Kenya's 2011 drought response: corruption risks in food assistance programmes

Over the last decade, there has been an increased focus on corruption in emergency assistance. In recent studies, food aid has been identified as one of the most vulnerable sectors, along with cash programming and post-disaster reconstruction. Daniel Maxwell et al., ‘Preventing Corruption in Humanitarian Assistance: Perceptions, Gaps and Challenges’, Disasters, vol. 36, no. 1, January 2012.  In the 2011 drought response in Kenya, Transparency International Kenya (TI Kenya) launched a study examining the integrity, transparency and accountability of food assistance. Integrity is defined as ‘behaviours and actions consistent with a set of moral or ethical principles and standards, embraced by individuals as well as institutions, that create a barrier to corruption’. Transparency International, The Anti-Corruption Plain Language Guide, July 2009.  The main question explored was the extent to which different types of food assistance instruments (in-kind aid, cash and vouchers) posed different risks, and the standards different assistance actors applied to ensure the integrity of these mechanisms, including the government of Kenya and non-government agencies. It also considered whether the policies and systems in place help to protect against corruption and ensure the integrity of assistance.

Food assistance in Kenya and the 2011 drought response

Recurrent failed or poor rains, sustained high food prices, a lack of migration options, insecurity and limited recovery time from previous droughts: all have strained coping mechanisms and exacerbated pre-existing chronic poverty, particularly in the arid and semi-arid lands (ASALs) of northern Kenya. Warnings of another drought emerged in November 2010, but the government and the international community were slow to respond. It was not until May 2011 that the government declared a national disaster, with 3.75 million people said to be food-insecure in rural Kenya. Government of Kenya, The 2011 Long Rains Assessment Report, Kenya Food Security Steering Group, 2011.  The worst-affected areas were the ASALs, where rates of global acute malnutrition in some parts exceeded emergency thresholds. In the district of Turkana North levels of malnutrition were higher than in some areas of South Central Somalia (over 37%).

Food assistance programmes have represented the largest component of humanitarian assistance in Kenya for many years, and have been the most consistently funded. The 2011 drought response was no exception: 84% of the UNled appeal for food aid was funded, compared to 29% for agriculture and livestock, 15% for health and 40% for water and sanitation. For its part, the government of Kenya announced that it would provide 9 billion shillings ($100 million) to buy food for victims of the drought. Kenya Cabinet Authorises $100 Mln To Fight Drought’, 14 July 2011, http://www.reuters.com/article/2011/07/14/kenya-foodidUSLDE76D15C20110714.

Corruption risks in food assistance

In the 2011 drought response in-kind food aid was considered to be most vulnerable to corruption and diversion, largely because of its scale and scattered dispersion and weaknesses in transparency and accountability mechanisms. It was diverted physically through transport and storage, and indirectly through manipulation of targeting and registration. Two food pipelines were run, one by the government and the other by the World Food Programme (WFP). The rationale for running two pipelines was partly to ensure coverage, but there were also wider political factors at work. Having its own pipeline gave the Kenyan government visibility and demonstrated its capacity to support its own people in times of need. These political pressures meant that the government had to be seen to be providing food aid, even though most stakeholders agreed that one pipeline would have been preferable, that WFP was better able to run it than the government, and that the government should have focused on more sustainable interventions. As one interviewee in the TI Kenya study noted: ‘there’s a double language in Kenya: the broad public message is that sustainable programming is most important but at the district level, MPs with a shorter political life want to be seen to be giving food’.

Partly due to these political drivers, the government struggled to ensure an accountable and transparent response. There was no official reporting in the public realm on how government funds were spent, or how food assistance was allocated. According to interviewees, monitoring of government food aid was almost non-existent, and the information channels between different line ministries and the district-level administration were very unclear. At district level, there is little understanding of the criteria used by the government in food allocation, specifically how much could be expected, how it compared between districts and how it should be prioritised. This was due to poor information management as well as considerable inconsistency in the pipeline, which meant that tonnages varied from month to month without warning. An important conclusion from the study is that, if there is no clear information regarding the number of actors involved, how much food aid is planned and the timeframe for receiving it, as well as clarity in the responsibilities of the various actors in the distribution chain, there are going to be larger opportunities for diversion. The lack of governance and accountability in the government’s food aid system was most acutely demonstrated in West Pokot County, where a District Commissioner was arrested in September 2011 and charged with stealing 280 bags of maize worth KSh1.2million (approximately $13,000). Employees of the National Cereals and Produce Board (NCPB) were also charged with selling relief food.

