90388 Global Review of Grievance Redress Mechanisms in World Bank Projects Acknowledgements This note was completed in February 2013 and summarizes the findings of a desk review of GRM usage and implementation issues in the Bank’s portfolio. The principal author was Marie Brown, with research assistance from Bruce Jenkins, Patricia Leon and Aaron Seyedian. It was prepared under the guidance of Amar Inam- dar, Manager, OPSOR. A detailed study on GRMs was also prepared for the LCR region by Juan Dumas; this paper includes some of the findings from the LCR report. The following staff provided advice and inputs or commented on earlier drafts: Sanjay Agarwal, Sadig Aliyev, Andre Bald, Christopher Bennett, Paul Berming- ham, Alexandra Bezeredi, Greg Browder, Jacques Buré, Sergio Coltrinari, Kevin Crockford, Ibrahim Dajani, Fei Deng, Juan Dumas, Benedictus Eijbergen, Soulemane Fofana, Parthapriya Ghosh, Nagaraja Harshadeep, Peter Johansen, Kelly Johnson, Markus Kostner, Pilar Larreamendy, Iftikhar Malik, Maria Elena Garcia Mora, Glenn Morgan, Grant Milne, Satya Mishra, Somin Mukherji, Knut Opsal, GNV Ramana, Cordula Rastogi, Julie Rieger, IUB Reddy, Sona Thakur, Miguel Vargas-Ramirez, Harvey Van Veldhuizen, Chaohua Zhang. Global Review of Grievance Redress Mechanisms in World Bank Projects ii Global Review of Grievance Redress Mechanisms in World Bank Projects Contents Abbreviations and Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v INTRODUCTION AND BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 REVIEW METHODOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 RECOMMENDATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 ANNEX 1: Summary Table of GRMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 ANNEX 2: Case Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Contents iii iv Global Review of Grievance Redress Mechanisms in World Bank Projects Abbreviations and Acronyms AFR Africa Region IFC International Finance ARD Agriculture and Rural Corporation Development sector INT Integrity Vice Presidency CAO Compliance Advisor IPP Indigenous Peoples Plan Ombudsman IPPF Indigenous Peoples Planning CPS Country Partnership Strategy Framework CSI Core Sector Indicators ISDS Integrated Safeguards Data Sheet DPL Development Policy Loan ICR Implementation Completion DRP Dispute Resolution and Report Prevention unit (OPCS) ISR Implementation Status Report EAP East Asia and Pacific Region LCR Latin America & Caribbean ECA Eastern Europe and Central Region Asia Region MENA Middle East and North Africa EDU Education sector Region ENV Environment sector OPCS Operations Policy & Country FPD Financial and Private Sector Services Development ORAF Operational Risk Assessment GAC Governance and Anti-Corruption Framework GRM Grievance Redress Mechanism P4R Program for Results Loan HNP Health, Nutrition and Population RAP Resettlement Action Plan IBRD International Bank of SAR South Asia Region Reconstruction and SDV Social Development Development SP Social Protection ICT Information and WBI World Bank Institute Communication Technology IDA International Development Association Abbreviations and Acronyms v vi Global Review of Grievance Redress Mechanisms in World Bank Projects INTRODUCTION AND BACKGROUND 1.01 Effectively addressing grievances from people impacted by World Bank projects is a core component of managing operational risk and improving a project’s results. Grievance redress mechanisms (GRMs) can be an effective tool for early identification, assessment, and resolution of complaints on projects. Understanding when and how a GRM may improve project outcomes can help both project teams and beneficiaries improve results. However, there is little data available on the prevalence, quality, or impact of GRMs in existing World Bank projects. This note provides a snapshot of current usage of GRMs in World Bank projects, a qualitative assessment of selected GRMs, and recommendations for improved risk management via GRM implementation and design. WHAT IS A GRIEVANCE REDRESS MECHANISM? 1.02 A grievance redress mechanism is a locally based, formalized way to accept, assess, and resolve community feedback or complaints.1 They are increasingly used to improve project outcomes through creating more predictable, timely and results-oriented responses to citizen concerns. There are several development and operational benefits of a well designed GRM: • Improve project outcomes at lower cost: GRMs focus on corrective actions that can be implemented quickly and at a relatively low cost to resolve identified implementation concerns before they escalate to the point of harm or conflict. • Help to prioritize supervision: Using citizen feedback, GRMs are a channel for early warning, helping to target supervision to where it is most needed. • Identify systemic issues: As part of a management system, GRMs can be used to CAO Advisory Note “A Guide to 1  identify some systemic implementation issues and trends that need to be addressed. Designing and Implementing Griev- • Promote accountability: Because most GRMs rely to some degree on local people ance Mechanisms for Development and institutions, an effective GRM can help improve local ownership of devel- Projects.” opment projects. INTRODUCTION AND BACKGROUND 1 1.03 Many countries and municipalities have their own GRMs which can receive feedback on a broad range of issues, for example through an Ombudsman office or grievance mechanism tied to a line ministry. “Project-specific” GRMs can be linked to existing institutions or stand-alone entities. They are designed to receive feedback on issues related directly to a project’s design and/or implementation and generally refer grievances on unrelated issues elsewhere. This note assesses project-specific GRMs and does not review local or national grievance entities. THE BUSINESS CASE FOR MANAGING GRIEVANCES 1.04 A recent Goldman Sachs report found that seventy percent of 192 major infrastructure projects around the world were stalled because of land issues. Similarly, the Compliance Advisor Ombudsman (CAO) at the IFC found that “land, water, and labor are the source of the majority of complaints” on IFC projects. 2 Factors such as demographics, the spread of democracy, and advances in commu- nications contribute to increasing demand for accountable, participatory processes —especially on projects with significant impacts on local communities. 1.05 The costs of ignoring such conflicts—or responding too late—are high. Research from the extractives industry found that the greatest costs of conflict with communities are the delays in project execution, the most frequent costs are lost opportunities, and the most overlooked cost is management distraction.3 A core characteristic of an effec- tive grievance mechanism is the ability to identify minor community incidents before they escalate into unmanageable disputes. In many instances, the grievance process can provide the opportunity for resolution via independent mediation or alternative dispute resolution, versus a lengthy court proceeding or compliance investigation. At present, mediation is rarely used to facilitate resolution of disputes on World Bank 2  “CAO at 10” projects. However, it can be a powerful tool under the right circumstances to achieve 3  Davis, Rachel & Daniel Franks, rapid, low-cost, and lasting development solutions. (2011), “The costs of conflict with local communities in the extractive 1.06 When linked to existing country institutions, GRMs can have lasting impact industry.” In: David Brereton, Dante Pesce and Ximena Abogabir, that continues even once Bank engagement ends. Building and strengthening exist- “Proceedings of the First Interna- ing country systems for managing grievances allows for greater impact, improved tional Seminar on Social Responsi- sustainability and an increase in potential value to the Borrower and beneficiaries. bility in Mining, First International Simply using existing systems however, does not automatically strengthen them. Seminar on Social Responsibility in Mining,” Santiago, Chile, (1–13), The decision to use a local or national GRM structure to capture concerns on a Bank 19–21 October 2011. project requires a credibility assessment and, in certain instances, targeted capacity building. The goal is to create stronger, more credible institutions capable of managing risks and conflicts in many different areas. 2 Global Review of Grievance Redress Mechanisms in World Bank Projects 1.07 The Bank has taken steps to strengthen operational grievance management by creating The Dispute Resolution and Prevention Facility provides the Dispute Resolution and Prevention (DRP) in-house support to project teams in the following areas: team within OPCS. DRP was created to strengthen the Bank’s work on grievance redress and support • Practical advice on approaches to better predict and project teams in their supervision and implemen- resolve citizen grievances addressed to the Bank —includ- tation efforts.4 Recognizing some of the findings ing through the use of borrower systems and project on project delays noted above, DRP’s objective mechanisms; is to help the Bank to better support countries • Mediation services through its extensive regional roster to achieve improved development outcomes via of independent experts; • A knowledge platform and skills development program to lower cost, more rapid resolution of operational improve the capacity of project teams to better respond complaints. DRP’s work is complementary to to citizen feedback and complaints. existing initiatives within the Bank, including the GRM-related work of the Social Development anchor, GAC agenda, WBI and others. MANAGING GRIEVANCES ON BANK PROJECTS 1.08 Borrowers and Bank staff focus on grievance redress during both project preparation and implementation. During preparation, the goal is to work with the borrower to better predict and prevent potential disputes before they arise. Risk assessments, structured collaboration, dialogue and stakeholder consultations are an essential part of the toolkit for building trust and credibility. Identifying project beneficiaries (and potential losers) or vulnerable groups is critical to achieving results. Some of this work is built into Resettlement Action Plans and Indigenous Peoples Plans, but there is scope for much broader risk assessments that take into account the concerns of impacted communities, whether related to our safeguards or not. The Forest Carbon Partnership Facility is a good example where the Bank allocates resources to identify issues during project preparation and build capacity with coun- tries before major project investments start to flow. 1.09 Not all complaints can be predicted, but the Bank can design more pre- dictable systems to help staff and borrowers manage potential problems during implementation. Grievance redress mechanisms help the borrower as well as external 4  For additional information stakeholders to have a more systematic and managed approach to addressing questions on DRP’s mandate and work, or concerns. This helps ensure that project-affected people have a place to register see “Addressing Grievances in a complaint or grievance—and helps the project sponsors and government work to Bank Operations: A Risk-Based Approach” circulated to the Board resolve issues early on. GRMs give the Bank and the borrower real-time feedback May 24, 2012. (SecM2012-0245) on issues and concerns and allow for better service delivery and ultimately better INTRODUCTION AND BACKGROUND 3 results. This concept—of using GRMs to enhance results, strengthen accountability, and improve risk management—is embedded into the theory behind Program-for-Re- sults and is increasingly advocated by senior management. Our findings, however, indicate there is room for improving implementation in this area. “GRMs can help not only identify and address implementation problems in a timely manner, but also be the conduit to receive suggestions to improve agency services or project imple- mentation, and can significantly contribute to improving program or project outcomes.” —Ede Jorge Ijjasz-Vasquez, Sector Director, SDN LCR 1.10 Due to the scale and complexity of the Bank’s work, some disputes and conflict between stakeholders are inevitable. Ensuring that access to low cost, relatively rapid processes for dispute resolution, including mediation, facilitation or other informal problem solving, can make a difference between project success and failure. Mediation has been mainstreamed into the operations and legal contracts of many large private sector developers as well as other international financial institu- tions (the IFC and regional development banks all have a complaints handling office that encourages the use of mediation to resolve disputes) but still remains a relatively unknown tool inside the World Bank. Unlike in the private sector, the active promotion of mediation and/or alternative dispute resolution (ADR) as an option for resolving disputes is not part of our standard bidding documents, policies, or procedures. 1.11 Increasingly, project teams working in difficult regions or sectors are looking for independent experts who can provide a neutral forum for resolving project-related disputes. The Natural Resources Management Project in Kenya is just one example where the project team brought in a regionally-based mediator to work with commu- nities and the borrower to agree an action plan for resolving some controversial and challenging issues. Given the increased focus in some regions on fewer, larger-scale projects—especially big infrastructure projects—we anticipate increased demand for ADR options to keep projects on schedule and delivering results. While mediation does not work in all situations, it can be a powerful tool for resolving grievances and creating greater ownership over projects. 