (Baumgartner and Jones 1993; Mahoney 2000). These periods of relative stasis in the social contract can be punctuated by episodes of rapid and sudden changes because of major societal evolution or external shocks. Causes of stress to an existing social contract can be caused by endogenous movements within civil society, because of changing expectations, or heightened perception of unfairness, aided by increased information or exposure to outside models. Changes in the state’s capacity or legitimacy, due to fiscal constraints or exposition of corruption scandals, can also trigger disequilibrium. External shocks, such as climate change, conflict, migration, or technological disruption, or changes in market outcomes, can be a source of imbalance, especially if they are not adequately mitigated by evolving policies (World Bank 2018). These “critical junctures” in the social contract are moments of break in the path dependence of sociopolitical and economic institutions. During these periods of flux and imbalance in the social contracts, key players face a broader than usual range of feasible options for change and their choices among these options are likely to have a significant impact on the subsequent path-dependent development of the social contracts (Capoccia and Kelemen 2007). References Acemoglu, D., and J. Robinson. 2008. The Role of Institutions in Growth and Development. Washington, DC: World Bank. Baumgartner, Frank R., and Bryan D. Jones. 1993. Agendas and Instability in American Politics. Chicago: University of Chicago Press. Capoccia, Giovanni, and Daniel R. Kelemen. 2007. “ The Study of Critical Junctures: Theory, Narrative and Counterfactuals in Institutional Analysis .” World Politics 59 (3): 341–36. 60 Appendix B Conceptual Framework DfiD (Department for International Development). 2011. The Politics of Poverty: Elites, Citizens and States: Findings from Ten Years of DfID Funded Research on Governance and Fragile States 2001–2010. London: DfID. http://www.dfid.gov.uk/Documents/publications1/evaluation/plcy-pltcs-dfid-rsch-synth- ppr.pdf. Ianchovichina, Elena. 2018. Eruptions of Popular Anger: The Economics of the Arab Spring and Its Aftermath. MENA Development Report. Washington, DC: World Bank. Keefer, Philip. 2011. “Collective Action, Political Parties, and Pro-Development Public Policy.” Asian Development Review 28: 94–118. Mahoney, James. 2000. “Path Dependence in Historical Sociology.” Theory and Society 29 (4): 507– 48. McCandless, Erin. 2018. “Reconceptualizing the Social Contract in Contexts of Conflict, Fragility and Fraught Transition.” Wits School of Governance Working Paper Series, University of the Witwatersrand, Johannesburg. Mouhoud, El M. 2012. “Political Economy of Arab Revolutions: Analysis and Prospects for North-African Countries.” Mondes en développement 2: 35–50. https://doi.org/10.1111/j.1467-7717.1996.tb01033.x. North, D., J. Wallis, and B. Weingast. 2009. Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History. New York: Cambridge University Press. Rijkers, Bob, Leila Baghdadi, and Gael J. R. F. Raballand. 2015. “Political Connections and Tariff Evasion: Evidence from Tunisia (English).” Policy Research Working Paper, no. WPS 7336, World Bank Group, Washington DC. (Paper funded by the Knowledge for Change Program and the Strategic Research Program SRP.) http://documents.worldbank.org/curated/en/828841468179081001/Political-connections- and-tariff-evasion-evidence-from-Tunisia UNDP (United Nations Development Programme).2016. Social Contract in Situations of Conflict and Fragility. Oslo: UNDP. World Bank. 2016. The Role of Political Economy Analysis in Development Policy Operations. Independent Evaluation Group Learning Product. Washington, DC: World Bank. ———. 2017. World Development Report: Governance and the Law. Washington, DC: World Bank. ———. 2018. Citizen Engagement. http://www.worldbank.org/en/about/what-we-do/brief/citizen- engagement. 61 Appendix C. Case Study Summaries Guatemala Social contract framing has improved the diagnostic approach underlying country engagement. The Systematic Country Diagnostic (SCD) of 2015 explicitly embraces a social contract approach. It provides added value by encompassing a variety of seemingly disparate development challenges in a single analytical approach. Specifically, it frames key concerns such as institutional deterioration, the rise of corruption, a lack of state effectiveness in delivering social services, emigration, and drug trafficking not as causes of Guatemala’s development problems but as symptoms of something deeper: a broken social contract. The SCD identifies the root problem as state capture by a self-serving elite with little interest in increasing the fiscal and administrative capability of the state to deliver social services to the rest of the country. The social contract approach has led to a limited departure from business as usual. The Country Partnership Framework (CPF) for FY17–20 adopts the language and perspective of the SCD. However, the World Bank has not been able to radically change its development approach in Guatemala and adopt the “big push” recommended in the SCD. Given that the social contract in Guatemala is rather static, there is little opportunity for the World Bank to drastically change its approach. Instead, the project portfolio envisioned in the country program implicitly adopts a “do no harm” approach. In other words, the suggested portfolio of projects does not exacerbate existing inequalities, but neither does it address fundamental social contract issues. Although the SCD calls for a concerted, systematic, and large-scale push to bring Guatemalan state- society relations to a new equilibrium, the CPF explicitly espouses selective country engagement. Moreover, a large part of the current CPF indicates continuity with projects proposed or implemented in earlier periods, rather than a significant departure. The gap between the scope and ambitions of the SCD and CPF and the portfolio of operation calls into question the feasibility of implementing a social contract approach, both regarding counterparties and time frame. The CPF uses a Venn diagram metaphor to explain selection criteria for its proposed portfolio of projects. Only issues that are both government priorities and areas where the World Bank has a comparative advantage are included. Of a wide range of potential projects, only four loans and eight technical assistance projects made the cut, while others did not count on enough buy-in from the executive branch and the parliament to win approval. Entire development areas such as education were left out in this CPF period. The selectivity of the World Bank portfolio in Guatemala raises some doubt regarding the feasibility of aligning a four-year country program with an analysis as profound and encompassing as a social 62 Appendix C Case Study Summaries contract diagnosis. Moreover, it is hard to imagine the current beneficiaries of the status quo would endorse projects that aim to help transform it. As an interviewee put it poignantly: “if the government is captured by a self-serving elite, can the World Bank really push for more inclusive institutions as long as it works with that government?” Social contract framing has neither helped nor hindered the effectiveness of country engagement. It was not a necessary condition for effective engagement, as a considerable share of World Bank Group projects over the past 25 years has targeted excluded parts of the country and society. It was not a sufficient condition either: only a fraction of the World Bank’s loan portfolio is being implemented due to political developments beyond those identified in the social contract analysis. In particular, a large part of Guatemala’s elite is under corruption investigations initiated by the International Commission against Impunity in Guatemala, a UN-sponsored panel. The resulting antiforeigner sentiment of the local elite has also affected proposed World Bank–financed projects that failed to win approval in parliament. In sum, in an environment where the social contract is relatively static and a government that represents the status quo, the World Bank appears to have a very slim window of opportunity to change a deeply flawed social contract. The Guatemalan case shows the absence of an inflection point in state-society relations, in spite of a change in government. This situation calls into question the degree to which a social contract lens can be coherently translated into a CPF, and the conditions with regard to scope under which World Bank engagement can be legal, legitimate, and transformative. In this situation, nonlending activities can show unexpected results. In a country where the World Bank is very constrained in what it can achieve through lending, nonlending activities can be used effectively to create change at the margin. For example, regional (trust-funded) activity can be successful at reaching excluded parts of society. Specifically, indigenous rights activists are engaged in the Forest Investment Partnership (FCPF, REDD+) and the World Bank–supported Indigenous Forum of Abya Yala, a key venue for peer learning among indigenous communities about how to hold their respective governments accountable. Moreover, indigenous representatives helped the World Bank Presidency revise safeguards in indigenous policies, and they have received capacity building assistance from headquarters. Nepal Although the Bank Group did not adopt an explicit social contract framing in Nepal, it addresses several key social contract themes in its diagnostic work. The Bank Group has abstained from explicit use of a social contract lens but the 2017 Risk and Resilience Assessment and the 2018 SCD assess equity and inclusion issues. The reports note that 63 Appendix C Case Study Summaries elite capture represents a risk to human resource development, and they recommend the strengthening of mechanisms for citizen engagement and representation. The