Water supply for refugees and their host communities in protracted situations: costs and financing options for sustaining services in Ethiopia and Uganda

Globally, water supply systems in refugee and host community settings face significant challenges related to long-term sustainability. In protracted displacement situations, humanitarian water interventions, designed for short-term relief, are often ill-suited for the extended periods that refugees reside in these areas. More than 60% of refugees remain displaced for decades (UNHCR/UNICEF 2020), yet water systems in these contexts are rarely designed for long-term sustainability. As part of the international response, frameworks such as the Comprehensive Refugee Response Framework and the Global Compact on Refugees (United Nations 2018) emphasise the importance of integrating refugees into national systems, including water infrastructure. However, ensuring consistent and affordable water services in these settings requires a comprehensive understanding of life-cycle costs and financing mechanisms that support both refugee and host communities (Huang et al. 2023). This need is increasingly recognised in the context of the global humanitarian crises and the growing importance of resilience and sustainability in water infrastructure management.

In Ethiopia, by 2023, over 900,000 refugees were hosted, with 383,000 South Sudanese refugees residing in the Gambella region, predominantly in three camps near Itang (UNHCR 2023a). The influx of refugees into this area placed a significant strain on local water systems. In Uganda, the Rwamwanja refugee settlement hosts over 90,000 refugees from the Democratic Republic of the Congo, alongside nearly 40,000 people from the host community (UNHCR 2023b). In both situations, the existing host community's water service systems were initially overwhelmed by the influx of displaced persons, triggering tensions and conflict between communities.

National governments in Ethiopia and Uganda, with the support of the United Nations High Commissioner for Refugees (UNHCR), United Nations International Children's Emergency Fund (UNICEF), and other agencies and partners, have made concerted efforts to accommodate these large refugee populations. Both countries have made efforts to integrate refugees into their national systems, but financial sustainability remains a major challenge, as water systems designed for emergency response must evolve into long-term solutions to meet the needs of both refugees and host communities.

Despite these recognised challenges, there remains a significant gap in understanding how to finance water systems in refugee contexts sustainably. Refugee populations, often economically disadvantaged, cannot support the operational costs of water systems through tariffs alone, placing a burden on governments and international organisations.

The objective of this paper is to present an analysis of life-cycle costs for water systems in two refugee and host community settings – Itang, Ethiopia, and Rwamwanja, Uganda. By estimating these costs across different phases of water service provision, the study aims to provide a clearer picture of the financial resources required to sustain water services in these protracted refugee situations. Additionally, the paper seeks to explore how these cost estimates can inform discussions on financing mechanisms that are essential for long-term sustainability.

This paper hypothesises that documenting and analysing the life-cycle costs of water systems in refugee and host communities can reveal key insights into how financing infrastructure in these contexts can inform broader global practices. Understanding these costs and the associated financing needs is critical for developing policies and strategies that ensure the sustainability of water services in refugee settings, contributing to resilience in humanitarian contexts worldwide.

This study employs a Life-Cycle Cost Analysis (LCCA) framework (Fonseca et al. 2010) to assess the financial sustainability of water service provision in two refugee and host community contexts – Itang, Ethiopia, and Rwamwanja, Uganda. The analysis focuses on distinct phases of system development, comparing costs over time and exploring financing mechanisms necessary for sustainable service delivery. The methodology consists of four key components, as follows:

The study primarily relied on secondary data from a wide range of sources, supplemented by interviews with key informants from relevant agencies and utilities. A significant effort was made to collate historical data, which were dispersed across more than 150 reports, contracts, maps, datasets, evaluations, and utility meeting minutes from organisations such as UNICEF, UNHCR, and local utilities. Many of these reference materials date back to 2014 and were sometimes difficult to locate. Only sources considered reliable were included in the analysis. Interviews with key informants helped interpret the data and provided insights into the operational and financial challenges faced during different phases of system development.

To understand the evolution of water infrastructure and service delivery models, the study identified phases of development at each location, including emergency water provision, expansion and investment in permanent infrastructure, and long-term utility management. These phases reflect the transition from humanitarian interventions to utility-based water services, with changes in cost structures linked to the infrastructural developments. The analysis of the collated reports and interviews supported the identification of these phases, and the transitions were visualised in schematic diagrams to provide clarity on the system changes over time.

