Issue 71 - Article 10

Using social protection mechanisms to respond to urban shocks

March 27, 2018
Herma Majoor and Larissa Pelham
Women receiving treatment at Gonoshsthaya Community Health Center outside Dhaka, which provides health care and health insurance to underserved populations in Bangladesh.

Social protection programmes have long contributed to reducing poverty and overcoming chronic and transitory vulnerability and food insecurity, but they are relatively new to emergency response. This article documents lessons from an attempt to use social protection approaches in a simulation exercise involving a large urban emergency in Dhaka. The simulation revealed a gulf in understanding between social protection and humanitarian practitioners about each others’ interventions and ways of working in using social protection to address humanitarian crises. The government of Bangladesh is progressive in this area and has plans to extend safety nets in its social protection system to large-scale shocks. However, much more research is needed to understand how and whether social protection can play a role in urban humanitarian crises. The significant investment in time, capacity and financing that this will require means that it is still unclear whether social protection can be responsive enough to meet the needs of large-scale, rapid-onset shocks in urban areas.

Social protection in Bangladesh

In May 2017, under a DFID research learning project, The Urban Crisis Learning Project was a two-year research project financed by the UK Department for International Development (DFID). The consortium comprised Habitat for Humanity, Oxfam, ODI and University College London (UCL). See Habitat for Humanity and Dhaka Ahsania Mission through Oxfam hosted a three-day simulation of an earthquake response in Dhaka. Bangladesh is located in an earthquake zone, and it has long been anticipated that such a crisis might hit Dhaka. One of the objectives of the simulation was to explore the viability of using social protection during a humanitarian crisis. Over 50 people attended, representing local and national government, international and national NGOs and UN agencies. The aim was to learn lessons to improve humanitarian response and preparedness to major disasters in an urban metropolis in the food security, shelter and water, sanitation and hygiene (WASH) sectors. A scenario was presented to participants detailing the impacts of a fictional earthquake on people, infrastructure, markets and government services. From this, participants had to develop an emergency response action plan focused on each intervention area. The simulation focused specifically on how using cash, market analysis and social protection in these sectors might deliver a faster, more effective response to affected populations.

Bangladesh is one of the countries where using social protection channels to respond to urban emergencies could be an option. However, despite recent growth in the coverage of government social protection for the rural population (from 16% to 30%), only 9% of the urban population is covered by formal social protection programmes. See Urban social protection programmes are smaller and more fragmented than in rural areas, and there are neither beneficiary registries nor a widespread infrastructure that might assist a rapid scale up in response to a large-scale urban crisis. Informal social protection – credit and savings groups, funeral societies – also tends to be weaker in urban areas, leaving urban households with even fewer sources of support.

In 2015, the government launched a National Social Security Strategy (NSSS) (2015–2026). The strategy is intended to improve the effectiveness and coverage of social protection in both rural and urban areas. Ensuring that the social security system supports an effective disaster response system, including urban populations, is highlighted in the NSSS as a priority. By the end of 2020, the NSSS plans to improve management information systems and electronic cash delivery channels, as well as establishing mechanisms to identify areas most affected by crises. Small-scale electronic cash transfers have already begun, and would enable the use of social security schemes to deliver emergency payments if a crisis hits. The government also expects to establish shortterm humanitarian assistance schemes in the form of food, clothing, temporary shelter and medicines. The plan is to ensure that social security schemes for the elderly, children, vulnerable women and people with disabilities are expanded so that urban residents have the same access as people in rural areas. New proposals for childcare will initially benefit urban residents more than rural ones, since more women in urban locations work outside the home.

The main lesson from the simulation was the gap in knowledge between humanitarian and social protection practitioners. This made it difficult for participants to utilise social protection in action planning for earthquake response: opportunities to engage do not exist, and humanitarian practitioners are unaware of existing safety nets that they might be able to make use of in an emergency response. Relationships need to be formally established: if social protection is to be used for humanitarian response, both social protection and humanitarian practitioners need opportunities to meet, and to participate in disaster planning and design meetings, cash working groups, cluster meetings and assessments. This needs to happen at almost all stages, from emergency preparedness planning to implementation, and in the design and revision of safety nets. Such coordination is not yet in place, in Bangladesh or elsewhere.

