A more proactive UN role in the security of NGO staff?

The recent and very tragic deaths of UNHCR workers in West Timor and Guinea have once again focused attention on the precarious security circumstance under which humanitarian relief work is so often conducted.

At the UNHCR ExCom meetings in Geneva in 2000, opening remarks from High Commissioner Ogata, Secretary-General Annan and WFP Executive Director Bertini each highlighted concern over the security of humanitarian aid workers. Just a week earlier, directors and senior managers from 25 major American and European agencies met together for a two-day, OFDA/InterAction-sponsored workshop on staff security. Indeed, security of aid workers has been a growing concern and priority over the past few years and is only punctuated when situations such as West Timor and Guinea bring to the forefront the sobering realities of the security environments in which the UN and NGOs work.

In recent years there have been notable inter-agency efforts to address security concerns. Working under funding from OFDA, in 1996 InterAction established an NGO Security Task Force comprised of representatives of a number of American and European NGOs. The Task Force and subsequent Security Working Group went on to design a comprehensive curriculum for training NGO workers on security. The curriculum has since been picked up by the British NGO RedR which, with funding from OFDA and Britain’s Department for International Development, is offering the week-long course from regional centres around the world over the next two years.

The Humanitarian Practice Network at the Overseas Development Institute has recently published Good Practice Review #8: Operational Security Management in Violent Environments,(1) authored by Koenraad Van Brabant who has also been an active participant on the InterAction Task Force and Working Group. The book is an important reference for NGO security planners, bringing together input from scores of NGOs and establishing both a common language and conceptual framework for understanding security and strategies for protecting aid workers. Not only have NGO contributions to the book been impressive but major governmental donors have also come forward to support the effort: publication was supported by the governments of Britain, Denmark, Ireland, the Netherlands, Sweden and the US.

VOICE’s Humanitarian Safety and Protection Network (HSPN) has also made an impressive effort to establish a mechanism for tracking and analyzing security incidents. Prior to the HSPN project - with the notable exception of ICRC - our understanding of security incidents, qualitatively and quantitatively, has been largely rooted in anecdotal reporting.

The UN is also taking significant steps towards improving security for its field operations. The Secretary-General’s recent report on ‘Safety and Security of United National Personnel’(2) overviews the scope of the problem faced by the UN in the field and the shortcomings of a strategy developed for the situation as it appeared twenty years ago. The report sets forth a number of proposals to improve security. These would include: appointment of a Security Coordinator at the Assistant Secretary-General level, establishment of a more reliable mechanism for funding Field Security Officers and increased resources for the office of the UN Security Coordinator (UNSECOORD) which would provide for enhanced staff training, security assessments, counseling and stress management.

Coordination with the NGO community

The Secretary-General’s report outlines improvements that are a step – if not a leap – in the right direction toward enhancing the security of UN personnel. However, beyond recognizing that NGOs face the same challenging operating environment, the report makes no mention of the need to enhance coordination and joint security efforts with the NGO community upon which the UN is increasingly dependent. This is a significant oversight, not only for NGOs but for the security of UN field operations.

UNHCR, as a case in point, relies heavily on NGO implementing partners to achieve its mandate(3) yet has done little to clarify its role with NGOs in the realm of security. In the NGO-UNHCR meetings prior to this year’s Executive Committee meetings in Geneva, UNHCR’s Chief of Safety, Roland L’Allier, indicated that there is no formal relationship but rather “ad hoc” arrangements that vary from situation to situation. Mr L’Allier also indicated that the ad hoc arrangement is “flexible” and “works”. Indeed there is flexibility. Talking to NGO field staff around the world about what role UNHCR plays in security issues, one gets replies evocative of the parable of the blind men and the elephant where each man describes the overall appearance of the animal based upon the one appendage that he is able to touch: security management in each situation is so dramatically different that it is difficult to grasp the overall vision. It is less clear whether or not this “works”.