The analysis focuses on the volume of safely managed water supplied to refugees and host communities as the key indicator of service levels. Estimates for water production and supply were based on available datasets, including borehole yields, water trucking records, meter logs, and non-revenue water estimates. The following additional service indicators were considered:

• Total number of water facilities in service, disaggregated by type of facility and technology;¹

• Percentage of functional, partially functional, and non-functional water supply facilities (specific to Rwamwanja);

• Average quantity of water supplied (litres per person per day) across different phases of system development;

• Time and distance travelled for water collection by end users (sourced from affordability studies);

• Water quality, based on hydrological studies and water safety plans; and

• Reliability of service, measured by reported instances of breakdown and repair.

Using the LCCA framework, cost estimates were calculated for each location and service delivery phase. This included the following:

• 1. Capital Expenditures (CapEx): Costs of infrastructure based on reports and contracts, adjusted to 2022 USD.

• 2. Operating Expenditures (OpEx): Ongoing costs of system operation, including energy, staff, and maintenance, also adjusted to 2022 USD.

In the results table, cumulative CapEx values are presented, with a key assumption that 50% of Phase 1 capital costs were transferred to Phase 2 to account for infrastructure benefits realised later. Detailed data on tariffs, cost recovery, and affordability for users were also collected from utility and other stakeholder reports.

All costs were adjusted to United States dollars with a base year of 2022 to facilitate comparison across the phases. The life-cycle cost data were analysed to highlight the impact of different financing mechanisms, including external support and cross-subsidies, on long-term sustainability.

Three phases were identified in the development of the water system in Itang, Ethiopia, which directly impacted costs and service delivery.

Prior to 2014, the communities in Itang and Thurpham towns relied on wells with handpumps and surface water. The arrival of approximately 120,000 refugees (compared to a local population of 18,000) required a rapid water supply response. The cost structure during this period was dominated by CapEx for the development of surface water abstraction and trucking operations. Phase 1 initially involved surface water abstraction from the Baro River (located 15–18 km from the two newly established camps) with emergency water treatment² and trucking, followed later by the development of four boreholes and their water abstraction complemented by continued trucking of treated surface water. By the end of Phase 1, seven boreholes had been installed and were supporting a large-scale water trucking operation for the camps.

OpEx was high due to reliance on trucking and diesel-powered water distribution. Service levels were basic, with an initial focus on emergency supply rather than sustainability. Water production was limited, averaging 1,000 m³/day across all boreholes, and trucking operations introduced significant inefficiencies, which increased operational costs. The CapEx during this phase totalled $3,452,030, but only 50% of this was attributed to Phase 1, as much of the infrastructure contributed to later phases. With an assumed lifespan of 2 years for the initial setup, the CapEx per year was $1,726,015. The operational costs were even higher, with OpEx totalling $2,427,009 per year, resulting in a combined CapEx and OpEx per year of $4,153,024, or a cost of $11.38 per cubic metre (m³) of water. This phase was characterised by unsustainable costs, driven by the reliance on water trucking and temporary infrastructure. The operation during this phase was managed by UNICEF and UNHCR in collaboration with Non-Government Organizations (NGOs) including the International Rescue Committee.

As the refugee population increased to 190,000, the infrastructure was expanded to include a 20 km pipeline from boreholes near the Baro River to the camps and host communities, reducing the need for water trucking. Water production increased significantly to 3,590 m³/day. The CapEx for this phase was $16,957,213, with an assumed lifespan of 20 years, resulting in a CapEx per year of $847,861. Operational costs were reduced, with OpEx falling to $1,259,436 per year, leading to a combined CapEx and OpEx per year of $1,847,861, or a cost of $1.61 per m³ of water.

The transition to piped systems and borehole-driven supply reduced operational inefficiencies and costs, although reliance on diesel generators kept OpEx relatively high. Treatment was limited to chlorination given the reliance on groundwater sources. The shift towards piped systems improved service levels (77% of water going to refugees and 23% to the host community), but the cost recovery was limited, with the host community contributing only a small share of the system's revenues. The Itang Town Water Utility was established in 2017, and there was a gradual transitioning of the management of the system from NGOs to the utility. Overall, maintaining functioning generators and boreholes was a key challenge in this period.

With 235,000 refugees and 26,000 host community members, Phase 3 marked a critical shift towards sustainable service delivery, and the establishment of the Itang Town Water Utility led to a restructuring of cost recovery mechanisms. Water production increased to 4,113 m³/day as additional boreholes were drilled and electrical grid connections were established. The CapEx during this phase rose to $30,557,213, with a lifespan of 30 years, resulting in an annual CapEx of $1,018,574. OpEx fell significantly to $522,763 per year owing to the more efficient management and reduced reliance on diesel. This led to a combined CapEx and OpEx per year of $1,541,337, or a cost of $1.03 per m³ of water. The shift to utility management reduced both operational inefficiencies and costs.