World Bank identifies improving public institutions, including federal transition and increased inclusiveness in the political process, as the most important constraint to development progress. Earlier work by the Bank Group and the U.K. Department for International Development in turn provided the key concepts that have helped transform the social contract in the country. In 2002, the World Bank initiated a comprehensive Gender and Social Exclusion Assessment in cooperation with the U.K. Department for International Development. The report provides an assessment of the social structure, social inclusion, and relative empowerment of different religious and ethnic groups, as well as the impact of gender on these attributes. It identifies caste and gender as key predictors of the variation in empowerment and inclusion levels. The report concludes that attempts to address gender exclusion would invariably have to tackle caste and ethnic exclusion simultaneously. The Bank Group work on gender and social exclusion had transformative impact at an inflection point in Nepal’s history. It was published in 2006, in the same year that a peace agreement ended a civil war that was driven by grievances about exclusion and inequality. The social contract lens underlying the report can be understood as part of a Trojan Horse strategy designed to raise awareness of social contract issues among key stakeholders—with success. The government of Nepal took ownership of the report and formulated various policies to address inequality and exclusion along gender, caste, and regional lines. Other development partners have embraced a common framework that builds on the World Bank study. In 2017, the International Development Partners Group under the leadership of UN Women followed up on the World Bank assessment and formulated a common framework for Gender Equality and Social Inclusion (GESI) that now guides country engagement. The government has also taken credible steps to mainstream GESI among government institutions, including affirmative action policies, hiring of GESI specialists, and adoption and monitoring of sectoral GESI policies. GESI has thus provided a policy framework and the ideological and institutional basis for redefining the social contract. Country engagement prematurely moved past social contract concerns but is returning to the issue in the future. The current CPF embodies a strategic redirection from postconflict reconstruction to longer-term development assistance. Recognizing that political economy issues are still key challenges, the World Bank commissioned a GESI assessment to help design the next CPF and ensure that the World Bank strategically 64 Appendix C Case Study Summaries supports Nepal in the renewal of its social contract, political settlement, and state organization. Niger Niger’s social contract can be characterized as lopsided but relatively static, with a powerful governing elite, a largely disconnected and powerless citizenry, and a barely existent third pole of nongovernment players. Even though 80 percent of the population lives in rural areas, more than 90 percent of the government’s budget is spent at the central level, mostly concentrated in the capital, where more than 65 percent of government employees are based. Both supply- and demand-side respondents from field interviews conducted for the Niger Social Accountability brief estimated that life priorities and realities of governing elites and citizens are very distinct. The dynamics of Niger’s “political settlement” are arguably the most important factor in Niger’s stability. This political settlement involves multiple actors balancing each other’s interests (and power) in a dynamic but relatively stable equilibrium. The ruling coalition changes regularly, coopting new power brokers and sidelining others, leading to high levels of turnover in the administration. Since gaining independence in 1960, Niger has had no less than seven republican regimes and four military coups. Among the core power players are the military, a group of lifetime politicians, powerful traders, and selected traditional leaders, sheikhs, and religious leaders. This “political class” is sustained by a very poor, uneducated, and uninformed population that seeks protection against poverty and shocks. Consequently, politics is at times turbulent but rarely violent, and the social contract is in relative status quo. The 2017 SCD contains an in-depth and refreshingly frank analysis of the state-citizen relationships, marking a clear departure from previous diagnostics. It documents public sector incompetence and corruption issues along with a candid assessment of what is called “Niger’s political settlement” and of its consequences. The report links the analysis to development issues such as the concentration of expenditures in the capital and a patronage-based promotion/demotion system that contributes to low civil servant morale and bureaucratic ineffectiveness. In consultations for the CPF, World Bank and central government representatives made concerted efforts to reach out to relatively excluded parts of society. The Country Management Unit pushed for high-level representation in a joint delegation that would travel to several less-developed regions and provide local stakeholders with an unprecedented opportunity to speak out. The World Bank also provided an analysis of vulnerabilities, including issues related to social contract bargaining. World Bank teams prepared an implementation note to provide the background and rationale for granting Niger access to the International Development Association (IDA) 18 Risk Mitigation Regime. This analysis focuses on key 65 Appendix C Case Study Summaries drivers of fragility, conflict, and violence from within the country as well as cross-border terrorist threats. The note also goes beyond analysis to recommend specific projects that could or should contribute to reducing fragility, conflict, and violence risks. Country engagement represents a sharpening of focus rather than a significant departure from business as usual. Even though the CPF FY18–22 states that business as usual is no longer an option, the portfolio of operations embodies continuity rather than the sharp departure from business as usual called for in the diagnostic. The relative stasis of the social contract in Niger and the fact that the government is a guardian of the status quo do not provide much room for the World Bank to adopt the drastic change in its development approach invoked in the SCDs and CPF. There a few exceptions of World Bank lending that seek to directly address citizen-state relationships and counter the Niamey bias of the current social contract. For example, the new series of development policy loan operations envisaged in the CPF and now under way has a strong focus on rural areas, where the bulk of the poor live, and supports the establishment of e-vouchers and e-extension as well as increased connectivity for farmers. This is a departure from the previous development policy loan series, which focused on improving the public finance management and public investment management systems and on public irrigation. The development policy loan series encourages the transfer of mining revenues to communities, thus addressing an important fragility driver and contributing to restore territorial equilibrium in the current social contract. Given the constraints on the lending program, the World Bank team has sought to create change through its nonlending activities. Interviews with Nigerien counterparts suggest that the World Bank is seen as evolving in three major ways: more attention to gender issues and women’s empowerment, greater openness to dialogue, and a push for citizen- centric approaches. For example, the World Bank is seeking to systematically embed citizen engagement in its activities and to leverage its Advisory Services and Analytics portfolio to strengthen its emphasis on women’s empowerment and citizen engagement. In addition to the external obstacles to translating a social contract diagnostic into operational practice, there are several significant internal challenges. First, given the client-centric model adopted by the World Bank, projects and official documents have to be approved by the government, so the incentive for World Bank staff is to privilege interactions with central state authorities. Second, the World Bank’s effectiveness also relies on its ability to maintain its privileged relationships with government counterparts and its reputation as a neutral, technocratic partner that does not interfere in domestic politics. Moreover, the Nigerien civil society is still too weak to represent a true counterweight to central power. Third, the staff skill mix and team composition can 66 Appendix C Case Study Summaries be an obstacle to translating a social contract diagnostic into operation. The profile of staff working on a given country matters enormously, and Niger is no exception. Those with a background in behavioral sciences or political science seemed more at ease with the concept of a social contract, more knowledgeable about ways to improve its quality, and more interested in engaging with a variety of actors (including at the grassroots) and seeking feedback from clients and beneficiaries. This is reflected at the sector level, with sectors such as social protection, social development, and governance more likely to integrate into their operations approaches and measures to give voice to regular citizens. Tunisia The World Bank has developed a thorough understanding of the social contract in Tunisia. The World Bank has built on the research underlying the strategy for the Middle East and North Africa (MENA), and on a series of World Bank analytical work that started in the early 2000s and gained attention in the aftermath of the Arab Spring, to identify the root causes of profound development challenges—including youth dissatisfaction with public