Gender equity in contract farming in Eastern Africa

Lead Research Organisation: University of Copenhagen
Department Name: Geography and Geology

Abstract

Contract farming is growing rapidly in Eastern Africa. This research project will evaluate innovations in contract farming operations designed to increase the benefits wives accrue from the contract. It also evaluates whether this improves the relationship between the household and the firm. It integrates a gender dimension into an otherwise gender-blind area of practice and research.

Contract farming is a form of vertical integration within agricultural commodity chains such that a firm has greater control over the production process and final product. Changes in the demand for and supply of agricultural products have increased the popularity of this form of vertical integration in many low-income countries and it is attracting considerable policy and academic attention. While academic work in the 1980s and 1990s offered a mixed assessment of the extent to which contract farming engaged with and benefited smallholders, recent literature offers a much more positive interpretation of smallholder participation.

That said, there are still considerable risks for both firms and farms such that contracting operations frequently collapse. For firms, there is a large risk of smallholders side-marketing both inputs and produce. There are also numerous risks for small-scale producers. Two risks are especially important for smallholders: contracting can contribute to a loss of autonomy and control over farm enterprises and a form of dependency on the contracting firm; and, second, the intra-household distribution of labour/income can be altered to the detriment of women's interests.

This research project focuses on the last of these risks through evaluating innovations that seek to both increase wives' benefits and to improve the relationships between the firm and the household. This is through adding gender-specific elements to the contract to increase the 'self-enforcement range'. By this we mean that by adding these extra investments within the contract we reduce the likelihood of either party breaking the deal due to a change in market or contextual conditions.

The overarching research question is:

To what extent and how do gender-specific clauses within contracts improve outcomes for both farms and firms?

The subsidiary research questions are:

To what extent and how do gender-specific clauses improve the benefits wives accrue from contract farming operations?
To what extent and how do such clauses influence farm production?
And to what extent and how do such clauses improve the stability and longevity of contract farming relationships?

In Malawi, we evaluate the inclusion of a contract for wives to grow and market groundnuts (a conventional crop for women to produce) alongside tobacco to examine effects on yields, prices and, most importantly, the intra-household distribution of labour/income. We do this by combining a randomized design - where members and clubs are randomly assigned the innovation - with a series of qualitative research methods including life history interviews and focus group discussions. The attrition rate (in other words, the number of households dropping from the scheme) will assess whether the farm/firm relationship has changed. In Tanzania, we utilise the same methodology and sequence of research methods but the precise contractual innovation will be determined during the first two months of the project.

In the second phase of research, we extend the project to examine gender-specific innovations in palm oil in Ghana and tobacco in Zimbabwe.

Planned Impact

Outreach, uptake and use strategy

This project has a very clear and achievable set of activities to maximise the dissemination, uptake and use of the research findings by academics, policymakers and practitioners. The strategy aims to engage and influence four specific communities. These are as follows:

Policymakers and practitioners in Lilongwe and Dar es Salaam

Output: To provide a rigorous examination of the costs of benefits of separate contracts for wives
How: Through the presentation workshop in Lilongwe, the set of research briefs, and online resources.
Impact: To influence on-going policy discussions on contract farming legislation, in addition to the agricultural development strategy and national development strategies.

Donors and other stakeholders in the agricultural sectors (for example, DfID, World Bank, Danida, NORAD, SIDA and NGOs)

Output: To provide an example of how development partners can structure innovative impact evaluations of interventions within contract farming schemes.
How: Through the presentation workshops, articles in daily newspapers regarding the impact of the project, research briefs and online resources.
Impact: To help inform donor country strategies, and improve the range of tools available to evaluation specialists and consultants.

Evaluation communities (for example, the African Evaluation Society and European Evaluation Society)

Output: To demonstrate how a qualitative research tool can be utilised within a quantitative research methodology.
How: Through working papers and journal articles.
Impact: The help reduce the gap between advocates of randomized control trials and mainstream evaluation communities.

Development community in the UK

Output: To contribute to the debate on contract farming in Africa, and the debate on the use of experimental methods in development research.
How: Workshop at Overseas Development Institute (ODI), London, research briefs, working papers and journal articles.
Impact: To help bridge the gap between qual and quant evaluation practitioners, and show how mixed methods can be conducted in practice.

The project will include web-based dissemination and outreach tools, which will host details of the study, its methodology, and later, anonymised datasets, visual trajectories and papers.
 