With the solarisation of the water system, further cost reductions are being achieved. Water production remains at 4,113 m³/day. The annual CapEx increased to $1,065,240, but OpEx dropped to $333,955 per year as solar power reduced energy costs. The combined value of CapEx and OpEx per year is expected to be $1,399,196, or $0.93 per m³ of water. The solarisation of the system marks a significant milestone in achieving long-term sustainability through reduced operational costs. Through total investments of $32 million, costs were reduced by 92% between Phase 1 and Phase 3 (Table 1) (Figure 1).

In Rwamwanja, Uganda, two main phases were identified, as follows:

In Phase 1, the water supply system was developed around a mix of technologies, including 79 boreholes fitted with handpumps, six small water networks supplied by motorised boreholes, and eight protected springs. The motorised facilities were primarily solar-powered, with standby diesel generators and one hydroelectric-powered facility. These decentralised systems produced an average of 900 m³/day. The CapEx for this phase totalled $1,104,820, with an assumed lifespan of 10 years. This results in an annual CapEx cost of $157,831 and a cost per m³ of $0.34. The OpEx during Phase 1 was high, in part due to diesel-powered operations, totalling $354,466 per year, leading to a combined CapEx and OpEx cost of $1.42/m³. The system relied heavily on community volunteers, and although water service was provided, cost recovery mechanisms were limited.

In Phase 2, Uganda's National Water and Sewerage Corporation (NWSC) assumed responsibility for water supply services to Rwamwanja, marking a shift towards utility-based management and more centralised control of operations. This transition is in line with Uganda's Water and Environment Sector Refugee Response Plan (Republic of Uganda. Ministry of Water and Environment 2019), which aims to integrate refugee water services into national systems. By September 2022, the system had expanded to include seven motorised boreholes connected to piped networks, with 653 connections, including domestic connections, public standposts, and institutional and commercial connections. Water production increased to an average of 1,130 m³/day. The CapEx for this phase increased to $1,809,722, with an assumed lifespan of 30 years. The annual CapEx was $452,431, leading to a reduced cost per m³ of $0.15. Operational costs dropped significantly, with an OpEx of only $59,493 per year, leading to a combined CapEx and OpEx cost of $0.29/m³. The introduction of tariffs and centralised management improved cost recovery, although challenges remain in balancing affordability for refugees and operational sustainability. Through total investments of $2,914,542, costs were reduced by 80% between Phase 1 and Phase 2 (Table 2).

The transition to a utility-managed model for water service delivery in refugee-hosting contexts, as seen in Itang, Ethiopia, and Rwamwanja, Uganda, required significant institutional and regulatory changes. In Ethiopia, UNICEF played a critical role in supporting the legal establishment and capacity strengthening of the Itang Town Water Utility (ITWU), alongside efforts at national and regional levels to reform policy and regulatory frameworks. This transition allowed for more sustainable service delivery mechanisms, including flexible management systems that can adapt to emerging challenges such as fuel shortages and security incidents. Similar institutional strengthening was seen in Uganda with the involvement of the NWSC, which improved water system management and operational efficiency.

The results demonstrate that despite increased service levels and reduced production costs achieved through significant investment, introduction of solar technologies, and transition to utility management, long-term financial sustainability cannot be achieved. These findings are consistent with other global research findings (WHO UNICEF, and World Bank 2022). This relates to not only capital expenditures but even the operation and maintenance of systems. Subsidies will remain critical for long-term sustainability, as confirmed by other experiences (Huang et al. 2023). The challenge is exacerbated by the fact that the majority of refugees and their host communities are economically disadvantaged and unable to bear the burden of tariffs required to support long-term cost recovery. Well-targeted subsidies could be designed to direct more spending to those who can least afford water supply services (Andrés et al. 2019). The benefits such as easing tensions between the refugee and host communities with better services believed to be contributing positively to the social cohesion between refugees and their hosts in these fragile situations are hard to monetise locally.

In Itang, the utility's revenue to cover costs is critically dependent on invoicing UNHCR on behalf of the refugees. Revenue collection from the host community is relatively low, as the population is small in comparison to the total beneficiary population, and their ability to pay is limited. In February 2024, the Itang tariff was reduced to $0.42 per m³ following partial solarisation of the system and further tariff reductions are expected based on additional solarisation investments. User tariffs in Rwamwanja were $0.29 per m³ for public standposts and $0.96 per m³ for domestic connections.