services, joblessness, and heavy state regulation and elite capture. The 2015 SCD applies this diagnostic to Tunisia, skillfully incorporating perception data and integrating historical and sociological factors in ways that reveal the nature of state-society relations far beyond standard economic analyses. The rich data and analysis embedded in the SCD were followed by sector-specific political economy analysis and research to refine the World Bank’s diagnostic and facilitate the translation of this diagnostic into operations. However, the World Bank missed the opportunity of disseminating its social contract diagnostic broadly within the country. The World Bank’s analysis of Tunisia’s social contract is not well known among government and civil society’s stakeholders, nor among other development partners. All stakeholders interviewed during the case study mission thought that by not disseminating its analytical work on social contract widely, the World Bank missed an opportunity to demonstrate its depth of understanding of the country’s social fabric, to improve its image among skeptical civil society organizations and citizens, and to build policy reform coalitions. Greater dissemination efforts may have helped foster alliances to improve the implementation of reforms. The World Bank’s country program and operations in Tunisia are well aligned with its social contract diagnostic. The CPF FY16–20 is a faithful translation of the World Bank’s social contract diagnostic. It focuses on job creation, social protection, and social inclusion programs for the most vulnerable parts of society, and on the development of lagging regions. It aligns well with the country’s priority to implement a new social 67 Appendix C Case Study Summaries contract, which was negotiated and agreed on in 2013. The World Bank’s projects in Tunisia are well targeted to lagging regions, which many stakeholders considered made the World Bank’s country program particularly relevant to Tunisia, compared with other donors whose targeting is more diffuse. Several projects in the World Bank’s portfolio in Tunisia include all the principles of social contract renewals laid out in the MENA strategy. These principles include making private sector jobs resistant to elite capture, bringing informal jobs into the formal economy, targeting lagging regions, and empowering youth and women. According to government and civil society stakeholder interviews, Tunisia’s youth economic inclusion project, or Moubadiroun, is a good example of how these principles successfully infiltrated MENA’s project designs. First, the project targets disadvantaged men and women between the ages of 18 and 35 who live in lagging regions with employment opportunities. This stands in contrast to the employment programs of the “old social contract” that were prone to elite capture. Second, the project eliminates barriers under the “old social contract” that prevented small enterprises from entering the formal economy and reaching national and international markets. This was accomplished by developing value chains and better linking businesses in the country’s lagging regions to established markets. Third, the project includes a scorecard mechanism for program participants to rate the quality of service providers. However, engaging on social contract issues carries efficacy and reputation risks, which the World Bank team navigates daily. Despite the political astuteness of the World Bank’s country team, key reforms that it supports are currently at a standstill or have had to be aborted, including reforms of the pension system, education sector, subsidy regime, and state-owned enterprises. Challenges include (i) navigating a very complex political economy situation and reaching consensus on reforms that remain unpopular with the powerful Tunisian Labor Union or some parts of the civil service, (ii) overcoming powerful vested interests linked to the “old social contract” that resist opening the economy, (iii) undoing the moral hazard created by the series of budget support that occurred despite a wave of new recruitments and salary increase in the public sector that went against the World Bank and International Monetary Fund recommendations, and (iv) continuing to gain traction in policy dialogues despite having reached its prudential limit on lending. In addition, several aspects of the World Bank’s operating model constrain the effectiveness of the World Bank’s work on social contract issues in Tunisia. First, staff involved in operations that seek to address core social contract issues emphasized that design and implementation schedules are frustrated by the long and sensitive political bargaining that requires World Bank technical staff to engage in contentious 68 Appendix C Case Study Summaries negotiations among political factions and divided civil society groups. Second, interviewed government counterparts also expressed frustration with the disbursement metrics by which the World Bank judges the performance of these interventions and argued in favor of alonger-term approach to programming. Third, the type of radical changes in political settlement, state organization structure, and socioeconomic transformation that Tunisia has embarked on can take decades to come to fruition and can be at odds with the World Bank’s emphasis on producing measurable outcomes within a relatively short time frame. Uzbekistan The Bank Group has responded quickly to what may be an inflection point in the social contract of Uzbekistan. The country was ruled by one man, Islam Karimov, for a quarter century. Since 2017, Karimov’s successor has embarked on an ambitious reform course that includes the marketization of the economy and significant reform in how the government interacts with society. The 2018 Performance and Learning Review embraces something akin to social contract thinking both in its analysis and portfolio of proposed actions. It makes ample reference to the government’s National Development Strategy, which in turn focuses on increased civil society participation, and it elevates citizen engagement to one of three focus areas of the updated and extended CPF. The Bank Group has led transformative changes in state-society relations long before the new government came to power. The Bank Group Board responded to a 2013 Inspection Panel claim by instituting a Third-Party Monitoring mechanism to curb forced and child labor in cotton harvesting. This policy has helped engender changes in state-society relations much beyond its original purpose. The monitoring mechanism has provided civil society representatives with a means to obtain legal status and connect with their peers at the local level. The new government has adopted and expanded its communication channels. There are encouraging signs that the World Bank can leverage the current window of opportunity opening in the country to contribute to social contract transformation in Uzbekistan, but significant external and internal constraints exist. The current opening is carefully controlled by the government and the World Bank is responding to a still tightly circumscribed government demand for assistance in citizen engagement. Reform of state-society relations is a top-down process driven by the central government, and actual social transformation is perceived as slow and prone to reversals. This is also reflected in the World Bank portfolio. The number of projects expected to integrate elements of the National Citizen Engagement Framework is expected to rise from 0 to 4 69 Appendix C Case Study Summaries by 2021, a rather modest share of the current total of 17 loan-financed projects (plus 10 ASA). In addition to insufficient human resources, internal World Bank incentives may not be aligned to scale up citizen engagement. Currently, the Country Management Unit in Uzbekistan has no social development expert on staff. In addition, touching on sensitive issues requires more sophisticated and thus costly preparatory work, which clashes with (perceived or real) budget constraints. Moreover, staff (even those working on opinion polls) regard their remit as chiefly technocratic and focusing on improving the objective economic conditions of citizens. Finally, citizen engagement often gets short shrift when country managers engage in budget negotiations with each Global Practice. 70 Appendix D. Coherence Analysis Objective The team conducted an analysis of World Bank Systematic Country Diagnostics (SCDs), Country Partnership Frameworks (CPFs), and Country Engagement Notes (CENs), as well as lending and advisory portfolios, to contribute toward two key questions of the evaluation (see table D.1). In short, these were (i) how the World Bank understands and diagnoses social contracts; and (ii) how coherently the World Bank incorporates diagnosed social contract reform areas into country strategies and portfolios. Table D.1. Contributions of Coherence Analysis to Overall Evaluation Questions Contributing to Evaluation Question Unit of Analysis 1a: How does the World Bank understand and Systematic Country Diagnostic; Middle East and diagnose disequilibria in social contracts and how North Africa regional strategy does it explain social contracts’ influence on development outcomes? 2a: In countries where the World Bank has identified Country Partnership Framework; Country imbalances in social contracts through Systematic Engagement Note; Country Diagnostics, to what extent and how has Lending Portfolio; this diagnostic shaped its country engagement, that Advisory Services and Analytics is, Country Partnership Framework, lending and advisory portfolios? Source: Independent Evaluation Group. Methodology Data Collection Countries included in this review were those with an SCD that mentioned the keywords “social contract” and/or “social compact” at least five times. An exception was countries in the Middle East and North Africa (MENA) Region, which were included even if they did not have a unique SCD (or if the SCD did not mention the keywords). For these 12 countries, the 2015 MENA regional strategy was used as the diagnostic document because of the regional strategy pillar 1, which aimed to renew social contracts. 