Ideas to Action:

Independent research for global prosperity

A Syrian woman at a sewing machine
CGD Notes

Five Ways to Improve the World Bank Funding for Refugees and Hosts in Low-Income Countries and Why These Dedicated Resources Matter More than Ever

Sarah Charles , Cindy Huang , Lauren Post and Kate Gough
November 5, 2018
Download PDF

The International Rescue Committee (IRC) and the Center for Global Development (CGD) have been closely following the rollout of the IDA18 sub-window for refugee hosting nations, particularly in East Africa (Ethiopia, Uganda), West Africa (Cameroon, Chad), and Asia (Bangladesh). Drawing on original research and analysis, as well as field visits, this note identifies a few clear successes and areas for improvement. It also outlines recommendations for the implementation of the IDA18 sub-window moving forward, and makes the case for a sub-window in the IDA19 replenishment.

For too long, the international community has accepted the convenient fiction that refugee crises are temporary. In 2016, as waves of refugees came ashore in Europe seeking a life free from violence and conflict for themselves and their children, a spotlight was cast on the hopelessness that exists in even relatively well-funded refugee contexts—where there are few, if any, opportunities for meaningful livelihoods and education—and the very real struggle developing countries face under the strain of hosting large numbers of refugees. This spotlight shined brightest on Jordan, Lebanon, and Turkey, which collectively host more than five million Syrian refugees.

The World Bank has been a leader in responding to the fact that the traditional humanitarian system was not designed to meet medium or longer-term needs, and has begun to play a significant role by providing multiyear financing and engaging in policy dialogue on refugee rights and opportunities. The Bank also led the way in recognizing that the dynamics that were so apparent in the Middle East had long existed around the world. In response, in late 2016, the World Bank made an important and significant decision to raise a dedicated $2 billion in its IDA18 replenishment (July 1, 2017–June 30, 2020) that is available as additional concessional financing to low-income countries (LICs) bearing the burden of hosting large numbers of refugees.

With this decision, the World Bank jumped feet first into contexts that were long dominated by humanitarian actors. It brought a set of resources, expertise, and credibility and relationships with host governments that were new to refugee response—including a deep bench of experts and best practices for how to strengthen service delivery and infrastructure and improve socioeconomic outcomes for poor and vulnerable populations. It also brought lessons learned from the implementation of the Global Concessional Financing Facility, a similar model that provides additional financing at concessional rates for middle-income countries (MICs) hosting large numbers of refugees and that has been piloted in the Middle East. In creating the IDA18 sub-window, the World Bank catalyzed the so-called humanitarian-development nexus from mostly rhetoric to real test cases.

Over the last two years, there have been some critical and consequential changes as development and humanitarian actors joined forces to respond to protracted refugee crises. Staff throughout the World Bank have had to grapple with how to advance and adapt the Bank’s toolbox to accommodate the unique needs of refugees and approaches of new partners, particularly the UN Refugee Agency (UNHCR). As World Bank-financed programs start to move from paper to implementation, there is promise of real improvements in the lives of those adversely affected by displacement. Still, remaining uncertainties and challenges need to be addressed if the Bank’s financing is to have its intended impact for refugees and host communities. While the Bank brings many unique assets to address protracted displacement, its model of host governments as clients and the priority placed on disbursing funding can sometimes limit its capacity and willingness to engage robustly on policy and protection issues. In this note, we discuss early achievements and challenges of the IDA18 refugee sub-window, and offer five recommendations to improve its impact.

The IDA18 midterm review, to be held in Livingstone, Zambia from November 13–15, serves as a critical moment for the World Bank and its stakeholders to take stock of progress and challenges associated with the refugee sub-window, and put mechanisms in place to correct course where needed. It is also an opportunity to begin thinking about the goals and structure of the IDA19 replenishment, and consider how the Bank can best support LICs that are fragile, conflict-affected, and/or hosting large numbers of refugees.