Description What were the most significant achievements from the award?

Overall, the impacts from formally including wives within contract farming operations depends on how wives are included, the crops in question and the context. Data from Malawi and Tanzania presented below are based on pilot clustered randomized designs with regression estimates adjusting for non compliance. Further studies should be conducted at scale to ensure power calculations hold.

In Malawi, formally including wives within a tobacco contract to grow soya led to some counterintuitive results: wives withdrew labour from soya as the crop now fell outside of their domain, contributing to shortfalls in soya yields and production. Husbands revealed concerns about household maize production from the inclusion of wives in the contracting scheme. As reflected in the strong preference for separate contracts expressed by both spouses, integrating wives within an existing contract farming scheme led to non-cooperation between spouses and an inefficient and sub-optimal result.

In Tanzania, including wives within a cotton contract to grow hybrid maize led to wives applying more labour to cotton and significantly greater cotton yields. It also led to less child labour on cotton and other crops. An overwhelming 96 per cent of husbands and 94 per cent of wives interviewed supported the distribution of seeds other than cotton seeds by the firm (mainly for food crops). Whether encouraging wives' participation in contracts through including seeds for staple food crops brings benefits to farms, firms and families depends heavily on contextual factors. Further replication studies at scale and are needed to corroborate or disprove these suggestive findings in different contexts.

Our survey in Tanzania also allowed the project team to assess the demand for and access to private sector extension services by contracted cotton producers. Our bonus findings show that only 22% of contracted smallholders accessed any extension services in the previous agricultural season, mainly from the public sector. Wealthier smallholders received most, the poorest least. The private sector only provided extension to 5% of contracted growers.

Practically all smallholders were willing to participate in private sector extension and reported they were willing to spend an average of 3 days per month at extension meetings offered by the firm. The tension regarding huge demand for and very limited supply of private sector extension could be resolved by involving third parties within the supply of extension from the ginnery. Donors could play a role here by facilitating the inclusion of third party providers of cotton-specific extension advice to farming groups, the cost for which could be recouped from any productivity increases achieved and collected by the firm.
Exploitation Route The funding provided by the grant, less than GBP100,000 for both countries, was only sufficient to complete two pilot clustered randomized control trials. The findings presented above need to be conducted at scale to ensure power calculations hold. Donors should support this. In particular, donors could collaborate with firms operating contract farming schemes in Africa to support gender-specific forms of innovation within their programmes., Specifically, donors could offer to bear the first tranche of losses on innovations which include wives within contracting schemes bearing in mind the main findings from this grant - that the impacts from formally including wives within contract farming operations depends on how wives are included, the crops in question and the context.

Donors could also facilitate the inclusion of third-party providers of extension advice to contracted groups, the cost for which could be recouped from any productivity increases achieved and collected by the firm.
Sectors Agriculture, Food and Drink,Government, Democracy and Justice

 
Description The findings from the grant have led to a host of engagement activities, one publication being accepted and a book contract secured from Routledge with the manuscript initially scheduled to be delivered in July 2019. Due to the PI being assaulted in London in late April 2019, and suffering multiple skull fractures, the delivery of this manuscript has been delayed. Specifically, the publisher has agreed to the submission of the manuscript in December 2020. Please note, the PI now works at the Green Climate Fund, South Korea, and is completing this manuscript in addition to the everyday workload this job required. It is hard to trace the direct impact of project activities and outcomes. Engagement with Alliance One International and Philip Morris International show interest from the private sector. Alliance One International Malawi has changed some of its practices: it is now including food crop seeds, specifically groundnuts and pigeon pea, to 7,000 smallholder households as a direct consequence of this project. Alliance One International is also engaging with crop trials by the Soya Innovation Lab Malawi, funded by USAID, as a consequence of this project. Engagement with IFPRI colleagues shows this work is state-of-the-art and of interest to leading global agricultural research institutions. The main challenge to achieving impact was the size of the pilot trials were just too small to make strong claims about the impact of the intervention. Similar interventions need to be conducted at scale with sufficient funding.
Sector Agriculture, Food and Drink
Impact Types Economic,Policy & public services

 
Description Alliance One International now includes extra food crop seeds to 7,000 contracted households as a direct consequence of the pilot conducted in this project
Geographic Reach Africa 
Policy Influence Type Influenced training of practitioners or researchers
 
