A team of 4 people carried out the review: the global CPWG Coordinator (Katy Barnett from UNICEF), a representative of INGOs (Misty Buswell of Save the Children), a child protection in emergencies and separated children specialist (Pernille Ironside from UNICEF HQ), and an independent consultant as team leader (Joanna Wedge). The team conducted 32 in-depth phone interviews with key informants in French or English and the team leader met with over a dozen CPiE actors for individual interviews or focus group discussions during a brief mission to Haiti in May. Many thanks are due to all key informants, who generously gave their time and reflected with honesty and often impressive clarity on the Haiti response.
A call was made for all written documents such as emails, meeting minutes or handover notes, which might provide relevant insights, and the small number of such documents received proved to be useful sources of information and ideas.
The specificities of the Haiti earthquake, which pose a challenge to the review team coming new to this context and also limit the transferability of some of the lessons learned and promising practice, must be stressed. The scale of the earthquake, hitting at the heart of the country’s financial and administrative centre and the fact that it debilitated both government and humanitarian agencies in country are each uncommon challenges in their own right. Additionally, as is the case in all contexts, there were unique cultural, linguistic and socio-economic aspects in play, as well as the nature and level of national capacity. Of particular interest in this context were the pre-existing issues of the widely accepted practice of forcibly placing children with other families for indentured domestic work (the restaveks), the linked phenomena of widespread use of orphanages by families unable to care for their own children, and the vast number and on-going close links of both the diaspora and foreign well-wishers.
The review was intended to have been conducted almost entirely in country; however, due to constraints within the UNICEF country office in Haiti, the methodology was reworked to become chiefly based on telephone interviews. Only the team leader visited Haiti, for a short period and at a different time than originally planned. Additionally, logistical challenges in country and the limited ability of the team leader to move around the city resulted in less focus group discussions and interviews being carried out than originally planned. These factors made for a restricted amount of exchange between the review team and the members of the child protection sub cluster in country, via the Haiti-based CP coordinator. This and the limited number of interviews possible with national actors may mean that some areas of enquiry were less well explored than they could have been – for example, the coordination at sub national level, and the views of national child protection actors who did not engage fully with the child protection sub cluster.
The learning role of the global Child Protection Working Group
The global level CPWG is situated within the Global Protection Cluster and unites some twenty key actors in child protection in emergencies at the global level. The responsibilities of the group are consistent with those of a global level cluster, and include building the capacity of the CPIE sector, setting and promoting standards in CPIE programming, the development of CPIE related tools, global level advocacy, and supporting the establishment and efficient running of national CP sub-clusters.
In 2008 and 2009, the global CPWG carried out a series of reviews of coordination of CPIE at the field level. These included two multi-country reviews based on telephone interviewsvi and two in depth examinations in Myanmar and Uganda respectively, which were carried out in each case at the field level by a small interagency team from the global CPWG. This review should be seen as the latest in this series of learning exercises undertaken by the global CPWG. It differs from previous reviews in that, in response to requests from its members, it covers a key programme area (separated children) as well as overall coordination of the response.
The cumulative learning from these reviews of CPIE coordination carried out by the global CPWG is complemented by learning from exit interviews which have been undertaken with outgoing CP coordinators wherever possible since the start of 2010. Key findings from all these sources include the need for dedicated child protection coordination capacity, and the importance of direct representation for child protection coordination groups in inter-cluster and humanitarian country team meetings (rather than via the protection cluster), along with many other lessons and indications of good practice. These findings are used to influence policy at the global level, to shape the training and technical assistance provided to CPIE coordinators, and to inform the work plan of the global CPWG.
