CHAPTER 1
Introduction: What do we really know about the impact of reconstruction?
Theo Schilderman
The need for a longer-term approach to reconstruction and recovery
The Philippines were hit by a typhoon in November 2013, the intensity of which they had not experienced before. According to the International Disaster Database, EM-DAT (2014), the number and intensity of hydro-meteorological disasters is increasing steadily. Financial damage attributed to disasters also shows an upward curve. Disasters and development are closely interlinked. On the one hand, disasters regularly wipe out years of development (Schilderman, 1993; IEG, 2006), an impact we strive to reduce. On the other, development gone wrong creates the vulnerabilities that turn natural hazards into disasters (Wijkman and Timberlake, 1984; Grunewald et al., 2000; IEG, 2006). Even in recovery, people sometimes end up struggling more with underlying development problems such as a declining economy or poor governance, rather than simply rebuilding houses (Cosgrave et al., 2009).
Disasters are not neutral; they are known to affect the poor more (Guha-Sapir et al., 2013; Arnold and Burton, 2011). Reconstruction and recovery can be very difficult for the poor and may further increase pre-disaster inequalities; those longer-term impacts, though, often remain hidden from the outside world because the attention of donors and the media has shifted elsewhere (IEG, 2006; Arnold and Burton, 2011), and can vary hugely. A recent report by a special rapporteur to the United Nations (2011) concluded that reconstruction and recovery too often focus on physical structures (houses), whereas for those affected the recovery of livelihoods, social networks, or services is frequently a higher priority than houses. Predominant approaches to reconstruction and recovery seem to be short-term and reactive, where they ought perhaps to take a longer-term view and pro-actively seek to reduce pre-existing and post-disaster vulnerabilities, not just of houses, but of affected people.
The full impact of recovery and reconstruction on people's livelihoods and resilience can only be truly understood in the longer term. Yet, agencies rarely return to locations where they were supporting projects. Much of our knowledge about impact comes from project evaluations or end-of-project reports produced shortly after projects ended. There have been far fewer longitudinal studies of impact. This chapter is largely based on a literature review that aimed to take stock of such longitudinal studies. Of the 99 documents reviewed, only 15 were written more than five years after the end of reconstruction, 40 within four years; the remaining often covered several projects and a more variable timespan, but were rarely found to include long-term lessons.
For us to be able to develop more effective long-term approaches to reconstruction and recovery in the housing sector which consider their contribution to general development and to a reduction of vulnerabilities, especially of the poor and marginalized, we need to learn more lessons from the past. That is why the Building and Social Housing Foundation (BSHF) and its partners undertook the research that forms the basis for this book.
Developing the research
BSHF was established to identify innovative housing solutions with a proven track record, through research and a competition, and to promote their transfer to potential users worldwide. Its sources of funding are limited. We, therefore, could not opt for a quantitative approach, e.g. interviewing hundreds of households. Instead, we adopted a more qualitative method, interviewing small numbers of people as individuals and in groups, but verifying the reliability of their answers through triangulation. Our aim was to begin to understand what issues and factors had the greatest impact on the long-term resilience of people in disaster areas and their houses, illustrate them with some real-life examples, and communicate these to reconstruction agencies. Hopefully, this would then help them to design better projects in the future, identify indicators to measure success during and after the life of projects, and identify gaps in understanding to be addressed by additional research.
Through its annual World Habitat Award (WHA) competitions, BSHF has developed resources, including databases of excellent projects dating back to 1985, and of professionals involved in housing, many of whom have been involved with us in networking, exchanges, or collaborative research. A number of these dealt with reconstruction or disaster mitigation of housing. For most, we also had a project summary that had adequate information on the end-of-project situation and the lessons and impacts at that stage, which we could use as a 'baseline' with which to compare the current situation of each project. Fieldwork could provide evidence not only for changes in housing and livelihoods over time, but also stakeholder perspectives on factors influencing those changes. The 'baseline' and the fieldwork could then be used to write a story of change for each project, ultimately becoming chapters in this book. We contacted a number of agencies who had implemented reconstruction and mitigation projects that had either previously won a WHA or were finalists and asked them whether they would like to partner with BSHF in looking back to establish the projects' impact over time; seven were happy to do so. Three agencies then offered to investigate additional projects: in one case, a reconstruction project was added to a mitigation project in the same country; in another case, two more projects were investigated in neighbouring countries; and in the last case, a different office of the same international agency, on another continent, offered another study. None of these were on our database, but we believed all to be good projects, of which we were able to obtain 'baseline' information and could draw lessons. That produced 11 case studies with a good worldwide spread: five in Latin America; three in South Asia; and three in South-east Asia.
