Information exchange for humanitarian coordination in the Horn of Africa
by Ben Watkins June 2003

The benefits of free flows of humanitarian information are immense, but so too are the challenges, both institutional and technical, involved in establishing exchange mechanisms.

This article looks at some of the constraints facing humanitarian data and information exchange in the Horn of Africa, and reviews possible solutions. It suggests that, at national and regional levels, there is a need to build up the voluntary but formal institutional arrangements for information exchange. Technical services or ‘warehouses’ should store, compile and disseminate data, while providing basic technical support to humanitarian partners.

The benefits of information exchange

Humanitarian data and information includes conventional early-warning information, data on planned and ongoing programmes, baseline socio-economic, geographic, health and environmental data, and project and programme evaluation reports. Numerous organisations – NGOs, government bodies, the UN and donors – collect this material, but no single entity gathers enough to meet all its needs. The rest is generally acquired through free exchange.

Effective exchange of data and information between organisations and countries has a number of benefits. First, organisations can increase the impact of their programmes by working in locations and sectors not covered by other organisations. This kind of coordination can only be achieved if everyone knows what everyone else is doing. Exchanging programme information also ensures that programmes reinforce, rather than undermine, each other. For example, a free food aid distribution can suppress market prices, with a negative impact on, say, a project to increase farm income from wheat production. The impact of nutrition programmes is likely to be increased by other programmes for disease control, hygiene education and the provision of safe drinking water. Second, information is expensive. When data and information are useful to numerous organisations (such as baseline or early warning information), sharing it reduces the total information costs of operations. Third, information exchange increases the knowledge of the NGO community as a whole, and at least in theory encourages learning about different approaches, evaluations and experiences.

Information exchange in the Horn

In most of the Horn of Africa, systems to collect data and information relevant to humanitarian operations are unreliable, and it will take many years to develop independent and high-capacity processes. That said, valuable resources do exist, albeit under-utilised. With better sharing mechanisms, these resources could be exploited, helping to plug some critical gaps at relatively low cost.

The information weaknesses in the Horn are extensive. There is, for example, no comprehensive and compiled information on the organisations working in the humanitarian sector, what they are doing, and where they are. Up-to-date information on humanitarian expenditure by sector, country or organisation is hard to find. Most operations require, but do not have, detailed logistics maps, although the kind of information necessary to generate them is available for most of the region.

There are several critical constraints to information-sharing in the Horn. First, the public-information culture is marked by secrecy, manipulation of humanitarian information and deficient accountability. Often, the scale of a disaster is played down for political reasons; in May 2001, for instance, the Sudanese government vehemently challenged an NGO’s claim that famine was looming in the Nuba Mountains. Conversely, governments can be slow to release information that might lead agencies to scale back their operations.

The second constraint is the complexity of the humanitarian community itself. Over 100 humanitarian NGOs work in Ethiopia, Kenya, Somalia and Sudan, together with at least nine UN bodies. Several government ministries can be involved in aid interventions, but cooperation between them is usually weak. The donor community too has become more complex, with the entry of new players from South-east Asia and the Arab states. Competition for resources and media coverage discourages networking between agencies, and more particularly between NGOs and donors. Exchanging information between the region’s states is also difficult, particularly with sensitive cross-border issues like security and migration. Regional groupings like the Intergovernmental Authority on Development (IGAD) – which includes Djibouti, Eritrea, Ethiopia, Kenya, Somalia, Sudan and Uganda – lack the institutional capacity and political will to change this. Several IGAD information initiatives, such as the Regional Early Warning System, have collapsed, despite substantial donor assistance. Finally, there are basic technical constraints, particularly in government departments and the smaller NGOs. Data can be lost or in incompatible formats, and exchanging it, for example digitally, can be difficult.

Towards a solution?

Several promising initiatives hold out the prospect of tackling these institutional and technical difficulties. National institutional structures for humanitarian coordination, including an infor-mation sharing component, are beginning to emerge. In Ethiopia, the Early Warning Working Group (EWWG) is a relatively free forum for exchanging information and methodological collaboration. The Kenya Food Security Working Group (FSWG) has made impressive progress, as has the Somalia Aid Coordination Body (SACB). Joint assessment processes in most countries, including government, donors, NGOs and the UN, are also helping to create a culture of transparency and exchange.

There are, however, some drawbacks to general humanitarian-coordination forums. First, broad gatherings that cover a range of issues in addition to data and information exchange can lose focus, and become unwieldy and time-consuming. Second, broad coordination forums are often attended by managerial rather than technical staff, so technical issues can be overlooked. It is generally advisable to establish a working group structure, as a spin-off of the main coordination structures, specifically tasked with developing protocols and methods for information exchange. ‘Soft’ institutional networks must be comple-mented by formal structures for managing, compiling and disseminating information: a ‘data warehouse’ in the jargon. The most advanced structure of this kind is UNDP Somalia’s Data and Information Management Unit (DIMU). In Kosovo, the Humanitarian Community Information Centre (HCIC) performs a similar service see Humanitarian Exchange 18.

