Academic and intellectual case

Research question and objectives

Despite government commitments to land reform, impacts have not been accurately assessed. By collecting detailed empirical data, comparing data across different sites in three national settings, and synthesising findings, the project will:

The current state of knowledge

Given growing policy concerns around land reform and rural livelihoods in southern Africa, this project comes at a critical juncture. There has long been recognition of the importance of land reform as a key condition for rural economic growth and poverty reduction, especially where the distribution of land is highly inequitable (Lipton, 1974; Binswanger et al, 1995; Toulmin and Quan, 2000; Palmer, 2000; Deininger and May, 2000; Deininger, 2003). The colonial inheritance in southern Africa resulted in some of the most skewed rural land distributions in the world. This has created a dualistic pattern of agrarian relations where large-scale commercial farms occupy fertile land alongside small-scale ‘communal’ farming areas, often of poor agricultural potential, where the majority of the rural population live. This agrarian structure is characterised by high levels of inequality, poor economic linkages, and low levels of growth, resulting in deep-rooted poverty (Cliffe, 2000; Bernstein, 2002; Berry, 2002; Peters, 2004).

Beyond political arguments about redressing historical injustices (de Klerk, 1991; Moyo, 2000), there remains an array of unresolved questions about the impacts of large-scale land redistribution. Some claim that land redistribution does not result in significant improvements in rural economic growth and has limited impacts on poverty reduction (Sender and Johnson, 2004). Others dispute this, showing how impacts are positive because of the efficiencies of small-scale farming (Eastwood et al, 2004) and the multiple livelihood strategies of rural people (Lipton, 1993, 1996; Andrew et al, 2003). Studies from southern Africa have shown how positive impacts may only emerge over time (Kinsey, 1999, 2000), and if accompanied by other support measures (van den Brink, 2003; Hall et al, 2003). Given the paucity of other options in economies under stress, there is a clear need to boost and sustain rural economic growth, and so secure livelihoods and reduce poverty (DFID, 2005). Given the rapid rate of economic decline in Zimbabwe and controversy over the role of ‘fast-track’ land reform, there is an urgent need for research findings to inform policy debates within the region.

In southern Africa many agree that land reform must form part of efforts to reduce poverty and inequality. But what approaches are appropriate for what contexts? What mechanisms for institutional, technical and financial support are required? Remarkably, despite important but isolated empirical studies (e.g. Hall, 2004; Mtethwa et al, 2004; Wegerif, 2004 - South Africa; Kinsey, 1999, Moyo, 2000; Chaumba et al, 2003a,b – Zimbabwe; Werner, 2003, 2004 - Namibia), there has to date been no systematic assessment of the poverty reduction and livelihood impacts of land reform in southern Africa. This proposal aims to fill this important data gap, developing appropriate and replicable methodologies for such an assessment.

The conceptual challenge

A pivotal issue at the centre of the land reform debate is the often vaguely-specified issue of viability of new land-based livelihoods (see, for example, MALA, 2001; GoZ, 1998). Many questions arise: is the provision of land to new settlers going to result in beneficial impacts? Are new settlers capable of using the land they obtain in a productive manner? Are the necessary support services in place? Are new settlers likely to achieve food security in the short term?  Will the scheme, project or programme be sustainable in the longer term?

However, a deeper and conceptually well-informed examination of what is meant by ‘viability’ is often absent in policy debates. It must be asked: viability for whom? At what scale? Over what time period? In relation to what criteria? Answers to these questions must guide impact assessments and frame trans-disciplinary methodologies used for assessment, monitoring and evaluation. Interrogating particularly economic notions of viability is therefore central to achieving a better understanding of the impact of land reform efforts in southern Africa and what this implies for future strategies. To date debate has largely focused on the scale and profitability of production, based on conventional farm management planning, business models and the notion of a ‘minimum economic unit’ (cf. Aliber, 1996). The resulting models often carry with them assumptions about the tenure arrangements required (van Zyl, 1996).

