GHD and funding according to need
by Andre Griekspoor, WHO March 2005

The Good Humanitarian Donorship (GHD) initiative can be seen as the donors’ equivalent of agency initiatives such as the Red Cross/NGO Code of Conduct, which aims to improve the quality and accountability of humanitarian responses. In the GHD, donors have committed themselves to a set of principles and good practice for humanitarian action, including the provision of flexible and timely funding in proportion to need. This article looks at one aspect of GHD: allocating humanitarian funding ‘in proportion to needs and on the basis of needs assessments’.

Needs assessment in the CAP

In 2003, the Inter-Agency Standing Committee sub-Working Group (IASC sWG) for the Consolidated Appeals Process (CAP) started work on improving the needs assessment aspects of the CAP, with a view to establishing a stronger foundation for the Common Humanitarian Action Plan (CHAP).

The starting-points for strengthening the assessment process in the CAP were:

• to bring together existing needs assessment information;
• to organise the information in a systematic and transparent way; and
• to provide a platform for colleagues in the field to discuss needs, severity and risks.

This, combined with an analysis of who is most vulnerable, where they are and what capacities are available, will assist in setting priorities and in joint programming. If done well, the process and its findings will increase confidence in the priority-setting process for both donors and agencies, and enable more evidence-based resource allocation. A needs assessment framework and matrix (NAFM) was designed to structure the description of the humanitarian situation. It soon became clear that the GHD initiative needed such an objective overview of needs if it was to make progress on its commitment to funding according to need.

Supporting Burundi and DRC in the assessment process

The NAFM was piloted in support of the assessment process for the 2005 CHAPs for Burundi and the Democratic Republic of Congo (DRC). This was a logical choice given that these countries were also the pilot countries for the GHD. In Burundi, the NAFM was considered useful; in DRC, however, it was generally rejected as being too complex and not user-friendly. Most criticism focused on technical aspects like what the measuring units should be (populations, or specific vulnerable groups or geographic areas?), or on levels of aggregation. While demonstrating the potential value of the process, the piloting exercise also showed the constraints. Donors were supportive, but special funds were not made available.

Substantial progress was made only in sectors where additional staff had been deployed (despite the fact that they also found the matrix unhelpful). Staff visited all relevant partners in capitals and in the field, to ask for their analysis and to discuss findings. Time was required to gain trust and confidence, for people to understand what they were contributing to, and to convince people of the added value of the exercise for the entire sector, including their own organisation. A wide variety of sources needed to be covered, and it took time to put things together in a systematic way and to synthesise findings. One of the conclusions of a quick lessons-learned exercise by OCHA was that to do this well, funding and staff time need to be dedicated to the process. It needs to be properly planned, and agencies need to include a specific reference to this process in CAP programmes, so that donors understand the resources required. This confirmed the finding of HPG’s work on needs assessment, namely that assessment has to be recognised as a key activity in its own right.

In general, the results in terms of describing needs fell short of expectations. Perhaps the NAFM document got too much attention. Rejecting the tool meant mostly that nothing else was done. It may have been misinterpreted as being an assessment tool, while the essence was that it would merely structure the process to bring the available information together.

Nonetheless, our experiences in Burundi and the DRC were constructive in indicating how to plan the assessment process better, and how the NAFM tools can be made more user-friendly. Through the IASC sWG for the CAP, an improved version was expected to be ready by the end of February 2005, and preparations are being made to extend support to five or six countries in preparing the CHAP for 2006. Once the process is more firmly established, it is expected that it will be less labour-intensive, and assessment overviews will help to improve monitoring and review processes.

More challenges ahead

The experience of piloting the NAFM highlighted how difficult it is to arrive at an objective overview of a humanitarian situation in a way that would be useful for overall strategic decision-making. Assessing needs, and developing estimates of the resources required to meet them, is a complex problem. It involves many judgments at every stage, and is not a straightforward, rational process. Depending on the principles one uses to make these judgments, different outcomes will result.

Interpreting findings

The findings of an assessment are often compared to benchmarks such as the situation prior to the crisis, or trends over time. In practice, these benchmarks and the subsequent interpretation of needs differ depending on the context. Needs may be understood very differently in Burundi and Kosovo, for example. Others argue for a rights-based approach, and an interpretation of needs based on a comparison with a universal desired minimal level, like the Sphere standards.

Mortality rates are one example of this problem. It seems logical to compare rates to the situation prior to the crisis, or the average in the region. But this is meaningless in crises that have lasted for 20 years. There is also something fundamentally unethical, from a humanitarian perspective, in talking about mortality rates reaching emergency thresholds when they are more than double the baseline rate. Average mortality rates in Sub-Saharan Africa are already more than double what they are in other parts of the world; to declare an emergency in Sub-Saharan Africa, they have to double again! If different absolute emergency rates are applied between contexts, resulting in inequitable aid responses, what does this say about the universal value of human life?

