Sustainable Livelihoods: A Case Study of the Evolution of DFID Policy

Research reports and studies
May 2003
William Solesbury

This paper is a case study of the influence of research on a particular shift in policy for the Department of International Development (DFID). In the 1997 White Paper on international development, DFID made the 'sustainable livelihoods approach' (or SLA), a core principle of its strategy for pro-poor policy making. The concept of SLA had first appeared in research literature in the 1980s, and its inclusion in the White Paper marked its transfer to the policy domain. This Working Paper offers a descriptive narrative of this progression, identifies major events in the story, and analyses this successful transfer from research to practice and policy through the framework of context, evidence and links.

The paper forms part of the Overseas Development Institute's Research and Policy in Development (RAPID) programme, which seeks to learn more about linkages between development research, policy and practice. The main questions addressed are:

    * How did the idea of the Sustainable Livelihoods Approach come to be adopted, and

    * What was the role of research in this process

The principle events of the evolution of the SLA framework were as follows:

    * 1987 The World Commission on Environment and Development publishes the 'Brundtland Commission report'

    * 1988 IIED publishes 'The Greening of Aid: Sustainable Livelihoods in Practice'

    * 1990 UNDP publishes the first Human Development Report

    * 1992 UN holds Conference on Environment and Development; IDS publishes 'Sustainable Rural Livelihoods' by Chambers and Conway

    * 1993 Oxfam employs the SL approach

    * 1994 Care adopts household livelihoods security as a framework for relief and development

    * 1995 UN World Summit for Social Development; UNDP adopts Employment and Sustainable Livelihoods as one of top five priorities; IISD publishes 'Adaptive Strategies and Sustainable Livelihoods' by Singh and Kalala; SID launches Sustainable Livelihoods and People's Everyday Economics project

    * 1996 'Adaptable Livelihoods' by Davies is published; DFID invites Sustainable Livelihoods projects; 'Participatory Research for Sustainable Livelihoods' by Rennie and Singh is published

    * 1997 New Labour publishes White Paper on International Development 'Eliminating World Poverty: A Challenge for the 21st Century

These events have been mapped onto the RAPID framework of context, links and evidence, to explain the interactions of research, policy and practice. The case of SLA shows the framework to be useful in organising events and processes which influenced the policy shift in DFID, but suggest it must be refined to guard against determinism.


In a number of ways, the sustainable livelihoods approach (SLA) was well aligned with its political and institutional context. Firstly, the SLA was in tune with wider shifts in approaches to development through the 1980s and 1990s; towards a focus on human-wellbeing and sustainability rather than economic growth. Crystallised in the Brundtland Commission Report in 1987 and the first UNDP Human Development Report in 1990, NGOs and supportive researchers had negotiated this shift over the preceding decades. The new perspective was welcomed in the mid 1990s by DFID as it strove to redefine its role and mark the change of government in 1997 with a distinctive and timely approach to international development. The sustainable livelihoods approach succeeded in winning the attention of key policy-makers in donor institutions in the early 1990s, DFID in 1997 and the Natural Resources Department, away from the competing knowledge and theory which key individuals have been exposed to during the course of their careers. This attempt succeeded then because of the collision of two factors: a broad international climate which favoured people-centred approaches, and a specific need to mark out a new phase of development practice in DFID.


A number of individuals worked as 'testers, developers, champions, communicators, interpreters and advocates' of SLA to facilitate its adoption within DFID. More often than not, these individuals performed more than one of these roles, and many roles were performed by several people. In the development of SLA, core researchers, policy-makers and practitioners tended to know, or at least know of, each other. This was especially true within DFID, which fostered mutual respect, awareness and trust and eased the quality of ideas exchanges and the path of SLA. The creation of the Sustainable Livelihoods Resource Group lent a boost to the development of this community on an international level. Flows of ideas between individuals were reinforced by flows of individuals themselves, as critical agents moved their jobs and took their expertise and perspectives from one organisation to the next. The movements of individuals such as Conway between Imperial College, IIED, Ford Foundation, Sussex University and the Rockefeller Foundation, illustrate the non-linear trajectory of research, policy and practice in the development of SLA.

In the SLA case, the conventional view of research informing policy which frames practice: Research - Policy - Practice could be better represented as a triangle where all components inform each other.


As it was accumulated over the decade preceding the 1997 White Paper, the concept of SLA had been tested both intellectually and empirically by researchers and practitioners in dialogue. Evidence with these qualities was particularly attractive to DFID, who were not persuaded by a single campaigning group, but through their interactions with diverse sources and media. The numerous strands to SLA, and the fact that it was communicated many times by many agents, lent resilience to SLA as a core concept which would not be undermined by a single piece of counter evidence or failed argument. The very personal means by which SLA was transferred between colleagues, often through face to face discussions, also fostered resilience as the concept could be tailored to engage with the expertise and interests of specific individuals. However, of equal importance to the multifaceted nature of SLA, was its clarity and ability to be recognised and referred to. Since Chambers and Conway coined the phrase in their 1992 research paper, 'sustainable livelihoods' has expressed a complex set of relationships and ideas with great economy. The Sustainable Livelihoods Support Office produced a range of literature for both lay and political audiences which presented SLA clearly, with a range of case studies and practitioner guides. These have provided some centre of gravity around which more diffuse and wide-ranging debates around SLA can revolve, and have ensured that - whilst being worked from many angles - the concept remained coherent.

Refining the Context/Links/Evidence Framework

There are two elements of the case of SLA which fall outside RAPID's context/links/evidence framework: time and chance. In terms of time, a decade passed between the conceptual formation of SLA and its adoption in the 1997 White Paper. This time was necessary in part for the conjunctions of context we have outlined, but also in a less perceptible way for ideas to be internalised and embedded. The context/links/evidence framework falls short of tracking this process. As for chance, a number of lucky encounters, overlapping diaries, and external decisions set up the chronology of SLA. Arguably, the web of relationships that characterise the community of policy-makers, researchers and practitioners, was sufficiently close and well developed that had these incidents not taken place, others would have emerged in their place, but this cannot of course be verified.

Thus, in order to counter the potential for determinism from the context/links/evidence framework we must recognise a separation between necessary and sufficient conditions. Without necessary conditions there would be no chance of a successful impact of research on policy, but these conditions do not guarantee change on their own. Sufficient conditions lead to actual impact. In the case of SLA, these sufficient conditions were time and chance encounters. In other instances of research influencing policy, the conditions may be very different.

Read the research