Influenced by past history, post-independence policymakers throughout the region have been heavily biased towards models drawn from the large-scale, capital-intensive, freehold sector. These suggest particular perspectives on land subdivision, the setting of farm size ceilings, property regimes, concentration and linkage between enterprises and economies of scale (van Zyl, 1996; Cliffe, 2000). Strong continuities are therefore apparent with the past, with colonial-era assumptions often framing present-day approaches (Lahiff, 2003). At the centre of much debate around land reform – and implicit in most conventional methodologies used to assess impacts – are therefore a set of normative assumptions about what it is to be a rural dweller pursuing a largely land-based livelihood, and so in turn assumptions about what it means to be a successful ‘farmer’ (Lipton, 1996; Scoones and Wolmer 2003). 

Interrogating understandings of viability and exploring methodologies for impact assessment thus goes to the core of the land reform debate. A central proposition of this research is that if land redistribution is to have an impact on poverty reduction, land reform models and assessment techniques must go beyond narrow conceptions of ‘viability’ – and associated assumptions about property regimes, settlement models etc. - based on existing norms established for the large-scale commercial sector. Instead, new approaches require a focus on the wider livelihood consequences of land reform, including social, institutional and environmental impacts (Lahiff and Cousins, 2005). Rather than seeing land reform solely in narrow economic terms, it opens up a more comprehensive basis for assessment, one that examines the complex, dynamic relationships between land and livelihood sustainability. To understand the poverty impacts of land reform, therefore, a dynamic livelihood pathways approach is required, which traces how the transfer of one key livelihood asset – land – affects livelihood options and poverty reduction outcomes more broadly (Cousins, forthcoming). This requires an approach that goes beyond a simple input-output focus to one that examines the dynamic, sometimes unpredictable, social, political and institutional mediation of livelihood change.

A dynamic livelihood pathways approach presents therefore some major methodological challenges. What are the appropriate measures for assessing livelihood/poverty reduction impacts? What livelihood outcomes are important for different groups of people? What are the institutional underpinnings of new forms of livelihood opportunity? What time-frames and scales should be used to assess impacts? How are prospects for land reform conditioned by existing agrarian structures, and how will land reform affect these? Despite the centrality of these issues, there remains remarkably little empirical material with which to answer these questions in South Africa, Namibia or Zimbabwe (especially post-2000). Political debate, media commentary and much policy discussion thus often rest on untested and politicised assertions. Building on substantial past research by the applicants on these themes (see www.uwc.ac.za/plaas; www.ids.ac.uk/slsa), this project will examine how land reform has changed livelihood and poverty outcomes and affected the wider agrarian economy, and what post-land reform support is needed.
 
Research design and analytical approaches

The research will focus on livelihood impacts in South Africa (Limpopo Province), Zimbabwe (Masvingo Province) and Namibia (Kunene, Otjozondjupa, Oshikoto and Omaheke regions). These have been chosen because they are i) the focus for on-going land redistribution efforts; ii) are broadly comparable in terms of agro-ecology, existing support infrastructure and livelihood profiles and iii) are areas where the applicants have extensive field contacts and past research experience.

Within each province or region a series of field sites will be chosen, representing a range of different land redistribution settings, such as (i) low-input, dryland agriculture and livestock production; (ii) joint venture arrangements for high-value irrigated crops such as horticulture and sugar; (iii) wildlife or tourism-based enterprises. The research focus will be on (i) as the dominant setting for redistribution in all three country cases, with comparative cases drawn from other transfer models.

Within each site, data will be collected by a mix of qualitative and quantitative methods at two levels:

 

This two-level assessment will be diachronic, tracing changes and comparing the situation ‘now’ (based on original field data collection in two seasons (06/07 and 07/08, with retrospective data collection from 04/05 and 05/06), with the ‘pre-settlement’ situation. Pre-settlement data will be derived from: (i) largely qualitative retrospective data on individual households’ livelihood status before land reform, including micro-histories of livelihood building and biographies of change of individuals and households since redistribution; (ii) data on production and returns from the land in its former use prior to land redistribution (triangulating between available farm records, interviews with former land owners/farm workers, government surveys etc.). In addition to looking at  ‘before’ and ‘after’ impacts, the project will trace the often unpredictable pathways of change that occur in livelihood patterns as a result of land reform, developing innovative methodologies for such a dynamic livelihood pathways analysis.

