In brief 

The humanitarian system often struggles to measure and understand the difference it is making for the people it serves. However, there is now stronger evidence of the system’s effectiveness in achieving outcomes and improving the well-being of crisis-affected people. Over the past decade, the system has invested in technical capacity, programme quality and evidence gathering, and this appears to be paying off. More recent efforts to improve accountability and participation may also make a difference, as engagement was found to play a significant role in the effectiveness of aid.

A fundamental aim of humanitarian assistance is to save lives. In this study period there was some limited positive evidence of humanitarian action reducing excess mortality in crises, but a lack of data makes it difficult to assess its precise contribution. Out of a sample of 29 countries with humanitarian responses only 4 had mortality data available consistently year on year.

There was renewed attention to the importance of effective protection and improvements at the global and country-level were evident but relatively nascent. Leadership on protection was strengthened and within the available, yet patchy, evidence of effective programming the best examples came from child protection, SGBV and multi-sectoral approaches.

Across multiple other sectors, humanitarians paid more attention to the quality of programming, referring frequently to the Sphere minimum standards in humanitarian response in their monitoring and increasing their adherence to the Core Humanitarian Standard (CHS) to track and evaluate performance.

While the past four years have brought new expectations and possibilities for timelier, and therefore more effective, humanitarian support, the system is not as fast as it could be. This is primarily because the system’s operational capacity to respond early is not matched by an increase in well-timed and flexible funding. When the system does act quickly, it must rely on smaller pots of money to do so, meaning that a fast response is often an insufficient one.


‘Does it work?’ is perhaps the most common performance question asked of international aid, yet also one of the most challenging to answer. Due to the difficulties inherent in monitoring and evaluation in a crisis, the humanitarian system generally has a poor track record in demonstrating its effectiveness beyond outputs and numbers of people reached. Data on the outcomes of humanitarian action is not a niche measurement issue; it is central to progress. Evidence gaps limit the system’s ability to use the most cost-effective programme designs, prioritise across sectors and scale innovations. As a result, the 2018 SOHS observed that ‘It is more difficult than it should be to say whether humanitarian activities are effective.’367 

Answering the ‘Does it work?’ question is made more difficult by the lack of consensus on what humanitarian action should achieve. Over the past decade, the system has seen shifts in these expectations, due to a range of factors including a greater emphasis on dignity pushing the system to address a wider range of needs, a rise in protracted crises straining humanitarian capacity, and a greater diversity of perspectives on what humanitarian action should look like in the 21st century. All of this has changed understandings of what humanitarian effectiveness looks like, from saving lives to supporting people’s resilience against future shocks; from a technical exercise in meeting needs to a holistic goal shaped by culture and context.  

Traditionally, effectiveness is evaluated in terms of whether humanitarian programmes have achieved their stated objectives.368 When we understand effectiveness in this way, it is striking how little has changed since the 2015–2017 period; many of the findings in this review repeat those of the 2018 SOHS. Around half of aid practitioners surveyed think that their sector is ‘Good’ or ‘Excellent’ at effectiveness (51%, compared with 52% in the 2018 SOHS). Yet there also continues to be confusion over how the system defines effectiveness and evaluations indicate no meaningful progress since 2018 among humanitarian staff on connecting activities clearly to the overall objectives and outcomes of a response or programme 

Defining effectiveness in terms of meeting objectives offers a limited perspective on the achievements of humanitarian action, as aims are often poorly stated and tend to focus on what agencies do in the short term (people reached, outputs delivered) rather than what they achieve, either in the short or long term.369 For these reasons, we focus in this SOHS edition on the outcomes achieved by humanitarian action over the period 2018–2021. We look first at the two most basic outcomes that can be expected of humanitarian action: that it saves lives and protects people from harm. We then review evidence for other outcomes, such as reducing need or achieving well-being. The final section examines two issues often linked to humanitarian effectiveness: the quality of assistance and its timeliness, the latter of which was raised by aid recipients as a priority topic for the 2022 SOHS.  

Does humanitarian action save lives?

A fundamental aim of humanitarian assistance is to save lives, yet it is difficult to understand the precise contribution of humanitarian assistance to reducing excess mortality. One key challenge is the limited availability of mortality data: a review undertaken for the SOHS shows that, out of a sample of 29 humanitarian responses, little more than half (15) produced any death rate estimates. Available estimates almost invariably only covered pockets of the affected population; only four countries had mortality data available consistently year on year, and availability of records substantially declined over 2020 and 2021, likely owing to the impact of COVID-19 on data collection. Data sets may also be incomplete or inaccurate. In Yemen, for example, understanding the effectiveness of the cholera response was complicated due to an over-diagnosis of cholera and under-reporting of deaths, making the low case fatality rate unreliable.370  

Out of a sample of 29 humanitarian responses, only four had consistent mortality data year on year.

Even when mortality data is available, it is hard to establish what constitutes ‘excess’ mortality due to the lack of wider historical population statistics in many humanitarian crises; most of the poorest and most vulnerable countries are also data poor.371 Using a standard emergency threshold – such as 1 death per 10,000 people per day – for mortality is unreliable and potentially misleading.372 It is even more difficult to establish the causal effect of humanitarian assistance on excess mortality, due to the lack of pre-crisis baselines and the difficulty in establishing a counterfactual. While some approaches may provide some actionable evidence,373 more rigorous estimation of lives saved will likely require greater investments in expertise and data collection and availability – of which there is currently little sign. 

There was limited clear or positive evidence of humanitarian action reducing excess mortality in crises over the study period. Original research for the SOHS across three recent responses suggests a weakly positive effect of assistance on mortality in Nigeria,374 some evidence – albeit less clear – of a beneficial effect in Somalia, and no clear trend in South Sudan. Other evaluations of responses in South Sudan, Yemen and Sierra Leone attributed a fall in the mortality rate below emergency thresholds as a success for humanitarian action,375 although the use of emergency thresholds is contested, and robust methods were not used to make this link.  

Reductions in mortality may also be indirectly linked to humanitarian action through other outcomes, for example through reductions in morbidity or malnutrition. Given the strong link between undernutrition and child death, reducing undernutrition was often taken as indirect evidence of a reduction in excess mortality.376 When sufficiently resourced, humanitarian nutrition programmes were generally found to be effective at reducing severe and moderate acute malnutrition, thus likely contributing to a reduction in deaths of children under five. WASH programmes were also linked to reductions in disease, potentially leading to lower mortality rates.

Does humanitarian action protect people from harm?

Humanitarian protection is concerned with reducing the risk of physical and psychological harm facing people in crises. There are many factors that make humanitarian protection challenging – the breadth of scope has led to confusion over what protection looks like operationally, and protection outcomes are often poorly defined and shaped by factors outside humanitarian agencies’ control. The features that make protection difficult to do also make it difficult to evaluate. As a result, previous SOHS reports have noted that the evidence of protection outcomes is weak.  

