Hillside settlements in Port-au-Prince, Haiti Hillside settlements in Port-au-Prince, Haiti Photo credit: Michael Hooper
Humanitarian standards in urban, post-disaster contexts: a study of Sphere shelter standards in Haiti
by Martha Pym and Michael Hooper June 2017

In the face of rapid urbanisation, population growth and increasingly intense and frequent natural disasters, vulnerability in urban areas is growing. At the same time, humanitarian organisations are increasingly acting in urban contexts. This raises questions regarding the effectiveness of humanitarian engagement in these settings. One response to broader questions around the accountability and effectiveness of humanitarian organisations has been the development of a set of minimum standards for humanitarian assistance, created as part of the Sphere Project. This article examines how humanitarian organisations have worked with these standards, particularly those relating to physical aspects of shelter, and asks what can be learnt from their implementation in urban environments.

Interviews with representatives of 14 humanitarian organisations who responded to the Haiti earthquake in 2010 reveal two opposing forces shaping adherence to Sphere standards: pressure from donor organisations, which promotes adherence, and the spatial realities of working in an urban environment, which constrains it. Interviewees reported four approaches that organisations have adopted to resolve this tension. Two – innovative technical solutions and donor education – allowed them to work in urban environments while also meeting donor demands for compliance with the standards. In contrast, the other two approaches – avoiding working in cities and creatively reinterpreting the standards – provided a way of sidestepping what were often seen as the insurmountable challenges of achieving Sphere adherence in urban areas. Based on these findings, the article proposes that any review of Sphere standards should actively incorporate feedback and lessons from organisations that are seldom included in such processes, including smaller organisations and those based in the global South with experience of urban disaster response.

A new urban context for disaster response

The rapid growth of urban settlements in disaster-prone settings, combined with the increasing frequency of natural disasters, has increased the vulnerability of urban populations. This reality has been reshaping humanitarian approaches to disaster management and response. The last ten years have seen an increase in reports and discussions regarding lessons learnt from responses to urban disasters. This is reflected in the findings of the 2010 World Disasters Report, which states that ‘rapid urbanisation and population growth are combining to create enormous new challenges for the humanitarian community and are pushing us out of our comfort zone to deal with a strange new urban world’.+D. McClean, World Disasters Report: Focus on Urban Risk (Geneva: IFRC, 2010).

This article questions the extent to which the Sphere standards, first published in 1998, are applicable within urban contexts. While there are a number of established codes of humanitarian practice, Sphere constitutes the most commonly used and most widely known set of standards. The standards seek to ‘improve the quality of humanitarian response in situations of disaster and conflict, and to enhance the accountability of the humanitarian system to disaster affected people’.+Sphere Project, Sphere 2020: Strategic Plan 2015–2020 (Geneva: Sphere Project, 2015). Close examination of the Haiti case allows us to better understand how Sphere standards have been implemented, and whether the urban context creates barriers to adherence.

The Sphere Project and physical shelter standards

The Sphere Project is a voluntary initiative governed by a board comprising representatives of 18 humanitarian agencies.+Sphere Project, Sphere Handbook: What is New? (Geneva: Sphere Project, 2012). The Sphere Handbook, the project’s key document, comprises the Humanitarian Charter and Minimum Standards in Humanitarian Response. The impetus to introduce humanitarian standards came from two sources. The first was the proliferation of humanitarian organisations during the 1980s and 1990s and, as a consequence, the increasing diversification of approaches to humanitarian response. The second was the Rwandan crisis of 1994. The scale of displacement, intense media scrutiny and the perceived failure of the humanitarian community in Rwanda led to questions about the accountability of humanitarian actors. The 1990s has been described as humanitarianism’s coming of age, and the drive towards standardisation resulted from, in part, a desire to professionalise the field and codify practice. That the standards were created through a consensus-focused process is an important feature of the project.

To better understand the application of the Sphere standards in urban areas, this article focuses on Sphere’s physical shelter standards. Shelter is of vital importance in disaster response, providing security and protection against the elements. Sphere’s physical standards for temporary planned or unplanned camps, which form part of the broader chapter on shelter, are described in detailed guidance notes. These include guidelines that the site gradient not exceed 5% and that the site’s lowest point should be at least 3 metres above the water table (Figure 1). The standards also call for a minimum usable surface area of 45m² for each person. When communal services are provided elsewhere within the area this can be reduced to 30m². The minimum covered floor area per person should be 3.5m² and the minimum ceiling height should be 2m. To prevent fire, there should be a 2m space between each structure and, for every 300m of development, a 30m firebreak.

Haiti: a case study of Sphere implementation

Haiti is one of the largest urban tests to date of existing approaches to humanitarian response. The earthquake, which struck on 12 January 2010, left 1.5 million people displaced and destroyed over 250,000 homes. The shelter response in Haiti was coordinated through the UN cluster system, which was created in 2005 to enable collaboration and information exchange between humanitarian organisations. The shelter cluster in Haiti sought to coordinate the work of organisations involved in shelter issues. However, the cluster system faced considerable challenges and often paid too little attention to the perspectives of smaller and local organisations.+M. Hooper, ‘Priority Setting Amid the Rubble: Organisational Approaches to Post-disaster Reconstruction in Haiti’, International Journal of Disaster Risk Reduction, 2014.

Haiti’s proximity to the United States resulted in a massive influx of responders, including at least 980 non-governmental and civil society organisations.+V. Ramachandran and J. Walz, Haiti: Where Has All The Money Gone? (Washington DC: Center for Global Development, 2012). In Haiti, 59 organisations were listed on the UN’s shelter cluster website, though this number probably does not include all those involved in the sector. Despite the considerable involvement of small and local organisations in on-the-ground implementation of post-disaster response and reconstruction, most organisations listed on the shelter cluster site are large and based in the global North.

