Uganda
Better Data, Better
Resilience: Lessons in
Disaster Risk Finance from
Uganda
Region: Sub-Saharan Africa

Authors: Barry Patrick Maher,
Richard Andrew Poulter
The World Bank Group

A growing number of low- and middle-income
countries are investing in social safety nets to
improve the lives and livelihoods of their poor and
Results
vulnerable residents. According to the World Bank
(2015) report The State of Social Safety Nets, more ●● The El Niño event in 2016 caused a widespread absence of rain on
than 1.9 billion people in 136 low- and middle- the African continent, contributing to a humanitarian crisis in many
income countries are now beneficiaries of social countries. The early warning system developed by the government
safety net programs. In Africa alone, the number of Uganda (GoU) captured the drought in the north and triggered an
of countries setting up such programs has doubled automatic scale-up of the Third Northern Uganda Social Action Fund
over the past three years, and rigorous evaluations (NUSAF III), a safety net project that is supported by a US$130 million
prove that these programs work to reduce poverty. World Bank lending operation.
●● To finance this scale-up, US$4 million was rapidly drawn from a US$10
But around 55 percent of the world’s poor, or 773 million reserve fund set up under the Disaster Risk Finance (DRF)
million people with acute needs, still lack safety component of the NUSAF project. It is expected that the scale-up
net coverage. Moreover, even where safety nets will provide disaster assistance to about 25,000 households (125,000
are in place, the gains they make possible can people) in Karamoja—that is, to about 20 percent of households in the
be undermined by disasters, which tend to have affected region. This is in addition to the core beneficiaries of about
the highest impact on the poorest. Recent studies 5,000 households (25,000 people) already receiving assistance.
show that disasters drive over 26 million people
into poverty each year (Hallegatte et al. 2016).
●● he DRF component of NUSAF III is
Over the life of the project, t
estimated to finance the cost of scaling up assistance to 84,000
additional households.
I
n Uganda, NUSAF III makes use of social safety nets to invest in the livelihoods of poor households, and it specifically includes a disaster
risk finance (DRF) subcomponent that scales up protection in response to disaster shocks. Uganda’s predominantly rural population, which
consists mostly of smallholder farmers subject to various production constraints, has limited capacity to cope with recurrent shocks. The
country has made significant gains in reducing poverty—the share of people living in poverty fell from 62 percent to 35 percent between 2003
and 2013—but the poorest 40 percent of households remain exceptionally vulnerable to drought and price shocks (Hill and Mejía-Mantilla
2017).
NUSAF III is designed to lessen this vulnerability and “to provide effective income support to and build the resilience of poor and vulnerable
households in Northern Uganda.”1 A US$130 million International Development Association (IDA) lending operation, it builds on findings by
the World Bank Group’s Social Protection and Labor Global Practice showing that safety net systems can provide additional support in times
of crisis, help to defend the welfare of vulnerable households, and enable them to develop strategies to build their resilience (Monchuk 2014).
NUSAF III has a strong focus on labor-intensive public works (LIPW), and through the DRF subcomponent it is developing a mechanism
for scaling up assistance to poor and vulnerable households immediately following disaster shocks in Northern Uganda.2 The DRF
subcomponent is activated temporarily and automatically in response to crisis or shocks, primarily climatic shocks such as drought. Once a
predefined trigger has been reached, LIPW activities are scaled up and coverage extended to additional beneficiary households. The ability
to automatically scale up LIPW is expected to prevent households’ consumption from dropping after drought periods and to protect their
livelihoods and assets, leading to a more rapid post-crisis recovery.
This DRF mechanism, which is being piloted in Uganda’s Karamoja region, became live in July 2016. A full scale-up of LIPW activities was
triggered in the first year of operation, when Karamoja suffered an El Niño–driven drought. The lessons learned from this experience—
described below—relate specifically to the expansion of public works, but some lessons may apply more broadly to scalable social protection
initiatives where different disbursement mechanisms are used (for example, unconditional cash transfers).

