Evidence on corruption and humanitarian aid
BY PAUL HARVEY
Paul Harvey is a partner with Humanitarian Outcomes, a consulting firm providing research evidence and policy advice to inform better humanitarian aid. Prior to that he was with the Humanitarian Policy Group at ODI and an aid worker for different NGOs in places including Sierra Leone, Somalia and Kosovo.
Corruption in humanitarian action remains an under-researched issue. Humanitarian aid agencies devote significant time, energy and resources to internal efforts to combat corruption but remain reluctant to openly discuss and share learning due to fears that talking about corruption could undermine public support for aid in donor countries, as well as impact local aid relations and increase tensions within societies.
But without talking about corruption more openly and tackling it better, the Core Humanitarian Standard (CHS) theme that, “Principled humanitarian response builds trust and facilitates access” won’t be realised. Corruption undermines trust at multiple levels – disaster-affected people don’t trust aid agencies to provide assistance fairly, local authorities don’t trust principled rhetoric around impartiality, and donor publics don’t trust that their money is reaching people who need it most.
There is a small but growing body of research and evidence on corruption challenges. The Humanitarian Policy Group at ODI produced a short think-piece for DFID on managing corruption risks. This led to studies with Transparency International (TI) on mapping corruption risks, further research in partnership with Feinstein and seven large NGOs on preventing corruption and a TI good practice handbook. Field based case studies have been conducted in Afghanistan, [Liberia] (http://www.odi.org/sites/odi.org.uk/files/odi-assets/publications-opinio...), Uganda and Kenya. There has been great research by aid agencies themselves, for example a CARE report looking at issues around sexual exploitation and food aid in Burundi.
Surveys and qualitative research gathering the views of people in crises consistently highlight corruption as a key concern. The preliminary findings from a Humanitarian Outcomes study on Secure Access in Volatile Environments found that in south central Somalia and Afghanistan corruption was cited as one of the largest impediments to receiving aid. People in these contexts perceive an unclear or unjust rationale for who receives aid and who doesn’t, and there are disincentives, stemming from a sense of resignation or disempowerment, to report instances of corruption and abuse of power. Similarly, a recent NRC study of people's views about humanitarian aid in the Central African Republic found a frequently voiced complaint that, “the process of identifying the most vulnerable people was often deeply flawed”. It also noted that people had little trust in local staff or local partners with frequent accusations of them profiteering from positions.
In WHS consultations in Jordan, refugees talked about complaints mechanisms that were not credible and favouritism as a common practice. Corruption was also identified by people in humanitarian crises as a major obstacle to receiving the aid they needed in the latest State of the Humanitarian System report (2015).
The evidence from all these studies that corruption affects humanitarian assistance is unsurprising. Humanitarian action generally takes place in difficult and insecure contexts where corruption is often deeply embedded in societal norms (as it is in developed contexts). At the same time, there is also recognition that whatever is done to tackle corruption needs to be done in a way that continues to enable humanitarian actors to channel aid to reach difficult-to-access vulnerable populations. Solutions need to be balanced with a view to maintain assistance to those in need (not to cut off assistance) and to enhance its integrity.
The TI authored chapter in the 2015 Humanitarian Accountability Report summarises this body of evidence well. It calls for multi-pronged anti-corruption strategies because transparency and financial accountability alone will not be enough to reduce corruption. This includes preventative standards and controls, access to information, improving communication and addressing feedback from affected communities and, most importantly, integrating the lessons learned from all forms of monitoring and evaluation into new programming.
Tackling corruption better matters because:
- corruption limits the scarce amount of aid reaching people who desperately need it;
- corruption is one of the main factors preventing better and more direct funding to humanitarian actors in developing countries;
- the perception of corruption undermines support for aid in donor countries.
The first point is about the aid reaching those, and doing what, it was intended for. Any aid that is corruptly diverted is not reaching the vulnerable people who need it most. The second point requires some elaboration. Implicit in the current model of international humanitarian action, where funding overwhelmingly flows through international aid agencies and not directly to host governments or civil society, is an assumption that local actors are less able to combat corruption risks, and may even play a part in them. This has been made explicit in interviews with donor government humanitarian officials (although largely undocumented in official policy). The problem is that there is a lack of evidence about whether or not this is actually true and, because it is implicit, it is very difficult for local NGOs or governments to challenge the stereotype. The final point, that humanitarian aid agencies are reluctant to talk openly about corruption because they fear it will undermine public support, is misguided because polling in donor countries suggests many people already see corruption as a major problem for aid. Aid agencies need a better public narrative about how they are addressing corruption. Over 50% of people polled by YouGov in the UK agree or strongly agree that “Most overseas aid does not get to the intended recipients”.
The problem with the way corruption is currently being tackled is not due to a lack of policy or a lack of internal organisational risk management processes. There is no shortage of rhetorical commitments to combatting corruption (zero tolerance policies are constantly promoted, audits and other control mechanisms are regularly undertaken, and toolkits and guidelines exist). But these tend to be narrowly organisational and miss or deliberately steer away from some of the main ways in which humanitarian assistance is being abused and diverted. They also risk being ignored as zero tolerance is an unlikely goal. Given this, it seems that there are strong organisational and institutional incentives within the current system making it difficult to more effectively tackle corruption.
One of the few examples which takes a more collective approach is TI-Kenya’s work in drought-affected northern Kenya, where it is piloting, in partnership with some 40 state and non-state organisations, an integrated mechanism to facilitate joint collection, referral and follow-up actions for corruption complaints from affected communities, as well as a social accountability project supporting communities to monitor implementation of aid projects and to report suspected cases of corruption. More of this type of work needs to happen.
In order to make progress, more needs to be done at the system wide level. There is a need to move from agency specific work to combat corruption within their own organisations to independent and rigorous analysis, reporting and monitoring on corruption risks across responses. I’d therefore like to see (in addition to the recommendations in Chapter 9 of the Humanitarian Accountability Report):
- More systematic surveying and in-depth qualitative work with disaster affected populations. And for this work to be independent and look across organisations, programmes and sectors.
- More rigorous, independent and system wide monitoring of humanitarian action. This would examine the quality of humanitarian action across organisations and sectors on a regular basis. It should involve the establishment of independent monitoring units in all major ongoing and new crises.
- More use of creative methods and triangulation to get at difficult issues including corruption in monitoring and evaluation processes.
- A focus on targeting, including the ways in which selection and registration processes can be treated as corrupt abuse of power. Action research and shared evidence on new and innovative targeting approaches to minimise risks is needed.
- Further research that compares corruption risks across different types of actors and providers – asking for example, where local NGOs and governments are directly responsible, how do corruptions risks compare?
- Action research on whether there are ways of directly funding host governments and NGOs in ways that meet donor requirements for rigour and accountability.
Without this sort of system wide action, trust will remain in short supply. Donors won’t trust local governments or NGOs, people won’t trust aid agencies and western public support for humanitarian aid will be undermined. The World Humanitarian Summit presents a wonderful opportunity for collective action to build that trust but I’m not holding my breath – the appetite for radical action within donor governments and humanitarian aid agencies seems pretty thin on the ground.
Read more about principled humanitarian action in the 2015 Humanitarian Accountability Report at: http://chsalliance.org/resources/publications/har/principled-action