4 Global Review of Grievance Redress Mechanisms in World Bank Projects REVIEW METHODOLOGY 2.01 The goal of the review is to provide project staff and managers with: • A quantitative overview of current GRM application in project design; • Qualitative assessment of GRM implementation issues; • Recommendations for improved risk management via GRM design and implementation. 2.02 For the quantitative assessment, DRP staff conducted a desk review of all new approvals in FY2011 (463 total projects; $44 billion in total commitments). The product lines include: IBRD, IDA, GEF, and Recipient Executed Trust Funds (RETF).5 For each project, the team reviewed the PAD and, where available, the Resettlement Policy Framework, the IPP (Indigenous Peoples Plan), or the IPPF (Indigenous Peoples Planning Framework). 2.03 DRP staff also solicited input on GRM case studies from regional managers, safeguards specialists, and SDV specialists and conducted qualitative reviews of 23 projects spread across the six regions. The qualitative reviews are based on interviews with the relevant TTLs and Social Development Specialists along with a desk review of the project documents. We asked a series of yes/no questions designed to measure the quality of GRM design and implementation; these responses are summarized in Annex 1 for all 23 cases. Written write-ups of 9 of these projects provide additional detail (Annex 2). The projects included in the qualitative reviews were approved between 2005 and 2011 to ensure that we captured projects with a sufficient track record of implementation. We used the interviews to understand how 5  Only RETFs over $5 million were GRMs worked in practice and what challenges arose during implementation. Where included, consistent with IEG’s available, we requested supporting documentation in order to verify input received methodology for assessing total from the interviews. We were not able to conduct field visits to verify information projects in a given fiscal year. received from staff or found in project documents. REVIEW METHODOLOGY 5 2.04 Based on the quantitative and qualitative data, we developed recommen- dations for project teams and managers to improve GRM implementation and design. Opportunities for improved project and portfolio risk management are also considered. Throughout this process, we have not considered GRMs as a cure-all for every project risk. Rather, we consider an effective GRM to be one tool among many, and the decision to include a GRM in a project should be risk-based. All the projects reviewed predate the creation of the DRP unit in 2012. This note will thus serve as baseline data against which progress can be measured in future years. 6 Global Review of Grievance Redress Mechanisms in World Bank Projects FINDINGS QUANTITATIVE RESULTS 3.01 GRMs are included in a much larger share of the portfolio than previously thought. Of all FY11 projects, exactly 50 percent include a grievance mechanism in the project design, either as an explicit reference in the project documents6 or because the project triggered the Involuntary Resettlement (OP 4.12) or Indigenous Peoples (OP 4.10) safeguards (see Figure 1). An earlier study based on a limited sample of FY08 approvals found that just 28 percent of projects contained GRMs, thus there has been a significant increase in GRM coverage in recent years.7 OPs 4.10 and 4.12 both require grievance redress mechanisms, either as part of an Indigenous Peoples Plan (IPP), Indigenous Peoples Planning Framework (IPPF), or a Resettlement Action Plan (RAP). Program-for-Results, which was approved in 2012, also requires an assessment of a borrower’s capacity for managing grievances. Figure 1.  World Bank projects with a Grievance Redress Mechanism 6  This includes the PAD, IPP, IPPF, Linked to and/or RAP where applicable. Safeguards 7  “Governance and Anticorrup- 78% tion: A Benchmarking and Learn- No GRM GRM ing Review” produced BY QAG in 50% 50% 2009. The QAG statistic is based on Not Linked a limited sample of FY08 approv- 22% als and the exact methodology is unclear. Therefore, the data are not strictly comparable but can be con- sidered a rough comparator. FINDINGS 7 3.02 However, the relatively broad usage of GRMs in projects is primarily linked to projects that trigger OP 4.10 or OP 4.12. 78 percent of GRMs are included because of these two safeguard requirements. Thus, it is still too early to say that grievance mechanisms are being mainstreamed into project design outside of the safeguards requirements. 3.03 Almost all Category A projects (93 percent, or 38 out of 41) have a GRM. The same number (38 out of 41) triggers either OP 4.10 and/or 4.12 (Figure 2).8 This is consistent with the increased corporate attention and budget devoted to Category A projects, both in the design and implementation phases, and consistent with the risk profile of Category A projects. Figure 2.  GRM Coverage by Categorization 3.04 72 percent of Category B proj- ects included a GRM in project 250 design (173 projects out of 240). Unlike Category A projects, which 200 generally entail “irreversible” risks, there is a much broader spectrum of Total Projects (#) 150 risks within the Category B portfolio. And unlike the Cat A projects, where 72% 100 GRM usage is required under certain circumstances, there is no consistent 50 basis for determining whether the 93% risks in Cat B projects merit a GRM or 22% 0 not. It is unclear whether this is the Cat A Cat B Cat C “right” number, i.e. whether GRMs GRM No GRM are appropriately targeting the right universe of projects based on risks. 3.05 Interestingly, 22 percent of Category C projects contained a GRM in the proj- ect design (14 out of 65). None of these Category C projects were required to have a GRM, and yet they did. To get a better sense for which types of projects included GRMs even when they were not a requirement, we looked at the full universe of projects (not just Category C) that had a GRM not triggered by a safeguards policy, 8  However, two of the projects with and compared the sectoral breakdown of this group with those GRMs linked to OP no GRM do trigger OP 4.12, thus they are technically out of compli- 4.10 and/or 4.12. We found that GRMs triggered by the safeguards were clustered ance with the Bank’s requirements. in infrastructure-related sectors (Figure 3). GRMs not triggered by the safeguards (Figure 4) also covered infrastructure projects (Transport, Energy, Urban, and Water made up 20% of the total), but there was significant coverage within social sectors (Social Protection and Education in particular), Agriculture and Rural Development 8 Global Review of Grievance Redress Mechanisms in World Bank Projects (17 percent), and Economic Policy (9 percent). This is consistent with anecdotal evidence that GRMs are being increasingly used in Community-Driven Development (CDD) projects, and especially cash-transfer projects that would fall under Social Protection or Education. Figure 3.  GRM Triggered by the Figure 4.  GRM not Triggered Safeguards: Where are They? by Safeguards thenot GRM Triggered by the Safeguard: Where are They? Urban 5% Water 5% Energy & Mining ICT Other ARD Social 7% 5% 15% 18% Prot. 21% HNP 7% Urban Transport 10% 16% ARD Econ 17% Policy Water 9% Energy EDU 11% ENV & Mining 9% 14% Other 11% 10% Transport 10% 3.06 In terms of compliance with Figure 5.  GRM Compliance the Bank’s safeguards policies, the data show broad compliance but 180 some gaps. We looked at breakdowns 160 according to Categorization (Figure 5) 140 and policy triggered (Figures 6 and 7). 120 Total Projects (#) 95 percent of the Category A projects 100 that were required to have a GRM did 80 indeed have one (36 out of 38). The 60 comparable number for Category B 40 projects is 89 percent (145 out of 163). 20 Compliance within the Category B uni- verse is clearly weaker than in Category 0 Cat A Cat B A, where almost all projects that were required to have GRM had one. Trigger w/ GRM Trigger but No GRM 3.07 We also wanted to understand whether compliance with the GRM require- ment varied across the two policies, OP 4.10 and OP 4.12. Figure 6 shows the share FINDINGS 9 Figure 6.  Figure 7.  Projects that trigger Projects that 140 OP 4.10 only trigger OP 4.12 only 120 12% 100 80 60 40 5% 23% 20 0 Projects that Projects that Projects that trigger OP trigger OP trigger OP All other Projects that 4.10 only 4.12 only 4.10 and projects trigger OP 4.10 OP 4.12 and OP 4.12 GRM No GRM of projects that trigger one or both policies; Figure 7 indicates compliance coverage within each sub-group. Projects that triggered both policies had the best compliance with the GRM requirement: only 5 percent of these projects had no GRM (Figure 7). Of the 131 projects that triggered only OP 4.12 (Involuntary Resettlement), 12 percent had no GRM. However, of the 31 projects that triggered only OP 4.10 (Indigenous Peoples), a surprising 23 percent had no GRM. The absence of GRMs in even a small number of the projects where their use is mandated indicates a need for further man- agement attention and better monitoring and tracking by project teams. 3.08 Many have asked whether grievance redress mechanisms are also being used in the World Bank’s budget support loans, i.e. DPLs. The Program-for-Results instrument (P4R) is also budget support, but it was approved in FY12 and thus is not 9  Mongolia – Development Policy Credit 2; Indonesia – 7th DPL; India part of our FY11 data set. DPLs are a significant part of the Bank’s portfolio: in the – E-Delivery of Public Services. FY11 portfolio, 15 percent of projects were DPLs (71 of 463), which equated to 26 10  These were listed as “key mea- percent of total lending by volume ($12 billion out of $44 billion total). Only three sures” or “benchmark actions,” i.e. of these DPLs included a grievance redress mechanism in their project design.9 Two actions that would get monitored during implementation but are not of these included specific actions on GRM design and implementation that would be required for disbursement of funds. measured during implementation.10 3.09 Are some regions performing better than others in terms of application of GRMs to their higher risk projects? This is a broad question, which the data can only partially address. Figure 8 shows the share of Category A and B projects by region, along with each region’s share of projects with and without GRMs. Some interesting points: Africa has the most Category A and B projects and MENA has the fewest, 10 Global Review of Grievance Redress Mechanisms in World Bank Projects however almost none of MENA’s Category A and B projects contain GRMs (only 2 out of 9) whereas 70 percent of Africa’s higher risk projects have a GRM (76 out of 109). EAP and SAR similarly “outperform” relative to their peers in terms of GRM coverage within their Category A and B portfolio. 3.10 Figure 8 of course only tells part of the story. What is harder to assess is whether the regions are targeting the “right” projects in terms of using grievance redress mechanisms, and how do regional staff determine what the these projects are. Most regions we interviewed had their own internal list of higher risk projects— based not just on categorization or sector but broader criteria. However, only some regions could identify the specific criteria that would place a project on their internal risk list, and there are no consistent criteria for assessing project risks that are used across the Bank. Triggering of specific safeguards is an important determinant—but not the only one. Staff and managers have significant discretion in determining which projects are deemed higher risk. Despite the increased focus on collecting beneficiary feedback—via GRMs and other means—we found few instances where this feedback made it back into the Bank’s internal system and was factored into the Bank’s risk management and supervision. 3.11 While it is appropriate for Bank clients to operate and manage GRMs, the fact that project-based GRM data rarely if ever reach the TTL illustrates a major lost opportunity for the Bank to improve its own supervision. We found one innovative TTL who identified this problem and designed a simple SMS system for projects that Figure 8.  Category A & B Projects by Region 120 100 Category A & B Projects (#) 80 60 40 20 – AFR EAP LCR SAR ECA MNA GRM No GRM FINDINGS 11 automatically sends all grievances received to not only the project manager, but also the TTL and other “Any GRM where the Bank doesn’t get World Bank staff in real time in an unfiltered format. the data isn’t worth it.” This allows the TTL to monitor and support response —TTL (EAP) and resolution of issues by the implementing agency. The data on grievances received and resolved is then input into the projects’ ISRs and ICRs and used to drive supervision attention. This system was produced quickly (2 months) and cheaply (under $5,000) and has been replicated in a number of projects in this TTL’s portfolio. However, this example is unique. 