10 Key Lessons

1 Establishing a scalable safety net requires cross-disciplinary engagement from
both the World Bank and government.
The critical building blocks for developing a scalable safety net are (i) flexible systems, process and procedures (including
registration, targeting, grievance redress, payment mechanism, etc.) to enable the safety net increase (and decrease)
assistance when appropriate; (ii) good-quality risk information to understand when a shock has occurred; and (iii) exante contingency planning
and financing to rapidly mobilize and disburse funds in the event of a crisis. The NUSAF III project drew on expertise from three Global
Practices (Social Protection and Jobs, Finance and Markets, and Social, Urban, Rural and Resilience) as well as one cross-cutting solutions
area (Climate Change) to deliver a comprehensive One World Bank Group solution addressing these key areas to the client. Furthermore,
the project required interministerial engagement within government, with teams working together from the Office of the Prime Minister, the
Ministry of Finance, Planning and Economic Development, and the Ministry of Gender Labour and Social Development. This cross-disciplinary
approach from both the World Bank and government is required to develop successful scalable safety net programs.

2 A DRF strategy needs to be developed in advance to ensure that once a scale-up
is triggered, funds can flow rapidly to it.
Safety net programs need to expand quickly to effectively protect household welfare following a crisis. If the scalable
component of the safety net does not establish a comprehensive DRF strategy in advance, there may be delays in implementing
the expansion, leading to an erosion of household welfare, greater reliance on humanitarian assistance (which can suffer from
delays), and ultimately a greater cost of providing disaster response.
In the case of the NUSAF III, the World Bank collaborated with the government of Uganda to use financial resources from the IDA, with
US$12 million specifically dedicated to the DRF subcomponent. A total of US$10 million was set aside to finance scale-ups and US$2 million
to invest in data collection systems over the five-year project period.
Financial resources do not have to be in the form of a dedicated credit line; contingent credit arrangements could fit this purpose, as could
a sovereign parametric insurance product if the trigger for a payout mirrors the trigger for scaling up the safety net. The important thing is
that funds are immediately available when a scale-up of the safety net requires them.

3 It is critical to have an objective data source to trigger the mechanism.

Determining when and to what extent a safety net program expands following a crisis requires a transparent, rules-based
approach. In practice this requires an objective indicator of the impact of the crisis on household welfare, with a predefined threshold to
trigger response. It also requires predefined guidelines on how many households will benefit from support, in what regions, for how long,
etc. This approach avoids politicization of response, which can lead to costly delays, and enables a rapid, transparent response. The greater
predictability of when and to what extent a scale-up will occur not only assists the target beneficiaries but also helps the government plan its
budget, thereby ensuring that resources are utilized effectively. In the case of the NUSAF III program, an index of satellite-based observations

1 This is the Project Development Objective; see World Bank, “Project Information Document (PID): Appraisal Stage,” http://documents.worldbank.org/curated/en/209271468116059684/pdf/
PID-Appraisal-Print-P149965-04-13-2015-1428956063313.pdf.
2 Ibid.
of ground vegetation was used as an indicator of drought conditions and was the trigger for implementing a scale-up.
Another innovation of NUSAF III is to have a secondary indicator of conditions in place that can be used alongside the primary indicator,
since a satellite index is only a representation of conditions on the ground and does not reflect the actual situation perfectly. In the case
of NUSAF III, the secondary indicator selected was the Integrated Food Security Phase Classification (IPC), which consolidates wide-ranging
evidence on food insecurity using data and evidence from several development partners. By August 2016, the scale-up threshold of the
primary indicator was met in six of the seven districts where the mechanism was operational. However, the secondary indicator showed
clearly that the conditions in the seventh district were very similar to those in the other six, despite its failing to meet the primary indicator’s
threshold for a scale-up. The secondary indicator was used as the basis for a scale-up in this seventh district—an appropriate step given the
very similar conditions in all districts across the entire region.