IDA18 regional sub-window for refugees and host communities: Five areas of success and challenge

There are five areas where we have seen important success in the implementation of the IDA18 sub-window, though these should be tempered with fact that there is also room for improvements. Moving forward, the World Bank and IDA Deputies should consider how to implement key changes to ensure that the sub-window delivers on its considerable promise. Running through our assessment are two central questions: First, is the IDA18 sub-window incentivizing policies and supporting programs with the highest likelihood of improving refugees and hosts’ lives and their prospects for self-reliance? And second, how will we know?  

1. Refugee-related policies: Prioritize policy changes that will directly and indirectly enable self-reliance among refugees and host communities.

Even in a tough political climate, where there has been considerable retreat among global leaders to share the responsibility of meeting refugee needs, the Bank has worked with some refugee hosting governments to achieve important policy changes that can enable refugees to more sustainably support themselves and their families—or achieve self-reliance. However, there are also examples where it appears the Bank prioritized the disbursement of funding over the challenging task of working with governments and partners to change refugee-related regulations. In its country operations, the financing sub-window arguably needs a clearer focus on achieving policy changes that can improve the lives of refugees, such as freedom of movement and access to quality schooling and decent jobs, to have real impact.

To access the IDA18 sub-window financing, host country governments are rightly required to have an action plan to meet the needs of refugees. Country commitments at the 2016 Leaders’ Summit on Refugees and Migrants have largely served as the baseline for these action plans. This has proven useful in some country contexts, less so in others—depending on the rigor and robustness of the pledges and the Bank’s interest and ability in holding governments accountable in achieving them based on the policy and political context. For instance, Ethiopia made nine distinct pledges, including notably to reform its 2004 Refugee Proclamation, to allow some refugees to move freely and legally work outside of camps. As part of the Bank’s agreement with the Government of Ethiopia, the Bank is holding off on implementing an employment package until the new Refugee Proclamation is ratified.

Compare this to Cameroon, where the government pledged to ensure that returns of refugees to their countries of origin would be free, voluntary, and only when security conditions exist. Despite reports of forced returns and little investigation into these cases by the government, the Bank has not halted financing or its programs, placing little pressure on the government to resolve the issue. Or, further, compare this to Bangladesh, where the Bank has approved two projects to directly support Rohingya refugees, but has not publicly pushed for the Government to improve its policies that would enable refugees to participate in formal, accredited schooling or to find work.

Although refugee and host populations often have acute needs and holding up programs is not ideal, the Bank’s financing should not be a blank check to governments, nor should the Bank use sub-window financing to become just another humanitarian donor. Otherwise, we can expect the situation of refugees to remain status quo. In the case of Bangladesh, for example, it may have been preferable to address nearer-term needs with other crisis or trust fund mechanisms to help build confidence and open space for policy discussions. The IDA18 sub-window financing could have then served as an anchor for international commitments toward a medium-term approach that includes a greater focus on refugee self-reliance. Working with host governments toward improved legislation and regulations—such as access to education and livelihoods and greater freedom of movement—is key to unlocking greater self-reliance and ensuring the Bank’s investments and role complement, rather than duplicate, those of traditional humanitarian donors.

The Bank has uniquely strong relationships with governments compared to other donors and development and humanitarian actors, and could use them more strategically to ensure policies—the backbone of lasting change—are a key part of agreements and accountability frameworks. As the 2016 pledges become a more and more distant past, and as new countries seek to draw on the sub-window, it will be imperative for the Bank to work with governments to determine what refugee-related policies need to change to see sustainable improvements in people’s lives. This may mean countries need to go above and beyond their initial pledges to align with current realities.

Given the retreat since 2016 of global leadership on supporting refugees, new tools, approaches, and research may be needed to support policy changes. For example, regular stock-taking of refugee policies, such as through a refugee policy index that evaluates host country policies and the legal status of refugees relative to host communities, could prove useful in determining critical areas for reform and motivating change. As IRC and CGD outlined in our earlier Refugee Compacts report, the Bank, in consultation with UNHCR, should explore the creation of such an index. In addition, research and analysis on the impact of certain policy changes can highlight areas where progress could yield benefits for refugees and hosts. For example, giving refugees greater access to the formal labor market can increase the productivity of local firms, boost tax revenue, and encourage hosts to upgrade to better-paying jobs. It can also reduce competition in the informal sector as some refugees obtain formal jobs, mitigating any downward pressure on wages that may exist. Finally, building on the compact model, more exploration is needed of how a set of financial and beyond-aid commitments (e.g., trade concessions and labor mobility opportunities) can open policy and political space by offering a triple win for refugees, hosts, and national development efforts.