71 Appendix D Coherence Analysis Using this methodology, 33 countries2 were identified as in scope, and SCDs (or MENA regional strategy) were reviewed as a starting point for relevant social contract information (approved between FY15 and FY18). This included how the SCDs diagnosed key elements of the current social contract, and how they articulated those areas in need of reform. SCDs served as the baseline because they are the instrument used by the World Bank Group to identify, through in-depth analysis and stakeholder consultations, the most important challenges and opportunities faced by a country. An SCD informs each new CPF. The team subsequently reviewed the corresponding CPF or CEN—the key strategy document guiding the Bank Group’s support to a member country. The CPF/CEN pillars were reviewed and extracted to facilitate the coherence analysis between the challenges diagnosed (in SCDs) and the reform areas formally agreed on (in CPFs/CENs). Finally, the lending and nonlending (Advisory Services and Analytics [ASA]) portfolios that had been approved since the disclosure of the reviewed SCD were downloaded from Business Intelligence to further track coherence. The abstracts were screened to identify those lending and ASA projects that explicitly addressed issues related to the social contract and recommended reforms that were articulated in the reviewed SCD and CPF. This was done for the full cohort of sampled countries. A coding template was developed and administered in Microsoft Access to consistently extract relevant data from the documents reviewed (see table D.2). 2Algeria; Bahrain; Belarus; Bulgaria; Chile; Costa Rica; Djibouti; the Arab Republic of Egypt; Guatemala; Guinea-Bissau; Haiti; Iran; Iraq; Jordan; Kuwait; Lebanon; Libya; Mali; Mauritania; Mauritius; Morocco; Oman; Paraguay; Qatar; the Russian Federation; Saudi Arabia; Sierra Leone; South Africa; Tajikistan; Tunisia; United Arab Emirates; West Bank and Gaza; the Republic of Yemen. 72 Appendix D Coherence Analysis Table D.2. Coherence Analysis Coding Template Variable 1. Country name and year of the SCD used as starting point 2. First year covered under CPF after the SCD 3. Definition of social contract according to the SCD 4. Key elements of the social contract and those in need of reform 5. Policy system that typifies the social contract 6. Historical legacy of the current social contract 7. Bargaining mechanisms through which the social contract can change (formal, informal institutions) and the extent to which citizens are able to participate 8. External or internal factors that caused a change in the social contract (or that require change) 9. Proposed areas for reform in the SCD 10. Groups specifically targeted in the SCD 11. Themes/pillars of the country CPF/CEN 12. Groups specifically targeted in the CPF/CEN 13. How the World Bank has been addressing social contract elements prior to its explicit diagnostic (Completion and Learning Review, Performance and Learning Review) 14. Reforms given as a recommendation in the SCD 15. Reforms given as a program recommendation in the CPF 16. Reforms included in project design in projects approved since the SCD (lending) 17. Reforms included in project design in projects approved since the SCD (Advisory Services and Analytics) Source: Independent Evaluation Group. Note: CEN = Country Engagement Note; CPF = Country Partnership Framework; SCD = Systematic Country Diagnostic. Data Analysis Data analysis was carried out in Microsoft Excel by thematically coding the data exported from the Access database, and graphically presenting various parts of the analysis. The subsequent sections of this appendix provide a summary of the findings organized as follows: (i) definitions of social contracts in SCDs, (ii) social contracts in transition, (iii) social contracts across contexts, (iv) diagnosed issues with social contracts, (v) groups targeted with social contract reform areas, (vi) overall coherence analysis, and (vii) dedicated MENA analysis (see Summary of Findings section). The sample sizes for the different analyses vary depending on the extent of data points included in the analysis. For instance, for analyses of social contract diagnoses (that is, social contract definitions, social contracts in transition, diagnosed issues with social contracts, groups targeted in SCDs) the sample size is 22, which includes all countries 73 Appendix D Coherence Analysis with a unique SCD (n = 21)3 and the MENA regional strategy. For the analysis of vulnerable groups in CPFs/CENs specifically, the sample size is 21,4 which are all countries with a unique CPF/CEN. For the coherence analysis, only countries that have both a unique SCD (or are associated with the MENA regional strategy) and unique CPF/CEN were included (n = 19).5 Limitations The coherence analysis was challenged by (i) the variation in focus and quality of social contract diagnostics; (ii) the variation in timelines of the SCD and CPFs sampled, ranging from FY15 to FY18; and (iii) analysis of the social contract alignment of lending and nonlending portfolios, which was not explicit in most cases. Summary of Findings Definitions of Social Contracts in Systematic Country Diagnostics Despite adopting a social contracts approach in their diagnostics, 45 percent of the SCDs sampled did not provide a definition of the concept (see figure D.1). Another 23 percent employed a state-focused definition that exclusively emphasized the role of the public sector to deliver public goods and services, provide social protection (social safety nets and subsidies), and create an environment conducive for growth. In contrast, 9 percent had a citizen-focused definition that emphasized greater citizen engagement as central to increasing long-term growth, as well as economic and social inclusion for peace and social stability. Finally, almost a quarter (23 percent) provided a more nuanced definition that combined—to some extent—the roles (and rights) of, and relationships between, the state, citizens, and other stakeholder groups. 3Belarus; Bulgaria; Chile; Costa Rica; the Arab Republic of Egypt; Guatemala; Guinea-Bissau; Haiti; Iraq; Jordan; Lebanon; Mali; Mauritania; Mauritius; Morocco; Paraguay; the Russian Federation; Sierra Leone; South Africa; Tajikistan; Tunisia. 4Belarus; Bulgaria; Chile; Costa Rica; Djibouti; the Arab Republic of Egypt; Guatemala; Guinea- Bissau; Haiti; Iraq; Jordan; Lebanon; Mali; Mauritania; Mauritius; Morocco; Paraguay; Tajikistan; Tunisia; West Bank and Gaza; the Republic of Yemen;. 5Belarus; Bulgaria; Chile; Costa Rica; Djibouti; the Arab Republic of Egypt; Guatemala; Guinea- Bissau; Jordan; Haiti; Lebanon; Mali; Mauritania; Mauritius; Morocco; Tajikistan; Tunisia; West Bank and Gaza; the Republic of Yemen. 74 Appendix D Coherence Analysis Figure D.1. Definitions of Social Contracts in Systematic Country Diagnostics Citizen-focused definition, 2, 9% State-focused definition, 5, 23% No definition, 10, 45% State-society focused definition, 5, 23% Source: Independent Evaluation Group. Note: n = 22; all Systematic Country Diagnostics and the Middle East and North Africa regional strategy. Social Contracts in Transition Some SCDs provide insights on historical legacies that have contributed to, or continue to influence, the status of the social contract. However, overall there is limited description and evidence on the dynamic of social contracts, what factors contribute to destabilizing or disrupting them, what actors are instrumental in creating or hindering change, what is the status of the bargaining mechanisms that may enable or hinder changes. Table D.3 presents (based on data from SCDs) short descriptions of change dynamics. Interestingly, a social contract framing has been employed by the World Bank in countries that are arguably in transition, as well as those that do not appear to be undergoing significant change. 75 Appendix D Coherence Analysis Table D.3. Social Contracts in Transition as Articulated in Systematic Country Diagnostics Country Brief Description Jordan Society’s original occupational segmentation between two different groups has given way to an amorphous middle class including public sector workers, especially in the administrative layer, employees, and proprietors of smaller firms in the business sector. Lebanon In the absence of effective state institutions, sectarian organizations have been providing key social services and welfare support to the most vulnerable groups, while privatization of service delivery has led to exorbitant prices. This has created steep inequalities in access to services. Mauritius Postindependence inequalities have since transitioned to responsive institutions that expanded public services for all and social protection programs for the most vulnerable. Shared economic growth has pulled most of the population out of poverty and created a vast middle class. Sierra Leone Neopatrimonial patterns of state power have proven resilient, surviving the transition to independence, the civil war, and the more recent institutionalization of electoral democracy, though a long-term commitment to the public good have yet to emerge. Costa Rica A long history of egalitarian ideology has resulted in Costa Rica’s notable welfare state. Guinea- Repeated cycles of fragility and weak institutions. Bissau Haiti Independence did not lead to inclusive institutions; the state has remained absent outside of the capital for decades. Mali Mali’s population has become increasingly disillusioned with the political realities of general long-term deterioration of governance. Paraguay The interplay between structural factors (an agrarian economy with natural resources endowments) and sociopolitical institutions (small state apparatus and a limited bureaucracy) has created several path dependencies that continue to shape the policy arena. South Africa High inequality and the legacy of exclusion fuel the contestation over resources, increasing policy uncertainty and deterring investment, while also undermining the financial stability of state-owned enterprises and their ability to provide quality public services. Tajikistan The civil war and weak governance have had widespread consequences on the effectiveness and accountability of public institutions, civil trust, and the legitimacy of the state. Source: Independent Evaluation Group based on Systematic Country Diagnostics. Social Contracts across Contexts The adaptability of social contract diagnostics is demonstrable in the 21 diverse SCDs in which the framing was applied. As shown in tables D.4, D.5, D.6, and D.7, these countries span four regions, include all income groups, fragile and nonfragile states, and resource-rich and resource-poor countries. 