4 Given scalable safety nets are an innovative concept, focus on capacity
development in the initial stages is important for long-term sustainability.
The World Bank team spent significant time and resources to develop the capacity of GoU on the key principals of DRF,
drawing on relevant examples from international experience (Kenya, Ethiopia) in the initial six months of the project. As the approach of
the DRF mechanism requires a paradigm shift in disaster response, with government playing a leading role, this development of capacity was
critical and ultimately paid off-GoU has now taken full ownership of the DRF mechanism, with scaling up of NUSAF III seen as a government
response in country.

5 Financial analysis is required to establish the rules for when the DRF mechanism
will trigger.
Assuming limited financial resources for funding scale-ups (whether over a one-year time horizon or across multiple years),
financial analysis is needed to inform key policy decisions regarding how often a scale-up is expected to occur and the level of scale-up
that is to be implemented. In the case of NUSAF III, the World Bank team produced an analytical tool to help the government understand the
financial implications of different policy decisions regarding the rules of the scalability mechanism. The government used this tool to set the
rules of the mechanism such that the annual cost was expected to be within a range aligning with the funds available.
The financial analysis can also provide insights into which assumptions have the largest impact on the cost of the mechanism.

In the case of NUSAF III the DRF mechanism expands the existing LIPW program at the onset of crisis conditions. There are therefore
several factors that determine the total cost of a scale-up:
●● The number of new beneficiaries enrolled in public works after a scale-up
●● The length of projects introduced following a scale-up
●● The number of days worked per month by each new beneficiary
●● The daily wage rate for all participants (both existing and new beneficiaries)
●● The nonlabor input costs of the new projects financed by the scale-up
These factors were important determinants of the overall cost. A financial analysis (including sensitivity analysis) allowed the government
to understand which of the factors had the largest impact on the cost, so that these factors could be emphasized when finalizing the key
policy decisions. In the case of NUSAF III, the tool developed by the World Bank showed that relatively small changes in the daily wage rate
had a very large impact on the overall cost, and so this factor became key in determining the overall cost of a scale-up, bearing in mind how
often a scale-up is expected to occur.

6 There needs to be sufficient absorptive capacity within the disbursement
mechanism for a scale-up to be effective.
A very large scale-up may be desirable in certain situations (e.g., a very severe drought), but this is possible only if there
is sufficient absorptive capacity within the disbursement mechanism. In the case of NUSAF III, where the disbursement
mechanism is public works, implementation was delayed while planning and preparation took place for the many new public works projects
to be implemented (in some districts, there was an increase of over 500 percent in the number of public works projects associated with
the DRF mechanism triggering). The additional planning and preparation involved in implementing such a significant expansion needs to be
considered at an early stage when determining the appropriate level of a scale-up. In order to ensure that funds for public works safety nets
can be disbursed rapidly in a crisis, significant planning must be done before a scale-up is triggered. This can ensure there are enough public
works projects preplanned to facilitate a large increase in available resources.

7 A pilot approach allows the government and development partners to learn from
the process before expanding to wide-scale rollout of the DRF mechanism.
The DRF mechanism in the NUSAF III project is being implemented on a pilot basis, targeting only one region of the country
(Karamoja) and one peril (drought). Following the full scale-up in the first year, several lessons have been learned that can
inform the refinement of the mechanism. A pilot approach is critical to identify key bottlenecks in the process and ensure that sufficient
capacity is built within government and development partners at both the national and local levels for rapid, efficient implementation. This
pilot is allowing the government of Uganda to clear bottlenecks and address capacity gaps. Moving forward, the government will be exploring
other ways to strengthen the delivery mechanism (for example electronic payments, a single registry of beneficiaries, etc.) as well as the
potential to use the DRF mechanism to respond to other crises (for example an influx of refugees).
8 Conducting both an impact evaluation and process review early in implementation
is important to inform the design of the mechanism going forward, as well as
contribute to the global knowledge agenda.
Following the scale-up of the DRF mechanism in 2016/2017, two types of studies are planned to understand its effectiveness.
Both are critical for learning from experience how to refine the mechanism:
●● An impact evaluation will determine how successful the mechanism has been in delivering timely assistance to affected households
and in defending household welfare. This has been embedded in the government-led monitoring and evaluation framework of the project,
and its results will be used to inform further investments in the DRF mechanism as an efficient and effective method of providing disaster
response.
●● A process evaluation will determine how effective the government and development partners were throughout the scale-up process,
from the observation of the primary and secondary triggers to the delivery of resources to beneficiaries. Such a study can identify
bottlenecks in the process and provide useful insights into how the process can be streamlined to ensure there are no unnecessary delays
in reaching the final implementation stage. One key finding from the review in Uganda was that the lack of pre-planned “on the shelf” LIPW
projects at the district level in the first year of the project delayed the disaster response to the El Niño event.