IDA Deputies and the Bank’s Executive Board, which set the terms of the sub-window financing, have an important role to play in ensuring that the Bank’s financing is leveraged to secure and support policies that will enable tangible, intended impact. Deputies and the Board should review and consider adjusting the current policy that requires financing to be allocated in 50 percent grants and 50 percent loans, with the implication being that loan-based funding is meant to be additive concessional funding for host communities and grant-based funding is meant to support refugee populations. A more flexible policy—for example, one that can deviate from the 50/50 rule during the initial period of a crisis or based on extraordinary circumstances—could give the Bank more leverage in negotiations with host government. For example, if an analysis finds that a sudden influx of refugees has had a sharp and concentrated impact on development progress in host communities, it may be reasonable to offer more grant financing for the host community than would typically be allocated in these arrangements. Importantly, this shift, if taken forward, should come with an even greater focus and emphasis on what outcomes the financing is aimed at achieving for which populations (i.e., refugees and host communities), and what refugee-related policy changes are required by host governments towards achieving them.

The Deputies and Board should also request that regular reports be published on the IDA sub-window. This should include how much of the $2 billion dollars has been allocated and disbursed, as well as explanation of the pipeline, reasons for progress and delay, and early lessons. The goal of such reporting would be to increase dialogue with the Deputies, Board, and other stakeholders on progress against the sub-window objectives, including advancing policy changes at the country level. Importantly, it should not be used to raise pressure to disburse funds quickly in the absence of commitment to the action plan associated with IDA18 lending.

2. Accountability for outcomes: Define a set of outcomes, targets, and indicators for the sub-window that can guide results frameworks for projects financed through the sub-window.

The World Bank is a key actor in helping countries achieve the outcomes set forth in the Sustainable Development Goals (SDGs), and is a leader in the development community for its rigor in measuring the impact of its projects. While the Bank has outlined clear objectives for the sub-window, such as to promote refugee welfare and inclusion, support livelihoods in host community areas, and support legal solutions, the results framework for the IDA18 financing for fragility, conflict, and violence fails to include goals, targets or indicators that measure sub-window level progress towards outcomes in the lives of refugees and communities. Rather, it focuses inward, assessing the Bank’s progress towards deepening its knowledge in fragile settings, improving staff effectiveness and flexibility, and promoting partnerships for a more effective response. Although these are important goals for the Bank, they do not gauge whether the IDA18 sub-window is doing what it intends to do: improve people’s lives.

Ideally, defining a set of outcomes would precede and inform decisions about country-level policies and project designs for sub-window financing, as well as reviews of the sub-window’s performance. An explicit, agreed on set of development outcomes are key to negotiations between the Bank, governments, and other partners around which policies and interventions are required to improve the lives and wellbeing of refugees and their host communities. Targets that benchmark progress should then serve as the basis for reviews of the sub-window. The forthcoming Independent Evaluation Group report and the internal assessment completed by the IDA team can provide a summary of what has been done and where there is room for improvement, but neither process has a set of outcomes or targets against which to base their early assessment of impact.

At the same time, Bank teams implementing projects financed by the sub-window should include a set of outcomes, targets, and indicators for both host and refugee populations in their results frameworks. Among projects approved so far, there are promising signals, but this is not yet standard practice. For instance, the social safety nets project in Cameroon seeks to reduce levels of poverty, and adjusts the indicators to include refugees and disaggregate results by nationals and refugees. However, the results framework and disbursement-linked indicators in the reform support project in Cameroon is less robust, listing primarily output-level targets (e.g., distribution of teachers, access to textbooks, enrollment) versus outcome targets (e.g., improved math, literacy, or psychosocial skills) and very few indicators focus on or appear to be disaggregated for refugee populations.