There has been system-wide action to address concerns that protection was not being prioritised or well-defined, most significantly in the IASC’s 2016 Protection Policy, which outlined key requirements to support system-wide efforts in protection.377 In the study period, protection was increasingly prioritised in responses – at least on paper. For example, USAID now requires that all proposals include protection indicators and in DRC, all clusters are required to report on several protection indicators.378 However, a major new review of the IASC Protection Policy has found important limitations in implementation. These included a lack of common understanding; weak leadership and accountability; lack of collective ownership; an overly complicated coordination infrastructure; the prioritisation of activities over outcomes; and limited connections with the multiple actors – within and outside the humanitarian system – required to support effective protection.379 

One aid worker interviewed in Cox’s Bazar echoed the overarching sense that protection was not a ‘central pillar’ of humanitarian response: ‘It’s kind of been side-lined and it’s lost its place, which needs to be addressed … As the UN and INGOs in an emergency response, we have a mandate to bear witness and to flag humanitarian and protection issues… we haven’t done that. We have allowed protection issues to go on unchecked.’

As the UN and INGOs in an emergency response, we have a mandate to bear witness and to flag humanitarian and protection issues… we haven’t done that. We have allowed protection issues to go on unchecked.

Protection advocacy 

There were some positive steps in protection advocacy over the study period, with variation across different aspects of protection. Evaluations cite some positive examples, including successful advocacy by UN-Habitat and the Global Land Tool Network for Yazidi land rights in Iraq and more equal ownership for women,380 and by the Health and Protection Clusters around attacks on health facilities in Mali.381 On the global level, collective advocacy by multiple agencies within the Call to Action on Protection from Gender-Based Violence in Emergencies, led by successive donor governments, influenced the IASC’s decision to include GBV responsibilities in Humanitarian Country Team (HCT) compacts.382 There were good examples of protection advocacy by INGOs, but these remained largely ad hoc, short-term and scattered.383   

The more successful advocacy efforts targeted multiple stakeholders, including donors, government and development actors, and had strong individual leadership.384 There has been some progress in strengthening leadership from the UN, NGOs and HCTs to prioritise protection in responses and advocate for the safety of crisis-affected people.385 While outcomes are not yet clear, protection leadership improved in some contexts by the end of the study period, including in DRC and Somalia, with improved links to other clusters and stronger collective protection monitoring systems, respectively. However, strong protection leadership was not evident in several responses in this study period, including during the COVID-19 pandemic, leading to what were deemed foreseeable and avoidable rises in protection risks.386 As we have seen in the ‘Focus on COVID-19’ section the protection impacts of the pandemic in refugee camps were considerable and the lack of effective action was notable.387 

Protection leadership improved in some contexts but remained absent in others, including during the COVID-19 pandemic.

Globally, strategies to protect civilians have focused increasingly on improving engagement with parties to conflict, conflict-affected communities and UN missions and country coordination structures, with mixed success. Some form of engagement has been undertaken in several conflict-affected contexts, yet protection of civilians remains an area of weaker global progress compared to protection against GBV and child protection. Chapter 11 explains some of the failings in global advocacy during active conflict in this period. Engagement with parties to conflict remains challenging, including in DRC and South Sudan,388 and there are limited examples of improved outcomes. Exceptions to this include the demobilisation of hundreds of child soldiers in Mali, which was attributed to advocacy by child protection committees, and healthcare access for older people in Aleppo during the COVID-19 pandemic, attributed to advocacy by community volunteers.389  

Protection programming 

It has been hard to know exactly where protection programming is succeeding or falling short. In part, this is due to the difficulty of attributing responsibilities between humanitarian and other actors, but it is also because progress tends to be monitored and assessed at the level of activity (e.g. number of children reached), rather than outcome (e.g. reductions in risks, vulnerabilities and people feeling unsafe).390 During the study period, there were notable efforts to strengthen the monitoring and evaluation of protection, including by the Directorate-General for European Civil Protection and Humanitarian Aid Operations (DG-ECHO) and InterAction, which is experimenting with proxy indicators, risk equations and outcome mapping techniques.391 

Within the evidence available, the best examples of effective protection programming came from child protection and protection against sexual and gender-based violence (SGBV), such as working with community-based committees; providing safe, sex-disaggregated spaces for homeless children, and developing the child protection capacities of local police and judicial actors.392 Multi-sectoral approaches that integrated protection with other services were found to be effective in the short term. For example, Global Affairs Canada’s provision of shelter, nutrition, hygiene and sexual and reproductive healthcare in DRC helped reduce the vulnerabilities of at-risk girls in the short term, but gains were difficult to sustain beyond the project period.393  

Cash and voucher assistance has also been effective at achieving several protection outcomes.394 Linking cash to life skills workshops for girls facing SGBV was also found to improve protection outcomes.395 However, as discussed further in the following section on CVA effectiveness, the effects of some of these programmes were short-lived.396 Evaluations also point to gaps where specific protection risks from CVA programming have not been effectively considered.397 For example, women have been targeted when collecting their transfer money from ATMs and have been subject to GBV in the home.398 

Available evidence of effectiveness is more mixed for the use of referral mechanisms for SGBV.399 The results depend heavily on whether staff are adequately aware of the existence of referral mechanisms and how to use them safely.400 This was difficult where referral systems were overly complicated: one interviewee in DRC explained that there were three national referral networks in place, each covering a different protection need but with different responsibilities depending on the age of the survivor.401 Additionally, in some refugee contexts, effective referral processes for SGBV survivors were hindered by a lack of trust in government authorities, with humanitarian agencies fearful of survivors being stigmatised or deported.402 

The sheer scale of protection needs in conflict-affected locations and displacement contexts limited progress. For example, research in Yemen found that protection had been integrated into several programmes, but the scale of need far outstripped provision. Some interviewees questioned whether initiatives such as hotlines could really protect against attacks on schools, child marriage and child labour, sexual exploitation and abuse or the recruitment of children into militant groups. Focus group discussions and interviews in DRC revealed that, despite mainstreaming of protection into WASH and nutrition programming, people displaced by conflict still face rape, attacks and sexual violence against children. According to one interviewee in DRC, ‘We are witnessing the violation of human rights because there are rapes in camps, rape of minors; for what, because there is no help, as there is no help the children give themselves to sexuality to have food’.403 (See ‘Focus on: Active conflict’ section)

What other outcomes does humanitarian action achieve for people affected by crisis?

The effectiveness of humanitarian action in achieving outcomes varied widely between sectors, as well as between contexts. Generally, the strongest evidence for effectiveness came from the food security, nutrition and education sectors, as well as in cash modalities and in the COVID-19 health response. Yet, sufficiency of funding, COVID-19 restrictions and the degree of engagement afforded to aid recipients all affected the delivery of outcomes for crisis-affected people. While a key issue influencing the effectiveness of humanitarian action in this period, the varied impacts of COVID-19 on humanitarian action and the results of the humanitarian system's attempts to respond to the pandemic are predominantly covered elsewhere in this report. 