Lessons learned from the review of the Coordination of CPiE and the RESPONSE TO Unaccompanied and Separated Children in Haiti
Capacity for coordination
In contrast with prior emergencies, there was immediately a recognized need for dedicated capacity to be allocated to coordination of child protection in Haiti. Within the first 10 days of the emergency, UNICEF deployed a full-time child protection coordinator who took over from an in-country staff member. That staff member went on enforced leave which contributed to the overall break in institutional memory and lack of local technical and coordination knowledge. One month after the earthquake, a full time coordinator and assistant from the Save the Children standby team were seconded to UNICEF for a 3 month period. Although there was significant staff turnover in the first two months, the coordinator’s position remained filled throughout the response. Lessons learnt from Haiti show that perceived and actual seniority of a Coordinator can be critical to ensuring good working relations and the requisite authority with the government and other sectors for carrying out coordination activities; this is particularly true for Child Protection which is often dismissed as not a life saving intervention, “soft” or “just” a sub-cluster.
The dedicated capacity was essential in achieving any kind of coordination in a context where there were a large number of actors, many of which were new to child protection and/or emergencies and with their own staff overstretched by their own programme demands. However, in a disaster of this scale, it became apparent that the coordination capacity was not adequate for all the tasks that were required and the small size of the coordination team affected the quality, speed, and scope of the child protection response. The absence of an information management officer in particular was a key factor in the inability of the cluster to effectively gather and analyse assessment and response data – something which several other clusters and cross cutting issues did relatively well thanks to the appointment of full time national information managers.
The approach of the WASH cluster was to start with two dedicated coordinators and build up to a team of 14 by the end of the third month, including 5 information managers. In this model, functions which were normally fulfilled by WASH cluster members, such as tankering, were taken on by the cluster, and cluster members were not expected to take on coordination responsibilities. In Child Protection, staff of cluster members did take on some of the coordination responsibilities at the sub-national level, and for working groups under the national sub-cluster, combining this with their own agency programme responsibilities. Whilst this responsibility-sharing is well viewed by the cluster membership, in some cases the coordination roles and expectations were poorly defined and understood, resulting in tensions between organizations and delays in the adoption of common programme approaches.
Coordination is time-consuming, particularly when the sub-cluster has several working groups (WG), there are additional sub-national clusters and there are significant child protection concerns in other clusters, requiring a child protection cluster member to attend those meetings. Currently, there is no acknowledgement and guidance on staffing implications, especially for INGOs and government.
The global CPWG needs to further develop surge capacity for coordination in child protection, including for information management and ensure that there is a pool of potential coordinators at the correct seniority and with a diversity of language skills.
As a standard, capacity for both information management and assessment design and implementation should be included in Child Protection coordination teams for large scale, sudden onset emergencies. Standard Operating Procedures should be developed that clearly define the number of coordination staff required depending on the scale of the emergency, and other procedures for getting a cluster up and running.
The global CPWG should undertake consultations within its membership - both UN and INGO - to develop clear policies for engaging in the child protection sub clusters in emergencies, expectations and commitments to the cluster and possible staffing implications.
Template ToRs should be developed for working groups and sub national level clusters that clearly define the roles and responsibilities of sub working groups and their coordinators and the relationship with the national level sub cluster and coordinator.
The global CPWG should continue to advocate both within UNICEF and the donor community, on the importance of allocating sufficient human resources to child protection coordination in large scale emergencies.
Co-lead arrangements should be considered at the field level in order to share the burden and responsibility of coordination, and template ToRs should be developed clearly spelling out each co-lead agency’s role and responsibilities.
Child Protection sub cluster membership
The diversity and breadth of membership of Haiti’s CP sub-cluster was striking. However, it raised a number of challenges. Foremost, it highlighted the small number of local organizations participating in the sub-cluster and the challenge of engaging these organizations in a sustained and meaningful manner. Experiences showed that seemingly ‘small’ matters – such as transport to a meeting, location of the meeting and security clearance, timeliness of minutes when people could not attend, language of the meeting and/or simultaneous translation – need to be resolved rapidly so as neither to lose momentum nor alienate potential participants (particularly community-based organizations). Indeed, there were many early meetings when only a tiny number of participants had any prior experience in country. (See below –section 4 – for discussion of government relations).