Potential difficulties surround the extent of government ownership of the structures that are established. On the one hand, participation is vital, but on the other there is a risk that basic principles of free exchange and objectivity will be compromised. Joint ownership structures are probably the best solution. A promising example is the Information Coordination Centre (ICC), hosted by the Eritrean Relief and Rehabilitation Commission (ERRC) and supported by the UN. To guard against the risk of manipulation, ‘parallel’ data repositories in a more neutral body, like the UN, are advisable, and the principle of open data exchange needs to be enshrined in project agreements.

Regionally, the main challenge is to achieve strong inter-governmental participation, while avoiding over-dependence on the region’s weak geopolitical structures. In the current climate, the best option seems to be voluntary, ‘virtual’ networks, in which the UN, NGOs, donors and intergovernmental bodies are equal partners and stakeholders. The technical servicing of such a network (including the provision of a data hub and maintenance of an Internet-based discussion and data-exchange forum) is beyond the capacity of IGAD; on government salaries in the region, it is simply impossible to hire appropriately-trained professionals. An inter-agency initiative is underway under the Data Exchange Platform for the Horn of Africa, to perform this function through a regional data and information ‘warehouse’. Proposed management structures include representatives from across the humanitarian community. It is hoped that a small physical facility will be staffed and resourced by a spread of organisations, in order to maintain a philosophy of joint ownership.

In the Horn, services and products need to be basic. State-of-the-art Internet technologies offer no panacea; in all the IGAD countries, local Internet Service Providers (ISPs) are unreliable and connec-tivity among government and NGO partners is only partial. Moreover, the lack of capacity to maintain and upgrade sophisticated applications means that investing in their development is wasteful.

Rather than attempting to develop a set of industry-standard tools, the focus should be on providing the humanitarian community with low-cost technical services. These might include:

  • help-desks and technical facilities to assist national and regional partners in maintaining and upgrading information technologies, and building up staff capacity;
  • a data hub or storage facility for ensuring that mission-critical information is not lost, and core database applications to address routine infor-mation needs; and
  • common basic data standards, particularly for labelling geographic data.

In terms of content, there is a clear bias towards information on food security, and much work in the humanitarian information sector has focused on early-warning indicators and methods, as well as supply-chain management systems. Broadening the information base – and the instruments of humanitarian response – has two elements.

First, other specialised systems (for example for epidemiological and conflict monitoring) must emerge. Second, there is a need for more general information products that form the basis of humanitarian interventions in any sector or phase in the emergency cycle. Developing a core of basic information products would be straightforward, relatively cheap to do and useful throughout the humanitarian community. Yet the products described below – the targets of the proposed ‘data warehouse’ – are available for only very limited parts of the region. The first product is ‘base maps’ for humanitarian planning. These can be compiled from existing maps and remote-sensing imagery, ideally on a scale of 1:100,000 or higher. DIMU has developed a series for Somalia, but elsewhere these fundamental products are lacking. Map themes may include geo-physical characteristics, transport and communications infrastructure, distribution of basic services, settlements and population densities, and administrative boundaries. Second, agencies need a comprehensive and easily-accessible database of the humanitarian community, and complete databases covering the ‘who, what and where’ of current humanitarian activity in the region. Third, data on planned projects and programmes, which currently covers only the larger international NGOs and the UN, needs to be expanded to encompass smaller, local initiatives. Fourth, resource tracking systems to compare funding requirements with inflows are required for follow-up and advocacy purposes. Again, this data is available for the main humanitarian agencies, but not for small projects, and again, information about assistance other than food aid tends to be patchy.

Given the scale and frequency of forced and voluntary migration in the Horn, timely information on the numbers and location of migrant commun-ities is essential. Agencies have real problems knowing where and when to deliver relief assistance, social services, and projects. Migrant tracking is perhaps more developed for conventional refugees and returnees, though these people represent only a fraction of the vulnerable migrant populations. A more holistic reporting and data management network would be beneficial. Normative materials, including technical manuals on how to implement specialised types of humanitarian intervention and conduct assess-ments, and evaluations/lessons learnt of previous interventions are available internationally, but inaccessible to people without Internet connections. Repositories of these documents in walk-in documentation centres or on CD would be useful for a number of small organisations. A compendium listing available humanitarian data and information resources, along with instructions on where and how to get it (‘metadata’ in the jargon) is an essential complement to a data ‘hub’.


There is a strong case for exchanging humanitarian data and information between involved organisations and countries. To make the most of the weak information base, we must address the daunting institutional and technical challenges in the Horn of Africa. On the institutional side, cross-community discussion networks and forums are beginning to emerge. Now sub-group forums are needed to deal with the specific technical issues of data and information exchange. Such forums cannot function without the support of joint structures or ‘warehouses’ for managing data. Nationally and regionally, warehouses should focus on enhancing the capacity of partners to manage data and information, and should promote the exchange of this information. Facilities must avoid the siren calls of state-of-the-art applications, and concentrate instead on providing a few simple services and products that are feasible, relatively cheap and useful for everybody.

Ben Watkins is a consultant for UNOCHA in the Horn of Africa. The views expressed in this article are the author’s, and do not necessarily reflect the views of UNOCHA.