What is self-recovery? – The challenge for humanitarian agencies

This was first published on CARE Insights in June 2018.

Last year CARE and Habitat for Humanity published a study of six agencies’ ‘support to shelter self-recovery’ programmes after the devastating Typhoon Haiyan, which hit the Philippines in late 2013. But what do we mean by self-recovery, and what does support to self-recovery look like?

The humanitarian sector, like many others, loves inventing new terminology. Increasingly post-disaster shelter programmes are being described as self-recovery, or support to self-recovery. However, as so often with new terms, it is not well defined and is differently understood by different actors, which makes it difficult to support it in practice.

In previous years the shelter sector has advocated the use of so-called ‘owner-driven reconstruction’ approaches, while more developmental actors have pursued policies of ‘self-help’ housing development. Is support to self-recovery just a re-branding; a result of short memories? Are we re-inventing the wheel?

Formal and informal support for post-disaster recovery

The term self-recovery developed from the hypothesis that the majority of disaster-affected people have to arrange, or rebuild, their own post-disaster housing, without any significant support from the formal humanitarian system. This system is complicated, and is made up of myriad governmental and non-governmental agencies and donors. The system is itself poorly defined, and it is very difficult to say where formal assistance ends and informal assistance starts.

It’s clear, for example, that remittances play a major part in many people’s post-disaster recovery, and are not part of the formal humanitarian system. It is likely many do in fact receive some form of external support, and perhaps the hypothesis that most rebuild without support is flawed. Debt, migration, unpaid labour, changes to gender roles and many other sources of money and capacity contribute to people’s recovery and rebuilding efforts.

Nonetheless, support to self-recovery is now a key approach in post-disaster shelter responses by humanitarian agencies (and governments).

Owner driven approaches

After a disaster large portions of the affected population may receive emergency items, such as plastic sheeting, blankets, etc. In most disasters, very few will receive a completed house from the responding agencies or other institutions. While it’s extremely hard to measure the amount of assistance people get from what sources, it’s reasonable to draw the conclusion that the provision of post-disaster housing by the humanitarian system is patchy, of inconsistent quality, and insufficient scale. In too many cases the number of houses built by external responders number in the 100s and 1000s while the numbers destroyed number in the 10,000s and 100,000s.

Even when agencies do build complete houses for affected people, the outcomes have often been less than hoped for, with houses left unoccupied, or sold, because they don’t meet people’s needs. It’s long been understood that owner-driven approaches tend to lead to better outcomes. But in practice such approaches have remained focused on achieving high quality products, and have not relinquished much control to affected people. They have rarely, if ever, actually been driven by affected people.

Affected people are often required to build to standard design, with careful control of construction quality. There are cases of this approach being extremely successful, but they do require high levels of oversight and organisational capacity, and sufficient resources to allow people to build to the required standards. Where this is not in place, the most vulnerable affected people may be unable to complete the buildings, and may become trapped in limbo, unable to complete a liveable home, and unable to access additional resources.

Control & agency

For CARE, the difference between owner-driven approaches as they have been applied in practice, and self-recovery approaches, is about the level of control exerted by external responders, and the level of agency affected people have to take decisions about their own recovery. Whether support to self-recovery is just re-branding what agencies already do, or actually something different, depends on whether humanitarian agencies and their donors actually change their approaches and expectations. 
Two definitions of (supporting) self-recovery that have previously been proposed are:

  1. Parrack et al (2014) describe self-recovery as “households rebuild or repair damaged or destroyed homes using their own assets.
  2. Maynard et al (2017) describe ‘supporting self-recovery’ as projects which “provide one or a combination of material, financial and technical assistance; during the relief and/or recovery phase; to enable affected households to repair, build or rebuild their own shelters themselves or through using the local building industry.”

The first definition, just of self-recovery itself, focuses on the disaster affected households and the use of their own assets, but does not explicitly include households being able to make decisions for themselves, or exercise agency.

The second definition, of support to self-recovery, focuses on the programmes that humanitarian agencies deliver, but similarly does not mention agency of affected people. It is hard to distinguish between this definition and the established owner-driven reconstruction approaches already discussed.

Self-recovery = recovery with agency

Both definitions focus on the output of a shelter, or house, being built. But is the objective of the shelter sector to provide physical shelters, or is it to contribute through processes of sheltering to making sure vulnerable people are safe, can realise their rights, and have a chance to realise some, if not all, of their aspirations? Seeing the shelter as the ultimate goal, and not as a means to an end, has limited the meaningful adoption of owner-driven approaches, and threatens to co-opt the term self-recovery into another agency-driven, agency-owned thing. Recent research and thinking, including our new study of projects aiming to support self-recovery in after Typhoon Haiyan in the Philippines, has highlighted the importance of affected people’s own choices and decision-making in making self-recovery a meaningful concept.

For CARE, self-recovery means recovery which is in accordance with the choices and decisions of the individuals and groups who are recovering. Self-recovery is recovery with agency.

The role of humanitarian agencies

The role of organisations like CARE is then to support that self-recovery, so that it is not just recovery with agency, but recovery with (1) agency, with (2) resources, and with (3) the best possible enabling environment. Shelter self-recovery is the aspect of this process which is specifically concerned with people’s ability to achieve post-disaster shelter in accordance with their choices. Projects which aim to support shelter self-recovery must therefore maximise the agency of the people they aim to support, ensure they have sufficient resources (money, materials, knowledge and information) and engage in the enabling environment to overcome any barriers preventing their recovery.

A range of interventions and support

Projects may therefore provide material, financial and/or technical assistance as in definition 2 above, but could also provide legal assistance, governance interventions, livelihoods interventions, or other assistance or interventions in the wider enabling environment (such as lobbying to change land and property laws or practice) which allow people and communities to achieve the results they desire as much as possible on their own terms.

To do this, any assistance must be provided with as few restrictions and limitations as possible so it, when pooled with individuals’ and communities’ own resources, ensures sufficient capacity to achieve a recovery in the way those individuals and communities choose for themselves.  Unconditional, unrestricted cash must be a key component of this in practice.

For CARE, having been investigating the promotion of safer buildings after disasters, and support to self-recovery, for several years now, it is becoming increasingly apparent that if we are to make significant changes to the way disaster-affected people are enabled to rebuild after disasters we, and our donors, need to relinquish control over timescales, over location, and over the end product. We need to get better at listening, better at analysis, and better at adapting what we do to what we hear.

Relinquishing control – but not responsibility

The key challenge for CARE, and other shelter actors, is relinquishing control over the recovery and rebuilding process to the affected people and communities, without transferring all the risk and all the responsibility for the outcomes of our programming onto those very same communities.

Self-recovery cannot mean we, or other actors (including governments), absolve ourselves of responsibility for what happens to people affected by disasters. The poorest and most vulnerable must not be, cannot be, expected to achieve a safe, dignified and adequate recovery entirely by themselves. We must still offer serious and substantial assistance. But we must learn how to do that on their terms, not on ours.

If we can do that, then perhaps supporting self-recovery might be something truly different, and not just a rehashing of old ideas.