Emergency preparedness in South Africa: lessons from the Zimbabwean elections

The presidential elections in Zimbabwe in May 2002 took place within a context of political violence, economic crisis and increasing food shortages. As conditions deteriorated and the possibility of a mass population exodus from Zimbabwe increased, neighbouring countries started preparations to deal with the possibility of a mass influx of refugees across their borders.

For South Africa, this was the first time it had faced such a challenge. (1) After a three month contingency planning process, however, by the eve of the elections the existing preparations only catered for 1000 people for three days, and there was no effective mechanism to provide the identified refugee reception camp with potable water, electricity, toilets, tents or food on short notice. If a mass influx of refugees had indeed occurred, South Africa's response would have failed to meet the basic needs of refugees.

Emergency preparedness for the Zimbabwean elections

Preparedness for emergency humanitarian assistance should be seen as a fundamental part of fulfilling a country's duty to provide refugee protection and levels of assistance that comply with international standards. The legal framework for refugee influx preparedness in South Africa is given by the Refugees Act of 1998, which includes a section on the reception and accommodation of asylum seekers in the event of mass influx (Article 35). The institutional emergency response structure centres the National Disaster Management Centre which works with the police, defence forces and the National Intelligence Agency as well as such other government departments as Health, Public Works and Education as needed.

In spite of the fact that political violence was widely predicted for the election period and food shortages were already developing, the South African emergency response institutions did not initiate a contingency planning process until encouraged to do so by UNHCR in December 2001. The Priority Committee on the Possible Mass Influx of Refugees was established by a Cabinet decision in early February 2002 to coordinate preparations at the national level. International and domestic NGOs were invited to participate from mid-February, ie only one month before the elections. Provincial and municipal contingency planning meetings only commenced in late February.

The final, very limited and incomplete contingency plan was presented to the provincial and the municipal planning meetings at the extremely late stage of 7-8 March: one day before the election weekend. What led to such late and incomplete preparedness?

Invisible influx?

First, there was lack of agreement among emergency response actors on what characterised the crisis at hand. Because of the official South African position that there was no crisis in Zimbabwe, government departments were reluctant to admit the necessity for contingency planning. In addition, the expectation of a 'mass influx', even though this was never officially defined in terms of numbers or timeframe, diverted attention from an alternative possibility - an 'invisible influx'.

Official government figures suggest that there was no significant increase in border crossing from Zimbabwe into South Africa before or around the election. These figures have been used to suggest that no South African preparedness or response was, and is, needed. However, the statistics are contradicted by interviews from the border area with Zimbabwe where there is a consensus that a significant number of Zimbabweans have indeed crossed into South Africa illegally. They are probably not crossing en masse, and not across the main border stations, but rather in small groups all along the rest of the border thus remaining invisible to the South African authorities. An invisible influx is difficult to confirm empirically but given the consistency of reports from the border and strong concerns about the validity of government migration statistics because of their composition and methodology(2), the scenario is quite probable. If an invisible influx is in fact taking place, the presence of additional illegal migrants could be placing considerable pressure on the South African social system without the support of organised humanitarian assistance or international help.

In assessing the need for further emergency preparedness activities, it is crucial to take into account all probable characteristics of the potential humanitarian crisis, including scenarios such as an invisible influx, so that appropriate interventions can be designed.

Clear decision-making structures

The second main constraining factor in the contingency planning process was a lack of clear leadership and decision-making structures. As determined by international law, the 'receiving' government has the duty to take the lead in providing protection and assistance in cases of refugee flows. In the context of the African continent, South Africa has very good pre-conditions to be able to fulfil this duty. South Africa itself is at peace and it has a formal national disaster management structure that has been tried and tested through domestic and regional interventions (although not mass refugee situations). However, government leadership was hampered in this case by political considerations, lack of clarity on a formal lead department, 'passing the buck' among different levels of government, and the pre-eminence of the military in South Africa's emergency response processes.

Government action in disaster preparedness is always highly political and was so in this case as well. Lack of political will to acknowledge the crisis brewing in Zimbabwe prompted top government officials to describe preparations as interference in the internal affairs of a neighbouring sovereign state that would create tensions bilaterally and within the Southern Africa Development Community (SADC). Although the OAU Convention specifies that granting asylum (and by extension emergency preparedness) is a "peaceful and humanitarian act and shall not be regarded as an unfriendly act by any Member State", political considerations vied with logistical imperatives and led to a distinct vacuum of political leadership for effective preparedness.