76 Appendix D Coherence Analysis Regions Table D.4. Regional Distribution of Systematic Country Diagnostics Using Social Contract Diagnostics Middle East and Europe and Central Latin America and the North Africa Region Africa Asia Caribbean Egypt, Arab Rep. Guinea-Bissau Belarus Chile Iraq Mali Bulgaria Costa Rica Jordan Mauritania Russian Federation Guatemala Lebanon Mauritius Tajikistan Paraguay Morocco Sierra Leone Haiti Tunisia South Africa Source: Independent Evaluation Group. Country Classifications Table D.5. Country Classifications of Systematic Country Diagnostics Using Social Contract Diagnostics Low Income Lower-Middle Income Upper-Middle Income High Income Guinea-Bissau Egypt, Arab Rep. Belarus Lebanon Chile Haiti Mauritania Bulgaria Mauritius Mali Morocco Costa Rica Paraguay Sierra Leone Tunisia Guatemala Russian Federation Tajikistan Iraq South Africa Jordan Source: Independent Evaluation Group. Fragile Situations Table D.6. Distribution of Systematic Country Diagnostics Using Social Contract Diagnostics between Fragile and Nonfragile countries FCV Countries Non-FCV Countries Guinea-Bissau Belarus Jordan Sierra Leone Haiti Bulgaria Mauritania South Africa Iraq Chile Mauritius Tajikistan Lebanon Costa Rica Morocco Tunisia Mali Egypt, Arab Rep. Paraguay Guatemala Russian Federation Source: Independent Evaluation Group. Note: FCV = fragility, conflict, and violence affected. 77 Appendix D Coherence Analysis Resource Endowment Table D.7. Resource Endowment of Countries with Systematic Country Diagnostics Using Social Contract Diagnostics Resource-Rich Countries Resource-Poor Countries Chile Morocco Belarus Lebanon Egypt, Arab Rep. Russian Federation Bulgaria Mauritius Guatemala Sierra Leone Costa Rica Paraguay Iraq South Africa Guinea-Bissau Tajikistan Mali Tunisia Haiti Mauritania Jordan Sources: Independent Evaluation Group and Natural Resource Governance Institute. Note: “Resource-rich” countries include those that are included in the Natural Resource Governance Institute’s Resource Governance Index. The Resource Governance Index assesses the quality of natural resource governance in 81 countries that together produce, among other commodities, 82 percent of the world’s oil, 78 percent of its gas, and 72 percent of all copper. https://resourcegovernance.org/analysis-tools/publications/2017-resource-governance-index. Diagnosed Issues with Social Contracts In the 21 SCDs and the MENA regional strategy reviewed where a social contract framing was used to diagnose and explain development challenges, the most frequently cited challenges were (i) social exclusion and low citizen voice, (ii) lacking jobs and private sector, (iii) low transparency and accountability, and (iv) poor service delivery (see figure D.2). Figure D.2. World Bank–Diagnosed Issues with Social Contracts in Systematic Country Diagnostics Social exclusion & low citizen voice 19 Lacking jobs & private sector 13 Low transparency & accountability 12 Poor service delivery 10 Insufficient social protection 8 Fragility, crime & violence 6 Weak resource mobilization & management 6 Overcentralized state 2 0 2 4 6 8 10 12 14 16 18 20 Countries (no.) Source: Independent Evaluation Group. Note: n = 22; all Systematic Country Diagnostics and the Middle East and North Africa regional strategy. 78 Appendix D Coherence Analysis Groups Targeted with Social Contract Reform Areas Women, the poor, and youth are the most frequently cited vulnerable groups in the sampled SCDs and CPFs/CENs. Lagging regions were also targeted in three CPFs, all of which were in the MENA Region. Figure D.3. Target Groups Referenced in Systematic Country Diagnostics Youth 10 Poor 9 Women 5 Target group Lagging regions 4 Minorities 2 Children 1 Elderly 1 Indigenous people 1 Landless 1 Slaves/former slaves 1 0 2 4 6 8 10 12 Systematic Country Diagnostics (no.) Source: Independent Evaluation Group. Note: n = 22; all Systematic Country Diagnostics and the Middle East and North Africa regional strategy. Figure D.4. Target Groups Referenced in Country Partnership Frameworks and Country Engagement Notes Women 7 Poor 6 Target group Youth 5 Lagging regions 3 Conflict affected 1 Minorities 1 0 1 2 3 4 5 6 7 8 Country Partnership Frameworks and Country Engagement Notes (no.) Source: Independent Evaluation Group. Note: n = 21; all Country Partnership Frameworks and Country Engagement Notes. Coherence Analysis Coherence varies significantly depending on the type of reform recommended (see figure D.5 and D.8). Typical public sector issues—such as improving service delivery and enhancing the enabling environment for jobs and the private sector—are 79 Appendix D Coherence Analysis consistently recommended and financed, particularly in lending operations. Conversely, social contract issues that aim to reinforce societal capacities for bargaining power are less frequently translated into operations. Although issues of transparency and accountability, social inclusion, and citizen engagement are often flagged as needing reform in SCDs, they rarely translate into dedicated World Bank operations. This is particularly true for dedicated social inclusion and citizen engagement activities. Issues around tax and financial management are also notably absent, occurring as a recommendation in only one-third of SCDs, and translating into even fewer lending or ASA operations. 80 Appendix D Coherence Analysis Figure D.5. Coherence across Reform Areas Given as Given as program Included in lending Included in ASA recommendation in recommendation in portfolio since SCD portfolio since SCD SCD CPF Given as recommendation in SCD Given as program recommendation in CPF Included in lending portfolio since SCD Included in ASA portfolio since SCD 16 14 14 14 12 12 12 Number of Countries 12 11 10 10 10 10 9 8 8 7 7 6 6 6 6 6 5 4 4 4 3 3 3 2 2 1 1 1 1 1 0 0 0 0 Source: Independent Evaluation Group. Note: n = 19; only countries with SCD and CPF. CPF = Country Partnership Framework; SCD = Systematic Country Diagnostic. Table D.8. Coherence across Reform Areas Coherence Coherence Coherence between SCD between CPF between CPF SCDs and CPF and Lending and ASA Reform Area (no.) (percent) (percent) (percent) Service delivery n = 14 86 100 42 Jobs and private sector n = 14 79 73 64 Transparency and accountability n = 12 75 67 67 Social inclusion and citizen n = 10 100 10 10 engagement Social protection n = 10 70 57 43 Tax and financial management n=6 100 50 33 Decentralization n=4 75 33 33 Land n=1 0 0 0 Source: Independent Evaluation Group. Note: n = 19; only countries with SCD and CPF. ASA = Advisory Services and Analytics; CPF = Country Partnership Framework; SCD = Systematic Country Diagnostic. 81 Appendix D Coherence Analysis Middle East and North Africa Analysis There is strong coherence in MENA countries between social contract issues identified in SCDs, and those that were sequentially integrated in CPFs and translated into operations and ASA. This trend was consistent across most reform areas, except for social inclusion and citizen engagement, which was recommended in multiple countries, but thus far only had dedicated activities in Tunisia (see table D.9). Table D.9. Coherence across Sectors in Middle East and North Africa Countries Tax and Financial Social Protection Jobs and Private Decentralization Service Delivery Social Inclusion Accountability Transparency Management Engagement and Citizen Sector and Country Egypt, Arab Rep. (2015) ◼◼◼◼ ◼◼◼◼ ◼◼◼◼ ◼◼ ◼◼◼◼ ◼◼◼◼ ◼◼◼ Iraq (2017) ◼◼◼ ◼◼ ◼ ◼ Jordan (2016) ◼◼◼◼ ◼◼◼ ◼ Lebanon (2015) ◼◼◼ ◼◼◼◼ ◼◼ ◼ ◼◼ ◼ Morocco (2018) ◼ ◼◼ ◼ ◼◼ Tunisia (2015) ◼◼◼ ◼◼◼◼ ◼◼◼ ◼◼◼ ◼◼◼◼ West Bank and Gaza ◼◼◼◼ ◼◼◼◼ ◼◼◼◼ ◼◼ (MENA Strategy 2015) Yemen, Rep. ◼◼◼◼ ◼ ◼◼ ◼ ◼ (MENA Strategy 2015) ◼ Given as ◼ Given as program ◼ Included in lending ◼ Included in ASA recommendation in recommendation in CPF portfolio approved portfolio approved SCD pillars since SCD since SCD Source: Independent Evaluation Group. Note: ASA = Advisory Services and Analytics; CPF = Country Partnership Framework; MENA = Middle East and North Africa; SCD = Systematic Country Diagnostic. When comparing overall coherence between MENA and non-MENA countries, there is significantly improved portfolio coherence in MENA (see figure D.6). This indicates that anchoring the MENA strategy around the social contract concept promoted coherence across reform areas, particularly also in those reforms that received less attention and exhibited reduced coherence in the overall analysis, for example, tax and financial management, social protection, and decentralization. 82 Appendix D Coherence Analysis Figure D.6. Overall Coherence in MENA Countries Compared with Non-MENA Countries MENA countries (n=6) non-MENA countries (n=14) 96% 100% 80% 73% 74% 65% 58% 60% 40% 29% 20% 0% Coherence between SCD and Coherence between CPF and Coherence between CPF and CPF Lending ASA Source: Independent Evaluation Group. Note: MENA countries (n = 6): the Arab Republic of Egypt, Iraq, Jordan, Lebanon, Morocco, Tunisia. Iraq does not have a CPF. Non-MENA countries (n = 14): Belarus, Bulgaria, Chile, Costa Rica, Djibouti, Guatemala, Guinea-Bissau, Haiti, Mali, Mauritania, Mauritius, West Bank and Gaza, the Republic of Yemen, Tajikistan. All have an SCD and CPF/CEN. CEN = Country Engagement Note; CPF = Country Partnership Framework; MENA = Middle East and North Africa; SCD = Systematic Country Diagnostic. 