The humanitarian system was generally effective at addressing food insecurity and nutrition – when sufficient resources are available. This is important, as food remains a priority need for crisis-affected people; it was the most-cited need in both the 2018 and 2022 SOHS aid recipient surveys. In Yemen and Venezuela, the system was able to scale up to meet growing food insecurity, despite the extraordinary challenges of operating through COVID-19. Some evaluations called for a greater emphasis on the prevention of malnutrition rather than treatment, a theme that resonates with an overall push towards more anticipatory humanitarian action.404 

Efforts at sustaining children’s access to education through crisis were also effective. Over the study period, education in emergencies programming provided access to and increased schooling for boys and girls, with this extending to host populations in some refugee contexts.405 Educational outcomes were also achieved through other sectors. For example, school feeding programmes – typically classed under food security – reduced absenteeism, encouraged attendance and contributed to improved educational opportunities in several contexts. Meanwhile, in Bangladesh, livelihoods support helped to increase school attendance for girls by enabling families to increase their income, which they then used for school fees.406 A notable exception to the achievement of educational outcomes was the lack of sufficient programming targeting early childhood development and pre-school education, particularly during the COVID-19 pandemic, when educational services had to move online.407 Indeed, attempts to deliver other education activities remotely during the pandemic has mixed results.408 Where possible, support was provided to help children return to school.409 

As Chapter 5 shows, cash-based programming has steadily increased in humanitarian response and, with it, consistent evidence of its effectiveness. Because agencies have had to prove the case for cash over the past decade, it tends to be better monitored than other modalities, and there is a strong evidence base for outcome-level effectiveness. Between 2018 and 2021, cash and voucher assistance was highly effective at achieving positive outcomes for crisis-affected people in the short term. These included improved access to shelter,410 education411 and health services,412 lower morbidity for children under the age of five,413 improved food security and diet diversity,414 a decline in child labour and early marriage,415 increased feelings of dignity,416 improved livelihood opportunities417 and overall improvements in living conditions and well-being, as reported by aid recipients.418  

However, these benefits are generally short-term; few studies measure the longer-term effects of cash transfers, but those that do so suggest that results are not sustained over time.419 While this may be expected for what are often intended to be short-term emergency cash payments, the general insufficiency of humanitarian funding has in some cases required the system to shift towards approaches that can support resilience and reduce the likelihood of individuals needing repeated assistance. With this in mind, experiments with providing cash earlier to prevent greater livelihood losses may be promising (see ‘Timeliness of funding’). Providing cash assistance through social protection systems can potentially deliver longer-term benefits, though there is often tension between humanitarian actors and governments over the amount of cash payment and targeting.420 The effectiveness of cash is therefore not universal: its efficacy depends on design and on the level of resources available. For example, with respect to food security and nutritional outcomes, in some settings multi-purpose or unconditional cash was preferred by aid recipients and more effective, whereas in others in-kind aid and vouchers were found to be more reliable in improving food consumption and dietary diversity.421 

The effectiveness of cash is well evidenced but not universal: it depends on design and sufficient resource and its effects tend to be short term.

Was humanitarian action timely and of good quality?

Humanitarian action is deemed effective not only by what it achieves, but also by its quality and whether it arrived at the right time.  


An area of continued improvement is attention to the quality of programming: agencies increasingly refer to Sphere minimum standards in their monitoring and there has been a noticeable uptick in the use of the Core Humanitarian Standard (CHS) to track and evaluate performance. Increased community participation, which is used as an indicator of compliance with CHS, was also credited with improving quality in other areas.422 In the SOHS survey, 56% of aid recipients felt that aid was of sufficient quality, a slight increase from the 54% reported in the 2018 SOHS. 

In the SOHS survey, 56% of aid recipients felt that aid was of sufficient quality, a slight increase from the 54% reported in the 2018 SOHS. 

Changes in the quality of humanitarian programming over 2018–2021 were dominated by the themes of contextualisation and sector-specific multi-agency initiatives to improve quality. On contextualisation, there were efforts by local and national actors to adapt training and standards to their contexts. This was supported to some degree by the 2018 revision to Sphere, which focused on broadening the standards to be more adaptable according to context. While this was a positive move (we know that technical standards and tools have little impact unless they are contextualised), it also created challenges for field staff, who often lacked guidance on how to adapt the new standards.423 

In the WASH sector, a new global initiative sought to address long-standing quality concerns.424 Led by Oxfam, Solidarités International and the Global WASH Cluster, the Accountability and Quality Assurance Initiative aimed to improve quality monitoring with the piloting of a new framework in 2022. The concerns that prompted this initiative emerged repeatedly in our research, including latrines frequently failing to meet basic Sphere standards, such as locks and lighting.425 There were also challenges with using more sustainable modalities, such as supporting and maintaining water infrastructure, as opposed to water trucking. The overuse of branding by humanitarian agencies contributed to a lack of local ownership in water infrastructure, which meant it fell into disrepair.426 A main contributor to these challenges was the absence of engagement and consultation with communities in WASH programmes.427 

Shelter was another sector that faced frequent challenges in quality, particularly when materials needed to be provided at short notice and there was a lack of prepositioning. The quality of shelter is closely linked to the success of early recovery, the longer-term sustainability of humanitarian support and protection outcomes428 – shelter experts in recent years have suggested that this makes quality shelter programming more complex than initially recognised and, similar to the WASH sector, there are efforts underway to improve learning and share best practices more widely.429 

Finally, as we have seen in Chapter 3, insufficient funding is frequently cited in evaluations as affecting the quality of humanitarian assistance, particularly in the nutrition sector. In response to limited funds, agencies tended to reduce the quality and frequency of rations, which in one instance was linked to an outcome of higher rates of anaemia.430  

Early and timely action 

It stands to reason that aid that comes on time saves more lives and livelihoods and prevents suffering from becoming acute. When compared to where the system was a decade ago, or to development actors’ engagement in crises today, the humanitarian system appears to be considerably faster.431 Evaluations tend to be positive about the timeliness of response, even when aid takes over a week to reach recipients, on the basis that it is not possible for agencies to be faster given the context or the way the system currently functions. But, for people in crisis who have reached the limits of their coping capacities, waiting days for support can still be too long. Humanitarians at the country level, observing the causes and impacts of delays, have a dim view of their own timeliness. Recent work on anticipatory action suggests that it is possible for humanitarian assistance to be much faster than it is currently, reaching affected populations before they are impacted by a crisis event, and this study period was marked by a major new focus on acting as early as possible – both by better preparedness for and anticipation of crises.