3.12 As with the regional data, there were some points of divergence when the data is broken down by sector (Figure 9). Of the 22 Category A and B projects clas- sified as Health, Nutrition, and Population (HNP), for example, more than half had no GRMs. On the other end of the spectrum, Water sector projects had very good GRM coverage (21 out of 24 Category A and B projects, or 88 percent). There were only three Governance projects in the sample, but none of them had a GRM—especially interesting given that GRMs themselves are a key tool of demand-side governance in projects. Similarly, HNP bears closer scrutiny: one hypothesis is that HNP projects are not triggering OP 4.10 or 4.12—and therefore are less likely to have a GRM—but still involve other risks that are not being captured by safeguards. Figure 9.  Category A & B Projects by Sector 60 50 Category A & B Projects (#) 40 30 20 10 0 ARD Transp. Energy Urban Water HNP ENV EDU FPD ICT SD SP Governance Econ Policy GRM No GRM 12 Global Review of Grievance Redress Mechanisms in World Bank Projects 3.13 The sectoral and regional breakdowns don’t provide the full picture, and there are other types of projects that could trigger concerns, disputes, and con- flicts—for example CDD projects with large numbers of beneficiaries or budget support loans (i.e. DPLs or P4R) in countries with limited capacity to manage risks. It is important to remind ourselves that it’s not just projects with negative impacts (i.e. involuntary resettlement) that can trigger conflict, but also projects where the direct impacts are overwhelmingly positive (i.e. cash transfers) due to concerns and disputes over who should receive benefits. Similarly, budget support loans that support contested reforms or programs can exacerbate pre-existing conflicts if not managed well; conversely, they also can provide opportunities to strengthen domestic capacity to manage such conflicts. QUALITATIVE REVIEW 3.14 We solicited input for case studies from sector directors and managers, safe- guards advisors, and other staff. We encouraged submission of cases that reflected a wide range of experience, with the goal of identifying and learning from the more challenging issues. In practice, however, most of the cases referred to us by managers were above-average in their GRM implementation and many had already received prior recognition as ‘best practice.’ We thus supplemented the cases referred to us by managers with a random selection of additional cases from the FY11 database. When we randomly selected cases from the 2011 sample, we found that many of the GRMs existed on paper but not in practice. This was consistent with the findings of a review on GRMs commissioned by the LCR Social Development network,11 which found a much wider range of implementation experience. We received over 30 examples and completed interviews on 23, with at least one project from every region. Interviews were conducted with either the TTL or the Social Development Specialist, or both. We solicited cases that were not triggered by OP 4.10 or 4.12 but did not receive any. While the case studies are not a representative sample, several trends emerged from our discussions that are worth highlighting. 3.15 Most project teams understood that strengthening grievance redress capacity at the country level is preferable to creating stand-alone or duplicative grievance systems. In most instances, the Bank team either relied completely on existing grievance structures or paired a Bank-designed grievance system with existing capacity. Both are consistent with two principles of effective grievance redress, i.e. (i) keeping the 11  The review covered 70 projects process as close to the source (i.e. the complainant) as possible and (ii) using culturally in the LCR SDV portfolio approved appropriate systems. However, we found no examples where the Bank worked with since 2008 and was completed in the client to assess the credibility of pre-existing grievance systems against objective 2012. FINDINGS 13 criteria and then used this assessment to help prioritize how to best strengthen griev- ance capacity.12 Many staff acknowledged that existing systems were weak, but their responses to this tended to fall in one of two extremes: either (i) relying on existing systems completely but with limited or no effort to improve them; or (ii) setting up an independent, stand-alone system that reflected best practices but was ultimately not used because it was not linked to existing systems. We found few instances where staff worked with the client to either assess credibility of existing systems or try to link up existing systems to a Bank-designed GRM.13 3.16 Political and cultural context matters. Not surprisingly, we found that in countries with strong domestic support for grievance redress (i.e. via legislation or existing institutions), Bank projects had better GRM design and implementation. India is a good example of this; grievance redress is mandated by law as part of the gov- ernment’s public service delivery responsibilities and all Bank projects in India have a grievance component. Some borrowers are skeptical of the value of a GRM—they don’t want to hear about complaints. Presenting a GRM not as a forum for complaints but rather as an opportunity to improve service delivery helped in some instances. Some TTLs said that when grievance redress is a tough sell to the borrower, they introduced the issue only after trust had been established working on other issues. On other projects, TTLs created stand-alone GRMs in order to ensure some of the basic principles of transparency, fairness, and accessibility. These, however, tended to be less used by the locals. 3.17 In all regions—even those with GRM issues mainstreamed into project 12  DRP has prepared a two-page design—GRM implementation, monitoring, and evaluation is sketchy at best. diagnostic tool, “Evaluating a Only 7 of the 23 projects covered in Annex 1 had data on grievances received—one GRM,” to help clients and staff indication that GRMs may be included in project documents but not put in place. discuss how well existing grievance Staff acknowledged this is a problem and voiced a number of possible causes: (i) mechanisms are working. 13  The latter is happening in sev- Bank budgets are allocated primarily to project preparation, not implementation; (ii) eral projects that have received GRM implementation is rarely included in results frameworks and therefore is not Inspection Panel complaints and monitored; (iii) beneficiaries still strongly rely on informal country systems versus are working to improve grievance Bank-designed GRMs; (iv) Environmental and Social Specialists—who tend to be handling on the ground: Hondu- ras Land Management is a good better versed in GRM issues—are hired by TTLs and are not always asked to join example. supervision missions, and (v) GRMs are just not a priority for clients. 3.18 When a GRM is fully operational in a project and its value is undisputed, the TTL will prove knowledgeable about it and sufficient documentation will be easily available to account for its operation, even if it is not fully developed. On the contrary, when a GRM is not considered essential for the project, the TTL is likely to know little about it and rely on the Social Specialist to meet with policy require- 14 Global Review of Grievance Redress Mechanisms in World Bank Projects ments. In these cases, some additional documentation may describe the GRM but it is not readily accessible. We found many champions of GRMs internally, but little institutional support or guidance for those seeking implementation support (versus support in the design phase). 3.19 In the same vein, projects with thorough descriptions of specific GRMs in their IPPF, IPP, RFP, or RP turned out to have good mechanisms working on the ground. Those that only offer a description—no matter how extensive—of all the options available to citizens in the national legislation to exercise their right to seek redress but do not describe any institutional arrangements to apply them to the project, do not necessarily have good GRMs on the ground. These projects may have been able to “check the GRM box” during the preparation stage but cannot provide infor- mation to understand how many complaints have been filed about the project, who they have been handled, and how the project has improved thanks to this interaction. 3.20 Generally, GRMs appear to be set up only to meet the requirements of OP 4.10 and 4.12. Good GRM practice found in only a few projects seems to stem more from the individual conviction of the TTL or the Social Development Specialist, or from the already existing practices of the borrower, than from an institutional com- mitment of the Bank. SUMMARY FINDINGS 3.21 The following points summarize our key findings from the interviews and group discussions with key Bank staff and managers. In many instances, issues raised during the interviews echoed the quantitative data or provided a more nuanced picture of the Bank’s current approach to grievance redress. GRMs in Project Preparation: 1. Overall usage of GRMs has increased in recent years; currently almost half of all projects include a description of a GRM in the project documents. There is increased management attention to beneficiary feedback systems, including GRMs, and their ability to provide real-time information on project risks and effectiveness. 2. GRM usage among Category A projects is almost universal (93 percent), whereas it is much lower among Category B projects (only 73 percent have GRMs). Category A projects tend to have better documentation of risks at approval and almost all contain a GRM in the project design. GRM references in Cat A projects are explicit and often detailed whereas most Cat B projects do not FINDINGS 15 include explicit references.14 Aside from the two safeguards policies, there are no risk-based criteria used to determine whether a project may benefit from a GRM. 3. GRMs are primarily tied to projects that trigger OP 4.10 or 4.12. Most (but not all) staff viewed GRMs as something you do only if your project triggers OP 4.10 or 4.12. There is limited consideration of either the development impact or cost implications of using a GRM to address other types of risks. In some instances, mandated usage turns the GRM into a “box-checking” exercise that is not truly owned by the Bank or the client. 4. Project design documents emphasize reliance on a country’s existing grievance systems but do not explicitly identify the strengths and weaknesses of those systems. Assessing credibility to the users is not something the Bank has artic- ulated or attempted to document in a systematic way. The absence of a capacity assessment is often reflected in implementation, for example in projects that underestimated social risks or fail to learn from (and link to) existing traditional ways of managing grievances. GRMs in Implementation: 1. GRMs exist on paper but not always in practice. Many PADs reference a GRM, but the findings from our random sampling indicate that many have not been put in place. Only 7 of 23 projects in the qualitative sample were able to provide data on grievances received and resolved. Reasons vary: lack of dedicated financial 14  We defined a project as having resources, project staff overburdened and/or uninformed, GRMs are rarely in an “implicit” reference to a GRM results matrices and therefore are not monitored, etc. We found that a key driver if it triggered OP 4.10 or OP 4.12 but had no explicit reference in the of whether a GRM moved from paper to reality was whether both the TTL and PAD to either a grievance or feed- the borrower are convinced of the GRM’s value to the project. When the business back mechanism. An explicit ref- case is not clear to those in the driver’s seat, the GRM often falls by the wayside. erence means the PAD or support- 2. Of the GRMs in existence, many are not being used. This was most evident ing documents described the GRM, irrespective of whether OP 4.10 or in projects that set up stand-alone GRMs and didn’t link them to existing struc- 4.12 was triggered. tures (i.e. village chiefs, local councils, etc). In Vietnam and China, for example, 15  There is a Core Sector Indica- we found GRMs in place but most project beneficiaries with concerns sought tor (# grievances received/# griev- redress through existing channels such as village leaders versus the ‘formal’ ances resolved) that is an option- al metric staff can track in certain project GRM. projects. When selected, CSI data 3. The Bank does not systematically document or measure GRM implemen- are intended to be incorporated tation. There is a wealth of literature on GRM design principles, best prac- into a projects’ results matrix and tices, and “how-to” notes. However, we found no accessible documentation tracked in ISRs and ICRs. As of early 2013 however, no staff had on experience with GRM usage and impact, either in Implementation Status opted to include the CSI on griev- Reports, Implementation Completion Reports, Country Partnership Strate- ances received and resolved. gies, the ORAF, or the Corporate Scorecard. GRM data is not systematically tracked at either a project level (via ISRs and ICRs) or a corporate level. 