When UNHCR chooses to undertake a coordinating role in security, the results can be most impressive. Coordinating NGOs – often likened to herding cats – is notoriously difficult. Yet, when UNHCR calls a security coordination meeting for the NGO community, the NGOs come. When UNHCR establishes a common communications network or frequency, NGOs participate actively. When UNHCR offers technical advice or training to NGOs on anything from site security to convoy operations, NGOs are generally keen to tap that expertise. These are roles that UNHCR is uniquely positioned to offer. Individual NGOs rarely have the resources to hire their own security officers – and even if they had, NGOs can benefit enormously from a common UN/NGO forum to discuss and coordinate security matters. A common understanding of the security environment emerges, along with a common language and coordinated responses. Information is shared which benefits not only the NGOs but UNHCR as well. UNHCR is uniquely situated not only for creating this forum but also for playing a larger representational role on security matters. One of the pillars of good security derives from the diplomatic relations between the humanitarian community and the local or regional powers. Again, in most situations, UNHCR is far better positioned than most NGOs to approach national and regional authorities – or major donors – at the highest levels to advocate for humanitarian access and the security of aid workers.

Compromise and competition

If UNHCR is so ideally situated to play a central role in security for NGOs, why can this role not be made more formal? One problem is that UNHCR’s effort is generally centreed on the appointment of a Field Safety Advisor (FSA). The FSA is a security professional with the expertise and mandate to advise on matters of security. The appointment of this position is not automatic, even in the most precarious of security environments: only 60 out of 80 high-risk posts have assigned security officers. For UNHCR, the decision to appoint an FSA rests solely with UNHCR’s Resident Representative, for whom the decision is largely one of resources and priorities. The UNHCR country office must be able and willing to fund the position out of the country budget using funds that might otherwise be used to support programme activities. In an atmosphere of diminishing funding, the decision can force difficult compromises. Yet, in the absence of an FSA, UNHCR’s ability to play a coordinating role for its NGO implementing partners is all but eliminated.

Even when an FSA is appointed, there is no guarantee that UNHCR will actively take on a coordinating role with NGOs. In the absence of formal guidelines or policies spelling out the role of UNHCR vis-à-vis the NGOs, the relationship between the FSA and the NGOs is largely a product of personalities and competing demands on the FSA’s time. As such, the relationship is indeed ad hoc, varying significantly from country to country and FSA to FSA. If the Secretary-General succeeds in his proposal to fund these positions out of general funds, then the job itself should be redesigned to include liaison and coordination with NGOs.

Even if UNHCR is not able to broaden the advisory services of its FSAs, it should be more forthright about what it can and cannot do for its implementing partners. Typically, UNHCR will offer vague verbal reassurances to NGO representatives. This tends to mask the fact – surprising to many NGO workers – that the UN has no formal responsibility whatsoever for NGO security. Such miscommunication of capacity and intent has left NGOs behind in more than one evacuation, and would no doubt alter NGO security planning if there weremore clarity from the beginning.

UN/NGO Memorandum of Understanding

The UN has attempted a more formal security agreement with NGOs. In 1996, UNSECOORD drafted a Memorandum of Understanding (MoU) in an effort to establish a framework for a security relationship between UN organizations and their NGO implementing partners (‘implementing partners’ are those NGOs with a contractual relationship with a UN agency for implementing a specific project under UN funding). Under the terms of the MoU, the general responsibilities of the UN include unspecified “protection of international staff”, inclusion of “relevant information” about international staff in the UN’s security plan, keeping the NGO informed about security developments and measures being implemented by the UN and, “to the extent possible”, provision of travel assistance on a reimbursable basis in the case of emergency. “Where possible”, the UN also agrees to represent the NGO’s security concerns to the authorities of the host government.

In exchange, the MoU requires that signatory implementing partners “fully follow the instructions of [the UN] regarding security matters”, yet the NGOs are to “assume all risks and liabilities related to the security of its staff” and “deal with all claims as may be brought against the UN arising from the extension under the Memorandum . . . to its international staff”. This surrender of authority has been a central issue for the NGOs, particularly in view of the murky promises of protection in return. One wonders what organization would agree to surrender authority over security-related decisions yet retain responsibility for the consequences of those decisions. The MoU goes on to require that the NGO “ensure that the [UN] is at all times informed of the whereabouts and movements . . . of international staff . . .”. The MoU also requires that the NGOs “lend, when possible and to the extent feasible, on a reimbursable basis, travel assistance to [UN personnel]”.