Preparedness for timely response 

Preparedness, presence and partnerships were critical to the timeliness of response over 2018–2021, both in rapid-onset disasters and in protracted crises and conflict. The response to Cyclone Idai in 2019 demonstrated this clearly: agencies that pre-positioned staff and stock before the cyclone were able to reach households within the first few days of the disaster, while for others a lack of existing supplies, staff capacity and partnerships caused considerable delays.432 In Yemen, preparedness was credited with contributing to the ‘eradication of cholera within a month of the disaster event despite limited access to the affected communities during the initial phase of the response.’433  

Preparedness has to have the right measures in place at the right scale to be sufficient. In the case of Cyclone Idai, the pre-positioned stock met only 10% to 20% of estimated needs in the immediate aftermath.434 Cash-based assistance has the potential to offer more sufficient, timely aid, but its timeliness depends largely on the pre-existence of registration lists and distribution systems. When these are in place, cash is faster than other modalities, but when they are not, it can be slower. In-country presence is also key, when it is appropriately staffed: in contexts like Mozambique, Malawi and Indonesia, some dual-mandate agencies struggled to take advantage of their country office presence due to the inexperience of development staff in undertaking a crisis response. Of course, the ‘presence’ of international agencies is increasingly shaped by partnerships with local NGOs and host governments, the quality of which were key to timeliness in many responses. In Haiti, for example, lack of pre-existing relationships between international organisations based in Port-au-Prince and local actors based in remote earthquake and tropical storm struck areas in the south led to delayed international aid for those communities outside, and ongoing humanitarian programmes were hampered during the COVID-19 pandemic by a ‘scramble’ to find local partners when international agencies had to withdraw staff.435  

Data- and technology-driven innovations played an increased role in improving the preparedness of the system over the study period. WFP’s development of a database related to natural hazard events to estimate numbers affected by crises was credited with improving the agency’s capacity for timely response; elsewhere, humanitarians used drones to gather data for early warning systems.436 Yet, despite greater investment in contingency plans and early warning systems, these have had little overall impact on the speed of humanitarian response. Some agencies struggled with speed due to outdated or inadequate contingency plans and poor use of early warning systems. Contingency plans for some organisations in Mozambique and Malawi were less useful as they did not account for a Category 3 cyclone, an important lesson for dual-mandate agencies operating in traditional ‘development’ contexts that may face increased disaster risks due to climate change. There were also practical problems at all levels of existing early warning – from triggers being poorly tied to concrete actions and early warning data being inaccessible or poorly analysed, all the way to poor donor response to analysis, which means that responses were delayed even when early warning data is available.437  

Surge mechanisms facilitated timely humanitarian response, but these also had their limitations. Despite their aims, surge mechanisms did not always bring in skilled expertise,438 with one evaluation noting that the skillsets of surge staff were increasingly out of step with the agency’s expanding operational ambitions.439 While attempts were made before 2018 to develop inter-agency surge rosters, these did not appear to be making a difference to the quality or timeliness of response over 2018–2021. Even when capacity contributed to a timelier response, there were trade-offs in terms of the continuity and stability of programming; one country office described the impact of surge staff as ‘chaotic’.440 

Similar trade-offs are evident more widely as humanitarian agencies struggle to find the balance between systems that enable fast response and those that support quality, relevance and engagement. Consulting with crisis-affected populations and taking appropriate steps to engage host governments were cited frequently by humanitarian staff as in tension with timeliness. The reliance on pre-positioned stock for speed means that agencies are delivering what they have, rather than what affected populations necessarily say is of most priority for them – with anticipatory or rapid cash payments being an important exception.  

The three main impediments to timely response were funding (discussed below), import delays and access. Both international and local procurement delays have increased since the beginning of 2021, but international procurement has seen the biggest increase, with logisticians citing import delays as a key risk to humanitarian supply chains.441 Poor-quality roads and infrastructure were a major constraint to the arrival of humanitarian assistance. Echoing a major theme of this edition of the SOHS, bureaucratic impediments and government blockades were also key constraints to timeliness. In Tigray, where the government imposed a blockade on aid in June 2021, 18% of aid recipients said that aid was timely, in stark contrast to other countries, where 70% or more were positive about the timeliness of assistance. Security challenges also caused access constraints. For example, in Haiti humanitarian actors had to negotiate with gangs to secure road access to deliver goods to the earthquake- and storm-affected south of the country.442  

Anticipating and preventing crises 

Until recently, the promise of anticipatory action – that supporting vulnerable populations ahead of a crisis event is more effective, efficient and dignified – has been largely hypothetical. Despite numerous studies on the return on investment of acting early, these rely on models rather than empirical data, and the experience and evidence for anticipation in the humanitarian system has otherwise been weak.443 However, much has changed in recent years and the rise of anticipatory and early action is one of the most significant shifts observed in the system in this edition of the SOHS. 

The rise of anticipatory and early action is one of the most significant shifts observed in the system in this edition of the SOHS. 

Over the period, agencies at the forefront of the anticipatory agenda – the German Red Cross, IFRC, members of the Start Network and WFP – continued or expanded their pilots of forecast-based financing and anticipatory action. In 2019, the joint WFP-START anticipatory funding mechanism, the ARC Replica, was triggered for the first time to provide early action to drought in Senegal, leading to the largest payment to date for a civil society actor for early action ($10.4 million to six agencies).444 Anticipatory action accounted for 18% of all START-funded humanitarian action in 2020, and in 2021 the IFRC’s Disaster Response Emergency Fund allocated over 1 million Swiss francs ($1 million) to forecast-based financing.445 

The high-level commitment to anticipation by the Emergency Relief Coordinator, Mark Lowcock, was a potential game-changer for anticipatory action in the humanitarian space, elevating the issue and turning OCHA’s pooled funding mechanisms into one of the largest pilot schemes for anticipatory financing in the humanitarian system. Between 2019 and 2021, $140 million was allocated through the Central Emergency Response Fund to 13 country pilots. The new Centre for Disaster Protection, and the creation of a multi-agency Anticipation Hub, made significant contributions to learning on early action through original research and the consolidation of resources and evidence.446 

There is evidence for the effectiveness of anticipatory action, but the more critical question – is anticipatory action more effective (or cost-effective) than traditional post-crisis response? – has been difficult to answer due to the inability to make any meaningful comparison across crises or responses. Since 2018, two studies have made progress in this area. Pre-crisis payments made in anticipation of flooding in Bangladesh (by the Red Cross and Red Crescent in 2017 and by WFP in 2019) resulted in improved household food consumption, reduced debt and improved employment rates for households receiving the payment in comparison to households that did not.447 In the WFP programme, even a one-day difference in when the cash payment was received had an impact on well-being outcomes.448  

Whether anticipatory action in the humanitarian system will extend beyond pilots to changing how the wider system does its job remains to be seen. Acting before a crisis is subject to criticism that it is at odds with the classic humanitarian model and, since no contingency plan is perfect at predicting the scale or nature of a crisis, there are questions as to how much time humanitarians should devote to these.449 Others perceive anticipatory action as potentially diverting limited funding away from humanitarian response, and question whether anticipatory action can ever truly be ‘no regrets’ if a crisis fails to materialise as expected and assistance goes to people who are not severely affected.450 In reply, supporters of anticipatory action point to the chronic and extreme vulnerability of people targeted for humanitarian assistance as justification for the ‘no regrets’ approach. One key informant claimed: ‘If you think in Bangladesh you give $50 to a family that in the end was only flooded up to their hip instead of up to their neck, I’m happy to go in front of [the donor] and explain why we did this.’ 

If you think in Bangladesh you give $50 to a family that in the end was only flooded up to their hip instead of up to their neck, I’m happy to go in front of [the donor] and explain why we did this.