This was reflected institutionally as disagreement on which government department would be the lead institution in overall charge of the contingency plan. The consequence of this national level indecision was the devolution of decision making to provincial and municipal levels but without matching authority and funds. It was primarily for this reason that the concrete contingency plan was delayed for so long, and was limited to what could be financed from existing municipal funds (1000 people for three days). The South African Refugee Act allows for a great deal of political discretion in key areas of the emergency preparedness process, such as the precise role of the lead institution and relations between national, provincial and local government offices. Consequently, lack of political will and commitment can easily jeopardise timely and effective decision making. South Africa needs to minimise its reliance on political discretion and strengthen autonomous systems and detailed frameworks.

A further leadership issue concerns the prevalence of the armed forces within disaster response in South Africa. The military was the main South African actor involved in the high profile Mozambican flood relief in 2000, and military responses to disasters were the norm during the apartheid regime. Due to the ambivalent role of the security forces (being key players in both refugee protection and the deportation process), their participation in the emergency preparedness process should be circumscribed to their comparative advantages and availability of resources to fulfil specific and limited roles within a civilian-run plan. The widespread perception - both among provincial government officials and defence force personnel - that all Zimbabweans are simply 'taking advantage' of the crisis situation to take South African jobs conflicted in this case with the legitimate right to protection.

A final crucial issue which constrained early decision-making was the lack of provision for emergency preparedness funding. While emergency response financing is provided for by various acts and funds, its release depends on the formal declaration of an emergency. Government departments, especially at the provincial and municipal levels, were very reluctant to spend money on preparedness, without the assurance that this would be reimbursed, thereby slowing down and limiting actual physical preparedness.

Effective coordination

Literature on disaster preparedness and response stresses the need for smooth cooperation among actors. In the South African case, relationships among actors are shaped by the twin legacies of international isolation and domestic political struggle, which have, respectively, led to a lack of awareness and expertise in international humanitarian and refugee law, standards and best practice, and a distrustful relationship between governmental and non-governmental actors.

Because of their general lack of operational expertise in humanitarian response, the role of indigenous NGOs within the emergency preparedness process was largely limited to observation and monitoring. This role was carried out both at the national and provincial planning stages, as well as during the actual election weekend and following weeks at the border post. While monitoring was very effective in the traditional NGO areas (such as legal support for asylum seekers, information dissemination to government and border officials, and advocacy for government action), it could not be fulfilled consistently in mass influx-specific ways, since there were not enough financial and personnel resources to monitor actual government actions along the border beyond the main Beitbridge border crossing. The realisation was expressed by all NGOs interviewed that they did not have sufficient experience, resources or contingency plans to be prepared for such an emergency. Concrete interventions offered on standby were mainly in the form of translators and volunteers for registration and distribution in the planned camp, all of which were not, in the end, called upon.

Part of the operational limitations was the strong local NGO focus on refugees' legal rights. While this is clearly an important issue, it is only part of the wider context of the crisis situation to which the emergency preparations needed to respond. The implication is not that specialised advocacy NGOs should diversify their perspectives and activities, but rather that all stakeholders should work to bring NGOs with more service- and welfare-oriented experiences and capacities into the emergency preparedness network.

The main role of UNHCR and the international humanitarian organisations involved (IOM, Save the Children UK, Oxfam, ICRC and JRS) was to transfer emergency expertise to South African actors. In addition to seconding experts and providing information materials, many of the international agencies committed some limited resources to ensure that emergency supplies were available. However, all international humanitarian actors made clear that they saw their own contributions purely as additional support in areas that the government was not able to cover.

Unfortunately, these efforts were not integrated into the government-led contingency plan. International agencies were perceived as a kind of fail-safe mechanism by some government actors, with the expectation that they would not let the situation deteriorate but they were only included in planning and information sharing on a need-to-know basis. From the perspective of the international humanitarian actors, their predominant focus tends to be crisis situations where the state is either exceptionally weak, extremely poor or has virtually ceased to function at all. Since none of these characteristics apply to South Africa, most international organisations found it difficult to justify to their own donors the necessity of spending resources for an intervention in this case.


Reports by international agencies have described the appalling extent and severity of the famine looming in the Southern African region. Assessments on the ground suggest that some 12.8 million people in the region are at risk of starvation, and nearly half the at-risk population live in Zimbabwe. Within this context, it is unfortunate and incongruous that the preparedness process in South Africa was broken off a few weeks after the elections, on the basis that 'nothing had happened'.

