3ie Funded Evaluation, TW4.1018. A link to the completed study will appear here when available.
By employing a series of interventions built around demonstration plots and warehouse receipt system, the study seeks to induce the adoption of ISFM (Integrated soil fertility management) technologies in Malawi in order to increase the agricultural productivity and income of smallholder farmers.
In Malawi, agriculture accounts for 35% of Malawi’s gross domestic product and employs 90% of the rural population. However, close to 51% of the population engaged in agriculture lives below the poverty line.
Hence, the Anchor Farm Model (AFM) of Clinton Development Initiatives (CDI), employs a multi-prolonged approach to increase agricultural production, income and food security through promotion of the adoption of yield-enhancing ISFM practices, particularly for soybean. The programme aims at improving its on-going interventions namely, increased access to credit, uptake of certain ISFM technologies, such as, fertilizers and inoculation of soybean, and extension worker training.
- What is the effect of adding ICT reminders to CDI’s traditional farmer club approach (club leader receives SMS reminders)? What is the effect of an individual-managed demonstration plot by the club leader, as opposed to the group-based approach?
- What is the conditional yield as a function of soil properties, climate and characteristics of the farmer, the farmer’s household, the farmer’s club and the village? What is the remunerative potential of ISFM technologies for smallholder farmers?
Clinton Development Initiative (CDI) primarily works with farmers in groups, organising farmers into clubs of 10 to 20 members. Each club elects a leader who is provided with inputs and trained at the anchor farm in soy production, ISFM techniques, post-harvest practices and receives information on commodity markets. To date, Anchor Farm Model (AFM) has reached 21,000 smallholder farmers. In Fall 2014, CDI plans to scale-up to cover an additional 30,000 farmers in Kasungu and Dowa district.
This cluster-randomized evaluation design assigns each of 250 villages, randomly, to one of five equal-sized groups:
(i) control I
(ii) demonstration plot
(iii) warehouse receipt system
(iv) demonstration plot (with input credit)
(v) demonstration plot (with input credit) and warehouse receipt system
The warehouse receipt system will be introduced only from 2015-16 onwards, which effectively results in the following design for 2014-15:
(a) control II (100 villages)
(b) input credit (50 villages)
(c) demonstration plot (with input credit) (100 villages)
Among group (c), the following, also randomized, variation for 2014-15:
(A) SMS reminders and club-managed demonstration plot
(B) individual-managed demonstration plot
(C) club-managed demonstration plot
The demonstration plots are set-up as follows. Ten interested farmers are invited to participate in a farmers’ club. A leader, elected democratically, is provided with inputs and training at the AF. All the club members are expected to contribute labour to the cultivation of the demonstration plot and jointly apply for input loans through a microcredit provider selected by CDI. As compensation for their labour, farmer club members receive a share of the output. In the first variation (iiA) we remind the club leader – through SMS messages – of the tasks ahead on the demonstration plot. In the second variation (iiB), the research team drop the group-based approach and invite just the club leader to manage and reap rewards from the demonstration plot. Each demonstration plot will feature the same three agronomic treatments each year. The exact treatments will be determined after soil testing, on-site consultation with farmers, extension workers and CDI field staff; all treatments will feature the following ISFM recommended practices: optimal planting date and density, incorporation of crop residues post-harvest.
The interventions are:
(i) Disseminating production knowledge through the use of demonstration plots, lead farmers and farmer field days;
(ii) Improving farmers’ access to input markets, in particular credit and seed markets through CDI’s contract with seed companies and intermediary role in the credit market;
(iii) Provision of access to structured output markets through its established relationship with international soybean buyers.
In addition, CDI also attempts to contribute to increase the cost-effectiveness of demonstration plots by exploring the role of reminder SMS messages and sole-managed plots.
Theory of change
Despite the potential benefits and its widespread promotion, ISFM adoption remains low. Hence, demonstration plots have emerged as a potential cost-effective alternative. Demonstration plots are a defined area within the village where a new concept is shown to farmers. They are often located for comparison next to a plot that demonstrates farmers’ standard agricultural practices. On the other hand, the warehouse receipt system offers farmers access to credit and an opportunity to engage in inter-temporal arbitrage, thus assisting in the alleviation of credit market imbalance during the critical post-harvest period.
The study is a randomised controlled trial (RCT) in 250 villages. Each village is randomly assigned to one of the five groups: (i) control, (ii) input credit only, (iii) warehouse receipt system only, (iv) demonstration plot + input credit, and (v) demonstration plot + input credit + warehouse receipt system. The credit sub-treatment and the credit component of the demonstration plot treatment will be introduced only from 2015-16 onwards.
The study also employs qualitative methods using structured and semi-structured interviews and focus group discussions. The qualitative research is expected to provide critical information on programme constraints, if any; factors determining farmers’ participation decision; farmer dynamics in the farmer club and programme’s socio-economic effects on farmers and their families. The information will be useful in designing context-appropriate questionnaires and interpreting quantitative findings.
The study will investigate the role of soil type, credit access, land and asset ownership, human capital and labour availability, and risk and time preferences on observed impacts. Heterogeneity will also be measured on the basis of gender intra-household labor and consumption effects.
This is an ongoing evaluation. This evaluation will assess the effectiveness of different interventions implemented under the Anchor Farm Model (AFM): demonstration plots, the warehouse receipt system and in-kind credits
In addition, it analyses:
A1) Heterogeneity of effects (labour constraints, gender, time and risk preferences, soil properties)
A2) Knowledge spillovers
A3) Short versus long term effects
To contribute to the understanding of demonstration plots as tools of agronomic training and dissemination three models will be compared in terms of plot fertility and yield, in addition to the farmer-level outcomes. The evaluation will address:
(T1) What is the effect of adding ICT reminders to CDI’s traditional farmer club approach? (club leader receives SMS reminders)
(T2) What is the effect of an individual-managed demonstration plot by the club leader as opposed to the group-based approach?
Using the plot-level data, the research team will establish an Integrated Soil Fertility Management practices (ISFM) yield function for soy and maize, and will try and address: What is the conditional yield as a function of soil properties, climate and characteristics of the farmer, the farmer’s household, the farmer’s club and the village? What is the remunerative potential of ISFM technologies for smallholder farmers?