Timeliness of funding

Outside of anticipatory pilots, humanitarian funding is often delayed or uncertain at the onset of a crisis, leaving agencies to take on the risk of responding to a crisis with internal funds, while being unable to recoup the costs from a donor further down the line. A study of 10 crisis responses that took place between 2015 and 2019 found that only 41% of total response funding had been committed after six months and, of what was committed, only 64% was disbursed 18 months post-crisis.451 Donors were generally quick to disburse funds for COVID-19, but these were slow to reach frontline responders. The pandemic also affected the timeliness of funding in other ways. The move to remote meetings, for example, led to delays in the allocation of pooled funds, and local organisations reported delays in payment due to the closure of in-country financial institutions.452  

At country level, the lack of timely funding is widely felt and contributes to humanitarian practitioners’ poor assessment of their own timeliness. Only 48% of practitioners felt their sector was ‘Good’ or ‘Excellent’ at responding in a timely way453 – an improvement on the 2018 survey (41%) but still one of the worst-performing areas according to the practitioner survey. Although affected populations were somewhat more positive about timeliness, with 57% saying they were satisfied with when aid arrived, this was a decline from 69% in 2018. Local and national NGOs are worst affected by funding delays and challenges in securing funds: ‘Sometimes this causes problems because the community knows that the project is going to start but it takes time. In most cases, the assistance is always delivered late.’454  

Almost all examples of timely humanitarian action from the SOHS evaluation synthesis were supported through one of three types of mechanism: (1) pre-arranged funding agreements with donors; (2) pooled funds such as the anticipatory or rapid response windows of the CERF, country-based and NGO pooled funds, and the Red Cross/Red Crescent Movement’s Disaster Response Emergency Fund; or (3) agencies’ own internal contingency funds. There is clear evidence that, without these, humanitarian action would have been substantially delayed in many responses. For example, in Bangladesh, through its anticipatory funding window, the CERF made its fastest-ever allocation within four hours of the early action trigger being reached, disbursing $5.2 million to reach 43,000 households prior to the peak of the floods.455 In Ethiopia, where delayed bilateral funding was a major factor in the slow and inadequate response, the CERF was one of the main sources of funding for several months.456 While these types of funding mechanisms appear to make a difference, none of them is currently being used at scale, and this type of funding continues to occupy a small percentage of overall humanitarian spending.  

A similar story played out at the broader level, where donors have gradually increased the funding they report for disaster preparedness and disaster risk reduction (DRR), but the overall proportion of this funding remained the same given the increases in overall international humanitarian assistance. Over the study period, support for DRR and preparedness increased by 50%, from $1.6 billion in 2018 to $2.4 billion in 2020. As a percentage of overall official humanitarian aid, however, it remained stable at 4.2% in both years. The top supporters for preparedness have also largely remained the same: the UK, US, EU, Japan and Germany, with Japan increasing its preparedness and DRR funding significantly in 2020. Germany has been a leader in funding the specific area of anticipation and early action, pledging to provide a minimum of 5% of its funding to anticipatory mechanisms by 2023.457 However, overall funding to date for such anticipatory action or preparedness remains too piecemeal to fully maximise their benefits to people experiencing or threatened by crises. For example, despite the increase in CERF funding for anticipation, this still remains a small percentage of its overall allocation at only 3.8% in 2020 and 5.2% in 2021.

Figure 33: Largest donors to disaster risk reduction and preparedness, 2018–2020

Funding for DRR and preparedness increased by 50% over 2018 to 2020 but continued to comprise around 4.2% of all humanitarian funding. EU Institutions, Germany, Japan and the UK have consistently been among the top five donors to disaster risk reduction and preparedness from 2018 to 2020.

Source: This figure is compiled from a range of sources.

Notes: Figures are bilateral allocable ODA. Excludes ODA targeted to COVID-19. See methodology in Annex 3. Data has been rounded up.


Objectives are the aims and goals of an intervention as set out by an agency and can include a mixture of outputs and/or outcomes.


According to the SDG Indicators Data Platform, only 16% of countries have data available on goal 1 (ending poverty) and on average they only have it available for one year.


ALNAP, ‘Mortality Emergency Threshold: A Case for Revision’, ALNAP 1997-2022, 2 August 2018.


 Such as measuring death rates in real-time to monitor whether they remain within acceptable ranges and comparing these against real-time evaluations of the quality and appropriateness of humanitarian services, to infer an effect on mortality


Please see the methodology for the 2022 SOHS in annex 3.



J. Baker and I. Elawad, ‘Independent Evaluation of the UNHCR South Sudanese Refugee Response in White Nile State, Sudan (2013 - 2018)’ (Geneva: UNHCR, 2018); Grünewald and Farias, Cholera in Time of War; Watanabe, Reinforcing Institutional Capacity.


From global key informant interviews.


Global Protection Cluster, ‘Centrality of Protection in Humanitarian Action Review 2020’ (Global Protection Cluster, 2020).


Gemma Davies, Protection Advocacy by International NGOs in Armed Conflict Situations: Breaking the Barriers (London: HPG/ODI, 2021).


Davis, Protection Advocacy; A. Meral, et al. ‘Refugee Advocacy in Turkey from Global to Local’, ODI, 2021. Please also see Chapter 11 for a discussion on advocacy and humanitarian principles


 G. Davies and A. Spencer, ‘The Call to Action on Protection from Gender-Based Violence in Emergencies: An Assessment of the Role of Collective Approaches’, HPG briefing note (London: HPG/ODI, n.d.).


Global Protection Cluster, ‘The Centrality of Protection in Humanitarian Action Review 2019’ (Global Protection Cluster, 2019); Global Protection Cluster, ‘Centrality Of Protection In Humanitarian Action Review 2020’ (Global Protection Cluster, 2020).


 Itad and Valid Evaluations, ‘Joint Evaluation of the Protection of the Rights of Refugees During the COVID-19 Pandemic’, (London: Itad and Valid Evaluations, 2022).


Global key informant interviews


Global Protection Cluster, ‘Don’t I Matter? Civilians Under Fire - Global Protection Update’ (Global Protection Cluster, 2021).


Global Protection Cluster, ‘Centrality of Protection in Humanitarian Action Review 2020’.


DG ECHO 2021 Protection Mainstreaming Key Outcome Indicator and Monitoring Tool;  Embracing the Protection Outcome Mindset: We all have a role to play. A results-based protection briefing paper. (Washington DC: Interaction, 2020); Washington DC; InterAction, ‘MindShift: A Collection of Examples That Promote Protection Outcomes.’ (Washington DC: InterAction, 2021).


Global Affairs Canada, ‘Evaluation of International Assistance Programming in the Democratic Republic of Congo 2012–13 to 2018–19’ (Global Affairs Canada, 2020).


F. Bastagli et al., ‘World Food Programme Multipurpose Cash Assistance in Lebanon: Protection Outcomes for Syrian Refugees’ (London: ODI and CAMEALEON, 2021) N. Giordano, ‘Evaluation of the Effects of UNHCR Cash-Based Interventions on Protection Outcomes in Rwanda.’ (Geneva: UNHCR, 2020); S. Allen, ‘CVA for Protection: A Mapping of IRC’s Use of Cash and Voucher Assistance to Help Achieve Protection Outcomes.’ (New York: IRC, 2019).