15 16 Global Review of Grievance Redress Mechanisms in World Bank Projects GRMs and Internal World Bank Risk Management: 1. There is no link between GRM data received on projects and internal Bank risk management on those projects. Because borrowers ultimately implement projects, there is no automatic link that sends grievances filed on individual projects to the relevant Bank staff. While it is appropriate for the implementation agency to be the first line of response for most grievances, the Bank could use grievance data to better prioritize its supervision efforts. As one TTL noted, “any GRM where the Bank doesn’t get the data isn’t worth it.” 2. Basic knowledge of grievance mechanisms among staff is mixed. With some notable exceptions, TTLs often had only cursory knowledge of GRM design and implementation techniques and no understanding of how a GRM could help with overall risk management. Many TTLs relied on Social Development specialists to handle all aspects of GRM design. 3. Bank systems do not have the capability to quickly pull up data on GRM usage or implementation. Many of the Bank’s internal systems (Business Warehouse, Business Intelligence, Client Connections, etc) are hampered by inconsistent and poor quality data combined with complicated user interfaces. The ORAF is an attempt to correct some of these problems, but data quality is variable and the data is not searcheable. None of our existing systems (ORAF, Business Warehouse) or reports (ISRs, ICRs, or CAS) systematically track the existence or content of grievance mechanisms. With the exception of a small number of high profile projects,16 monitoring of GRM implementation is non-existent, and data on GRMs can only be found by manually searching individual project documents. This means Bank staff and managers can’t do quick portfolio reviews, systematic knowledge sharing on project implementation isn’t possible, and basic questions about regional or sectoral portfolios are hard to answer. AVAILABLE GRM GUIDANCE FOR STAFF There is some in-house material on GRM principles and best practices and a wealth of information from external sources. While there is a good amount of material on 16  Mumbai Urban Transport Proj- principles and best practices related to GRM design, we found limited guidance on ect is a good example, as a detailed monitoring report of the resettle- GRM implementation; DRP created the first three documents to respond to this need. ment process, including the GRM Key internal and external resources on GRM issues are listed below: structures, was completed by an independent consultant. Internal Resources: Items 1-3 can be found on the internal DRP website (furl “disputeresolution”) FINDINGS 17 Items 4-6 can be found at the Social Accountability and Demand for Good Governance (DFGG) internal website (furl “dfggdb”) 1. The World Bank’s Approach to Grievance Redress in Projects: A practical framework that can be used by clients and project teams to (i) assess anticipated grievances, (ii) review the credibility and capacity of existing grievance systems, and (iii) build or strengthen grievance redress capacity. 2. Evaluating a Grievance Redress Mechanism: A two-page checklist of questions to evaluate an existing grievance redress mechanism. 3. Grievance Redress Mechanisms: Frequently Asked Questions. 4. GAC How-to Notes: Feedback Matters: Designing Effective Grievance Redress Mech- anisms for Bank-financed Projects, Part 1 & 2. 5. Using Demand-Side Governance Approaches to Identify and Manage Risks in Projects 6. Beneficiary Feedback and Third Party Monitoring in Bank-Financed Projects. 7. IFC CAO: A Guide to Designing and Implementing Grievance Mechanisms for Development Projects. 18 Global Review of Grievance Redress Mechanisms in World Bank Projects RECOMMENDATIONS Based on our findings, we offer the following recommendations to enhance the impact of grievance redress mechanisms and, more broadly, improve service delivery and risk management on projects. Many of the recommended actions were proposed by Bank staff and managers in our discussions with them; others are our own. The Dispute Resolution and Prevention team is already working on a number of immediate issues and is prepared to provide support on other items, as needed. 1. Create diagnostic tools to support GRM implementation and strengthening of existing country capacity in a credible manner. The strong push to support and strengthen existing country institutions for grievance redress is a good start, but staff needs practical diagnostic tools to do this in an informed way. Products that staff said would be useful include: case studies of GRM implementation, evaluating existing GRMs, linking a GRM to existing country systems, and a man- ual of basic GRM principles and procedures. (Action: Completed – DRP Team) 2. Improve risk assessment within the Category B portfolio in order to better prevent and avoid conflict. We need to assess the portfolio—Category B projects in particular—and risks that could lead to conflict beyond those required in OP 4.10 and 4.12 (for example, projects that take place in an area where land tenure is weak or CDD projects with large cash transfers in areas with weak governance). The objective would be to identify where GRMs could add value. (Action: Regional and Sector Managers, with DRP support) 3. Use feedback received via GRMs to prioritize supervision. Having real-time information from local citizens on project implementation is an obvious tool to help Bank teams target and prioritize limited supervision budgets. This is espe- cially relevant for geographically-dispersed projects and/or projects with large numbers of beneficiaries. However, very few TTLs receive GRM data—even in summary form—that would allow them to make small course corrections during implementation. (Action: Regional Managers) 4. Create incentives for monitoring and improving GRMs during implementation. At present there is no systematic monitoring or evaluation of GRM implemen- RECOMMENDATIONS 19 tation at the project or corporate level. One relatively easy option would be to require all projects with a GRM to track the existing CSI related to grievances in project ISRs and ICRs. At present, CSIs are optional selections for project teams; as of early 2013, no team had chosen to track the grievance-related CSI (Action: OPCS, Regions). 5. Improve our own internal handling of complaints on projects. The Bank could improve the predictability around its own handling of complaints by encouraging all complaints to be routed to the TTL, setting clear timetables for responses, 17  A similar procedure is in place and putting in place corporate tracking to ensure all grievances are responded to for the handling of complaints relat- ed to procurement on Bank proj- and addressed (if not ultimately resolved) in a timely manner.17 (Action: OPCS, ects (BP 11.00). Regions). 20 Global Review of Grievance Redress Mechanisms in World Bank Projects ANNEX 1: Summary Table of Grievance Redress Mechanisms ANNEX 1: Summary Table of Grievance Redress Mechanisms 21 Tanzania: Zanzibar Urban Project name Services Project Uganda: Electricity Sector Development Project Region AFR AFR Country Tanzania Uganda Project number P111155 P119737 Approval date February 2011 June 2011 Closing date June 2016 February 2017 World Bank commitment ($ mil) $38 $120 Total project amount ($ mil) $38 $153 Instrument SIL SIL Environmental categorization A A Safeguards triggered EA, PCR, IR EA, NH, PCR, IR, F Sector Flood protection, sub- Transmission/Distrib electricity, public admin national govt admin, solid waste mgmt World Bank contact(s) Andre Bald Somin Mukherji Questions – Grievance Redress Mechanism Is there a register or database that records Yes Not specified complaint handling? How many complaints have been received? Data not available Data not available Are there rules or procedures to ensure timely If not resolved in 14 days, Village GC to respond within 3 days. If not resolved, District Land responses? referred to Regional Tribunal to decide w/in 1 week. Manager of Resettlement Coordination Unit who seeks resolution w/in 21 days. If no resolution, recourse to courts. Is the GRM appropriately advertised and PAPs provided notification Yes communicated to project-affected people? of GRM Do multiple uptake channels exist? Yes, through local leaders Verbal and written forms specified (Shehas) or grievance officer. Form not specified Can anonymous grievances be filed? Not specified Not specified Is there a system to categorize, assign priority, Not a formal system Yes, Resettlement Unit classifies as local (route to Village GC) or central and route grievances to the appropriate entity? (to Resttl Unit) Are complaints acknowledged in writing? Not specified Not specified Does the acknowledgment outline the GRM Not specified Not specified process, provide contact details and indicate how long it is likely to take to resolve the grievance? Are there clear timetables that are publicly Yes available? Is action taken on every grievance? Yes Not specified Are there indicators to measure grievance Database is to record Indicators not specified but indep monitoring required to look at overall monitoring and resolution? resolution and time required status of PAPs for resolving each complaint Is there right to appeal? If yes, are GRM users Unresolved disputes May appeal to District Land Tribunal and, if not resolved, to courts. informed about this right? forwarded to Resettlement Weakness of Land Tribunals noted. Coordinating Unit. PAPs informed of right to appeal to courts. Innovations/Addl Info Utilizes a 3 tier grievance process all resting on local/national systems. Village GCs primary, District Land Tribunals serve as 1st referred instance, then courts. Project acknowledges some weaknesses in existing country systems (for example, complainant generally needs to pay for travel, food, allowance for District Land Tribunals to hear complaint). 22 Global Review of Grievance Redress Mechanisms in World Bank Projects Chad: Local Development Indonesia: Upper Cisokan Lao PDR: Nam Theun 2 Program Support APL II Cameroon: Sanitation APL Lake Victoria Phase II, APL 2 Pumped Storage Hydropower AFR AFR AFR EAP EAP Chad Cameroon Burundi, Rwanda Indonesia Lao PDR P113030 P117102 P118316 P112158 P076445 March 2011 June 2011 June 2011 May 2011 March 2005 June 2015 August 2015 June 2017 December 2018 2017 $25 $30 $30 $640 $20 mil (IDA grant); $50 mil (IDA PRG); $200 mil (MIGA guarantee) $77 $39 $30 $800 $1,450 APL APL APL SIL IL B B B A A EA, PM, PCR, IR, PIW EA, IR EA, NH, F, PM, PCR, IP, IR, EA, NH, PCR, IR, SoD EA, NH, PM, CP, IR, IP, DS, Dams, PIW PID, PDA, F Agriculture & fishing, Sanitation, public admin. Water sanitation and flood Large hydro Large hydro health, prim edu, water protection, agriculture and supply, public admin. fishing, public admin. Soulemane Fofana Miguel Vargas-Ramirez Nagaraja Rao Harshadeep Peter Johansen (TTL) Lis Nainggolan (SDV) Not specified Not specified. Notes Grievances to be registered Yes Yes that Commission records but unclear at what stage complaint and sends to Local and by which entity Resettlement Committee None to date Data not available Data not available None to date Data not available Not specified Not specified No, "response time depends Yes, 7 days to classify and refer to Yes on issue," but should be agencies which respond w/in 14d, addressed efficiently GTF responds to complainant w/ in 21 days Not specified Not specified PAPs directly informed but Yes unclear re broader outreach and comms Not specified No, RPF states complainants Multiple channels noted Yes, in writing, phone, email, fill out form and send to (local chief, PPCT, PAP SMS Commission Committee), but process and relationships could be further clarified. Form of complaints (verbal, written, etc) not specified Not specified Not specified Not specified Not specified Not specified Not specified Not specified Not specified, but implied in Yes role of PAP Representative Committee (inform PPCT of serious issues, refer cases to GRC) Not specified Not specified Not specified No (notice sent after decision if issue not addressed on the spot) Not specified Not specified No acknowledgment No initial acknowledgment specified Not specified Not specified Yes Not specified Not specified Not specified Yes Yes Not specified Not specified Not specified Yes Yes, both internal and external monitoring Yes, to courts. May appeal to courts only after Yes, if not satisfied by Not specified, may go to court Yes, to District Grievance rejection by Commission decision of local chief, may Committee, then to appeal to courts Provincial Court In addition to reference to Project provides technical local and traditional bodies, 2 assistance for mediating structures noted: Commission committee, legal counseling, on Findings and Evaluation independent monitoring agency, and a Local resettlement external compliance monitoring Committee. (continued on next page) ANNEX 1: Summary Table of Grievance Redress Mechanisms 23 (continued) Project name China: Hubei Yiba Highway Project Vietnam: Trung Son Hydro Region EAP EAP Country China Vietnam Project number P101258 P084773 Approval date March 2009 April 2011 Closing date December 2015 December 2017 World Bank commitment ($ mil) $150 $330 Total project amount ($ mil) $2,194 $412 Instrument SIL SIL Environmental categorization A A Safeguards triggered EA, NH, PCR, IR EA, NH, PM, PCR, IR, IP, DS, PIW Sector Roads and highways Power World Bank contact(s) Fei Deng (current TTL), Pilar Larreamendy Chris Bennett (former TTL) Questions – Grievance Redress Mechanism Is there a register or database that records Yes Yes, maintained and reviewed by an independent complaint handling? entity