In order to focus the research we first discussed the information all of us wanted to get out of it with the research partners involved. This was summarized into broad themes with a series of questions under each, that were subsequently discussed with around 150 members of the broader relief and reconstruction community, including practitioners, researchers, and donor agencies, at the UK Shelter Forum in Oxford and the Shelter Centre in Geneva. Those involved in all those discussions broadly agreed on five thematic areas of investigation:
• User satisfaction – what matters to beneficiaries, and how do projects achieve it? Is it a factor in people assuming 'ownership' of their house which could be an important indicator of the sustainability of the houses built?;
• Beneficiary targeting – do projects reach the right people, and are the solutions they offer appropriate in the long term? Do beneficiaries continue to benefit from their reconstructed home, particularly exploring why some did not occupy or abandoned their houses?;
• Replication – of the housing technology and implementation approach, by individuals or agencies, which is an even stronger indicator of appropriateness.
• Technical performance – assessing whether houses were durable in the environment and stood up to subsequent disasters, the weather and other external factors.
• Livelihoods – the link between livelihoods which generate a reliable income, vulnerability reduction, gender equality and future resilience to disasters, is well proven. The research sought to explore whether projects had increased the resilience of people through integrated livelihood activities, rather than focusing on building more resilient houses.
The full research methodology is provided in Appendix 1.
All research partners reported back to BSHF under the above thematic headings. They subsequently presented the main outcomes and lessons of their cases to each other and representatives of the broader reconstruction community at an international conference at Coventry University on 15–16 January 2014. This provided a true opportunity to share lessons from across the world and hear personal stories come alive from people able to revisit their projects. The discussions there and the people who contributed to them were influential in the formulation of the ideas developed in this book.
What do we know from past experience?
Most of what we know about the outcomes and impact of reconstruction was written shortly after the end of projects; documentation of their longer-term impact is much rarer. This chapter summarizes the available information on the identified themes, and may, therefore, be rather biased towards immediately visible lessons.
Over time the mistakes made by projects in their design or implementation become more apparent: leaking roofs or walls being affected by humidity, poorly cured concrete showing damage and rusting reinforcement, kitchens or toilets being in the wrong place, unused or moved. Outcomes that seem positive at the end of projects can seem less so later on, for instance because an agency has stopped supporting the activity or a temporary post-disaster construction boom has subsided. When that happens, people's opinions about their rebuilt house may worsen. Conversely, as reported by Duyne Barenstein (2013; Chapter 3 of this book), when residents have the skills and means to improve their houses at a later stage, their satisfaction may improve as well as the houses' performance.
Contextual changes over time can also affect people's opinions and desires. Rules, policies, and public opinion will evolve, for example, people becoming more conscious of their environment and saving energy. Economic growth and related increases in income, coupled with advertising of a 'modern' way of life, can create a desire to change. Guzmán Negrón (2010) reports that people in rural areas of Alto Mayo, Peru are still happily replicating the improved quincha (mud-and-pole) houses built after the 1990 earthquake, but in urban areas that had benefited more from Peru's economic boom, about a third of those were replaced by 'modern' concrete frame technology.
Looking back at projects, therefore, it is important to consider that not every change that occurred in housing or beneficiaries' livelihoods can be attributed directly to decisions made by the relevant reconstruction project teams; external influences may also impact people's opinions.
User satisfaction
If beneficiaries and those around them are happy with housing solutions provided, they are more likely to replicate them. We have to bear in mind, though, that happiness is subjective, may vary between residents and, as we have seen above, over time. Many researchers have indicated factors that contribute to the satisfaction or dissatisfaction of residents. They do not always agree with each other, but important trends exist.
Some sources maintain that owner-driven reconstruction (ODR) leads to greater user satisfaction than donor-driven reconstruction (DDR); e.g. Duyne Barenstein (2006) in post-earthquake Gujarat, and Lyons (2007, 2010) in post-tsunami Sri Lanka. But in a large survey of reconstruction approaches in Aceh, UN Habitat (2009) generally found little difference in satisfaction between these approaches, though some projects stood out, either positively or negatively. A key factor that distinguishes them is the degree of participation beneficiaries had in decision-making. Where they had little say, it often led to inappropriate houses (Sadiqi et al., 2011) and 'one-design-fits-all' solutions (Barakat, 2003). Lack of participation frequently produces culturally inappropriate house designs or settlements (Jigyasu, 2000; Kelly, 2010) or climatically unsuitable houses. Participation tends to promote some flexibility in design and, by considering beneficiaries' expressed needs, leads to more appropriate design and greater functionality (Wilches-Chaux, 1995; Duyne Barenstein, 2006; da Silva and Batchelor, 2010) and ultimately greater satisfaction.