Allen, ‘CVA for Protection: A Mapping of IRC’s Use of Cash and Voucher Assistance to Help Achieve Protection Outcomes.’


Bastagli et al., ‘World Food Programme Multipurpose Cash Assistance in Lebanon: Protection Outcomes for Syrian Refugees’; Giordano, ‘Evaluation of the Effects of UNHCR Cash-Based Interventions on Protection Outcomes in Rwanda.’


WFP, ‘WFP Evidence Summary Cash-Based Transfers Lessons from Evaluations’ (Rome: WFP, 2021); Tirivayi et al., School Meals Programme.


A significant caveat to this being that the evidence on referral mechanisms remains weak. The ability to measure the effectiveness of referrals can be low when agencies do not know the results of referrals to other actors and have no feedback on how the process was experienced by affected populations. A. Koclejda, G. Roux-Fouillet, and N. Carlevaro, ‘Afghanistan Shelter Evaluation Report’ (Norwegian Refugee Council (NRC), 2019).


I. Betzler and O. Westerman, ‘Evidencing the Value for Money of the CCI’s Cash and Legal Programmes’ (Mercy Corps, Oxfam, Norwegian Refugee Council (NRC), International Rescue Committee, Danish Refugee Council (DRC), 2018); Schenkenberg van Mierop, Wendt and Kabir, Joint Appeal to Rohingya; Hanley, Ogwang and Procter, SGBV in the Refugee Population in Lebanon.


Global key informant interviews


Key informant interview.


L. Poulsen, E. Donelli, and S.E. Yakeu Djiam, ‘Cameroon: An Evaluation of WFP’s Portfolio (2012-2017)’ (Rome: WFP, 2018).


H. Roxin et al., ‘Effectiveness of German Development Cooperation in Dealing with Conflict-Induced Forced Migration Crises’ (Bonn: German Institute for Development Evaluation (DEval), 2021).; WFP, ‘Evaluation of the Joint Programme for Girls Education (JPGE) with Financial Support from the Norwegian Government, July 2014 – October 2017’ (Rome: WFP, 2020),; J. Chaaban et al., ‘Multi-Purpose Cash Assistance in Lebanon’ (American University of Beirut and CAMEALEON, 2020),; Tirivayi, et al., ‘Evaluation of the School Meals Programme in Malawi with Financial Support from United States Department of Agriculture (USDA) 2016 to 2018’. ; J. De Hoop, et al., ‘“Min Ila” Cash Transfer Program for Displaced Syrian Children in Lebanon (UNICEF and WFP). Impact Evaluation Report Endline’ (New York: UNICEF and American Institutes for Research (AIR), 2018),; J. Murray, ‘Evaluation of the European Union’s Humanitarian Response to the Refugee Crisis in Turkey’ (European Commission ECHO, 2019),; DFAT, ‘Performance of Australian Aid 2017-2018’ (Australian Government - Department of Foreign Affairs and Trade, 2019),; S. Lister, ‘Ethiopia: An Evaluation of WFP’s Portfolio (2012-2017)’ (Rome: WFP, 2019),


Global key informant interviews


D. Nugroho et al., ‘It’s Not Too Late to Act on Early Learning: Understanding and Recovering from the Impact of Pre-Primary Education Closures during COVID-19’ (New York: UNICEF, 2021).


 UNHCR, ‘COVID-19 Evaluative Evidence Brief #2’ 2021.


UNHCR, ‘Phase II of the RTA of UNICEF’S Response to COVID-19 in Eastern and Southern Africa’ UNHCR, 2021.


J. Grasset and Q. Khattak, ‘Cash on the Move: Adapting Multi-Purpose Cash “Plus” Assistance to Support People on the Move. A Case Study from Peru.’ (London: Save the Children International, 2021); J. Hagen-Zanker, M. Ulrichs, and R. Holmes, ‘What Are the Effects of Cash Transfers for Refugees in the Context of Protracted Displacement? Findings from Jordan: The Effects of Cash Transfers for Refugees in Jordan’, International Social Security Review 71, no. 2 (April 2018): 57–77.


Hagen-Zanker, Ulrichs, and Holmes, ‘What Are the Effects of Cash Transfers for Refugees in the Context of Protracted Displacement?’; K Abay et al., ‘COVID-19 and Food Security in Ethiopia Do Social Protection Programs Protect?’ (Washington DC: The World Bank Development Economics Development Research Group, 2020); A. Hızıroglu Aygün et al., ‘Keeping Refugee Children in School and Out of Work: Evidence from the World’s Largest Humanitarian Cash Transfer Program’, Discussion Paper No. 14513 (IZA Institute of Labor Economics, 2021), W. Moussa et al., ‘The Impact of Cash Transfers on Syrian Refugee Children in Lebanon. Economic Research Forum’, Working Paper No. 1457 (Economic Research Forum, 2021); Daniels, Anderson and Yusuf Ali, 2017 Somalia Response; WFP, ‘WFP Evidence Summary Cash-Based Transfers Lessons from Evaluations’ Roxin et al., ‘Effectiveness of German Development Cooperation in Dealing with Conflict-Induced Forced Migration Crises’; Chaaban et al., ‘Multi-Purpose Cash Assistance in Lebanon’; T. Frankenberger, K. Miller, and T.C. Taban, ‘Decentralized Evaluation of UNHCR’s Livelihoods Programme in South Sudan (2016-2018)’ (Geneva: UNHCR, 2019)


Hagen-Zanker, Ulrichs, and Holmes, ‘What Are the Effects of Cash Transfers for Refugees in the Context of Protracted Displacement?’; Abay et al., ‘COVID-19 and Food Security in Ethiopia Do Social Protection Programs Protect?’; Moussa et al., ‘The Impact of Cash Transfers on Syrian Refugee Children in Lebanon. Economic Research Forum’; Daniels, Anderson and Yusuf Ali, 2017 Somalia Response; Chaaban et al., ‘Multi-Purpose Cash Assistance in Lebanon’


Doocy et al., ‘Cash and Voucher Assistance and Children’s Nutrition Status in Somalia’, Maternal & Child Nutrition 16, no. 3 (July 2020).