In short, the MoU provides for the exchange of security-related information and – “to the extent possible” – the evacuation of international staff in exchange for the NGO’s willingness to surrender its authority on security matters to the UN. The NGO also pays a fee to the UN, based upon the cost of the Security Provisions and the number of subscribers bearing those costs.


Above and beyond the overarching question of whether the vague provisions of the agreement are worth the costs, the MoU raises a number of questions requiring clarification. What exactly is an ‘implementing partner’: does the MoU pertain to NGO staff funded by other donors but working on UN-funded projects? Does the MoU extend to an implementing partner’s staff who are working on complementary programmes not funded by the UN? Is the MoU in effect when NGOs are implementing projects in good faith during the often protracted periods of time when the UN is processing proposals and agreements and, in a strictly formal sense, there is no agreement between the NGO and the UN? Do the provisions of the MoU relating to evacuation pertain to national staff brought in for the purpose of implementing a UNHCR-funded project? What if an NGO does disobey the security instructions of UNHCR – is the entire MoU revoked or are the recalcitrant NGO staff simply omitted from the related portion of the security plan, such as an evacuation?

Despite these and many other questions raised by the NGOs, UNSECOORD has been resolutely unwilling to alter the MoU, even to add clarity and even though not a single NGO has signed on to the MoU as a global agreement. (IOM, which is not generally considered an NGO and has a very different set of security concerns vis-à-vis the UN, is the only organization that has signed globally.)

Why is it unreasonable for the UN to expect NGOs to “fully follow the instruction of the [UN] regarding security matters”? Part of the answer surely relates to the culture of independence under which most NGOs operate, for better or for worse. However, there are other concerns as well.

First, and perhaps most important, an NGO’s response to a security environment must primarily be related to the NGO’s global mandate and local mission. An NGO implementing agricultural extension services should have a lower risk tolerance than an NGO undertaking life-saving medical services. It is unrealistic to expect both organizations to respond to security situations in the same fashion.

Secondly, the UN’s own response to security environments can be compromised by fiscal concerns that are unrelated to NGOs. For example, as noted above, if the Resident Representative is unwilling to prioritize funding from the country budget, there will be no Field Safety Advisor. In Uganda there is no UNHCR FSA and thus weak provisions for UNHCR staff working in the precarious environment in the North, where the Lord’s Resistance Army continues to wreak havoc. UNHCR staff are reluctant to visit field sites and, citing security concerns, have spent little more than a few days in the Achol Pii refugee camp so far this year. Nevertheless, UNHCR expects NGOs to carry out services in the camp on a daily basis. If NGOs were to follow UNHCR’s lead on security, there would simply be no services for refugees.

Funding is also involved in the determination of the UN’s security phases but not always as one might anticipate. According to the candid explanation of a senior UNHCR official in Hargesa, UNHCR in Somaliland remained at Phase Three long after conditions had improved, simply because of concerns over the impact of the elimination of a phase-related security allowance on an already demoralized staff. Given these issues, among others, it would clearly be unwise for NGOs to hand over security decisions to the UN even in the best of circumstances.

As a global document outlining the relationship between the UN and the NGOs in the realm of security, the MoU is significantly flawed. It demands that NGOs surrender authority over their own security affairs in exchange for unspecified protection and support for evacuation. At the same time, the MoU does not address the many critical security coordination issues that are so important to NGOs. It should be no surprise that no NGO has signed the document on a global basis. Interestingly, there are some instances – 14 in all - where NGOs have signed at a country level. The countries include Liberia, Tajikistan and Sudan. One might surmise that the usefulness of the MoU is greatly increased in such places where NGOs are unlikely to have the capacity to handle the logistics of evacuation or the diplomatic connections of securing – at the highest levels - humanitarian access and the protection of aid workers.

