Combined Methods Case study. Zambia PSIA: Land Reform

This study assessed the poverty and social impacts of the controversial land reform proposed by the Ministry of Land. In its Draft Land Policy, the government proposed the titling and conversion of some of the land under customary tenure (94% of the land is under customary tenure in Zambia) into state-owned land. This measure was seen to encourage investment, development and productivity through increased security of, improved access to, and incentives for better management of land – including increased rents and fees. The study informed the vivid policy dialogue, generated by the proposal, and the pro-poor focus of the new World Bank CAS.

The Zambia PRSP identified problems with the current land system, which date back to before independence. Numerous studies have shown the link between lack of access to and ownership of land and low productivity of labor and land, missed investment opportunities, and vulnerability and poverty. Past reforms have not been implemented as originally designed due to controversies and resistance. Consensual solutions have not had the intended poverty-reducing and economic development effects. Thus, it was to expect that the 2002 reform proposal would follow a similar path of political conflict and resistance.

The PSIA analyzed the potential impacts of the planned land titling, adjusted rents, and improved revenue collection, allocation of 30% of land to women, ethnic minorities and vulnerable groups, and strengthening of the Land Tribunal. It focused on i) small-holder productivity and incomes, the affordability of land and risks associated with using land as collateral; ii) access to land for farmers (small-scale, commercial, foreign investors); iii) access to land for women, ethnic minorities and vulnerable groups; iv) potential for improved access and equitable distribution within the current institutional framework; v) potential change in welfare given different scenarios related to land distribution; and iv) the viability of the current proposal to address problems, i.e. is the proposal solving the problems or addressing the consequences.

The study coordinators attempted to embed the PSIA in a participatory process in order to create ownership of the study process and findings amongst civil society, policy makers and donors. Civil society organizations, donors, government officials, private sector representatives and academics participated in selecting the policy as a priority, and continued to provide inputs into the data collection, analysis and policy recommendations. The government team was led by the Poverty Monitoring Unit of the Ministry of Finance and National Planning (MFNP) and the Ministry of Community and Social Development (MCDSD), and included representatives of other related ministries and agencies. Civil-Society Organizations (CSOs) and academics conducted the qualitative work (Participatory Poverty Research) , while local offices of UNICEF, FAO, ILO, other UN and bilateral agencies contributed background papers and participated in focus groups, interviews and discussions.

The PSIA team conducted series of focus groups, structured, semi-structured and opened interviews. The stakeholders consulted were from more than 15 ministries and agencies, 12 “big business” representatives, 11 local NGOs, 5 international NGOs, 2 Trade Unions representing the commercial farmers and the small-scale farmers, representatives of the traditional authorities, representatives of more than 11 donor organizations and 5 research and academic institutions. At this initial stage, most of the interviews were done with senior government officials, private sector executives and trade union leaders. Focus groups were used within organizations, combining technical and operational staff at similar level. This was done to avoid the influence of hierarchy. Field-based and Lusaka-based staff was interviewed separately as in most cases Lusaka based staff was senior and more likely to be politically connected. In the international organizations, international and local staff was often interviewed separately.

In parallel with the consultations, the team conducted a literature review. A key part of the review was the analysis of 10 years of Participatory Poverty Research (PPR) work in the same 6 communities already completed by Zambian academics. The data was at the group level but very meaningful as a starting point. Simultaneously, the team used the latest available household survey data to rerun a reduced form model of rural household expenditures to test the correlates with poverty. Based on all of the above, the team decided to analyze three reforms: land tenure, fertilizer subsidies and rural infrastructure.