Hagen-Zanker, Ulrichs, and Holmes, ‘What Are the Effects of Cash Transfers for Refugees in the Context of Protracted Displacement?’; Abay et al., ‘COVID-19 and Food Security in Ethiopia Do Social Protection Programs Protect?’; S. Kurdi, ‘The Nutritional Benefits of Cash Transfers in Humanitarian Crises: Evidence from Yemen’, World Development 148 (2021): 105664; Chaaban et al., ‘Multi-Purpose Cash Assistance in Lebanon’; Frankenberger, Miller, and Taban, ‘Decentralized Evaluation of UNHCR’s Livelihoods Programme in South Sudan (2016-2018)’


Moussa et al., ‘The Impact of Cash Transfers on Syrian Refugee Children in Lebanon. Economic Research Forum’; Roxin et al., ‘Effectiveness of German Development Cooperation in Dealing with Conflict-Induced Forced Migration Crises’; Chaaban et al., ‘Multi-Purpose Cash Assistance in Lebanon’; Frankenberger, Miller, and Taban, ‘Decentralized Evaluation of UNHCR’s Livelihoods Programme in South Sudan (2016-2018)’


WFP, ‘WFP Evidence Summary Cash-Based Transfers Lessons from Evaluations’; Betts et al., WFP Regional Response; Koclejda, Roux-Fouillet, and Carlevaro, ‘Afghanistan Shelter Evaluation Report’


Key Aid Consulting, ‘British Red Cross Final Evaluation: Nepal Earthquake Recovery Programme’ (British Red Cross, 2018); WFP, ‘WFP Evidence Summary Cash-Based Transfers Lessons from Evaluations’; Frankenberger, Miller, and Taban, ‘Decentralized Evaluation of UNHCR’s Livelihoods Programme in South Sudan (2016-2018)’


Hagen-Zanker, Ulrichs, and Holmes, ‘What Are the Effects of Cash Transfers for Refugees in the Context of Protracted Displacement?; Chloé Maillard and N. Minnitt, ‘Comparative Study of the Effects of Different Cash Modalities on Gender Dynamics and People with Disabilities Aleppo, Syria’ (Utrecht: Dutch Relief Alliance (DRA), 2021); Chaaban et al., ‘Multi-Purpose Cash Assistance in Lebanon’


Moussa et al ‘The Impact of Cash Transfers on Syrian Refugee Children in Lebanon’, (ERF, 2021).


P. Breard, ‘Evidence Summary on COVID-19 and Food Security’ (United Nations Evaluation Group (UNEG), 2021)


 e.g. Mutsaka et al., Real-Time Review; ADE, Response to the Rohingya Refugee Crisis; SANDE CONSULTORES LDA, ‘External Evaluation of Cyclone Idai and Kenneth Response in Mozambique’ (Oxford/London: Oxfam and DEC, 2021)


Maillard, Setyawan and Juillard, Real-Time Evaluation; Schenkenberg van Mierop, Wendt and Kabir, Joint Appeal to Rohingya; H. Daoudi, A. King, and R. Fransen, ‘Meta-Analysis of the Engagement of UNFPA in Highly Vulnerable Contexts’ (New York: UNFPA, 2018); H. Audi et al., ‘Lebanon, Livelihoods and Resilience Activities (2016-2019): Evaluation’ (Rome: WFP, 2019); FAO, ‘Evaluation of FAO’s Work on Gender’ (Rome: FAO, 2019); T Bene et al., ‘Strategic Evaluation of WFP Support for Enhanced Resilience’ (Rome: WFP, 2019)


ADE, Response to the Rohingya Refugee Crisis; SANDE CONSULTORES LDA, ‘External Evaluation of Cyclone Idai and Kenneth Response in Mozambique’; Maunder et al., Somalia: An Evaluation; Shaheen, Wichterich and Sardiwal, Norwegian Church Aid; R Ndhlovu, H. Radice, and M. Genene, ‘Evaluation of Oxfam’s 2017 Drought Response in Ethiopia’ (Oxford: Oxfam, 2018); D. Stone and K. Chowdhury, ‘Christian Aid’s Rohingya Crisis Response in Bangladesh’ (Christian Aid, 2019); Solutions Consulting, ‘Final Evaluation Report. Disaster Response in Yemen’ (Dublin/Paris: Concern Worldwide, Agency for Technical Cooperation and Development (ACTED), 2019)


ADE, Response to the Rohingya Refugee Crisis; Sansom, Leidecker and Gressmann, East Africa Crisis Appeal; Ndhlovu, Radice, and Genene, ‘Evaluation of Oxfam’s 2017 Drought Response in Ethiopia’; H. Sultan et al., ‘British Red Cross Bangladesh Population Movement Operation External Evaluation’ (Geneva: ICRC/IFRC, 2019); Stone and Chowdhury, ‘Christian Aid’s Rohingya Crisis Response in Bangladesh’; Solutions Consulting, ‘Final Evaluation Report. Disaster Response in Yemen’; Sanderson, Patnaik and Osborne, Nepal Earthquakes Appeal; IMPAQ International, LLC, ‘Mid-Term Evaluation of “Support for the Integrated School Feeding Program” in Côte d’Ivoire’ (Rome: WFP, USAID/OFDA, 2019)


D. Martin and j. Brown, ‘“Littered with Logos!”: An Investigation into the Relationship between Water Provision, Humanitarian Branding, Donor Accountability, and Self-Reliance in Ugandan Refugee Settlements’, Refugee Survey Quarterly 40, no. 4 (16 December 2021): 433–58; and country case studies for the SOHS


 S. Bourne, ‘User-Centred Design and Humanitarian Adaptiveness’, Case Study (London: ODI/ALNAP, 2019)


K. Sutton and E. Latu, ‘Tropical Cyclone Gita Response Program Evaluation’ (Geneva: CARE International, 2018); Koclejda, Roux-Fouillet, and Carlevaro, ‘Afghanistan Shelter Evaluation Report’; Sansom, Leidecker and Gressmann, East Africa Crisis Appeal; Sanderson, Patnaik and Osborne, Nepal Earthquakes Appeal; Key Aid Consulting, ‘Impact Analysis of Aid in Haiti - 10 Years On’ (Swiss Solidarity, 2020); Key Aid Consulting, ‘British Red Cross Final Evaluation: Nepal Earthquake Recovery Programme’


Global Shelter Cluster, ‘Shelter Projects 8th Edition. Case Studies of Humanitarian Shelter and Settlement Responses 2019-2020’ (IOM on behalf of the Global Shelter Cluster, 2021), 2019–20.


M. Saboya et al., ‘Evaluation of the Nutrition Components of the Algeria PRRO 200301. January 2013 – December 2017’ (Rome: WFP, 2018)


In the SOHS assessment of humanitarian action over 2010-2012, the humanitarian system was considered to be far too slow – delays in the system’s response to drought in the Horn of Africa feasibly led to hundreds of thousands of preventable deaths and aid recipients in the 2012 survey cited timeliness as the biggest improvement needed in the services provided to them.


A. Khan et al., ‘Learning from disruption: evolution, revolution or status quo?, (ALNAP,2021)


T Hanley and et al., ‘Evaluation of WFP’s Capacity to Respond to Emergencies. Evaluation Report Vol I.’ (Rome: WFP, 2019


UNICEF, Work to Link; FAO, ‘Evaluation of FAO’s Contribution to the Humanitarian–Development–Peace Nexus 2014–2020’ (Rome: FAO, 2021); Hanley and et al., ‘Evaluation of WFP’s Capacity to Respond to Emergencies. Evaluation Report Vol I.’; Steets et al., ‘Inter-Agency Humanitarian Evaluation of the Drought Response in Ethiopia 2015-2018’; FAO, ‘Evaluation of the role and work of the Sub-Regional Office for North Africa (SNE) 2017-2020’ (Rome: FAO, 2020), 2017–20


Baker et al., Response to Cyclone Idai; UNFPA, ‘Evaluation of UNFPA Support to the Prevention of, Response to and Elimination of Gender-Based Violence and Harmful Practices (2012-2017) Volume 1.’ (UNFPA, 2018); Daoudi, King, and Fransen, ‘Meta-Analysis of the Engagement of UNFPA in Highly Vulnerable Contexts’


Hanley and et al., ‘Evaluation of WFP’s Capacity to Respond to Emergencies. Evaluation Report Vol I.’