How many complaints have been received? < 300 Data not available Are there rules or procedures to ensure timely Yes responses? Is the GRM appropriately advertised and Yes Yes, leaflets include info on both the Bank-designed communicated to project-affected people? GRM and Vietnamese GRM Do multiple uptake channels exist? Yes Yes Can anonymous grievances be filed? Yes, web, SMS, in person Not specified Is there a system to categorize, assign priority, Yes, but none received Yes and route grievances to the appropriate entity? Are complaints acknowledged in writing? Yes Not specified Does the acknowledgment outline the GRM Not specified Yes process, provide contact details and indicate how long it is likely to take to resolve the grievance? Are there clear timetables that are publicly Yes Yes available? Is action taken on every grievance? Client says yes Not specified Are there indicators to measure grievance Yes No monitoring and resolution? Is there right to appeal? If yes, are GRM users Not specified Yes informed about this right? Innovations/Addl Info 24 Global Review of Grievance Redress Mechanisms in World Bank Projects Turkey: ECSEE APL#6 Azerbaijan: Highway 3 Azerbaijan: Highway 2 ECA ECA ECA Turkey Azerbaijan Azerbaijan P110841 P118023 P094488 August 2010 May 2010 January 2006 December 2015 March 2015 February 2014 $220 $242 $675 $240 $356 $1,028 APL SIL SIL B A A EA, IR EA, IR EA, IR Transm & distrib electricity Roads & highways Roads & highways Sergio Augusto Gonzalez Coltrinari Jacques Bure Jacques Bure Not specified. Appeals filed with court Yes Yes Data not available No complaints received so far as land acquisition 33 and civil works on sections with potential impact to population have not started yet 15 days to complain upon filing Yes, framework is defined in the RAP and also in Yes, framework is defined in the RAP and also in of expropriation lawsuit; 15 days the Presidential decree on Review and Solution of the Presidential decree on Review and Solution of to challenge court-specified Complaints. Complaints. compensation Affected persons notified of Yes Yes expropriation process and rights. Expropriations announced in local papers Complainants either directly negotiate Yes, there are number of available channels such as Yes, there are number of available channels such as or appeal to court directly applying to Road Agency or through Monitoring directly applying to Road Agency or through Monitoring Specialist and Supervision Engineer. There are also Specialist and Supervision Engineer. There are also many cases when grievances are directly submitted to many cases when grievances are directly submitted to contractors and resolved without escalating to higher contractors and resolved without escalating to higher level. level. No Yes Yes No. Only narrow range of grievances There is not a formal categorization system—but if There is not a formal categorization system—but if allowed, appeals filed with court solution of complaints depend on other agencies, solution of complaints depend on other agencies, the the road administration forwards the grievances road administration forwards the grievances accordingly accordingly and supervise till the issue addressed. and supervise till the issue addressed. Not specified Not specified Not specified Not specified Not specified Not specified Yes, 15 days to appeal Yes, complaints that do not require investigations are Yes, complaints that do not require investigations are replied to within two weeks. Complaints requiring replied to within two weeks. Complaints requiring investigation are replied to within a month. investigation are replied to within a month. Yes, on every appeal filed Yes Yes Not specified Yes, reports of a supervision consultant and monitoring Yes, reports of a supervision consultant and monitoring consultant consultant Yes, only to courts Yes Yes Complaints 100% through courts and International consultant was involved under the project Turkish legal system, no additional to support the road administration in designing GRM. GRM process. Local NGOs were also involved in channeling complaints to respective agencies. (continued on next page) ANNEX 1: Summary Table of Grievance Redress Mechanisms 25 (continued) Kyrgyz Republic: National Road Colombia: Rio Southern West Kazakhstan: South Rehabilitation (Osh Bogota Environment Bank Solid Waste Project name West Roads Project transport) Infrastructure Management Region ECA ECA LAC MENA Country Kazakhstan Kyrgyz Republic Colombia West Bank Project number P099270 P107608 P111479 P105404 Approval date April 2009 November 2009 December 2010 May 2009 Closing date June 2015 December 2014 June 2016 December 2014 World Bank commitment ($ mil) $2,150 $25 $250 $12 Total project amount ($ mil) $2,500 $31 $487 $26 Instrument SIL SIL SIL SIL Environmental categorization A B A A Safeguards triggered EA, IR EA, IR EA, NH, PM,PCR, IR EA, PCR, IR Sector Roads & highways/ Roads & highways/ Wastewater treatment, transportation transportation flood protection, public admin. World Bank contact(s) Jacques Bure (TTL) Cordula Rastogi Greg J. Browder Ibrahim Dajani (TTL) Lola Ibragimova (SDV) Questions – Grievance Redress Mechanism Is there a register or database that records Yes Yes Yes Yes complaint handling? How many complaints have been received? < 10; includes group 6 Data not available Multiple; exact data not grievances available Are there rules or procedures to ensure timely Yes Yes Yes, 15 days. If cannot Yes responses? address w/in 15 days, partial response Is the GRM appropriately advertised and Yes Yes Yes communicated to project-affected people? Do multiple uptake channels exist? Yes Yes Yes, website, Yes telephone, main office, neighborhood office Can anonymous grievances be filed? No Yes Not specified Yes but none received Is there a system to categorize, assign priority, No specific No (informal) Time bound routing Yes and route grievances to the appropriate entity? categorization and process to appropriate priority authority. Unclear on prioritization Are complaints acknowledged in writing? Yes Only complaints Not specified Not required; written involving significant or verbal confirmation funding acceptable Does the acknowledgment outline the GRM No Not specified Yes process, provide contact details and indicate how long it is likely to take to resolve the grievance? Are there clear timetables that are publicly Yes Yes Yes Yes available? Is action taken on every grievance? Yes Yes, requires follow up Yes on each complaint Are there indicators to measure grievance Yes, progress reports Yes Monthly status reports Yes monitoring and resolution? Is there right to appeal? If yes, are GRM users Yes, court appeals are Not explicit, can go to Not specified Yes, but only after informed about this right? typically used courts review and negotiation Innovations/Addl Info GRM is primarily Very proactive PIU Utilizes existing WB encouraged getting designed to channel and direct high-level grievance procedures political commitment grievances related to govt presence. State of client. upfront from the local the RAP although it Secretary has directly mayor AND neighboring has also been used to attended consultations. mayors. channel issues related to non-resettlement impacts from civil works 26 Global Review of Grievance Redress Mechanisms in World Bank Projects India: Mumbai Urban India: Vishnugad India: National Ganga River Transport Project (MUTP) Hydropower Basin Project India: Tamil Nadu Empowerment Project SAR SAR SAR SAR India India India India P113028 P096124 P119085 P079708 June 2010 June 2011 May 2011 July 2005 June 2015 December 2017 December 2019 September 2014 $971 $648 $1,000 $120m ($154m add) $430 $922 $1,556 $159 SIL SIL SIL SIL B A A B EA, IR EA, NH, PCR, IR, F, DS, PIW EA, PCR, NH, IR, IP, PIW EA, IP, PM, CP Railways, public Power, renewable energy Agriculture/social services administration Satya Mishra, IUB Reddy Parthapriya Ghosh Parthapriya Ghosh Kevin Crockford Yes Yes Yes Yes, at District/ State levels, not in each village ~ 3700 Data not available 100+ No central database Yes Yes, 2 weeks Yes, 15 days Partial Not avail Yes, word of mouth and brochures Yes Varied across villages; contact information clearly presented Written only Yes Yes Yes Not specified Yes, but none filed to date Not specified Yes Yes Yes Yes No (informal) Not specified Yes Yes After complaint addressed Yes Yes, the aggrieved person is informed Yes No initial ackn about the process and time required either by the NGO contract for RAP implementation or Senior Manager Social and Environment of THDC who is also the member secretary of the GRC. Not specified Yes Yes, 15 days Yes, in some villages; no at District/State levels Yes Yes and PAPs are informed in writing. Procedures say yes Unclear Yes Yes, M&E consultants keeps track of Process says web accessible Not formalized, but sig monitoring the complaints filed and decisions database taken. They also evaluate whether PAP is happy or not with the decision. Yes Yes Yes Not formal, but Yes Participatory mechanisms create space for addressing grievances without formal grievance procedures. (continued on next page) ANNEX 1: Summary Table of Grievance Redress Mechanisms 27 (continued) India: Second Madhya Pradesh India: Eastern Dedicated Pakistan: Floods Emergency Project name District Poverty Initiatives Freight Corridor Cash Transfer Project Region SAR SAR SAR Country India India Pakistan Project number P102331 P114338 Approval date June 2009 May 2011 March 2011 Closing date December 2012 June 2017 June 2013 World Bank commitment ($ mil) $100 $975 $125 Total project amount ($ mil) $110 $1,458 $580 Instrument SIL APL Environmental categorization B A C Safeguards triggered EA, PM, IP, NH, F EA, IR, PCR Sector Agro-industry/ agriculture Railways World Bank contact(s) Kevin Crockford Benedictus Eijbergen Kelly Suzanne Johnson, Iftikhar Malik Questions – Grievance Redress Mechanism Is there a register or database that records Not at village level but at PFT, Yes Yes complaint handling? district and state How many complaints have been received? Data not available Data not available 1.1 million appeals/ 5500 complaints (see case study) Are there rules or procedures to ensure timely Partial Yes Yes responses? Is the GRM appropriately advertised and Contacts made clear but GRM Yes Yes communicated to project-affected people? process unclear Do multiple uptake channels exist? Yes Yes Yes Can anonymous grievances be filed? Yes Not specified No Is there a system to categorize, assign No (informal) Yes Yes priority, and route grievances to the appropriate entity? Are complaints acknowledged in writing? After complaint addressed Yes Yes Does the acknowledgment outline the GRM No initial acknowledgment No Yes process, provide contact details and indicate how long it is likely to take to resolve the grievance? Are there clear timetables that are publicly No Partial No available? Is action taken on every grievance? Yes Yes Yes Are there indicators to measure grievance Yes Tracking, but unclear on Yes monitoring and resolution? indicators Is there right to appeal? If yes, are GRM users Not formal, but Yes Yes Yes informed about this right? Innovations/Addl Info Participatory mechanisms create Project-level Ombudsman space for addressing grievances serves as appeals body without formal grievance empowered to review and procedures. recommend resolution. Two- level grievance committees with CSO reps enhance accountability and legitimacy 28 Global Review of Grievance Redress Mechanisms in World Bank Projects ANNEX 2: Case Studies ANNEX 2: Case Studies 29 30 Global Review of Grievance Redress Mechanisms in World Bank Projects Kazakhstan SW Road Project Overview At the time of approval, this was one of the largest investment loans the Bank had ever done. The team was under pressure from the Borrower to finalize the project details and start implementation quickly. As a result, the initial EIA and RAP had some flaws, including a significant underassessment of social impacts and project affected persons. Bank management—with strong support from the MD on down—worked proactively with the Borrower to improve the documents prior to approval. How- ever, once implementation began, it became clearer that social risks—which had been deemed relatively minor at Project Details: first —were in fact going to be considerable. In particular, $2.125 billion IBRD Investment Loan to upgrade 1,062 the project required mobilization of a massive work force km section of the old Silk Road trade route linking (over 35,000 people employed directly or indirectly) which China to Russia and Western Europe. The total project generated significant issues that needed to be managed. cost is $7.5 billion. In addition, there were several requests for an Inspection Panel investigation. • Project Approval: April 2009 • TTL: Mr. Jacques Bure GRM Structure GRM Rationale During implementation, the Bank team realized they needed The project triggers OP 4.12. a much more structured grievance mechanism to handle the questions and complaints coming in on the project. They put in place a multi-level structure that encouraged immediate resolution of issues on the ground but created access to more senior authorities to handle issues that could not be resolved on the ground. The contractor and PIU focal points were encouraged to log all complaints on the spot and resolve immediately. Grievances that couldn’t be immediately resolved on the ground by the contractor or PIU focal points moved to the local multi-stakeholder committee. Issues that could not be resolved at that level would move to a national multi-stakeholder committee. ANNEX 2: Case Studies 31 Implementation Issues In practice the ad hoc local and national com- “You understand better because you have mittees were never formed. Instead, beneficia- information coming from the field with the real ries went to their local authorities since that concerns of the people.” is what they were used to doing. Some cases —Jacques Buré, TTL were referred to the local Kazakh courts. Thus, a key implementation challenge was creating the conditions necessary for locals to use the Bank-de- signed grievance system, and to appropriately link this system to the traditional practice of resolving issues through local authorities. The numbers of complaints received by the Bank-designed system (less than 200 total) indicates that this remains a challenge. National multi-stakholder cmte (NGOs, community members, borrower) Local multi-stakeholder cmte (NGOs, community members, borrower) Grievance focal point Grievance focal point (contractor on site) (individual within PIU; independent) However, once complaints did start coming in, both the Bank team and the client acknowledged the value of the feedback. According to the TTL, even the client, who was initially skeptical, took the GRM seriously. While only a small portion of the total complaints were lodged with the Bank-designed GRM (< 200 total), the TTL said the system was very effective at picking up many of the indirect impacts of the road on both the local communities and the more than 35,000 people indirectly employed by the project. Employee issues that were raised and addressed via the GRM included low salaries and sexual harassment. In addition, there were compliance concerns that needed to be addressed at a higher level, related to improper land acquisition on behalf of the government. 