9 The DRF mechanism needs a dedicated individual at the technical level to develop
DRF capacity within key government agencies, and to ensure the scalability
mechanism is mainstreamed in the project.
When a scalability component is introduced into an existing safety net, it is important to mainstream it into the larger project.
This will ensure that the surge in resources during disasters is considered by all project teams at the planning stage. The NUSAF III project
team required further capacity development on the needs of public works prior to a scale-up. This requirement calls for a dedicated individual,
with specific activities listed in his or her terms of reference, to perform the capacity development and mainstreaming. The individual will also
need to align the DRF component with the broader disaster risk management agenda in the country, as safety net scalability will form only one
aspect of government response to crisis.

10 The speed of the DRF component’s response depends on the perils covered.

Drought is the only peril covered for the initial phase of the NUSAF III DRF component. While the severity of dry conditions each
year is unpredictable, the timing of the dry (or lean) season is predictable year to year, and this is the period when vulnerability
is highest. The existing LIPW program aligns with the lean season, providing additional income support when households are most vulnerable.
As the distribution mechanism for the DRF support is the existing LIPW, the timing of disbursement is therefore predictable.

This approach would have to be amended for a DRF mechanism that aims to provide additional resources following different types of crisis
events. For example, a more flexible mechanism would be required to provide additional support to households affected by rapid-onset climatic
events such as floods. Because the timing of such events is unpredictable, there would not be sufficient time to plan and prepare additional
public works projects, and so an alternative disbursement mechanism—such as unconditional cash transfers to affected households, either
direct transfers into beneficiary bank accounts or mobile money solutions—would be preferable.

The World Bank Group’s Disaster Risk Financing and Insurance Program (DRFIP) is a joint program of the Finance and Markets Global
Practice and the Global Facility for Disaster Reduction and Recovery (GFDRR). As a leading provider of analytical and advisory services
on disaster risk finance, it helps governments, businesses, and households manage the financial impacts of disaster and climate risk
without compromising sustainable development, fiscal stability, and well-being.
Partnership with the Swedish International Development Cooperation Agency
In December 2016, the Swedish International Development Cooperation Agency and the Disaster Risk Financing and Insurance Program
of the World Bank, through the Global Facility for Disaster Reduction and Recovery (GFDRR), signed a partnership on DRF for Resilient
Livelihoods to increase the resilience of chronic and transitory poor households to disaster events.

References
Hallegatte, Stephane, Adrien Vogt-Schilb, Mook Bangalore, and Julie Rozenberg. 2016. Unbreakable: Building the Resilience of the Poor in the Face of Natural Disasters. Climate Change and
Development. Washington, DC: World Bank.
Hill, R., and C. Mejía-Mantilla. 2017. “With a Little Help: Shocks, Agricultural Income, and Welfare in Uganda.” Policy Research Working Paper 7935, World Bank Group, Poverty and Equity Global
Practice Group. January.
Monchuk, Victoria. 2014. Reducing Poverty and Investing in People: The New Role of Safety Nets in Africa. Directions in Development. Washington, DC: World Bank.
World Bank. 2015. The State of Social Safety Nets 2015. Washington, DC: World Bank. http://documents.worldbank.org/curated/en/415491467994645020/The-state-of-social-safety-nets-2015.

This work has been supported Information:
by trust fund resources from
the Swedish International Development Barry Maher
Cooperation Agency (SIDA) bmaher@worldbank.org