Baker et al., Response to Cyclone Idai; UNFPA, ‘Evaluation of UNFPA Support to the Prevention of, Response to and Elimination of Gender-Based Violence and Harmful Practices (2012-2017) Volume 1.’; J. Murray, F. Pedersen, and S. Ziesche, ‘Evaluation of the Global Cluster for Early Recovery’ (New York: UNDP, 2018); SANDE CONSULTORES LDA, ‘External Evaluation of Cyclone Idai and Kenneth Response in Mozambique’


CHORD, ‘The State of Logistics and Supply Chain in the Humanitarian Context. Global Survey Findings’ (Center For Humanitarian Logistics and Regional Development (CHORD), 2022).


Global key informant interviews.


L. Weingärtner, T. Pforr, and E. Wilkinson, ‘The Evidence Base on Anticipatory Action’ (Rome: WFP, 2020).


Start Network, ‘Annual Review 2020’ (Start Network, 2021).


This figure does not include additional investments by individual Partner National Societies in anticipatory action and therefore the total figure for the Red Cross/Red Crescent Movement’s spending on forecast based financing is higher than the DREF allocation.


‘Centre for Disaster Protection’, n.d., ; ‘Anticipation Hub’, n.d


While these studies did not directly compare households that received a pre-crisis payment to those receiving a post-crisis payment of equal value at a later point in time, they provided an accurate comparison between anticipatory payments and ‘normal’ post-crisis payments in so far as post-crisis payments still had not been made 80 days after the floods (and therefore comparison of an early payment to those who didn’t receive any payment 80 days after a crisis was valid). [email communication, Ruth Hill]


A. Pople et al., ‘Anticipatory Cash Transfers In Climate Disaster Response’, Working Paper (London: Centre for Disaster Protection, 2021).


Grünewald and Farias, Cholera in Time of War; FAO, DI and NRC, ‘Development Actors at the Nexus: Lessons from Crises in Bangladesh, Cameroon and Somalia’ (Rome/Bristol/Oslo: FAO, DI and NRC, 2021), 10.


E. Crossley et al., ‘Funding Disasters: Tracking Global Humanitarian and Development Funding for Response to Natural Hazards’ (Centre for Disaster Protection and Development Initiatives, 2021).


OCHA, ‘Somalia Humanitarian Fund Annual Report 2020’ (Somalia: OCHA, 2020).


36.3% said good and 11.5% said excellent


Key informant interview in DRC.


CERF Advisory Group, ‘Anticipatory Action Update’ (Central Emergency Response Fund (CERF), 2020).


C. Sozi, ‘Ethiopia Anticipatory Action Drought 2021’ (New York: UN Central Emergency Response Fund (CERF), January 2021).


Foreign Minister Heiko Maas, ‘Statement by Foreign Minister Heiko Maas at the High-Level Humanitarian Event on Anticipatory Action’ (German Federal Foreign Office, 9 September 2021).


The Integrated Phase Classification (IPC) is an internationally accepted five-phase scale which governments and humanitarian actors use to understand the severity of food emergencies. Phase 1, is minimal food insecurity, phase 2 is ‘stressed’, phase 3 is crisis, phase 4, emergency, and phase 5, catastrophe/famine/likely famine. Acute food insecurity is IPC phase 3 or above. For a full description see IPC, ‘Integrated Phase Classification’, Famine Early Warning System Network (FEWS NET), n.d.


FSIN and Global Network Against Food Crises, ‘Global Report on Food Crisis’ (Rome: Food Security Information Network (FSIN), 2021).


Although one reason for this increase is due to differences in RRPs were included in aggregate appeals data – see Chapters 2 and 3 for more.


Alex De Waal, ‘Social Nutrition and Prohibiting Famine’, World Nutrition 9, no. 1 (19 April 2018): 31.


OCHA, ‘Global Humanitarian Overview 2020’ (Geneva: OCHA, 2020).


ALNAP, ‘The State of the Humanitarian System.’, ALNAP Study (London: ALNAP, 2018); D. Hillier and B. Dempsey, ‘A Dangerous Delay: the cost of late response to early warnings in the 2011 drought in the Horn of Afrrica, Oxfam, 2012.


Maunder et al., Somalia: An Evaluation ; D. Maxwell et al., ‘Viewpoint: Determining Famine: Multi-Dimensional Analysis for the Twenty-First Century’, Food Policy 92 (2020): 101832.


UN Secretary General, ‘With 30 Million Facing Famine, Secretary-General Announces Prevention Task Force, Warns Security Council against Cutting Aid as Solution to Economic Woes’, United Nations, 2021.


Margie Buchanan-Smith, Jane Cocking and Sam Sharp, Independent Review of the IPC South Sudan (London: HPG/ODI, 2021).


Buchanan-Smith, Cocking and Sharp, IPC South Sudan.


Maxwell et al. (Ibid); F. Checchi et al., ‘Estimates of Crisis-Attributable Mortality in South Sudan, December 2013-April 2018’ (London: London School of Higiene & Tropical Medicine (LSHTM), 2018),


 Global Network Against Food Crises.


Global key informant interview.


Global Network Against Food Crises, ‘War in Ukraine and Its Impacts on Food Crises. A Review of Existing Analyses and Evidence’, Global Network Against Food Crises, 26 May 2022.


In the SOHS aid recipient survey, food continued to be the top need people cited when asked "What sort of aid was most needed?" at 38% of respondents.


WFP ‘Yemen Profile’, (Yemen, WFP, 16 Nov 2021).


Hanna, D. K. Bohl, and J. D. Moyer, ‘Assessing the Impact of War in Yemen: Pathways for Recovery’ (New York: UNDP, 2021).


Global Acute Malnutrition (GAM) is a measure of acute malnutrition in refugee children aged between 6 and 59 months. GAM provides information on the percentage of all children in this age range in a refugee population who are classified with low weight-for-height and/or oedema. It is obtained by combining the number of children in this age range who have moderate acute malnutrition and severe acute malnutrition.


OCHA, ‘Yemen Situation Report’ (Yemen: OCHA, January 2022).


 D. Maxwell et al., ‘Famine review of the IPC Acute Food Insecurity and Acute Malnutrition Analysis: Conclusions and recommendations for five areas in Yemen.’ (IPC, 2022).


D. Maxwell et al., ‘Famine review of the IPC Acute Food Insecurity and Acute Malnutrition Analysis’


J. Kim et al., ‘Sharing to Survive: Investigating the Role of Social Networks During Yemen’s Humanitarian Crisis’ (Washington DC: Resilience Evaluation, Analysis and Learning (REAL), 2022).


Oxfam, ‘Nearly 40 per Cent of Yemen Families Forced into Debt to Pay for Essentials’, Oxfam, 2021.