When designing the methodology the team had in mind not only the robustness and adequacy of tools but the following : i) the data needed to complement existing data so that it can be used to trace historic patterns and for purposes beyond the PSIA; ii) the methods had to be such as to maximize local capacity, to help build it where necessary and to be used in the country in the future; iii) the approach had to allow maximum stakeholders participation. Thus, the team decided to commission a PPR to up-date the already existing body of data, while extending the coverage for the purposes of the PSIA. The team also commissioned an anthropological survey to up-date an existing set of longitudinal anthropological data (ethnographies). Both studies were to inform the up-date of an existing rural household model designed specifically for Zambia and using existing quantitative data. All of the studies were to inform the institutional, stakeholders, scenario and risk analysis. Interviews and focus groups with key stakeholders were planned for the time when the research was completed and the model up-dated. Meanwhile, the literature review continued.

The PSIA employed a mixed-method approach that sequenced and combined qualitative and quantitative data. PPR, generating both qualitative and quantitative data was conducted in 10 communities and four agro-zones. This data was supplemented by longitudinal qualitative data generated by a repeat ethnographic study of the Kefa village, with results compared against a 25 year-old study that served as a baseline.

The rural household model to be used for the analysis was a stylized model of smallholder rural households that was prepared for the 1994 Zambia Poverty Assessment. The model is a single-period linear-programming model¹ that evaluates smallholder cropping patterns. The model combines census data (information on household composition, land holdings, economic activities), socio-anthropological data (time-use, gender distribution of labor for different activities, etc.), with farm budgets for different cropping activities to generate the economic variables of interest (crop production patterns, household income, shadow prices of land and labor). The model only considers incomes from on-farm agricultural activities.² As much has changed in Zambia since 1994, the model needed to be updated.

To understand the different changes³, to avoid the limitations of the available quantitative data and to better understand community, intra-household and gender dynamics, the qualitative research needed to be completed first. The PPR started before the raining season and run through it to provide information for time use, assets distribution, division of labor, access to markets and information, etc. The research covered communities in all the different agro-zones, ethnic groups, with different access to markets, roads and livelihoods. The separate in-debt study of a village where previous anthropological data was available provided great details regarding cropping patterns, local markets, time and infrastructure use, direct and indirect costs.

Using the PPR and anthropological data, the authors of the rural household model up-dated the scenarios and then used available quantitative data (census, household survey, agricultural statistics and other) to evaluate impacts of proposed policy changes on income, labor allocation, fertilizer use, land use, productivity and welfare. This informed the analysis of the potential direct and indirect impacts. The PPR and data collected purposefully at the national level, informed the institutional and stakeholders analysis. The PSIA team collaborated with other studies conducted at the same time and benefited from an IFC investment climate survey, poverty mapping done for the PVA (Poverty and Vulnerability Assessment) and a beneficiary assessment done for ZAMSIF (Zambia Social Investment Fund)

The data collected and the results of the model provided a useful blend of information. The qualitative data provided a detailed understanding of the situation in different parts of rural Zambia while the household model provided quantitative estimates of the benefits and costs, of stylized changes in policies, investments, technologies and cultivation practices on “representative” rural households’ incomes, labor allocation, input use and land cultivated. The contribution to the policy dialogue was significant, as the data was accessible for all stakeholders. The process of data- collection itself rose awareness and public debate. The participation of NGO representatives and civil servants in the PPR increased ownership and credibility. The stakeholders consulted initially and especially, the stakeholders key to the three analyzed reforms were revisited and provided with opportunities to comment and participate in formulating the policy recommendations.

The PSIA impacted on Bank operations, as its results fed into the Country Economic Memorandum, Country Assistance Strategy, Agriculture Sector Development Strategy and Poverty Assessments. The findings were also used by the IMF and DFID in their policy dialogue with the Government. The study impacted in-country dialogue, as the government and civil society used results for their debates and planning.

Source: World Development article

Contact: Abebe Adugna, aadugna@worldbank.org

¹ The household model analyzes a single 12-month period, disaggregated by monthly activities.

² Most smallholder households derive a majority of their income from agriculture activities. 

³ Some of these changes stem from policy reforms through the 1990s, while others stem from external factors such as conditions in international markets, recurring drought and the HIV/AIDS epidemic. As a result, farmers have adopted new technologies and production processes at different rates.