Although the child protection sub cluster in Haiti was given credit by many for conducting most of its meetings in French and/or providing interpretation, many challenges remained that prevented local organizations from participating fully. The location of meetings at the UN logs base was a particular challenge, both for local and international organizations, as transportation across Port au Prince was difficult and reaching the logs base could take more than an hour. Once there, it was difficult, if not impossible, for local organizations to get through UN security to enter the logs base. The meetings have since been moved to the Ministry of Social Welfare, and it is hoped that this will allow for increased participation of local organizations. Furthermore, future CPiE responses in well-developed settings would benefit from a coordinated strategy to engage with local civil society and thus may need a dedicated full-time or part-time assistant coordinator to ensure this is part of the overall approach from the earliest phase.
Another challenge was the issue of non-traditional organizations getting involved in child protection work for the first time. There were many groups present who had not previously worked in Haiti or in an emergency response; many of them needed an orientation to the cluster system, as well as to child protection principles, FTR systems and other areas of child protection. As a result, sub cluster meetings often focused on information-sharing requiring additional time of the cluster coordinators that would have been better spent on coordination of child protection activities. The establishment of Google groups was helpful as a forum for information sharing. However, if sub-clusters are to be effective, meetings must be a place for strategic planning and decision making with a sense of urgency and not primarily a forum for information-sharing. In addition, child protection is a sensitive area with frequent need for confidentiality, for example when discussing particular cases during sub cluster meetings. This became difficult in Haiti when new agencies, which were not aware of or did not adhere to agreed child protection principles, were present.
In fact in a crisis setting with dozens of actors, an explicit two track approach to coordination may be warranted: a smaller strategic decision making and planning group to include the traditional actors who are experienced in CPIE principles, programming approaches and the cluster system, and a wider group that meets less frequently with new entrants who need extensive educating, coaching, support and sometimes convincing. It must also be acknowledged that this approach is labour-intensive and would require additional coordination capacity.
Further suggestions for addressing this challenge include the development of an information sheet that could be distributed to new organizations containing basic information about humanitarian coordination, the cluster system and child protection; organizing basic CPIE training - 2 half days per week in country; and pre-vetting organisations against minimum standards for operation in emergencies – or some other means of promoting compliance with established interagency programme principles and approaches, such as ‘registration,’ -possibility in conjunction with the government.
Finally, there is a need ensure that coordinators are fully aware of and able to draw on the mandates and capacities of key actors in country, which might not exist at the global level (i.e. national government and DPKO mechanisms and strengths). Coordinators need to reach out early, at different levels and possibly multiple times - to all useful and relevant Ministries and government departments, especially when the governing structures have been severely affected by the crisis.
United Nations Peacekeeping missions and the various branches of the Red Cross movement are by definition emergency-response; thus, their staff and networks may be more resilient to crisis than development-oriented CP structures. It is important for child protection actors to be well-briefed on any relevant UN Security Council mandate and to build on what the mission has put in place, , as well as continuously engaging with the ICRC and national Red Cross to maximise the synergies that are present.
Cluster coordinators should ensure that there is a coordinated strategy for engaging with local civil society from the earliest phase of the response, including holding meetings in local languages, ensuring meeting locations are accessible and where possible building on existing child protection work previously done by local civil society.
The CPWG should develop and translate short messaging on CP sub-cluster TORs and ways of working that can be easily disseminated to local organizations and non-traditional child protection responders.
The CPWG should include in guidance for coordinators (including the revision of the CP cluster coordinator handbook) concrete tips on managing membership with options listed and examples of models that have been useful in emergencies of different scales and complexity.
As a routine measure for large-scale, sudden onset emergencies, the sector should deploy sufficient interagency capacity, either through coordination team or through sub cluster members, for ‘start-up’ training / briefing for new cluster members, which may also be beneficial for new staff of global CPWG members.