32 Global Review of Grievance Redress Mechanisms in World Bank Projects Lessons Learned The project team has identified a number of opportunities for improved outcomes: • Better up-front assessment of the potential social impacts and risks related to the workforce; • Need to do a more structured assessment of capacity and credibility of existing GRM mechanisms in the Kazakh system; • Train project staff in mediation and conflict-resolution skills prior to implemen- tation to better manage discussions with both the Borrower and the impacted communities; • Going forward, improve the link between the Bank-designed GRM and the tradi- tional systems for resolving disputes. One basic step is to ensure all complaints lodged through local authorities are logged and tracked, and that data on reso- lutions is made public. To To KAZAKHSTAN 60°E 65°E Tyumen‘ 70°E 75°E Novosibirsk 80°E 85°E SOUTH WEST ROADS PROJECT 55°N To Chelyabinsk To Kurgan Chistoe Karakuga To RUSSIAN Novosibirsk 50°E 55°E FEDERATION l NORTH Petropavlovsk Pnirtyshskoe Toby To Kaerak Samara RUSSIAN FEDERATION To KAZAKHSTAN lga Ufa Vo Kokshetau To Barnaul To Saratov To Orenburg Kostanay Ruzaevka Pavlodar Ob Pogodaeva Krasny ‘ AKMOLA PAVLODAR Aul l Ura Kamenka Zhaksy Ekibastuz Er 50°N 50°N Ural'sk ti km 102km 102 Shiderty s( Esil (Ishim) Öskemen KOSTANAI ASTANA Irt sh y ) Chapaevo (Ust-Kamenogorsk) Semipalatinsk Zhaiyk WEST Aktobe (Ura KAZAKHSTAN Karabutak Arkalyk Karagandy EAST Lake l) gh a i Zaisan Tor 273 273km km Turgay KAZAKHSTAN Vo Embi 215 km 215 km Ayakos ATYRAU lga m Zhe AKTOBE Shalkar To Astrakhan‘ Atyrau Zhezkazgan Kotyaevka Balkhash KARAGHANDY Aral To 45°N Leps 45°N i Dostyk Urumqi su Lake Saryshaghan ry Balkash ar Sa K at al To Urumqi Beyneu Baikonur Ile ALMATY Taldykorgan Beyneu Aral Kyzyl-Orda Shu KYZYLORDA (Ch 1062 km 1025 u) Aktau Sea Khorgos MANGYSTAU ZHAMBYL 205 km To Korla Ca Zhanaozen To Nukus rd Ta Sy las ar 480km 464 km Shu sp iya Almaty Ferry to Baku Turkistan ian Kordai 80 km Taraz 301km CHINA Bekdash Merke UZBEKISTAN To Naryn AZERBAIJAN Shymkent 203 km 123 km KYRGYZ Sea 40°N SOUTH Zhibek R E P. 80°E Zoly KAZAKHSTAN 40°N BORDER CROSSINGS MARITIME PORTS To WESTERN EUROPE – WESTERN CHINA CORRIDOR TURKMENISTAN Bukhoro To Dushanbe MAIN ROADS RAILROADS SOUTH WEST ROADS PROJECT TAJIKISTAN MAIN CITIES AND TOWNS SOUTH WEST ROADS PROJECT RESTRUCTURING OBLAST CAPITALS OTHER ROAD DEVELOPMENT CORRIDORS: This map was produced by the Map Design Unit of The World GALASY (CITIES WITH EBRD AND GoK FINANCING 0 100 200 300 Kilometers Bank. The boundaries, colors, REGIONAL STATUS) denominations and any other GoK FINANCING information shown on this map do NATIONAL CAPITAL MARCH 2011 IBRD 36239 not imply, on the part of The ADB, IDB AND GoK FINANCING 0 100 200 Miles AFGHANISTAN World Bank Group, any judgment on the legal status of any OBLAST BOUNDARIES REHABILITATED SECTIONS 55°E territory, or any endorsement or INTERNATIONAL BOUNDARIES 50°E 60°E 65°E 70°E acceptance of such boundaries. ANNEX 2: Case Studies 33 34 Global Review of Grievance Redress Mechanisms in World Bank Projects Pakistan Flood Emergency Cash Transfer Project Overview Over the course of July and August 2010, Pakistan Project Details: experienced the worst floods in its history, affecting $125 million IDA Emergency Recovery Credit to support Phase nearly 10 percent of its population across a vast II of the Government of Pakistan’s flood recovery cash transfer geographical area. 60% of flood-affected house- program. holds were without a main source of livelihood post-floods and 53% of households reported a • Project Approval: March 2011 76–100% decline in household income. • TTL: Andrea Vermehren & Iftikhar Malik GRM Rationale Given the infancy of Pakistan’s pre-existing social Large-scale cash transfer project in an emergency situation. safety nets, the government of Pakistan set up a rapid response cash transfer program focused on supporting flood-affected families—the Citizen’s Damage Compensation Program. Phase I of the CDCP provided immediate relief to 1.8 million families. The Flood Emergency Cash Transfer Project provided technical assistance and helped finance emergency cash grants for Phase II of the CDCP’s post-flood assistance. Actions to address the shortfalls of Phase I were included in the MOU signed by the government of Pakistan and its development partners for the project: the World Bank, DFID, USAID and the Government of Italy. Among the improvements agreed upon were the strengthening of CDCP grievance redress mechanisms and the institution of a robust public information campaign. The GRM Each of the program’s facilitation centers includes a grievance redress counter staffed by the National Database Registration Authority (NADRA). A public information campaign spread information about the grievance redress process through television, radio and print, as well as word-of-mouth communication facilitated by NGOs and community networks. In addition to the centers, the grievance redress system is able to receive complaints through text messages and phone calls. Depending on the nature of the grievance, different stakeholders are responsible for providing solutions: • Grievances related to Computerized National Identity Cards: National Database Registration Authority: NADRA handles the updating of CNICs to include changes ANNEX 2: Case Studies 35 in family status, address, family name, etc. Hotlines operate specifically for CNIC grievances. • Grievances related to eligibility/targeting: • Beneficiaries who believe they are incor- rectly listed as ineligible first check with their local authority (village level). • Appeals are run through basic filters in the national database (does the appellant live in a flood affected area? Have they already been included in the program?) and reviewed by a panel of local notables made up of respected, apolitical profes- sionals who were not involved in the original survey. • Eligibility status updates are then sent to the District Administration, which turns 685 people help to administer the grievance redress mechanism across the various service providers (Pakistani government, them over to the Provincial Disaster Man- World Bank, other donors). agement Authority. • The Provincial Appeals Secretariat of As of December 2012: the PDMA is the final decision-making authority. The PDMA informs NADRA • 49% of eligibility appeals and 85% of complaints have whether an appeal is verified for inclusion been resolved or disapproved. • 1.087 million eligibility appeals have been logged • NADRA enters the decisions into the case • 536,846 eligibility appeals have been resolved management system and formal lists for • 139,841 of these resolved appeals were accepted for issuance of debit cards are released. inclusion and issuance of cash transfer debit cards • Grievances related to payments: Partner com- • 5500 complaints (nonworking cards, requests for bribes, mercial banks respond to grievances related to etc.) have been logged payments, such as lost or nonworking debit cards, forgotten PIN numbers, etc. Partner banks operate offices and dedicated hotlines to address these grievances. • Grievances related to maladministration or unanswered complaints: Grievance related to inefficient service delivery, bribery or malpractice (termed ‘complaints’) are forwarded for resolution to the District Administration or NADRA depending upon the agency to which these complaints pertain (e.g there are complaints against NADRA or District Administration staff asking for bribes, etc.). If complainant remains dissatisfied with the resolution of complaints by the concerned agency, she can approach provincial or federal ombudsmen for redress. 36 Global Review of Grievance Redress Mechanisms in World Bank Projects Lessons Learned The project team has identified a number of points for improved outcomes: • Communication strategy is key – public information campaign critical to participation. • Developing standard practices – given the large and multi-nodal grievance redress structure, standard practices are essential for successfully administering the case load. • Training for all parties to the GRM – links to standard practices. All nodes of the grievance redress process must have the same understanding of the process. • Ownership of District Administration is essential for efficacious functioning of the committee of notables. ANNEX 2: Case Studies 37 38 Global Review of Grievance Redress Mechanisms in World Bank Projects Tamil Nadu Empowerment and Poverty Reduction Project (India) GRM Structure The project does not have a formalized, uniform GRM. Village committees, self-help groups, and other participatory bodies involved in project priority setting provide open fora for addressing grievances without for- malized grievance procedures. Village may inno- vate in grievance redress. For example, publicly Project Details: posted “problem tree” information boards provide $120 million (restructured with additional $154 million) IDA contact information for appropriate authorities, credit to improve the livelihoods and quality of life of the timelines for complaint handling, and to whom rural poor in Tamil Nadu (particularly women and other dis- appeals may be directed (see accompanying blog advantaged groups) through social, economic and democratic post). District-level Public Grievance Committees empowerment. This is a CDD project with over 581,000 target households in 2500 villages. may receive complaints and route to appropriate authorities. In addition, grievances may be filed • Project Approval: July 12, 2005 (Restructured Nov 18, 2010) at the project level, though the project website or • TTL: Kevin Crockford contact person. The Project Director, plus nominat- ed staff at state and district levels, may categorize GRM Rationale grievances and establish enquiry committees to OP 4.10 triggered investigate and report back. Implementation Issues Existing systems and structures: Grievance redress relies principally on established local and state structures. Villages utilize local committees and decision-making structures. Existing administrative processes for grievance redress, including those instituted to ensure implementation of India’s Right to Information Act, provide more formalized approaches. Participation and accountability: The participatory nature of the project provides multiple opportunities for identification and redress of emerging grievances within the community-driven structures of the project. Social accountability mechanisms and ANNEX 2: Case Studies 39 A “Problem Tree” Assures that Complaints are Quickly Addressed in Tamil Nadu Submitted by Kalesh Kumar on Thu, 07/07/2011, 19:31 The multi-colored ‘problem tree’ on the branch of a Banyan tree in Elamangalam Village in the Kadaloor district of Tamil Nadu grabs your attention. You see it as soon as you enter the village and English letters ending in @worldbank.org immediately piqued our curiosity despite our lack of knowledge of the local language. This poster, placed around the Village Poverty Reduction Committee (VPRC) and established under the World Bank supported Tamil Nadu Empowerment and Poverty Reduction Project (TNEPRP – “Vazhndu Kaatuvom”), in Elamangalam and other villages in Tamil Nadu gives the title, addresses and phone numbers of all the responsible project leaders from the government and the World Bank to help solve any complaints. This innovative Complaint Redressal System provides a timeframe within which a complaint is