83 Appendix E. Stocktaking of Governance and Social Development Analytical Frameworks Since the early 1990s, the World Bank has introduced governance frameworks with the aim to understand the political economy, governance, and institutional dynamics of member states. This stocktaking study examines the evolution of World Bank governance frameworks with the aim to identify (i) the framework’s influence on country engagement, lending, and projects; (ii) the factors that enabled their operationalization; (iii) the factors that hindered their operationalization; and (iv) key lessons learned from other donor experiences. The six governance frameworks studied include political economy analysis (PEA), Governance and Accountability Diagnostic (GAC), Poverty and Social Impact Analysis (PSIA), Institutional and Governance Review (IGR), citizen engagement, and Risk and Resilience Assessment (also known as Fragility Assessments). Introduction Starting in the early 1990s, the World Bank was influenced by external factors to shift the way it conducted business. The World Bank’s rhetoric moved away from focusing on macroeconomic structural adjustments toward implementing a lending approach that included a “good governance” approach. External factors that influenced the World Bank included the failure of shock therapy, failed structural adjustments in Latin America and Africa, and the successful East Asia Miracle. The World Bank was also influenced by civil society and pressure from member states for debt relief, as well as overall criticisms of the effectiveness of aid. Slowly World Bank senior leadership began to elevate and prioritize governance and corruption problems, including political analysis. The James Wolfensohn’s Cancer of Corruption Speech and the 2000 Reforming Public Institutions and Strengthening Governance Strategy (GAC strategy) signaled to World Bank staff new commitments to address good governance. According to World Bank staff interviews, it would have been impossible to talk about anticorruption measures prior to the Cancer of Corruption Speech. During these years, there was a shift in research focus, hiring of new staff and the creation of an institutional home for governance under the Poverty Reduction and Economic Management (PREM) group. But even with these important steps taken forward, there was a culture of resistance at the World Bank against adopting a more nuanced governance stand. 84 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks Governance analysis was part of the strategy to operationalize a good governance approach. The 2002 United Nations International Conference in Financing for Development in Monterrey, Mexico, encouraged donors to adopt approaches “to identify and address the political nature of governance constraints to aid effectiveness, and incorporate analyses of domestic political dynamics in operations.” The GAC strategy explicitly advocated for political analyses to be conducted routinely in World Bank’s operations with a focus on as assessing the political support for reform, the distribution of power among stakeholders, and whether there was political opening for reforms. It argued that through improved diagnostic work the quality of programming and effectiveness would increase. For example, it stated that by conducting diagnostic work that adequately identified “windows of opportunity” for reform, the World Bank would be better able to engage in policy dialogue without investing heavily. According to an IEG evaluation of the GAC strategy, the GAC strategy was a “cumulative effect of a decade-long evolution in frameworks and perspectives on the constraint effects of political processes, actors, and institutional development” (World Bank 2010, 20). Over the next two decades, the World Bank introduced at least six governance-related frameworks or analytical tools. Table E.1 highlights each framework’s objectives, dimensions to measure, and methodology. 85 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks Table E.1. Summary of World Bank Governance Frameworks Name of Dimensions Approach Start End Objective Measured Methodology Political- Ongoing An operational This lens helps Acknowledgmen economic lens that to unpack social t that several analysis (PEA) recognizes the actors’ interests, methods and important power relations, approaches can interaction that and their be used to politics relationships, to unpack the (governance and assess their relationship political factors) influence of between politics and economics support or and economics. can have on the opposition to a success or reform. Other failure of factors that may development be analyzed outcomes or include political reforms. incentives, distribution of political power, relations, vested interests, credibility, social divisions, institutional fragility, as well as the recognition of significant regional and subregional variation in the political- economic landscape. Problem-driven 2009 Ongoing The purpose of The framework Much like PEA, political the problem- also discusses problem-driven economy driven approach and provides PEA can be analysis was to focus on guidance on conducted using a specific issue, structural numerous challenge, or factors, methodological opportunity, as institutional approaches and opposed to variables, and data collection developing actors or tools. broad overviews stakeholders. It of political provides economy guidance on dynamics. how to capture Problem-driven historical legacy, 86 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks Name of Dimensions Approach Start End Objective Measured Methodology PEA was meant map formal and to generate informal useful findings. institutions, conduct stakeholder analysis, and evaluate resources, rents distribution, political stability and legitimacy. Poverty and 2002 Ongoing PSIA is a It assesses the A Social Impact political distributional multidisciplinary Analysis (PSIA) economy impact on approach that analysis welfare or well- combines instrument to being, including different assess the both income methods, types impact of and nonincome of data, and potential dimensions. Key disciplinary reforms on dimensions perspectives is poverty and include recommended. social outcomes identifying The design of According to the stakeholders, the World Bank understanding methodology website, PSIA is transmission and selection of an approach to channels tools is assess the (employment, dependent on distributional prices, assets, the nature of the (and social) access, transfer, reform problem impacts of and taxes) being policy reforms assessing addressed. It is on the well- institutions’ recommended being of effects on to use different groups outcomes, qualitative and of the analyzing quantitative population, impacts, data forms and particularly on contemplating timely the poor and enhancement approaches, and most vulnerable. and employ the right The instrument compensation analytical enables the measures, expertise to analysis of “both assessing risks, address the key the risks to a fostering policy problem. It reform process debate and includes gender and the political feeding back analysis, and net economy risks into policy effects of given of policy choice. reforms. The reforms to User's Guide of 87 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks Name of Dimensions Approach Start End Objective Measured Methodology equity PSIA 2003 outcomes.”[3]. includes a range of tools and instruments that can be used to analyze each of the dimensions. Governance and 1998 Early 2010s An institutional According to the Country-specific Anticorruption diagnostic tool internal GAC surveys that Diagnostics that identifies guidance the targeted survey (GAC) public key dimensions users institutions most that the (households, in need of diagnostic tools firms, reform, the type aimed to unpack enterprises) and of reform included (i) service providers needed, and a identifying (public officials) sense of the corruption by on key obstacles and type, (ii) vulnerabilities, opportunities identifying both including that may affect weak and strong corruption, reform. institutions, (iii) within a assessing the country's cost of each institutions. type of corruption on different groups of stakeholders, (iv) providing insight into the relationship between corruption and governance, (v) evaluating service quality and access and trust in public institutions, (vi) 88 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks Name of Dimensions Approach Start End Objective Measured Methodology identifying key determinants of good governance, and (vii) serving as a strong foundation for policy recommendatio ns and reform. Institutional and 1999 Mid-2000s An institutional Qualitative or Governance diagnostic tool quantitative Review that identified tools that can institutional measure weaknesses that institutional undermine reform. governance performance to prioritize reforms based on institutional and political feasibility. Risk and 2011 Ongoing Risk and Risk and Methodologies Resilience Resilience Resilience and approaches Assessments Assessments Assessments can are determined (Fragility identify the be conducted in by Country Assessments) underlying fragile states Management causes of and nonfragile Unit but may fragility and states. Other include desk violence (drivers domains that study review, of conflict), are studied country expert capacity of include triggers involvement, in- institutions to that can cause country manage them, fragility, conflict, seminars and and World Bank and violence interviews, actions to drivers, semistructured address the resilience factors interviews and identified to help a focus groups, drivers. country cope, World Bank and Group country opportunities to team feedback, address fragility, and discussion conflict, and with violence drivers. government and other stakeholders. 