Although not standard practice for Bank financing sub-windows, given the unique objectives of the sub-window to meet the needs of refugees and host communities, the Bank should establish a set of outcomes, targets, and indicators at the sub-window level, with a clear reporting cadence, to show overall progress. These outcomes can and should align with other global processes, such as the SDGs and forthcoming indicators to be included with the Global Compact on Refugees—processes which the Bank is helpfully already engaged on. The Board should also ensure during the approval process that there is alignment between the sub-window level outcomes and indicators and individual project level results frameworks so that project results can roll-up into higher level progress reports. In support of these ends, the Bank should move forward with utilizing its joint data center with UNHCR to collect and regularly publish relevant data.

3. NGO and civil society engagement: Standardize how the Bank engages with local actors, particularly beyond UNHCR, to include NGOs, civil society, other UN agencies, and local governments.

The World Bank has done an exceedingly good job at building a partnership with UNHCR at headquarter and national levels. This is evidenced in the advancement of the World Bank-UNHCR Joint Data Centre, and in joint missions in the lead up to program development in individual countries. However, an ongoing sticking point for other actors on the ground responding to refugee crises has been the Bank’s ad hoc and unclear consultation process with NGOs and civil society (as well as other UN agencies in some cases). While this critique is not unique to the IDA18 sub-window, stakeholder consultation is equally, if not more, important in refugee contexts given that refugees fall outside usual state-citizen accountability mechanisms. Therefore, broader consultation with civil society organizations and refugees themselves is critical.

In places like Ethiopia and Uganda, consultative processes have been complicated by the rollout of UNHCR’s Comprehensive Refugee Response Framework (CRRF), which some host country governments are financing with IDA18 sub-window funds. In these countries, it appears that the Bank has relied on UNHCR to deliver information about Bank projects to NGOs; however, this “telephone” arrangement has not always proven effective. Actors report being frustrated and confused by what the Bank is doing, and where and how they can contribute to needs assessments, decision-making, and implementation of programs. In Uganda, a clearer and more deliberate process for communication and collaboration with NGOs and civil society, as well as donors, could help prevent duplication of effort and better information sharing.

In Cameroon, where a CRRF structure is not yet being implemented, World Bank documents propose coordinating mechanisms at the national and regional levels, as well as coordination at the local level where implementation will take place. NGOs have recommended the Bank draw on local structures used and lessons learned in other contexts, such as in the Lake Chad Inclusive Economic and Social Recovery Programme. This type of input from NGOs and civil society can help the Bank identify structural and planning gaps early on in the process to improve programming and minimize the need for course correction later on.

Despite these challenges, a bright spot in the Bank’s staffing arrangements has been the deployment of Fragility, Conflict and Violence staff to oversee the regional and country rollout of the sub-window financing. These staff bring essential expertise in bridging the humanitarian and development divide, and have served as critical interlocutors for NGOs—especially based on IRC’s experience—at the country and global levels. However, it seems these staff could use more support to more regularly and meaningfully engage with a wider set of stakeholders. 

There remains a need to standardize how the Bank engages with local actors in humanitarian contexts, particularly beyond UNHCR. There should be a protocol for regularly sharing information and getting input on decisions about programming and policy goals. The World Bank should not wait until the procurement process to engage NGOs; rather, this should be done from the start of arrangements to ensure refugee protection is top of mind and that solutions will ultimately meet needs. A local advisory board, one that aligns with and is clearly connected to the CRRF process (in places where it is being rolled out), is one potential solution. Ultimately, project documents should seek to outline how Bank staff intends to engage with partners—a step beyond the current documents, which simply note that Bank staff will do so.