expected to get a response. If unsatisfactory, the plaintiff can appeal to a higher authority. Having clear time lines for escalation and resolution of problems is an essential cornerstone of good governance and social accountability in projects that are implemented at the grass root level. The last row of the poster has the name and email address of the project lead- er from World Bank and suggests 48 hrs as the time available for her to provide a response! The former project team leader confirmed to have received about 20 emails from across Tamil Nadu in her Washington office over two years reflecting the utilization of the system. The beneficiaries in other villages also had stories to tell about the effectiveness of this sort of complaint redressal systems. In Keerapalayam Village, 14 hand pumps were distributed without any loss. Feedback prompted water pipelines to be more evenly distributed across the village. Taking the process to the next level, in some villages we visited like Pathur, the external process monitoring (undertaken by competitively elected CSOs) strengthens downward accountability to project beneficiaries. Information tracking: Grievances filed through the public district authorities as well as at the project level are recorded and tracked. At the village level, it is unclear the 40 Global Review of Grievance Redress Mechanisms in World Bank Projects Social Audit Committee had also set up a complaint and suggestions box in a central location. TNEPRP is an empowerment and poverty alleviation project implemented by the Rural Development and Panchayat Raj Department of Government of Tamil Nadu with World Bank assistance. The project covers 2509 Villages in 70 Blocks in 16 districts. The target pop- ulation of this project are poor households, the most vulnerable including the physically challenged and the marginalized communities. The project follows the community driven development (CDD) approach where village communities identify their own needs, design and plan interventions and implement and monitor them. An operation manual was developed by the community and project staff, listing the principles of the livelihood approach to poverty reduction and the institutions that should be established or strengthened. It includes guidelines on how procurement should happen, how finances should be man- aged and how social accountability should be handled. This involvement of the community in generating user friendly guidelines helps ensure that the community knows what third party oversight entails and a series of workshops ensured that the community was informed. Communication materials targeted to rural households of illiterate and semi literate pop- ulace has been a hallmark of the project, pictorial posters (shown in pictures) are used to communicate community procurement procedures and processes. The project has thus far received much recognition for its pro poor approach and has set the stage for adaptation of its key principles in many grass root level initiatives of the state government. How can this be useful for your community? Source: Ending Poverty in South Asia blog, World Bank, at http://tinyurl.com/3esv37r degree to which grievances are registered. However, there is no centralized tracking of all grievances received under the entire project. Service standards: While the lack of uniform GRM procedures at the start of the proj- ect provided flexibility to address grievances through a variety of community-driven ANNEX 2: Case Studies 41 mechanisms/processes, the initial lack of standards (i.e., process guarantees regarding acknowledgements, response times, appeals procedures) at the village level could also have led to idiosyncratic approaches in addressing grievances. Subsequently, procedures regarding response times and appeals were incorporated. Lessons Learned CDD projects present a challenge given the geographic dispersion of the projects and the number of unique sub-projects. In this instance, the client could improve greater consistency and credibility among sub-project GRMs by applying the GRM Evaluation tool produced by DRP (available at furl “disputeresolution”). This is a short 2-page checklist of questions to ask of a GRM to assess it against some basic principles: legitimacy, accessibility, predictability, fairness, rights compatibility, transparency, and capability. It is not prescriptive but rather a tool that clients can use to strengthen existing structures and ensure they are credible and effective for the targeted users. 42 Global Review of Grievance Redress Mechanisms in World Bank Projects Upper Cisokan Pumped Hydroelectrical Power Project (Indonesia) GRM Structure Project Details: A two-level Grievance Task Force (GTF, with griev- $640 million IBRD Specific Investment Loan for con- ance units in the project area and a grievance struction of a dam and 1040 MW power generation center in Bandung) includes NGO and academic station, transmission lines and access roads; and design community development experts charged with and feasibility studies for an 880 MW pumped storage assisting complainants with filing grievances and power project. navigating the redress process. While part of the project team, the GTF is to operate independently. • Project Approval: May 26, 2011 Multiple uptake channels, including a hotline num- • TTL: Mr. Peter Johansen ber and SMS, reach the GTF which then classifies grievances and routes them to appropriate actors. GRM Rationale In addition, the GTF may assist complainants to OP 4.12 is triggered, with both land acquisition and displace- access additional support services funded by the ment of households expected. Pre-feasibility focus group discussions found that project affected peoples had insisted project, including a mediating committee and that a complaints handling mechanism be established. legal counseling Implementation Issues The GTF is consciously designed to be quasi-independent from the project team (its Terms of Reference were approved by the Bank team) and to serve as an advocate for project affected persons which may help drive complaints toward rapid resolu- tion. Tight timetables for grievance processing (initial action/referral within 1 week, response of competent office/authority within 2 weeks, completed handling within 3 weeks) also support quick resolution. The project also supports additional services (such as legal counseling and a mediating committee) to help complainants obtain satisfactory redress. The project engages an Independent Monitoring Agency to track implementation of project commitments. Grievance forms a central part of the monitoring process with grievance tracking forms and indicators for number of cases, meetings and field visits, ANNEX 2: Case Studies 43 Mechanism of Grievance Handling for the Upper Cisokan Pump Storage Project Project Affected Governor People (PAP) Head of District Grievances Taskforce Land, Physical Asset and Trees Acquisition Head of Village Head Sub Land Grievance Informal Leader District Acquisition Committee Resettlement Resettlement Policy Grievance Resettlement Formulation Implementation Team Team PLN Construction Hydroelectric Contractor Grievance Development of Java-Bali and satisfactory disposition of cases. Independent verification of the GTF’s handling of grievances should strengthen accountability and delivery of results. Monitoring indicators include (i) cases of land acquisition referred to court, pending and set- tled; (ii) number of grievance meetings; (iii) number of village-level meetings; (iv) number of field visits by PLN/RIT/RPFT officers; (v) number of cases resolved to the satisfaction of PAPs. 44 Global Review of Grievance Redress Mechanisms in World Bank Projects Hubei Yiba Highway Project (China) Overview This road construction project required signifi- Project Details: cant resettlement (around 10,000 people) and the $150 million IBRD Investment Loan for highway construction road went through a national park with sensitive through an ecologically sensitive area (Three Gorges Geological environmental features. The project documents Park). The total project cost is $2.2 billion. lay out a good basic GRM structure that includes • Project Approval: March 2009 multiple access points, an appeals process, and a • TTL: Ms. Fei Deng (current) centralized complaints database. Christopher Bennett (former) Implementation Issues GRM Rationale Once implementation was underway, it became The project triggered both OP 4.10 and OP 4.12, and the client requested Bank assistance in managing the environmental clear that the system had some considerable draw- impacts of the project. backs, namely that it was not being widely used and, when it was being used, not all grievances were being entered into the central database. Thus, the data that was received was less useful to the client (and the Bank team) since it wasn’t giving the full picture. The first indicator that something was not working right was the number of com- plaints received (fewer than 200) compared to the scale and scope of the project, which involved the resettlement of 10,000 people and went through a national park. Christopher Bennett, the TTL at the time, said “As a TTL, I should have been really happy with no complaints, but I just didn’t believe it.” The team was concerned that not all grievances were making it into the system. To assess whether this was happening, the TTL and Social Development Specialist surveyed impacted commu- nities to verify their concerns, and their review confirmed that not all project-related grievances were making it into the formal records. To address this issue, the Bank team hired a consultant to design an SMS system and web interface that would allow individuals the opportunity to directly input grievances and concerns. The design of the SMS system and accompanying web interface were ANNEX 2: Case Studies 45 simple, practical, and easy-to-use, but it was not embraced by the Chinese locals, who preferred to direct their complaints to local village authorities. The Bank estimated that over 95% of grievances were ultimately channeled through traditional mechanisms versus the SMS system. The chiefs recorded these grievances in writing, but there was no direct link to the Bank’s SMS system (or vice versa) that allowed for a comprehensive view of the concerns of the local communities. L–R: World Bank TTL Lessons Learned Fei Deng, resettled person, Mr. Tang, The team identified the need to assess existing grievance systems and incorporate director of Hubei these systems into the Bank-designed structure, to ensure all grievances are logged, Yiba Highway Project Construction, tracked and responded to. A shared database and Memorandum of Understanding with in charge of resettlement the local village authorities could be one option to ensure a clearer picture of local citizens’ concerns in real time. The database would include all grievances received by the chiefs as well as grievances channeled through the SMS and web. Training villagers to manage a centralized database and implementing regular performance reports on all grievances received could significantly improve the effectiveness and credibility of both systems. EXCERPT FROM PAD “Complaints are dealt with in up to four stages: • Stage 1: The DP’s may present their grievances either orally or in writing to the village committee or local resettlement office. If oral, a written record must be made and a clear response given within two weeks. If it is a serious matter, it must be reported to a higher level resettlement office and a reply received within two weeks. • Stage 2: If the complainant is not satisfied with the outcome of Stage 1, they may appeal to the higher level resettlement office within one month after receiving the Stage 1 reply. The county or district level resettlement office must reply within three weeks to this appeal. • Stage 3: If the complainant is not satisfied with the district or county resettlement office response they • may appeal to the regional resettlement office within one month of receiving the Stage 2 response. The resettlement office shall respond within one month. • Stage 4: If the matter is still not satisfactorily resolved the complainant may appeal to civil court within 15 days of receiving the response from the resettlement office. 