UNICEF should ensure that training, guidance, and pre-departure briefing of coordinators cover roles of key possible sub cluster members such as UN Peacekeeping missions and ICRC.
Haiti provided a clear example of how - particularly in a natural disaster -the CP coordination group is often faster moving, more operational, larger, created earlier and more organized than the Protection Cluster itself. Strong INGOs in the sector have a set program of child protection interventions and principles, considerable expertise and immediate capacity; in addition, UNICEF - as coordinator, implementer and donor - is able to move things ahead quickly. In contrast, the protection cluster may take longer to establish itself and determine the scope of its work. This dynamic can mean that early on there are demands on the CP coordinator to represent protection more broadly.
In Haiti, Child Protection greatly benefited from having a separate coordination body to protection and from side by side representation at the inter-cluster coordination and humanitarian country team (HCT) meetings and other such fora. However, several respondents felt that there had been limited outcomes in influencing the HCT. Occasionally, clear asks – such as for space for the call centre for the family tracing and reunification programme - were missing from these inter-cluster meetings, while opportunities to influence other clusters’ work or develop interagency messages for the national and international media were weak or slow.
It must be noted that protection, child protection and GBV run the risk of functioning very independently and need to ensure that they are integrating their work and jointly analyzing and addressing any protection gaps.
In a natural disaster, the child protection sub-cluster is often larger (in terms of both human and financial resources) than the overall protection cluster. Its ability to couple advocacy on better humanitarian practice with actual personnel to assist with improved operations is a key strength for the child protection sub-cluster. An example of promising practice in this regard was the sub-cluster’s work on the prevention of separation and other protection violations during the relocation of displaced families between camps in April.
3.1 UNICEF ensures through the CP coordination training and individual briefings that CP Coordinators have a good understanding of broader protection issues and opportunities.
3.2 While child protection Cluster Coordinators ideally should not represent protection on a sustained basis, and while child protection coordinators should have their own seat at inter-cluster forums in order to ensure that child protection is adequately represented, collaboration between Areas of Responsibility, including at global level in preparedness measures, is to be encouraged through the PCWG.
3.3 The child protection sub-cluster coordinator must consistently ensure that child protection issues are raised at the highest levels, including with the Humanitarian Coordinator (HC) and by participating in the Humanitarian Country Team (HCT) meetings.
3.4 In pre-deployment orientation, UNHCR should brief the Protection Cluster Coordinator on the role of the sub-clusters and the importance of a dual approach to representation of the CP sub-cluster within the humanitarian framework. ToRs and other information for Protection Cluster Coordinators should also include this information.
The role of government within coordination.
The earthquake in Haiti struck at the heart of the government, decimating both its physical structures and human resources. Understandably, the government – especially at central level – was dramatically affected and struggled to become operational again and to scale up to meet increased needs. There were varying reports from respondents regarding the interaction between the government and the Child Protection sub cluster. Some respondents felt that the government, particularly IBESR, was involved as much as it could be given the limited capacity after the earthquake. Other respondents felt that the cluster could have done a better job of reaching out earlier to ensure government ownership of the child protection response. One NGO felt that the sub-cluster should have continued working with the government department with which they had an established relationship prior to the disaster, but later realized they should have expanded their focus to additional departments. For its part, government representatives interviewed said that they would have welcomed better coordination within the international community – humanitarian, military and other actors – as it was confusing to have so many counterparts.
Ministries of Social/Family/Children Affairs are often significantly underfunded. It is important during an emergency, particularly a natural disaster, that adequate attention and deference be made to ensure that they can function as effective counterparts. This may mean providing physical space, office and transport equipment, counseling or emergency funds for stabilisation/continuity. In some cases, seconding the cluster coordinator into the relevant ministry has been found to be an effective way to ensure government ownership of coordination.