89 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks Name of Dimensions Approach Start End Objective Measured Methodology Citizen 2012 Ongoing A two-way Public hearings, engagement interaction polls, focus between citizens group and discussions, governments or citizen advisory the private bodies, citizen sector within the satisfactory scope of the surveys, Bank Group community interventions scorecards, that gives citizen report citizens a stake cards, grievance in the decision- redress making process mechanisms, with the citizen jury, objective of participatory improving the procurement intermediate monitoring, and final social audits, development participatory outcome of the public intervention. expenditure tracking, participatory planning, community management, community contracting, community sign-off, build citizen and government capacity. The Frameworks’ Influence on Country Engagement, Lending, and Projects Country Engagement There is strong evidence that suggests that governance analysis is included in Systematic Country Diagnostics (SCDs). As part of a 2017 IEG Evaluation on SCDs, 22 SCDs were reviewed. All 22 SCDs took into account aspects of political economy and governance in analysis and program recommendations. The SCDs also extensively analyzed poverty and shared prosperity. There is evidence to suggest that the World Bank undertook 90 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks broad citizen consultation to prepare SCDs. Finally, there were at least two examples where fragility assessments in the Democratic Republic of Congo and Cameroon led to an integration of fragility, conflict, and violence as a cross-cutting pillar or to focus programming on a certain geographical area. Country Partnership Frameworks (CPFs) also incorporate governance analysis. The 2017 IEG Evaluation on SCDs showed that CPFs considered the credibility and legitimacy of the government, incentives for pursuing public interest, and capacity of the public administration in implementing the policies. There was also good use of fragility assessments, when applicable and available. In fact, having fragility assessments appeared to have made it easier for SCDs and CPFs to integrate fragility into country strategies. Sixty-six percent of World Bank staff surveyed for a Fragility Assessment Lessons Learned document stated that the fragility assessment analysis fed into SCDs or CPFs, as well as Country Engagement Notes (World Bank 2018). A wide range of stakeholders were also consulted in CPFs. A 2018 IEG evaluation on citizen engagement found that all 46 CPFs approved between FY15 and FY17 included a reference to the consultation process with civil society, and increasingly with the private sector (World Bank 2018). Governance analysis is included in SCDs and CPFs, but it is not consistently applied. In a study of 10 SCDs, the political analysis was either sector- or problem-specific political economy analysis, or generic. Similarly, although the CPFs may have mentioned political analysis they did not discuss any analytical content or explore the issue in depth. A study of 27 CPFs found that 13 reports included no political economy analysis, while 9 of the 27 studies only discussed the topic superficially. There is also limited use of governance analysis in framing recommendations and programming in SCDs and CPFs. For example, when CPFs address a governance reform, the programmatic recommendations do not necessarily aim to improve the governance outcome. In other words, a CPF may support the creation of a new law and regulation, without building the environment to implement the law. Although citizen consultations are taking place, recommendations are not shifting the country priorities outlined in the CPFs, according to an IEG evaluation on citizen engagement. The engagements were also duplicative, with no subsequent feedback to stakeholders. The evaluation concluded that the World Bank was not using “citizen’s feedback sufficiently to adapt its strategies, portfolio composition, and project design and implementation in response to citizens’ demand.” 91 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks Lending Political economy analysis and poverty analysis is integrated into development policy operations (DPOs). An IEG evaluation on the use of PEA in DPOs found evidence that PEA was being integrated into DPOs, although there was insufficient analysis in sector- specific reforms. Similarly, PSIA analysis was included in some DPOs but not consistently integrated into all. For the most part, between FY11 and FY16, almost all the DPOs (57 out of a 60 sample) had some consultation with stakeholders during the preparation phase. The impact of governance analysis on the design and implementation of DPOs is mixed. There is evidence that strong PEA analysis contributed to the success of DPO implementation and that lack of PEA also led to weak implementation, even when the analysis was readily available. Similarly, while PSIA informed some policy dialogue, in other cases it was unclear how distributional analysis influenced operations. An IEG evaluation on citizen engagement concluded that citizen engagement, even though a requirement, “was not systematically used to support the broader citizen engagement agenda.” Moreover, feedback mechanisms were rarely used, with only 6 of the 40 DPOs mentioning activities that would engage with stakeholders at the end of the operation. Projects There is mixed evidence that governance analysis is being used to inform projects. On the one hand, analysis is being used, but on the other hand, the quality of analysis and report is not consistent. An IEG evaluation found that more than 90 percent of the projects reviewed were based on analysis of formal or informal governance factors. However, another IEG evaluation concluded that the political economy analysis was thin and inconsistently applied. Moreover, PEA was the least likely to be used by operational staff relative to other sources of material. PSIA analysis has been found to help teams reorient country team’s engagement, shift from broader sector to narrow focus, confirm commitments to work on existing efforts, and create more nuanced local criteria for program initiatives. However, Assessing the Use of PSIA concluded that there were issues related to quality. For example, “the section on poverty and social impact was quite broad. The [reports] did not systematically discuss whether support reforms were likely to have poverty and social impact and merit PSIA. They did not discuss whether relevant analytical work existed or there were gaps, and what measures the World Bank would take to mitigate the adverse impact of reforms.” Similarly, an IEG evaluation on citizen engagement found that it was not the norm to use citizen engagement processes and feedback loops to inform project design and trigger course corrections, including how services were being delivered. A staff survey found that 92 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks while fragility assessments were useful for strategic-level purposes and information, they were less useful at the portfolio or operational level. Factors That Influenced the Operationalization of the Frameworks Included There are several factors that ensured that governance analysis is integrated to a certain extent into country engagement, lending, and projects. These factors include the following: • The World Bank required, mandated, or suggested the inclusion of certain governance analysis into World Bank operational processes. For example, although a full formal analysis of political economy is not required or mandated, aspects of PEA need to be conducted to complete the following operational documents: Operational Risk Assessment, Poverty Reduction Strategy Papers, Systematic Country Diagnostic (SCD) Reports, and DPOs. Similarly, the Operational Policy 8.60 for DPOs recommends relevant PSIA analysis be included in all DPOs. Other World Bank products that require PSIA-type analysis include SCDs and Environmental and Social Frameworks. In International Development Association fragile and conflict-state situations, the World Bank made a commitment to conduct fragility assessments prior to all country strategies. Finally, the World Bank also made commitments to mainstream citizen engagement into its operations. The World Bank committed to incorporate citizen engagement processes in 100 percent of projects that can clearly indicate their beneficiaries, by FY18. Moreover, the Environmental and Social Framework introduced an environmental and social standard on stakeholder engagement and information disclosure. • Frameworks that collected data that did not have to be included in World Bank operational products have fallen out of practice. For example, GAC and Institutional and Governance Reviews (IGRs) focused on collecting data and analysis of corruption and institutional performance. These frameworks are no longer being used. Between 2001 and 2005, there were 24 country diagnostics that used the GAC framework. However, few GACs have been undertaken since then. Similarly, through FY05, 20 IGRs had been conducted, but no IGRs have been conducted recently. • Multidonor trust funds or continuous funding helped institutionalize certain frameworks by providing guidance, technical staff, and incentives for country programs to conduct analysis. The multistakeholder Governance Partnership Facility (GPF) supported the implementation of activities in pilot countries 93 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks related to PEA, strengthened demand-side governance and accountability institutions at the national and state levels, implemented global governance initiatives, and generated public