These household and scheme level data will be complemented by a district or provincial/regional level assessment of the wider economic and social impacts of land reform. This will look at the consequences for services, processing, input supply and markets, and the wider structure and social/market relations of the agrarian economy. This will be based on a tracing outwards from the case study assessments, complemented by interview survey work in major regional centres, focusing on an examination of (i) changes in market arrangements and structures and (ii) changes in the input supply and service sector. This work will thus gain qualitative insights into changes in ‘linkage’ and ‘multiplier’ effects, and so the connections between land reform efforts and the wider economy (cf. Delgado et al, 1998; HSRC 2003). Survey data will combine with an examination of available longitudinal data on the district and provincial/regional economies to produce a retrospective – and necessarily tentative - mapping of changes that have occurred. This, in turn, will be set within a wider assessment of the political economy of agrarian change, comparing the recent experiences of South Africa, Namibia and Zimbabwe, and how these have affected poverty impacts and livelihood outcomes (Lahiff, 2003; Cousins, 2004).

Case study sites

A number of case study sites have been provisionally identified for in-depth study. These have been chosen to highlight the variety of transfer models. Clearly each case has its own individuality, and inevitably none are will be truly representative. However, they have been selected purposively to cover the range options, with ‘typical’ cases of each chosen.

In Limpopo province, South Africa six sites have been identified. Four are cases of small-scale farm-based production, one is a mix of this and a wildlife/tourism enterprise, and one is a joint venture arrangement with a large citrus operation. In Masvingo province, Zimbabwe sites include three small-scale farming settlements (so-called Model A1), one medium-scale unit commercial farm settlement (Model A2), one mixed wildlife/farming scheme, and one commercial sugar plantation settlement. In Namibia regions have been selected where three different land reform models have been implemented, namely the Affirmative Action Loan Scheme, the Farm-Unit Resettlement Scheme, and Cooperative or Group resettlement (see Appendix 2).

Final choices of the case studies will occur at the beginning of Phase 2 (see below).The cases will be situated within a broader review from secondary material (including data and surveys which have not been analysed systematically) of the provincial setting occurring in the first phase of the project.

Disciplinary contributions and interactions

The methodological approach employed will necessarily be cross-disciplinary, drawing on inter alia livelihood assessments; biographical change profiles; agricultural/farm economics; social network analysis; social dynamics and gender analysis and environmental impact appraisal. Data collection techniques will be equally diverse, ranging from household surveys to participatory appraisal, ethnographic observation and simple crop/farm modelling. The research team reflects this range of methodological skills and disciplinary expertise, but, critically, will be able to combine approaches and tools to develop an integrated livelihoods impact assessment approach. The core applicants have extensive prior experience in such analysis (Cousins et al, 1996; Lahiff, 2000; Scoones, 1998), and have put together a cross-disciplinary team with the requisite capacities to develop and refine appropriate methodologies.

Work programme

The research will unfold over three phases, each comprising different activities and outputs:

Phase 1: Establishment of a research and user engagement strategy (6 months)
Activities: setting up field sites; field-testing of survey instruments; recruitment of PhD students; reviewing literature and preparing policy discussion papers; developing website, mailing list and network contacts.
Outputs: 3 literature reviews; 3 policy discussion papers; field site profiles; draft survey instruments; user engagement strategy; PhD research proposals; website launch.

Phase 2: Main field research period (24 months)
Activities: Field surveys; fieldwork methodology and analysis workshops; interacting with users; visiting fellowships at PLAAS/IDS for writing up/analysis.
Outputs: workshop reports; 4-6 working papers (for revision and submission to journals); 4-6 policy briefings; visiting fellowships at PLAAS/IDS.

Phase 3: Dissemination and policy-user networking (6 months)
Activities: travelling seminars for farmers/policy users/researchers across sites; co-funded regional conference.
Outputs: 6-9 working papers (for submission to journals); draft of project book; final field reports and seminars from PhD students; conference report.