In addition, while financing from the sub-window should aim to flow through governments to support inclusive country development plans, the World Bank should carefully analyze the capacity of governments to absorb financing and implement refugee-sensitive programs. In targeted cases, where this capacity is low, the Bank should consider alternative delivery options in the near-term that align with national plans and can contribute to building up national capacity. This may mean directly financing international or local NGOs to implement programs on an interim basis that can eventually be transitioned over to government ownership. The Bank should develop a clear strategy and protocol for working directly with NGOs in ways that facilitate meeting nearer-term objectives and that support longer-term plans and government capacity. For example, due to their experience, NGOs may be best placed to provide non-formal education closer to the start of the crisis. However, these programs should be designed as an interim bridge to formal options, which often take longer to implement because they require activities such as building and rehabilitating schools and curriculum development.

4. Development partners: Seek multiyear co-financing from other development actors to support and align with Bank projects that address the needs of refugees and host communities.

UNHCR expects funding for 2018 to meet roughly half of the $8.2 billion required to meet the needs of people who have been forcibly displaced from their homes. Over the last few years, more and more people have been forcibly displaced and become refugees (an increase from 65.3 million displaced—including 21.3 million refugees—in 2015 to 68.5 million displaced—including 25.4 million refugees—in 2018), and the funding shortfall has risen (from 42 percent of requirements underfunded in 2016 to 45 percent in 2018). As the conflicts and crises at the root of displacement burn on, the prospect of people returning home becomes an ever-distant prospect and the longer term needs of displaced people in host countries becomes more acute. The World Bank has rightly recognized that the traditional humanitarian system was not designed to meet medium or longer-term needs, and has begun to play a more significant role by providing longer term, multiyear financing. But it cannot do this alone. It will require co-funding from other development donors—donor governments, regional bodies, and other multilateral banks.

The World Bank, through its close partnership refugee-hosting governments and UNHCR, is essentially creating a much-needed roadmap that will enable countries to meet the needs of refugees within their borders, while at the same time achieving a country’s sustainable development goals. This provides a useful outline for other development donors to align with and support, complementing the work of the Bank, governments, and humanitarian actors. In some instances, other donors have followed the Bank’s lead: the UK’s Department for International Development (DFID), for example, is co-financing a portion of the Ethiopia employment package. However, this arrangement remains the exception not the norm.

Low- and middle-income countries often rely on financing from the World Bank to build up their public infrastructure as part of their national development plans; however, new or improved infrastructure is not always what—or the primary thing that—is needed to meet refugee and host community needs and enable their self-reliance. Building or rehabilitating schools and health clinics may be useful down the road, but only if the right programs to deliver services are put in place and people are able to access these services in the interim. Other development donors can bring in additional grant financing to help balance the portfolio of interventions to meet these intermediary needs. In addition to seeking co-financing from other donor governments, the Bank should lead the multilateral development bank platform for responding to forced displacement, which was announced in fall 2016 and formally established in spring 2017, but has not appeared to move beyond a memorandum of understanding.

5. Faster and standardized practices: Balance the need for a speedy process with the length of time needed to develop projects that reflect the experiences of refugees and host communities.

Refugee crises require development financing and tools to be deployed at or close to the speed of humanitarian response in order to more quickly achieve outcomes for refugees and host communities. There is an obvious tension here: needs assessments, stakeholder consultation, and careful negotiations about financing and program design all take time. And the Bank needs to be careful not to rush into topping-up existing programs without properly adapting them to meet refugees’ unique needs. However, the Bank needs to find a better balance between speed and the length of time needed to deploy resources. This could be accomplished through more standardized protocols, including common assessment and planning tools and procedures to quickly deploy surge staff. It could also include protocols around complementing sub-window financing with funds from the Bank’s crisis window in the earliest stages of a crisis. Doing so would allow the Bank to more deeply assess the situation and determine how best to use sub-window financing to advance the policy dialogue. In addition, frictions with local communities and humanitarian actors could be eased if the Bank was more transparent and communicated more clearly about its expected timelines.