Description Sustainable Agricultural Intensification Research and Learning in Africa (SAIRLA)
Amount £643,000 (GBP)
Organisation Government of the UK 
Department Department for International Development (DfID)
Sector Public
Country United Kingdom
Start 06/2016 
End 06/2020
 
Title CONGENIAL Malawi 
Description Qualitative and quantitative data on 600 individuals within 400 households in Kasungu District, Malawi 
Type Of Material Database/Collection of data 
Year Produced 2014 
Provided To Others? No  
Impact Not yet 
 
Title CONGENIAL Tanzania 
Description Interviews with approx. 1420 individuals in 710 households in Mara Region, Tanzania 
Type Of Material Database/Collection of data 
Provided To Others? No  
Impact None so far 
 
Description Alliance One 
Organisation Alliance One
Department Alliance One Malawi
Country Malawi 
Sector Private 
PI Contribution Research expertise and funds
Collaborator Contribution Team of field technicians, facilities and access to smallholders
Impact This is a longstanding collaboration which predates this grant. Our collaboration on this project led to the completion of brief containing descriptive statistics for Philip Morris International and the EU office in Lilongwe. It has also led to pilot innovations in the bundle of inputs the firm provides some smallholders and engagement with USAID. This is described in detail in the Narrative Impact statement. In addition, this project facilitated collaboration with Alliance One International Regional Office in Harare, Zimbabwe.
Start Year 2013
 
Description Mzumbe University 
Organisation Mzumbe University
Country Tanzania, United Republic of 
Sector Academic/University 
PI Contribution Research expertise
Collaborator Contribution Facilities and network of enumerators
Impact Completion of 1430 interviews, delivery of policy workshop
Start Year 2013
 
Description NASFAM 
Organisation National Smallholder Farmers' Association of Malawi
Department Research and Evaluation
Country Malawi 
Sector Charity/Non Profit 
PI Contribution Research expertise and funds
Collaborator Contribution Facilities and network of enumerators
Impact Completion of 600 interviews, delivery of policy workshop
Start Year 2013
 
Description S and C Ginning 
Organisation S&C Ginning Tanzania
Country Tanzania, United Republic of 
Sector Private 
PI Contribution Research expertise and funds
Collaborator Contribution Agronomists, facilities and access to farmers
Impact Findings were presented to stakeholders in the cotton sector in 2015 at a meeting in Mwanza, Tanzania. Subsequently, two research proposals were submitted to the Tanzania Gatsby Trust for wider collaboration on (i) agricultural extension, and (ii) the use of satellite NDVI images to improve crop estimates.
Start Year 2013
 
Description Briefing for EU office in Lilongwe 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Policymakers/politicians
Results and Impact Briefing for EU office in Lilongwe based on descriptive statistics
Year(s) Of Engagement Activity 2015
 
Description Briefing for Philip Morris International based on descriptive statistics 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact A briefing for Philip Morris International based on the descriptive statistics from the Malawi dataset
Year(s) Of Engagement Activity 2015
 
Description Conference 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact ESRC-DfID Power of Partnership Conference
Year(s) Of Engagement Activity 2018
URL https://www.theimpactinitiative.net/power-of-partnership
 
Description Meeting with IFPRI staff from Washington DC 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact The PI had a meeting with IFPRI staff in June 2015 including Noora-Lisa Aberman, Katrina Kosec, Valerie Mueller and Bjorn Van Campenhout regarding findings from Phase I and possible funding applications in the US for Phase II.
Year(s) Of Engagement Activity 2015
 
Description Presentation of findings at the Economics Association of Malawi conference, 2015 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Type Of Presentation workshop facilitator
Geographic Reach International
Primary Audience Policymakers/politicians
Results and Impact I was invited to make a presentation in Lilongwe at the Economics Association of Malawi conference in 2015

Interest from publisher
Year(s) Of Engagement Activity 2015
 
Description Presentation of findings to Tanzanian stakeholders 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact The Tanzanian findings were presented to stakeholders in the cotton sector at a meeting in Mwanza
Year(s) Of Engagement Activity 2015
 
Description https://www.mycongenial.com 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact A project website has been used to raise awareness of the grant and project: https://www.mycongenial.com. The website and datasets were also used as teaching material with undergraduates at Lund University 2015-2017.
Year(s) Of Engagement Activity 2015,2016,2017,2018
URL https://www.mycongenial.com