Despite our criticism of the MoU document, IRC is among the signatories to the MoU in Sudan where we are working together with the UN in the government-held garrison towns of Southern Sudan. Evacuation options are limited and communications are highly restricted. The Government of Sudan unofficially (if not officially) views the humanitarian effort as aiding and abetting its enemies. It is under these challenging circumstances - where evacuation options are dramatically limited and expensive and where high-level diplomatic relations with governments and warring parties may enhance humanitarian access - that a formal MoU may be most appropriate.


While this article may appear critical of the UN, its recommendations are built upon situations in the field where the UN has in fact stepped forward and taken a leading role in security:

Firstly, security coordination improves dramatically when a UN Field Security Officer is appointed. The decision to appoint a Field Security Officer should be made solely on consideration of the security environment, not upon the fiscal concerns and conflicting priorities of the Resident Representative. Accordingly, these positions – costing some $100,000 each - should be funded from a separate centrally managed fund. This, in essence, is among the UN Secretary-General’s proposals.

Secondly, NGO security coordination should be a formal responsibility of the UN Field Security Advisors. Included would be organization and facilitation of routine security coordination meetings, establishment of a shared security communications network, provision of threat assessments and exchange of pertinent security information. Participation in these activities would not (or could not) be required, nor would they imply a liability to UN. Nevertheless, experience shows that they would be well supported and valuable to NGO and UN staff alike.

Thirdly, the UN should make a concerted effort to embrace the language and conceptual framework that the NGOs have developed in recent years, as succinctly represented in ODI’s latest Good Practice Review and the RedR-InterAction security training course. Various UN organizations and the NGOs are increasingly accepting common ‘best practices’ under many sectors of programme operations, up to and including the Sphere Project. This is an excellent opportunity for the UN to recognize and embrace the substantial achievements of its NGO partners in the field of security.

Finally, the UNSECOORD MoU should not be abandoned but re-worked to add clarity and focused functions within specific contexts. It should be invoked when evacuation options are limited and where high-level representation and coordination on security are pivotal to ensuring humanitarian access. Portions of its text should be context specific, specifying the sites it covers and the roles that can be expected from the parties working in those sites. In these limited circumstances, such a tight security regimen should be offered to UN implementing partners and might even be required of them. In short, for the MoU to be useful, it is imperative that UNSECOORD show some heretofore-undemonstrated flexibility in re-crafting the MoU on a more context-specific basis.


The focus of this paper has been on how the UN might play a more active and useful role in enhancing security for their NGO implementing partners in the field. This should not be construed to imply that the NGOs cannot do more to address their own security concerns. The hope that the UN can play a more active security coordination role largely reflects the failure of the NGO community to fill this role for ourselves. Moreover, few NGOs have designated security officers at HQ or in the field; few have adequate security policy structures; few are adequately addressing security orientation and training; and few are adequately addressing the resource needs for enhancing field security. Clearly, we all have a long way to go.

The purpose of this article is to urge the UN to clarify, if not formalize, its unique and central role in enhancing security for all humanitarian aid workers. At the very least, it mayinspire new dialogue between concerned NGOs and UN agencies on how we might better work together to create a more secure working environment for our humanitarian mission.


Randolph Martin is Senior Director of Operations at the International Rescue Committee. Email: randy@intrescom.org


  1. Koenraad Van Brabant, Operational Security Management in Violent Environments. Good Practice Review 8. 2000. Overseas Development Institute www.odihpn.org/
  2. Kofi Annan, ‘Safety and Security of UN Personnel: Report of the Secretary-General’, UN General Assembly, Fifty-Fifth Session, 2000.
  3. In 1999, UNHCR budgeted nearly US$300 million through its NGO implementing partners.


Opinions in FMR do not necessarily reflect the views of the Editors, the Refugee Studies Centre or the University of Oxford.
FMR is an Open Access publication. Users are free to read, download, copy, distribute, print or link to the full texts of articles published in FMR and on the FMR website, as long as the use is for non-commercial purposes and the author and FMR are attributed. Unless otherwise indicated, all articles published in FMR in print and online, and FMR itself, are licensed under a Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. Details at www.fmreview.org/copyright.