We also need to make predictions for the future, and anticipate future needs. Defining needs and comparing severity within and between humanitarian crises is thus a complex process. Assessment findings do, however, make it possible to establish priorities within a sector. If a health assessment tells us that the majority of excess deaths are due to malaria, malaria control programmes will logically be among the top priorities. Moreover, even priorities within a particular sector, such as malaria control, may require work across sectors – in this case health, water and sanitation and shelter. At the end, the point is not whether water is more or less important than food or health, but how to achieve the optimal balance of sectoral inputs to achieve the priority goals. No practical models exist to make these decisions for us.

Developing programme approaches

There are usually different approaches to addressing needs. Where there is food insecurity one can give food aid, and also seeds or tools. Reconstructing a road may improve access to markets, and may have a longer-lasting effect on malnutrition than a selective feeding programme. Choices are determined by such things as the context, the degree of urgency, the balance between short- and long-term effects and the obligation to strengthen local capacities. The mandates of implementing agencies are also important.

Estimating how much funding is required

The choice of interventions and the approach to implementation has consequences for costs. Here again, there is no quick-fix formula. Some benchmarks exist, like costs per capita for medicines, or per metric ton for a specific food item. While there are some agreed levels for overheads, little is known about what would constitute an acceptable cost for coordination or quality assurance. There could be more transparency in the unit costs of common programmes, which could be adapted to the particular circumstances, for instance if goods need to be delivered to isolated areas by air.

Donors’ funding decisions

All the factors described above would need to be weighed against each other. To make rational funding decisions, cost-effectiveness analyses would need to be made. Work has been done on this for health interventions, but it has not yet been applied in complex emergencies as there are too many variables affecting both costs and effects. Moreover, cost considerations should never be the only criterion of choice. There are also continued problems in securing approval for programmes to reduce vulnerability, such as disaster preparedness and prevention work.

Even if such rational methods existed, donors’ funding decisions are also influenced by other factors: their confidence in the ability of an agency to deliver; the size of their budgets from the previous year; how budgets have been divided between sectors in the past; and official foreign aid policy. Available budgets are finite and almost always inadequate. The result is a second round of prioritisation, in which coverage and/or quality is reduced, or important activities have to be cut.

Funding based on needs: remaining challenges for GHD

If we really want to make progress towards needs-based programming and funding, all of the above will need to be considered. In the case of the pilots in Burundi and DRC, the starting point for donors was that funding would not be increased over the previous year; the intention was to see how available funds could be used more effectively. This constitutes resource-based planning rather than needs-based planning. In addition, the pilots in the two countries did not allow discussions on budget allocations between crisis-affected countries. We are still far from being able to compare DRC and Burundi, let alone all humanitarian contexts. To do this, we would need a ranking system for humanitarian crises based on multiple indicators, like the Human Development Index. Such rankings already exist; ECHO’s Global Humanitarian Needs Assessment (GNA), for example, ranks DRC and Burundi first and second. The GHD group could be used to bring the different ranking methods together, as a first step towards an agreed reference against which to develop resource allocation decisions.

To get to that next step, we also need more objectivity in how costs are estimated in response to needs. Only when we have such overall objective estimates will it start making sense to compare funds received to funds requested. OCHA’s Financial Tracking System has very limited value in determining whether funding needs are being met.

Adequate funding is needed if we are to make progress, and donor countries are encouraged to achieve the Monterrey target, which says that donors should allocate 0.7% of GDP to development aid to the poorest countries. New ways must be found to coordinate and invest humanitarian and development funds in fragile states, so that communities at risk can receive adequate support. At the World Conference on Disaster Reduction in Kobe, Japan, in January 2005 it was proposed that 10% of funds for responding to disasters be invested in strengthening preparedness.


With the GHD initiative, the role of donors is now also included in discussions around improving the effectiveness and accountability of humanitarian interventions. The Consolidated Appeals Process is the only mechanism for common programming, so that the whole can become more than the sum of the parts. It also has the potential to ensure that the limited funds available are used where they are needed most: that funds are allocated according to an evidence-based analysis of needs and priorities. It is not there yet on either, but both from the interagency and the donor side, commitments have been made to strengthen these processes. More work is called for to increase transparency and rationality in the complex process of assessing needs, and programming and allocating resources. The fact that the various stakeholders are finding ways to work together on these is promising.

Andre Griekspoor works in the Department for Health Action in Crises, World Health Organisation, Geneva. His career in humanitarian aid started with Médecins Sans Frontières in 1990, and has included field postings in Sudan, Ethiopia, Rwanda and Liberia. His email address is