Cash transferred by electronic mechanisms was perceived to be less prone to corruption due to the greater emphasis on accountability mechanisms and the use of technologies which tend to accompany cash programming, such as mobile phone transfers and smart cards. Cash transfers linked to broader social safety net programmes were however harder to scale up in the emergency, and a number of agencies chose to undertake direct delivery of cash. This posed considerable security and corruption risks, entailing substantially more administration and staff time. Generally, cash projects were all closely monitored, perhaps because of their perceived vulnerability to fraud.

Although vouchers are less common in Kenya, the European Commission’s Food Facility is one large-scale programme being implemented in two counties in the ASALs. The biggest risk to the scale-up of the voucher project during the emergency was perceived to be traders providing poorquality goods or short measures, and strong price, market and post-distribution monitoring was required to mitigate these risks.

Common risks

There were also risks common to all three food assistance instruments involving all types of actors – government, NGOs and UN agencies. In particular, there was evidence of political leaders, local elites and local relief committees manipulating targeting and registration. Most complaints from beneficiaries surveyed were about targeting and inconsistency of food distributions. There was political interference in recruitment matters and pressure to employ staff from certain ethnic groups, and the rapid scale-up of the response led to procedural short cuts. There were also risks regarding transportation of food aid, partly due to political influence in commercial trucking companies (as a form of rent-seeking). The fact that transportation costs were not adequately resourced within the government pipeline increased the likelihood of food rations being looted as a form of payment.

These risks were exacerbated in the 2011 drought response by broader challenges to ensuring the integrity of the humanitarian response. First, the failure to respond to early warnings led to a late, rushed and politically pressurised response. This meant that some organisations struggled to put strong procedures in place in the time available, heightening the risk of corruption and diversion. Second, the complicated aid architecture in Kenya, involving multiple ministries with differing responsibilities, meant that it was unclear which ministry was ultimately accountable for the humanitarian response. There was also a lack of clarity in policy. Disaster management policy is not formalised, and food assistance actors straddle the disaster management and social protection spheres, muddying the waters still further. Important reforms and policies have recently been enacted in the areas of governance, ASAL development and drought management, and these should lead to improvements in coordination at central and district/county levels, including the establishment of a National Drought Management Authority. However, it is not yet clear what impact this will have on general disaster management and overall accountability for disaster preparedness and response. Multiple actors are involved in food assistance in the drought response, with limited understanding between agencies as to who was doing what where.

The need for a collective focus on corruption

The study found that important investments have been made in beneficiary accountability mechanisms in the ASALs. The more creative and innovative approaches have longer-term objectives and a means of coordinating with other assistance providers, as opposed to single-agency initiatives or one-off post-distribution monitoring exercises. The value of a more coordinated approach includes improved programming and greater transparency and accountability in capturing the feedback of beneficiaries, as well as enhancing learning between organisations.

The study also argues that actors engaged in humanitarian response in Kenya should understand that corruption is a major public policy issue, and that addressing corruption risks requires political commitment and coordinated action among government and non-government aid agencies. Focusing on internal procedures, whilst important, is not sufficient. In particular, actors involved in food assistance – whether in-kind or cash support – should jointly analyse the potential corruption risks involved in different food assistance instruments as part of emergency preparedness, and collectively identify mitigation measures and processes for the joint monitoring and evaluation of food assistance activities. Agencies should also urge the government to provide more accurate and timely information to the public regarding the response effort, including transparent reporting, monitoring and evaluation of the government’s financial commitments to its own relief efforts. For their part, donor governments should promote transparent reporting of corruption cases and related challenges by partner organisations, and provide more resources to support field monitoring, downward accountability mechanisms and forensic audits.

Adele Harmer is a Partner at Humanitarian Outcomes. Nicolas Seris is Programme Coordinator at Transparency International Kenya. The Transparency International Kenya report Food Assistance Integrity Study, by Adele Harmer, Paul Harvey and Michael Odhiambo, can be found at http://www.transparency.org/news/pressrelease/20120321_drought_response.