To ensure long-term financial sustainability, external financing will continue to play a vital role. Subsidies from governments, international, or local organisations for the operation and maintenance of water systems are crucial to ensuring equitable access to water services, covering operational shortfalls, and protecting capital investments (George et al. 2024). Beyond external support, alternative financing mechanisms should be explored, including the following:

• 1. Cross-subsidies within utilities where wealthier customers or industrial users contribute more, helping offset costs for vulnerable populations.

• 2. Targeted subsidies for refugee-hosting regions, perhaps ringfenced specifically for water service provision.

• 3. Advocacy for national governments to dedicate budget allocations to sustain water services in these vulnerable regions.

The results currently emphasise costs in terms of cost per m³ of water supplied, but this metric does not fully reflect the financial burden on individual users. A more relevant metric could be cost per capita, particularly in refugee settings where income levels are low and tariffs can pose significant challenges to household budgets. The following table summarises the calculated cost per capita for water service delivery in both Itang, Ethiopia, and Rwamwanja, Uganda, across different phases of system development. The results demonstrate a significant reduction in costs, with the combined CapEx and OpEx costs being reduced by 82 and 25% in Itang and Rwamwanja, respectively (Table 3).

While this study adopts an LCCA framework, it is important to acknowledge the limitations of this approach due to incomplete financial data. A full Net Present Value (NPV) analysis of the systems' long-term costs over their entire lifespans was not possible due to the lack of detailed financial projections. As a result, the current results focus on comparing costs across phases without fully accounting for future replacements, expansions, or inflation adjustments. Future research should aim to model life-cycle costs more comprehensively by incorporating NPV calculations and projecting the financial sustainability of these systems over time. It is also important to note that the analysis does not consider the investments made in establishing or enhancing the capacity of the water and sanitation utility in Itang.

The findings from this study show that particularly in refugee-hosting contexts, where both refugees and their host communities are economically disadvantaged and unable to bear the full cost of water services, subsidies will remain necessary to ensure the long-term viability of water systems. These subsidies are required not only to cover capital maintenance expenditures but also, in the short term, to support operational expenditures until refugees and their hosts can afford to pay water tariffs. This external support also plays a critical role in safeguarding social cohesion between refugees and host communities by ensuring equitable access to water services.

While operational efficiencies and cost reductions (e.g., 82% reduction in cost per capita in Itang from Phase 1 to Phase 3) are promising, the study demonstrates that even with these improvements, long-term external financial support will remain essential. However, given the growing pressures on global humanitarian budgets, it is important for key stakeholders to explore sustainable financing mechanisms that extend beyond traditional humanitarian funding arrangements.

Alternative financing strategies for utilities that serve displaced populations should include the following:

• Cross-subsidies within utilities, where higher-income users offset costs for vulnerable users.

• Economic empowerment of both refugees and their host communities to enhance their ability to contribute to water tariffs.

• Advocacy for host countries to allocate specific subsidies or establish ringfenced financial facilities to cover water bills in fragile contexts.

• Exploring climate finance opportunities, as water service resilience is increasingly linked to climate change adaptation and mitigation.

Moreover, creating meaningful livelihood opportunities for refugees and host communities, such as by granting refugees the right to work and developing revenue streams, is crucial for increasing revenue generation through tariffs. These efforts are essential to ensuring that water services can be sustainably financed and maintained in these vulnerable settings.

We would like to express our sincere gratitude to all individuals and organisations who have contributed to this research paper. UNICEF East and Southern Africa Region (ESAR) Country Offices were instrumental in compiling a significant amount of historical data, and were supported by analysing results and inputs to the paper. IRC also acknowledges the support of both the Conrad N. Hilton Foundation and the Ministry of Foreign Affairs of the Government of The Netherlands in supporting IRC's global programme of learning and influence. The authors also extend specific thanks to Patrick Okello from UNHCR Ethiopia, Alemayehu Belay from UNICEF Ethiopia, and Simon Peter Odong from UNHCR Uganda. Numerous other UNICEF, UNHCR, and IRC WASH colleagues and peers provided valuable discussions, feedback, and encouragement during the course of this research. This work would not have been possible without the collective efforts and support of these individuals and institutions. Thank you all for your contributions.

UNICEF and UNHCR initiated and funded the research under the Regional WASH Programme for Refugees, IDPs, and Host Communities in East Africa (R-WASH) with funding from the German Government. IRC also benefited from programme funding from the Netherlands Ministry of Foreign Affairs Directorate-General for International Cooperation (DGIS) and the Conrad N. Hilton Foundation.

The study did not involve human participants in the research.

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.