Overhead cost allocation in the humanitarian sector: Chapter 2
What are organisations’ current overhead cost allocation practices?
On behalf of IASC Results Group 5 on Humanitarian Financing, DI, in partnership with UNICEF and Oxfam, conducted research on overheads which informed the development of newly endorsed IASC guidance.
International organisations take very different approaches to the provision of overheads to local and national partners. The results from the mapping of current practices of 22 UN agencies and international NGOs (INGOs) are shown in Annex 1. Key findings are detailed below.
Not all UN agencies and very few INGOs have policies in place regarding their local and national partners’ overheads
UN agencies vary in their approaches to providing overheads. Four of the eight agencies mapped have policies which set out a percentage – or percentage range – of indirect costs that local and national partners are eligible to claim: UNHCR (4%), UN Women (up to 8%), IOM (up to 7%, unless donor agrees to cover a higher rate) and UNFPA (up to 12%). This is provided to the partner to be used at their discretion and is calculated as a proportion of actual programme expenditure. Only UN Women include partners’ use of this funding in audits. Three of the agencies mapped – WHO, FAO, and UNICEF – have no policies on providing overheads to local partners.
Of the 13 INGOs mapped, the majority (11) have no global policies on the sharing or provision of overheads to local and national partners. The remaining two organisations (CAFOD and Christian Aid) do have policies, including splitting donor ICR 50-50 between themselves and their partners. For projects funded by non-grant income streams (such as public donations), both organisations agree a suitable rate with the partner to cover their indirect costs. This is generally around 10% for Christian Aid and 7% for CAFOD.
Despite their lack of policies, some INGOs still tend to share ICR in practice. For example, Kindernothilfe (KNH) provides partners with 10% in overheads as standard. Other INGOs have also begun to share ICR while organisational policies are under development. For example, Cordaid’s humanitarian team made the decision to begin sharing ICR with partners before it became the subject of an organisation-wide discussion. Indeed, as noted below, it is not uncommon for organisations to share some ICR, but this can vary considerably between projects.
This study looked at two pooled funds: OCHA’s Country-based Pooled Funds (CBPF) and the Start Network’s Start Fund. Both have clear equitable practices, with the CBPF committing a 7% overhead fund for both national and international recipients (in 2020, 26% of the CBPF budget was allocated to national organisations). As per the stipulations of the CBPF Grant Agreements, recipients must ensure that overheads are fairly distributed among any sub-grantees; proportionate to the project budget and activities undertaken by each party. The Start Fund also provides 10% in overheads to both international and national partners, although there are currently no policies on the pass-through of overheads from INGO recipients to national partners (see Case study 1).
Most ICR sharing is provided on a case-by-case basis and is not systematised
Most INGOs have decentralised grant management. This means that there are often discrete examples of ICR sharing within organisations where country offices have individually negotiated arrangements. Often country offices are at the forefront of innovative ICR sharing, which in some cases drives change at a head office level. This generally happens in contexts where there is a strong localisation initiative. For example, Trócaire in Sierra Leone decided to equally split the overhead funds received from ECHO grants (7%) between the country office, headquarters, and partners.
In affiliate organisations like Oxfam, approaches to ICR sharing often vary between affiliates depending on their donor partnerships and funding models (i.e., levels of unrestricted income). Some affiliates within larger network organisations demonstrate more progressive practices. For example, Oxfam GB in Myanmar has been sharing ICR as standard for several years. Start Fund Bangladesh has also been leading the Start Fund’s thinking on the issue of ICR sharing (see Case study 1).
Some UN agencies also gave examples of country offices negotiating overheads with long-standing partners, despite the lack of a global organisational policy. For example, WHO in Iraq has been providing overheads to local and national partners. The range of different approaches adopted means that the provision of overheads often varies widely within the same organisation.
What do ICR sharing models look like for INGOs?
The practices mapped in this study included various models of INGO ICR sharing. The following models listed below refer to the lead partner (the first-level recipient of funding from a donor – generally an INGO) and the implementing partner (sub-contracted by the lead partner – often a L/NNGO):
The lead partner shares the ICR received from the donor 50-50 with the implementing partner, regardless of the size of each partners’ budget. This model is used by INGOs such as Concern and Christian Aid. For example, a US$1 million grant with 10% ICR would mean each partner receives 5% (US$50,000).
The ICR is shared between the lead and implementing partner proportional to the size of each partners’ grant. For example, if an implementing partner is only responsible for US$20,000 of a US$1 million grant, they would only receive 10% overhead on their budget (US$2,000). The LIFT fund in Myanmar is one example of proportional ICR sharing (see Case study 4). The benefit of this model is that ICR is fairly weighted according to each partner’s level of implementation. However, there is a risk that this could discourage localisation, as INGOs might still incur the same costs while receiving less ICR. This may result in them choosing to implement more programming directly to retain a higher proportion of the overhead.