Academics and NGOs should concentrate efforts on collecting more data on actual illegal cross-border flows to complement official statistics in order to establish whether there is, in fact, significant famine-induced displacement into South Africa. South African defence forces should contribute to this effort by disclosing their official statistics to help monitor border crossing as the effects of the famine deepen. There is a need to consider the invisible influx hypothesis seriously, conduct relevant research to ascertain its probability, and develop responses to it. Within this context, the policy of continuing deportations of 'illegal immigrants' from Zimbabwe, at the same time as conducting emergency preparedness for a largely famine-based emergency, should be reconsidered. Although a mass influx has not occurred, individuals should enjoy protection and assistance based on need and regardless of the scale of the actual influx.

In relation to improved leadership, coordination and capacities for emergency response, there needs to be more regular communication between government and NGOs on disaster preparedness in order to increase mutual understanding. A new Disaster Management Bill was passed after the Zimbabwean elections. This is a valuable opportunity for the development of a detailed and practical National Disaster Management Framework that consolidates a clear division of responsibilities between national, provincial, and municipal levels to avoid 'passing the buck'; it also underlines the need for humanitarian expertise in individuals delegated to be responsible for emergency response in each department. Civil society should be active in developing recommendations to be incorporated within this embryonic framework.

International donors should recognise that, in spite of a relatively developed government infrastructure and economy, there are major capacity-building needs within the South African government and local NGOs in terms of emergency preparedness. South Africa has great potential to be a leader in quality emergency response for the region and on the continent. Donors should support international agencies working to build such a system in the country, as well as the South African government and local NGOs participating in emergency preparedness structures.

UNHCR, through its Emergency Preparedness and Response Section, should take the initiative to offer operation management tools and emergency management training to interested local NGOs, as well as financial assistance for intensive training and skills development, with a strong focus on comparative advantages of specific organisations. In addition to the provision of assistance within Zimbabwe, UN agencies and international NGOs should remain involved in working in South Africa, developing partnership arrangements with local NGOs working close to the Zimbabwean border to ensure standby agreements that can be activated at a moment's notice in response to imminent and actual emergencies.

Furthermore, local service provision NGOs (water, sanitation, child welfare, education, food distribution, shelter construction) which may have no emergency intervention experience but do have technical skills that could be used to this effect should be identified, offered emergency training and brought into communication and collaboration with the emergency preparedness structures.

Finally, on a regional level, information should be shared with other SADC and donor countries on an ongoing basis, and a coordinated strategy for emergency response should be developed. It is essential to incorporate international and regional burden-sharing models and principles into this strategy.

The way forward

Incorporating the lessons learned from the Zimbabwean elections of 2002 can give a new scope for emergency preparedness and conflict prevention in Southern Africa. The South African case highlights the need to strengthen risk assessment through shared and complementary information gathering, revise emergency decision-making procedures, streamline chains of command, determine financial models for funding preparedness, and establish a regular communication system among all stakeholders (governmental and non-governmental) to allow for familiarity, trust and smooth coordination.

As a leading partner in the region, South Africa's comparative advantages should be utilised for designing an effective response to the current crisis in the Southern Africa region. For all the efforts being made to address the crisis within Zimbabwe, including grain shipments and diplomacy, it is clear that a large sector of the population remains vulnerable to increasing hunger and malnutrition. Therefore, the lack of contingency planning for potential famine-induced displacement is a serious shortcoming. These lessons should be used to develop an integrated response based on a joint strategy between the South African government, international actors and civil society.


Hernan del Valle and Tara Polzer work at the Refugee Research Programme University of the Witwatersrand www.wits.ac.za/rrp. Email: witsrrp@mweb.co.za

RRP's Emergency Preparedness Newsletter links organisations working in emergency response in Southern Africa. To subscribe, send an email to witsrrp@mweb.co.za with the words 'subscribe Zimbabwe newsletter' in the subject line.

This article is a summary of a longer report available online at: www.wits.ac.za/rrp/emergency_preparedness_zim.pdf. The authors would like to thank the National Consortium for Refugee Affairs, the Foundation for Human Rights and the Ford Foundation, South Africa for their support. All opinions expressed in this paper are the authors' own.


  1. Although South Africa has been surrounded by countries in conflict for decades, in many of which the apartheid regime was a significant player, there has been only one mass influx of refugees onto South African territory in recent history. However, this influx of Mozambicans in the 1980s was not responded to according to international norms by the apartheid government, which left the task of receiving 300,000 refugees to the then homeland governments of Gazankulu and Leboa. In practice, therefore, the brewing crisis in Zimbabwe is the first time South African institutions are faced with the need to prepare for and potentially succour refugees.
  2. In practice that border is highly porous, and there are reasons to believe that unchanging immigration statistics may reflect limited SANDF capacity rather than static cross-border flows.


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.