Basic awareness-raising is required to define points of entry and promote understanding of the basic tenets of social protection and emergency response systems, for example using existing safety nets such as welfare lists or finance service or delivery providers during an emergency, rather than starting from scratch. Equally, the social protection sector needs to assess existing safety nets and explore how they might be used during large-scale shocks. Most formal social protection programmes in Bangladesh are small and fragmented and have weak administrative systems, which might make scaling up difficult during an emergency.

The simulation also highlighted the need to avoid jargon: the social protection and humanitarian sectors have their own histories, political drivers, vocabularies and standards, creating misunderstanding even if goals are shared. Humanitarians design interventions with reference to the Sphere standards, Do No Harm methodology, cluster systems, response analysis, minimum expenditure baskets and multi-sector cash. Social protection specialists couch programmes in terms of employment schemes, social assistance, registries, social protection floors, transfer values, appeals processes and means-testing. But there are also areas of alignment – for example, the need to embed social protection in a social contract closely fits with the humanitarian concern for community participation.

Both social protection and humanitarian practitioners need to routinely consider the informal sector – which is the first on hand to provide support in an emergency. This includes religious support, zakat, burial societies and loans and sharing. Remittances constitute a substantial safety net for many Bangladeshis, alongside other non-governmental support. During an urban shock, the urban–rural remittance chain might be broken, with significant impacts for rural communities. How these informal networks work, and how robust and flexible they are during a shock, is poorly understood. This is an area where NGOs, with their close engagement with communities, could help in understanding and finding ways to support these community structures, for instance by supporting the removal of transaction fees for remittances in the aftermath of a major shock, exploring safety nets at the community level or putting cash through burial societies and other community-based savings and loans groups.

Perhaps the key finding from the simulation is how few of the observations and conversations during the simulation apply to the urban context alone. Much applies to utilising social protection interventions for humanitarian response more generally. Areas for attention include:

  • A typology is needed for different urban contexts, from rapid-onset shocks to protracted displacement, to categorise how social protection may be used differently in urban contexts.
  • Protection and gender concerns in both access to and use of cash will differ in urban areas compared to rural contexts, for example how women access cash and employment and the protection risks they face.
  • We know that cash alone is not enough to support people: additional services and support are needed, for instance for small business development and other livelihoods support, which will require specific understanding of the urban context. Oxfam piloted an interesting flood insurance scheme in Bangladesh, insuring against loss of work in a rural emergency, and we need to see if such a scheme can translate to an urban context.
  • Design and operational issues will be vastly different between rural and urban settings: targeting methods, defining the household in urban settings, data collection, access to services, cash delivery methods, use of cash (such as for rent) and networks – informal, government, private sector – all need to be considered to see how existing systems might respond, or to support urban safety nets. Emerging evidence suggests that using such systems in urban contexts can help build social cohesion, as well as protecting against vulnerability or reducing poverty.
  • Social protection programming in the humanitarian space challenges the roles of all actors, including NGOs. The recent international conference on social protection in contexts of fragility and forced displacement, held in Brussels in September 2017, was framed predominantly as a donor–government relationship, steered by the UN. See Yet NGOs have a critical role to play in certain areas: ensuring that the core principles both of social protection and humanitarian standards are respected, engaging civil society, using appropriate targeting mechanisms and having in place appeals and complaints systems that are functional and that hold governments and donors accountable on both humanitarian and social protection mandates and standards.
  • Social protection may be a helpful exit strategy from urban humanitarian response by providing ongoing support to recovering households. This would also fulfil the social protection agenda for universal coverage of social protection, and act as a stepping-stone to recovery and resilience. See Overcoming the tension between the political considerations that can heavily affect social protection debates (such as providing cash handouts as a way of building political support) and the life-saving demands of emergency interventions will be tricky, but necessary. See

The simulation served as an initiation into social protection for many participants, and as such exposed the gaps between the humanitarian and social protection sectors, rather than the links connecting them. Bringing social protection into humanitarian response will require long-term commitment and sustainable finance from governments and donors. More systematic engagement is essential, and there are multiple other actors to consider in this mix, including the private sector, and the role of technology. NGOs have to clearly define their role as well. For actors in the social protection sector, the use of social protection mechanisms in humanitarian response does not fulfil their agenda to ensure provision for shocks throughout the lifecycle. Fully comprehensive social protection must demand both.

Herma Majoor is an independent consultant on gender, food security and nutrition. Larissa Pelham is Social Protection Adviser in Oxfam’s Global Humanitarian Team.


Comments are available for logged in members only.