Well intentioned post-disaster risk reduction initiatives have imposed land zoning and otherwise required relocation of communities. Studies have shown that often relocation affects, and may become a source of, dissatisfaction (Grunewald et al., 2000; IEG, 2006; Fallahi, 2007). Relocation has particular challenges but such projects are not universally causes of dissatisfaction for beneficiaries, e.g. Cronin and Guthrie (2011) report on an often flooded slum in Pune being relocated and inhabitants being happy with the result.
A major contributor to beneficiary unhappiness with relocation sites was the tardy installation of infrastructure (Grunewald et al., 2000; Hidellage and Usoof, 2010). At times, insufficient infrastructure can also cause problems to those reconstructing in-situ. If water or energy supplies are absent, building becomes difficult. Other components, like toilets, may be absent altogether when the focus of a reconstruction project is solely on the provision of permanent shelter. The review found a minority of beneficiaries were unhappy with some infrastructure in their projects, but where it was improved when lacking before that was appreciated (Basin South Asia, 2008; Cronin and Guthrie, 2011).
Good quality and durable materials and construction are highly valued (Wilches-Chaux, 1995), and the absence thereof regretted (Duyne Barenstein, 2006; UN Habitat, 2009; Guzmán, 2010). That appreciation can extend to the use of vernacular materials or technologies (Schilderman, 2004; Stephenson, 2008; Jha et al., 2010) and is often linked to achieving good disaster resistance (Barakat, 2003; van Leersum, 2008; Razavi and Razavi, 2013). An unfortunate consequence of quality improvement is sometimes that size is reduced to remain within budget. It is important to residents to have an adequately sized house and plot to accommodate household and livelihood activities, particularly to allow for future extensions (Wilches-Chaux, 1995; Duyne Barenstein, 2006; Maynard et al., 2013).
Both quality and size relate closely to affordability: beneficiaries like housing solutions that are affordable to them (IEG, 2006; van Leersum, 2008), even if these are based on vernacular technologies. Over time, residents do adjust to things they do not like. They will replace leaking roofs, or alter them to adjust better to the climate (Guzmán, 2010). They alter inappropriate designs (D'Souza, 1986), or change rooms into shops, kitchens, or areas of worship (Wilches-Chaux, 1995; Maynard et al., 2013). Some will add rooms (Wilches-Chaux, 1995; Schilderman, 2010). Where houses or toilets were inappropriate, some were turned into stores (Jigyasu, 2000; Salazar, 2002). If they cannot afford the resistant technologies implemented during reconstruction, they may revert to poor quality vernacular technologies that may make them vulnerable again (Jigyasu, 2000, 2004).
The choice of project staff can be crucial in determining the options beneficiaries have. Professionals involved often lack the experience and the education suited to traditional contexts, e.g. they may have a strong bias toward modern technologies over the vernacular, with which they are often unfamiliar and are, therefore, not confident to apply (Jigyasu, 2000, 2004); they may value quality and perceived greater disaster resistance over appropriateness and affordability; and they often view reconstruction from the narrow perspective of their own sector only, such as housing, which may explain the lack of integration in projects (United Nations General Assembly, 2011). A lack of consideration for people's livelihoods, e.g. in how houses are designed or where they are located, is a major contributor to user dissatisfaction.
Reconstruction can lead to health improvements, but the opposite happens more often. Evidence of improvement can be vague (Duyne Barenstein, 2006); Cronin and Guthrie (2011) claim it to 'have somewhat improved'. It may also be circumstantial, with authors stating that health risks are reduced due to the provision of better water, sanitation, and hygiene or occasionally because projects also built health facilities (Basin South Asia, 2008; Deprez and Labattut, 2010; Jha et al., 2010). More frequently though, residents' health is put at risk, particularly because of sanitation that is poorly built and maintained (Duyne Barenstein, 2006), leaking into the groundwater (Maynard et al., 2013), inappropriate and, therefore, not used (Salazar, 2002; Maynard et al., 2013) leading to reversion to open defecation (Jigyasu, 2000; Duyne Barenstein, 2006). In Aceh, inefficient drains cause health risks (Deprez and Labattut, 2010).
Beneficiary targeting
This review found many examples of beneficiaries not occupying houses allocated to them, or abandoning them after a while. Is this because projects did not target the right people? Or is it because inappropriate houses and locations were offered to the 'right' beneficiaries?