The case for an IDA19 regional sub-window for refugees and host communities

Violent conflicts have more than doubled since 2000, and more people—68.5 million, including 25.4 million refugees—have been forced to flee their homes than at any time since World War II. Twelve of the top 15 refugee-hosting nations are considered fragile states. Projections point to increased deprivation of people caught in crisis—people living in fragile and conflicted-affected states (FCAS), and people who have been forcibly displaced. New analysis from the IRC and Overseas Development Institute finds that by 2030, on current trend, unmet basic needs will be increasingly concentrated in fragile states; FCAS will be home to 85 percent of people who are extremely poor—342 million people. The health, education, livelihoods, and safety of hundreds of millions of people in fragile and conflict affected settings—especially displaced people—are on the line.

As the needs of the most marginalized people, including refugees and other displaced people, are likely to increase, and as poverty becomes more concentrated in already fragile or conflict-affected states, it is imperative that the World Bank remain a leader in supporting refugee hosting nations. In fact, if the World Bank intends to achieve the goal of eliminating poverty and supporting countries in achieving the SDGs, there is a case for the World Bank to increase its financing for these countries in its IDA19 replenishment. Building on its growing capacity and experience with forced displacement, the Bank should also consider how it can expand its role in addressing internal displacement through IDA19.

With only 18 months of implementation, the Bank’s sub-window for refugee hosting nations is still in its infancy, yet it has already started to demonstrate its importance and value-add. Host countries are, for the most part, signaling demand for these additional concessional resources to support their open door policy towards refugees and to meet the pledges they made in 2016. Projects developed so far have also demonstrated great promise for impact, from the Bank’s support on policy shifts to allow more refugees to access services and obtain jobs in Ethiopia to expanding refugees’ access to social services in Uganda and Bangladesh. And the Bank’s focus and engagement in refugee contexts has appeared to start to mobilize other multilateral actors to do the same; the African Development Bank, for instance, has recently announced resources to support refugee and host population needs in Zimbabwe and South Sudan and the Asian Development Bank has made an initial commitment of $200 million to address the refugee crisis in Bangladesh.  As with its refugee-related work with MICs like Jordan, the World Bank is engaging the private sector in IDA countries such as Ethiopia to drive more sustainable progress. This, too, is an important value-add to the usual set of refugee responders.

Since the IDA18 sub-window was created, new crises have generated large scale displacement—from the influx of South Sudanese into Uganda, to the mass displacement of Rohingya refugees to Bangladesh. In increasing financing for host states, the Bank can not only help meet needs directly, but also send a signal to other global institutions and donors that they, too, should prioritize their support for these countries—something that is direly needed in a world where we are seeing global leaders retreat from, rather than step up to, their responsibility to respond to refugee crises.

There should be a similar sub-window for refugees and hosts in the IDA19 replenishment, but with greater resources that reflect increased need. Such a window should consider the lessons from implementation of the IDA18 sub-window, and identify ways to improve the Bank’s processes and programs to ensure greater impact is achieved and adequately measured.

Specifically, an IDA19 sub-window should:

  1. Prioritize policy changes that will directly and indirectly enable self-reliance among refugees and host communities. This could be supported by a new refugee policy index, as well as greater flexibility for the terms of financing, to generate incentives among host governments.

  2. Define a set of outcomes, targets, and indicators for the sub-window that can guide results frameworks for projects financed through the sub-window. These outcomes should reflect changes in the lives of refugees and host communities. Progress against outcomes should be regularly measured and reported.

  3. Standardize how the Bank engages with local actors, particularly beyond UNHCR, to include NGOs, civil society, other UN agencies, and local governments. Also develop a protocol for how the Bank can directly finance non-governmental actors to deliver services as an interim bridge to delivery through government systems.

  4. Seek multiyear co-financing from other development actors to support and align with Bank projects that address the needs of refugees and host communities. One dimension of this should be through the multilateral development bank platform for responding to forced displacement.

  5. Balance the need for a speedy process with the length of time needed to develop projects that reflect the experiences of refugees and host communities. This may mean standardizing protocols (e.g., assessment tools and procedures for deploying surge staff), and/or complementing sub-window financing with funds from the Bank’s crisis window or other sources.

The authors would like to thank Masood Ahmed, Jeremy Konyndyk, and Scott Morris for providing their insight and inputs through review.

Rights & Permissions

You may use and disseminate CGD’s publications under these conditions.