To understand how organisations responding to the 2010 earthquake acted on the Sphere standards, the authors interviewed representatives of 14 organisations presenting a cross-section of those working on post-disaster shelter issues in Port-au-Prince. Using an annual budget of $10m as a dividing line, five small and nine large organisations were interviewed. The interviewees also included three organisations based in the global South and 11 in the North. Three had been involved in Sphere standard setting and 11 had not. This breakdown is broadly reflective of the organisations listed on the shelter cluster website. Organisations meeting the above criteria were selected randomly for interviews between December 2014 and February 2015. Interviews were structured around a set of standard questions, but also allowed for open-ended answers. We have kept responses anonymous to protect interviewees from any adverse consequences arising from their remarks.


Donor pressue increases the likelihood of adherence

Thirteen of the 14 interviewees, including representatives of all types of organisations, reported that donor pressure increased the likelihood that standards would be adhered to. As donors face pressure for accountability, they push this pressure downwards by requiring accountability from beneficiary organisations. This includes pressure to achieve Sphere compliance. Interviewees described how Sphere compliance was monitored in reports they were required to submit, in conversations with donors and through site visits and assessments by donors. Interviewees often mentioned the challenges this raised. Interviewees from organisations based in the global South remarked that, at some points, they were tempted to stop working with certain donors due to the pressure to comply.

Urban constraints make adherence to standards difficult or impossible

Nine out of the 14 interviewees said that the realities of the urban environment made Sphere adherence difficult or impossible. This was mentioned especially with regard to conditions before the earthquake and the plot sizes available after the disaster. A majority noted that pre-existing conditions often failed to satisfy Sphere standards. This feature of Port-au- Prince was exacerbated by the extensive damage caused by the earthquake, and by the influx of people into the city in its aftermath. Interviewees from smaller organisations and those based in the global South in particular highlighted what they saw as a mismatch between the standards’ spatial requirements and the reality on the ground.

Resolving tensions between standards and urban realities
Interviewees described four approaches their own and other organisations had adopted to resolve the tension between donor demands for Sphere adherence and the difficulty of adhering to these standards in an urban context.

  1. Avoid the city: Interviewees reported that, given organisational capacity, the reality of what could be achieved in Port-au-Prince and the expectations of many donors, some organisations exclusively targeted areas outside of the city. Interviewees argued that it was simply too difficult to operate in many urban sites.
  2. Reinterpret the standards: Another approach to resolving tensions around the standards involved their creative reinterpretation. One area where such reinterpretation occurred was around family size. Initial rapid assessments in Haiti suggested that the average family size was five. However, some organisations based their calculations of personal space on a family size of four. This reinterpretation allowed these organisations to be ‘Sphere compliant’ while working within the constraints of the urban environment.
  3. Educate donors: A widely reported response was the need for greater donor education on the urban context in which organisations were acting. It was hoped that greater understanding between organisations on the ground and donors would make demands for compliance more flexible.
  4. Technical innovation: The final response involved technical innovation. Multiple interviewees described how, given the challenging urban environment and the spatial constraints they faced, the solution was to build upwards. They proposed two-storey shelters (Figure 2). This innovation worked within the physical constraints of the city while also meeting Sphere standards.

However, only one organisation implemented this design solution. The others cited financial constraints and donor resistance as limiting factors. This approach connects with the need for donor education, as the higher cost of this modified design needed to be authorised and supported by donor agencies.

Figure 2. Two-story shelter in Haiti

Figure 2. Two-story shelter in HaitiPhoto credit: Rafael Smith


This research reveals a number of tensions around Sphere implementation in urban environments. The majority of interviewees argued that donor pressure increased the likelihood of standard adherence but that, simultaneously, working in an urban context made adherence difficult if not impossible. Managing this tension by avoiding urban contexts or reinterpreting the standards poses considerable challenges for advocates of humanitarian standards and those who hope for more successful humanitarian engagement with urban settlements. It also potentially prioritises Sphere at the expense of truly grappling with urban humanitarian needs.

Donor education and technical innovation offer greater promise for the future of Sphere and for efforts to effectively address urban disasters. While interviewees hoped to raise donors’ awareness concerning the challenges of implementing Sphere’s physical shelter standards, they also reported power dynamics that are likely to make such up-stream information flow difficult. This suggests that formalising feedback oppor-tunities for voices that are often marginalised, particularly those of smaller and developing country organisations, in the Sphere process will be important in developing a robust set of future standards. Technical innovation – while showing promise – likewise reveals challenges in how to incorporate such novel approaches into the repertoire of humanitarian organisations and donors. As the lack of interest in two-storey shelters shows, such innovations may not be supported due to their cost, complexity or unorthodox nature. Again, this points to the need for more robust means of providing feedback to those driving standard setting.

One possible challenge associated with the Sphere Project’s standard-setting process is that it has been consensus-driven. As the examples above suggest, true consensus concerning standards is unlikely to be achieved if it fails to take into account the perspectives of the wide array of organisations engaged in urban disaster response. To ensure the inclusion of often marginalised voices, some mechanisms for disagreement, debate and innovation appear to be necessary in revising the Sphere standards, and are likely to make the standards more robust than a consensus format would allow. While such a consensus may be achievable on paper, it is likely to leave critical voices unheard and lead to unrealistic standards that are unlikely to be achieved in urban post-disaster contexts.

Martha Pym is a Research Fellow at the Social Agency Lab, Graduate School of Design, Harvard University. Michael Hooper is Associate Professor of Urban Planning at the Graduate School of Design, Harvard University.