46 Global Review of Grievance Redress Mechanisms in World Bank Projects As mentioned above, the ‘Safeguards Compliance Monitoring System’ will be piloted as part of this project. It will allow for workers, residents and farmers to access project information as well as file grievances and concerns related to project actions, specifically issues related to resettlement, environment, or worker rights, using the text messaging feature on their cell phones, or through the internet—via e-mail or web-based form. Complaints can be filed anonymously or with contact information in order to receive a follow-up message. These complaints, which will be received by the Bank, the client, and an independent monitor, will be logged and addressed appropriately and in a timely manner.” The chart below indicates the grievance process and includes both the local tradi- tional systems (i.e. village leaders) as well as the Bank-designed SMS system that routs complaints directly to the resettlement monitor in the implementing agency: Follow up with If resolved World Bank Office Management Level If not resolved Follow up with If resolved Resettlement Monitor If not resolved Formal Complaint to Complaint sent to Client, Resettlement Monitor World Bank (via SCMS) If resolved Notify and Independent Monitor If not resolved Project Level Project If resolved Resettlement Monitor Province Level If not resolved If resolved District Leader District Level If not resolved Major Issues Local If resolved Resettlement Office If not resolved If resolved Complainant Local Leader Inform Local Level ANNEX 2: Case Studies 47 48 Global Review of Grievance Redress Mechanisms in World Bank Projects Chad Local Development Program Support Project 2 Overview The first phase of this CDD project focused on Project Details: establishing a decentralized, community-based Second phase of community-based development program approach to local development, including ensuring to improve access to infrastructure and basic social services. Bank committed $25 million of $77.25 million countrywide support from stakeholders and government offi- project. Local communities submit plans for micro-projects cials. Once that was achieved, the second phase (i.e., schools, health clinics) that are managed and monitored of the project was developed, and it was agreed by local committees. Some micro infrastructure projects may that it would focus on sub-projects in rural areas impact land holdings or assets, triggering OP 4.12 and the across the country. Chad is a fragile state with GRM requirement. severe poverty, ethnic diversity, and a history of conflicts and elite capture/corruption. • Approval: March 18, 2011 • TTL: Soulemane Fofana • Categorization: B The GRM Since individual projects were not defined at the GRM Rationale time of project approval, a Resettlement Policy Triggers OP 4.12. Framework (RPF) was developed to guide future projects. The GRM section in the PAD is fairly general, and states that grievance redress may be an informal consensus process, a formal process, or a court case. The PAD does not include guidance on what a GRM should look like in the event site-specific RAPs are required. The Bank team, however, has flushed out a general approach to developing project-specific GRMs. Projects will rely on existing traditional structures in the villages to handle grievances. The villages form committees (procurement committee, management committee) to oversee their respective sub-project. These committees are the first point of uptake. If they cannot resolve issues, grievances are escalated to the village chiefs. Although not described in the project documents, there is a grievance redress hierarchy: local committees, chiefs/elders, regional project implementation team, and a central team in the government ministry. There are no standard procedures, monitoring or tracking but the Bank team has emphasized quick redress at the local level to avoid conflict ANNEX 2: Case Studies 49 escalation. The Bank team has also encouraged greater use of local citizens versus government employees to manage grievances in order to reduce corruption. There is an outreach campaign, open meetings, public announcements, and public contracts to educate citizens of the process and reduce conflicts. As of early 2013, implementation was just getting underway and Bank staff was not aware of any complaints. No data was available to check this view. Lessons Learned The project is still in its early stages, so lessons are preliminary. This project rightly identifies the need to work with existing local systems for grievance redress, given the operating environment in Chad. The emphasis on on-the-ground resolution and simple systems is appropriate. Building a complicated GRM system in this instance would likely not be used, given that the projects will impact rural citizens who likely have limited/no electricity and may be illiterate. Working directly with the local committees is likely the best option for strengthening capacity to manage issues. This could include walking through the GRM Evaluation tool, which is a short checklist of questions to get clients thinking about ways to strengthen their existing systems. Similarly, putting in place timelines for responses and a basic monitoring and evaluation system (run by the villages) could improve outcomes for users in a “light-touch” manner. 50 Global Review of Grievance Redress Mechanisms in World Bank Projects Uganda Electricity Sector Development Project Overview The right of way and substation construction Project Details: requires land acquisition, affecting 13,596 persons. $120 million IDA credit (total project cost $153.2 million) to 1,152 persons will be resettled. As of early 2013, improve reliability and access to electricity in southwestern Uganda through construction of a 137km transmission line the project was still in the bidding process, and and related substations. no land acquisition or construction had begun. Thus, the GRM was not up and running but is • Approval: June 30, 2011 fairly well-defined in the project documents. • TTL: Somin Mukherji • Categorization: A The GRM GRM Rationale The proposed GRM uses existing systems and struc- Project triggers OP 4.12. tures and has three levels of redress. The Village Grievance Committees (with representatives of project-affected communities) are the first point of contact and seek to resolve complaints using cus- tomary rules. Where the village committee cannot resolve the complaint, complaints are referred to the District Land Tribunal. Where the Land Tribunals cannot solve the dispute, the aggrieved party is free to evoke their constitutional right to file a case with the court system. The courts take a long time to process cases so it is expected that most issues will get resolved at lower levels. A project-level Resettlement Unit, reporting to the implementing agency (Uganda Electricity Transmission Company, or UETCL), coordinates RAP implementation, including complaints processing, convening Village Grievance Committees, and resolving complaints that the Village Committee is unable to. Some service standards are specified. UETCL has experience with complex projects and grievances (Bujugali dam) and will hire a consultant to help manage the GRM. The RAP identifies that the second level of the proposed GRM structure is weak, since there is a lack of capacity in the Land Tribunals, and complainants must finance their own travel, meals, and allowance for tribunal members. The Bank team is emphasizing ANNEX 2: Case Studies 51 the Village Grievance Committees to resolve issues, with Resettlement Unit tasked with addressing unresolved complaints. The Finance Ministry, not the implementing agency, will do monthly reporting and external monitoring, and these reports will be shared with the World Bank in their entirety. Lessons Learned This project emphasizes on-the-ground resolution of issues, includes at least one level of escalation, and includes processes to ensure complaint-handling is predict- able and transparent. Furthermore, it appears that complaints will be tracked in a central database that is monitored by an independent entity—a key component that often is missing. Importantly, the Bank will receive the monthly monitoring reports which will allow for faster support to the client and early responses to developing risks. Suggested areas for improvement could include (i) ensuring anonymous complaints can be filed; and (ii) having the client evaluate the effectiveness of the Village Grievance Committees to identify potential weaknesses. Finally, given a project of this size, grievance procedures should be written down in a manual and shared publicly with local communities. 52 Global Review of Grievance Redress Mechanisms in World Bank Projects Zanzibar Urban Services Project (Tanzania) Overview As of early 2013, civil works had not yet com- Project Details: menced and key GRM elements (i.e., database, $38 million IDA credit to improve access to urban services in focal points, Resettlement Coordination Unit) had Zanzibar and conserve physical cultural heritage by provision of not yet been put in place. storm water drainage, sanitation services; and construction of a sea wall and promenade. Although no physical resettlement is anticipated, the project involves some land acquisition and GRM Structure temporary dislocation during construction. Project-affected persons (PAPs) are notified about the grievance mechanism and their legal rights to • Approval: February 24, 2011 file complaints with the courts (i.e., handed a letter • TTL: Andre Bald of notification that is verbally explained to those • Categorization: A who cannot read by a trusted intermediary). PAPs GRM Rationale may file complaints with either local community Triggers OP 4.12. leaders (e.g., Shehas) or with the community liaison and grievance redress officer appointed by the project’s Resettlement Coordination Unit (RCU). The liaison and grievance officer is responsible for receiving, recording, and processing of complaints. Local leaders who receive complaints must channel them to the officer within three days. PAPs may file grievances with community leaders at anytime. In addition, twice a month the officer informs the PAPs in a timely manner of the time and location he will be stationed (on site), and where the PAPs can lodge their grievances. If the dispute cannot be resolved within fourteen (14) days, it will be referred to the Regional Manager of the project resettlement coordination unit (RCU) where amicable resolution will be sought between the RCU and the PAP within another established period of twenty-one (21) days. However, if the dispute is not resolved, the aggrieved party is free to take the matter to a Primary Court of Law as their final resort. Compensation will be paid upon resolution of the grievance or dispute. The ANNEX 2: Case Studies 53 grievance team must keep a database of all grievance claims as well as the period if took to resolve the disputes which will be kept on record by the RCU. Lessons Learned While there is no implementation experience as of early 2013, this appears to be a well-designed, fairly “light-touch” grievance redress mechanism, consistent with the scale and scope of the project. It relies on local existing structures and keeps redress close to the PAPs, which is good practice. Timelines and processes are well-defined, along with appropriate escalation for more serious grievances. The design anticipates a centralized database of all grievances and requires monitoring. Some details remain unclear: whether anonymous grievances can be filed, whether all grievances are acknowledged in writing, the composition of the Resettlement Coordinating Unit, and how triage/prioritization of grievances occurs. Primary Court Resettlement Coordination Unit (RCU) 3 21 Days Days Grievance Redressal Officer Shehia-sheha 14 Days PAP’s Complaint 54 Global Review of Grievance Redress Mechanisms in World Bank Projects NamTheun 2 Hydroelectric Project (Lao PDR) GRM Structure A three-step grievance process was developed for Project Details the project. Complainants first raised complaints to $20 million IDA grant to support construction of 1,070 MW Village Grievance Committees (VGC) which sought hydropower facility, management of environmental and resolution through dialogue and discussion. If the social impacts, and monitoring and evaluation arrangements issue could not be resolved at the VGC, complain- ants could raise their grievance with District Griev- • Project Approval: March 31, 2005 (closed) ance Committees (DGC) which were comprised • TTL: Ingo Wiederhofer of the public District Justice Department, supple- mented by representatives of village and women GRM Rationale organizations. DGCs were required to respond to The project required resettlement of approximately 6,300 people in 15 villages, triggering OP 4.12. complaints within 15 days. If complainants were not satisfied with outcomes of the DGC, they could Steps in the Grievance Process Regarding Issues of Compensation and Resettlement PAP not satisfied with Compensation or the Process Appeal to the Village Grievance Committee of Elders Examination PAP satisfied – settled PAP still not satisfied Appeal to Project Grievance Committee and Court at the District Level NamTheun 2 Advice Advice Resettlement Examination Power Company Management Unit PAP satisfied – settled PAP still not satisfied Provincial Court ANNEX 2: Case Studies 55 appeal to the Provincial Court as a last resort. The project engaged a local NGO to act as an adviser and counsel to affected persons to navigate the grievance process. In addition to internal monitoring, an Independent Monitoring Agency was tasked with reviewing and reporting on grievance cases and their resolution. Implementation Issues Reliance on local grievance systems: The project relies largely on the local governmental dispute resolution processes, augmented with village and women organization repre- sentatives and supported by the project sponsor. The final report of the International Advisory Group (IAG, reporting to the World Bank President) found that the local institutions handled the complaints well if occasionally slowly. The IAG questioned whether reliance on local grievance structures would have worked as well if the government had been the direct project sponsor), an issue that may require attention with the handover of the Downstream Program (which has given rise to numerous complaints) from the Nam Theun Power Company to the Provincial Government. 56 Global Review of Grievance Redress Mechanisms in World Bank Projects