In many cases, INGOs negotiate a specific ICR sharing agreement which may or may not be split evenly, depending on the different partners’ needs. For example, a 10% ICR share might be split 7%-3% or it might only be applied to certain parts of the partner’s budget, rather than the whole project. Various examples of this were shared, including consortium arrangements where the grant manager may take a small overhead charge (e.g., 2%) with the rest of the ICR shared proportionately among implementing partners.
Increasing the ICR available
There are some rare examples of donors increasing the volume of ICR available so that both the lead and implementing partners can fully recover the indirect costs incurred through their delivery of a given programme. In this case, the lead partner does not have to share their ICR as an additional percentage is made available. For example, Danida’s new guidelines for Danish CSOs reference the provision of additional ICR for local and national partners (see section on donor-led change). There are various challenges to this approach, not least that donors may not wish to increase the proportion of total funding for overhead costs. On the other hand, such a model may better represent existing costs and is clearly linked to localisation commitments.
Another emerging idea (although this study did not identify any current examples) is an incentivised model. This is where the ICR is shared proportionately, but INGOs who channel funding through L/NNGOs are given additional ICR to incentivise localisation. This model offsets some of the challenges of the proportional approach to ICR sharing.
Organisations provide direct project support costs, sometimes in lieu of overheads
Most intermediary organisations are committed to covering the costs of their partners. If they are unable to provide a percentage overhead, some ask partners to translate overhead costs into direct project costs. Regardless of whether overheads are provided, most INGOs and UN agencies also provide funds for partner project support or administrative costs within the direct budget. For some organisations such as UNICEF, there is an expectation that all relevant costs incurred by partners can be included in the direct budget. FAO also allow a portion of direct support costs to be charged as a percentage of the total budget. While these funds are provided as a lumpsum, they must only be used for activities linked to project implementation.
For other organisations like UNFPA, all expenses associated with the project – including what has been defined here as project support costs – are included in the direct budget. UNFPA also pays an overhead-type cost to be used at the discretion of the partner for any costs that are not directly attributed to project activities. For example, if a partner is tasked with running a workshop, UNFPA will cover all the workshop costs as well as all the relevant support costs related to the following: personnel involved in the workshop, admin staff managing logistics, finance personnel making workshop-related payments, technical personnel developing workshop content and the coordinator overseeing the workshop via the direct budget. They will also provide an overhead as a mutually agreed upon percentage (up to 12%) of total incurred expenditures.
Localisation – including the fair provision of overheads for local and national partners – is a live issue for many organisations
Many of the INGOs mapped in the study see equitable sharing of overheads as key in actioning the localisation agenda. Most of these are in the process of developing policies on ICR sharing. Both Concern and Trócaire are currently establishing organisational strategies that include indicators specifically related to the development of an ICR policy. As donors do not generally grant sub-contracted partners additional ICR, INGOs must manage the loss of income involved in ICR sharing internally (see section on challenges). Some UN agencies are currently (re-)considering their overhead policies. For example, UNHCR recently increased the ICR allowance for local and national partners while WHO is developing a Localisation Strategy for its World Health Emergencies programmes that addresses the issue of allocating overheads. However, DI’s mapping suggests that there is more dynamism around this issue within INGOs than UN agencies.
Case study 1
Start Fund Bangladesh
The Start Fund is an NGO-managed pooled fund that provides rapid financing to combat humanitarian crises. In 2017, a national Start Fund was established in Bangladesh with the aim of directing more funding to national and local responders. Start Fund Bangladesh (SFB) started with 20 INGO members and now includes an additional 27 L/NNGO members who receive funding through INGO partners as well as directly from the SFB. The Start Fund allows all recipients to claim up to 10% ICR on project grants.
Despite L/NNGOs accounting for 83% of SFB project implementation costs in 2019, they received no ICR from the INGO partners that subcontracted them. These INGOs were only sharing ICR with other INGO partners. One of the reasons for this was that contracts were often held by INGO headquarters that tended to deduct ICR before the funds reached the INGO Bangladesh office. This meant that funding was passing through four to five different layers before it reached the local partner. By this point, there was no ICR remaining.
To limit these transaction layers, SFB changed the way in which funding is transferred. By pre-positioning funds at a country level and mandating that funds be transferred between national actors only (including INGO country offices), they reduced the role played by INGO head offices. This allowed INGOs to negotiate with their headquarters to pass on ICR costs to their local implementing partners. An ICR-sharing pilot was subsequently developed together with three INGOs: Save the Children, World Vision, and Care Bangladesh.
As a result, the amount of ICR transferred to L/NNGOs increased significantly with INGOs sharing 25% of total ICR received from SFB with implementing partners in 2020 – up from 0% in 2019. L/NNGOs can now also access the SFB directly and received 52% of the fund’s total ICR in 2020. SFB capacity assessment monitoring found that organisations receiving ICR saw a 16% increase in organisational capacity scores. Receiving ICR has also given L/NNGOs leverage in their ICR sharing negotiations with other partners. Furthermore, as more L/NNGOs receive funding direct from SFB, INGOs are adopting a more consistent approach to ICR sharing to avoid losing their partners. The Start Network are now developing global guidance on ICR sharing, in part thanks to the work of the SFB.