goods at the country level. PSIA was also supported by multidonor trust funds until 2016. According to World Bank staff interviews, the availability of the trust funds was an incentive for country programs to undertake PEA and PSIA studies. • Frameworks that did not receive substantial financial capital had less traction. For instance, the first IGR studies were funded by the anchor. The unit covered the time of staff to conduct the studies in Bolivia and Armenia. Once initial funding ran out, the IGRs were funded by the country teams. Similarly, after an initial investment from the U.S. State Department, GAC studies were funded by country programs. According to World Bank staff interviews, lack of funding also played a role in the decline of IGRs and GACs. • Once multidonor trust funds, earmarked and nonearmarked funding, dried up, the frameworks lost their technical champions that were tasked with producing guidance and providing technical support. For example, currently, PEA technical support is provided to country programs if staff time has not been allocated to other projects. PEA support is provided in an ad hoc manner by sharing terms of references, previous reports, or helping to identify consultants to conduct studies. The PEA and PSIA community of practice has become relatively dormant and has not been able to keep up with the ever-changing World Bank knowledge management processes. • Frameworks that are mandated to be included in World Bank operational process, despite the lack of trust funds, continue to be implemented and funded. For example, PSIA studies continue to be undertaken for all DPOs as mandated by the World Bank. Staff are assigned, and resources are allocated for each DPO to conduct PSIAs. Fragility assessments are also actively conducted with support from the Fragility, Conflict, and Violence Global Theme. • Given the diversity of member states, World Bank analytical frameworks needed to be conceptually and operationally flexible. The frameworks needed enough flexibility to be implemented across different contexts, countries, and sectors. This led to governance framework definitions that were intentionally broad and general and methodological approaches that were flexible. Governance frameworks could also be part of informal or formal processes, stand-alone products, or undertaken to inform the World Bank’s operational process. Guidance papers on different frameworks emphasized that the frameworks 94 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks needed to inform or create actionable recommendations and have the ability to scale up or down depending on a country’s needs, time, and resources. To add more specificity, PEA and citizen engagement have introduced sector-specific guidance, while GAC adapted its survey methodology so that it could be thematic and policy needs focused. Systemic Factors Impeded the Full Adoption of the Analytical Frameworks Although governance analysis has been integrated to a certain extent into country engagement, lending, and projects, there are several factors that impede the full adoption of analysis. • Political elites, who are often the World Bank’s clients, have few incentives to implement governance recommendations. Governance frameworks often unpack political economy incentives that are linked to the pursuit of political support or consolidation of power, such as maintaining political position, gaining or controlling lucrative access to sectors, winning or maintaining popular support of citizens, and means to strike deals with power holders. Governance recommendations that arise from analytical frameworks often go against the interests of political elites. As a result, recommendations for reform are often not implemented. • The World Bank is under pressure to maintain good relationships with the government, which is often made up of political elites. Raising politically sensitive topics can put a strain on relationships between country management teams and governments. For example, the inclusion of governance issues in the Egyptian SCD led to a 10-month delay in revising the SCD. Efforts to depoliticize IGR and GAC by focusing on institutions, as opposed to individuals or key stakeholders, were not necessarily effective at ensuring uptake. For example, a GAC Action Plan was not implemented due to a president’s opposition to the findings. Other Country Management Units that have pushed governments to implement governance reforms have been met with resistance. • The World Bank governance framework explicitly recommends limiting the distribution of analytical reports due to political sensitivities, which hinders uptake and use of findings. Fragility Assessments and PEA Guidance suggests a limited distribution of governance frameworks findings. Limiting the distribution of analytical findings, which is often left to the discretion of country management and sectoral staff, can prevent further uptake. 95 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks • Decentralized knowledge management systems within the World Bank makes it difficult to access internal documents. The World Bank has a decentralized knowledge management system, making it difficult to access documents, particularly if they are informal and not public. As a result, new staff are not aware of previous governance studies or diagnosis. • World Bank operational documents focus on identifying and clarifying “what” the World Bank should do as opposed to “how” it should do it. Operational documents, including the project appraisal document, primarily focus on what the World Bank will implement and achieve. Within operational documents, there is no obvious place to document the nuanced, and often political, details that arise from governance frameworks that can lead to project implementation success. World Bank staff interviews shared that there was no space in external documents, such as the project appraisal document, to document who to engage with, when, and why. • There is tension between conducting governance analysis for design purposes and analysis to ensure implementation is successful. An overarching challenge that all the frameworks face is the timing of production of the analysis. Some of the big investments in governance analysis are conducted at the country level for the design stage of an SCD, CPF, or certain project appraisal documents. According to a staff interview, these kinds of analysis often provide “building blocks” and are conducted every two to three years. Information in these analyses may or may not be useful for project-level implementation as they do not have the level of detail or information necessary to operate at the sector or project level. This finding is echoed by survey staff indicating that the fact that fragility assessments were not regularly updated was a limiting factor. According to World Bank staff interviews, quick and rapid sector-level or program-level analysis is hard to do. There is also limited time and budget for ongoing monitoring-style research on evolving country, regional, or community level dynamics. • Misalignment of a study conducted or timing with policy dialogue priorities. The governance analysis is only one input into the policy dialogue around reform. For the analysis to be useful, it needs to be part of an ongoing policy dialogue on reforms and public actions that are under consideration. Findings need to be timed to policy dialogue initiatives to have a more significant impact. • Small windows of opportunities for transformational change. Two World Bank staff interviewed spoke of the small and few windows of opportunity for 96 Appendix E Stocktaking of Governance and Social Development Analytical Frameworks transformational change. The staff stated that there were times that there was an opening for the World Bank to implement transformational programming, but these were few and rare. As a result, it is difficult to implement governance recommendations. Lessons Learned from Other Donors Implementing Governance Frameworks Although many donors have released frameworks and methodologies to assess different aspects of governance, few have been able to institutionalize the frameworks. For example, the Swedish International Development Cooperation Authority released the Power Analysis Tool in 2003, but as of 2012 only nine studies have been conducted using the framework. The U.K. Department for International Development mandated that Country Governance Analysis, which includes political economy analysis, be implemented when preparing country assistance plans. However, the business case model of project preparation and assessment does not have an explicit place to indicate governance issues. As a result, there was no real incentive to conduct PEA during the scoping stage. At the end, PEA studies were dependent on personnel’s interests, particularly of country management and project team leaders. The U. S. Agency for International Development has also made several attempts at institutionalizing political economy analysis. Its latest reincarnation seeks to attempt to create a cultural shift within the agency toward being aware and thinking strategically about political issues in its “Thinking and Working Politically” (USAID 2018). Much like the World Bank, donors also struggled with implementation of framework and quality of analysis. The quality of the donor’s governance reports and the influence on country programs varied. For example, an Overseas Development Institute evaluation of the U.K. Department for International Development’s Drivers of Change framework highlighted that the studies were (i) overly descriptive, (ii) focused on large- scale trends rather than dynamic medium-term factors, and (iii) limited participation and sharing of results. The evaluation also identified the need for staff trainings to draw out concrete program recommendations from studies, and the need to develop further guidance tools. 97 The World Bank 1818 H Street NW Washington, DC 20433