Even in dire circumstances, the CP sub-cluster must creatively and sensitively find ways to engage the government in the first two weeks of the emergency. If not, problems with support and ownership may occur. It is crucial to use pre-existing analysis of government interests and capacities to determine where strengths and weaknesses will likely lie and who to work with across a range of Ministries and departments. Where government cannot undertake the co-lead of the sub-cluster, it can be facilitated to chair every other meeting, be linked into meetings via telephone or video calls, be bilaterally briefed prior to/post-sub-cluster meeting (which worked effectively for MHPSS), be asked to participate in Strategic Advisory Meetings, etc.
The idea of an MOU between the WG on UASC and the relevant government department was discussed; its aim was to clarify roles related to FTR, interim care, reunification, and other child protection issues. Unfortunately, the government did not proceed with this approach, despite its potential to clarify support to existing orphanages and streamline the assistance that traditional and non-traditional responders were providing to the sector. When this approach did not work, the working group was slow to identify alternative strategies and approaches to working with the government that may have been more effective, such as building on bilateral relationships that existed prior to the earthquake.
From a global perspective, it can be seen that the CPiE response is increasingly systems-focused, rapidly mapping and assessing what structures existed prior to the crisis and what can be built on into the medium and long term response. Thus, the sector has good understanding of the concept of “Early Recovery” and can lay the groundwork for longer-term, broader-based social care of vulnerable children. How Child Protection engages with all of the relevant government departments is critical in this endeavor and was hampered in the initial months of the response in Haiti. Opportunities for strengthening the national child protection system were missed, owing to an inadequate analysis of pre-existing structures, strengths and weaknesses. It may have been possible to undertake much of the analysis out of country given the turmoil of the first few weeks.
In the first month, the CP Coordinator and senior management from the sub-cluster members should have bilateral meetings with key government officials to update them on sub-cluster work and clarify roles and responsibilities.
Coordinators should be encouraged to assess and provide the relevant government departments with logistical support and training to engage with sectoral coordination from the declaration of an emergency.
National Emergency Preparedness Plans must include government buy-in on Coordination mechanisms.
UNICEF HQ should draft two standard initial MOU for presentation to and negotiation with the national government in the first weeks of an emergency. The first should cover roles and responsibilities for coordination and the issue of registration procedures for all agencies working with children in the emergency. The second should focus on work on sexual violence, appropriate psycho-social approaches, basic principles on children facing any kind of separation relating to the emergency (abandonment, secondary separation, trafficking, associated with armed groups and armed forces, etc), and agreed top-line standards.
The support role of the global level child protection working group
The role of the global CPWG in complex, large emergencies has many facets. Its support to the Haiti response was largely well-received by colleagues on the ground, including in their pre-deployment phase.
The global CPWG undertook a number of roles. A troubleshooting approach – trying to resolve issues in a collegial manner if an organization is not performing to the level expected or if there are problems with the overall child protection response – was activated when the global coordinator and some key members became aware that the scale of the Child Protection response was not sufficient or effective, but unfortunately no improvement in staffing or programming seemed to stem from the bilateral or global CPWG teleconference interventions.
The global CPWG also had a role in advocacy, such as the development of common inter-agency messages that built on previous examples of best practice and relieved some of the burden from field level practitioners. In addition, its efforts to identify and adapt global standards for the Haiti context, which was done with the Interagency Guiding Principles on UASC, were deemed to be of great help in the field. In fact, technical advisors in government distributed and used those guidelines as talking points with both political and donor colleagues.
Other suggested roles have included linking field practitioners with the expertise that exists at the global level, as well as helping to develop messages that can be used with the media.
5.1 The global CPWG Coordinator should continue to convene teleconferences between global and field level members of CP sub clusters in onset emergencies and offer to provide support where possible.
5.2 A CP sub cluster “start up” package, containing contact information and the relevant tools and guidelines should be developed by the CPWG and distributed to field level coordinators and cluster members at the onset of a new emergency.