Providing a Bundle of Services to Improve Small-scale Farmers’ Adoption of Coffee Rejuvenation Practices in Rwanda

Adopting new agricultural technologies can increase farmers’ profits, but high upfront costs may make adoption burdensome, particularly for women farmers. Researchers are partnering with a large coffee buyer to conduct a randomized evaluation to test the impact of offering subsidized labor, access to loans, and digital delivery of personalized information on men and women farmers’ adoption of coffee tree rejuvenation practices in Rwanda.

Policy Issue

Farmers can increase their profits by adopting new agricultural technologies or upgrading their operations, but they might not be able to do so if such upgrades require high upfront costs. Rejuvenating coffee trees by cutting them down to the stump to promote new growth or uprooting old trees entirely and replanting new ones are proven methods to increase yields. However, both practices are labor intensive and result in lower yields for two to three years before the new trees start bearing fruit, which means that farmers face lower incomes in the short-term.These drawbacks may be more burdensome for women farmers, who tend to own less land and trees than men farmers and who are less likely to adopt productivity-boosting agricultural practices in general. Are men and women farmers more likely to invest in coffee rejuvenation practices if they are offered subsidized labor and loans? Does providing targeted, digitally-provided information increase farmers’ awareness and adherence to the program?

Context of the Evaluation

Coffee is an important part of the Rwandan economy. Over 400,000 small-scale farmers grow coffee, and it is the country’s second largest export crop, making up about a quarter of all earnings from agricultural exports. However, of major coffee-exporting countries, only Ethiopia has lower yields per hectare than Rwanda, and coffee trees in Rwanda tend to be very old, with older trees producing lower yields than young trees. Although farmers’ knowledge of coffee rejuvenation practices in the study area is high, only about one third of farmers reported having stumped or uprooted some of their coffee trees in the most recent agricultural season and fewer than 10 percent reported having replanted any trees.

Over 40 percent of Rwandan coffee farmers are women, which is a higher proportion than in most coffee-producing countries. However, women tend to have lower yields than men throughout the country. On average, women farmers in the study area own 15 percent less land than men and plant up to 25 percent fewer trees, which are often older than the trees farmed by men. Contributing to this disadvantage is the gap that women farmers experience in accessing resources and information. Partially due to their higher burden of household tasks, women are less likely to attend in-person training sessions hosted by a major coffee buyer and to interact with coffee washing station managers.

On average, farmers in the study owned two acres of land and planted coffee on about one quarter of their land. Smallholders in Rwanda typically sell their product to buyers that aggregate product across farms for processing and exporting. Farms are assigned to government-defined geographic zones for buyer-operated coffee washing stations, although this is loosely enforced, and buyers often compete for farmers’ product. The coffee buyer that researchers partnered with sources coffee from over 30,000 smallholder farmers across 30 coffee washing station catchment areas.

Details of the Intervention

In partnership with one of Rwanda’s largest coffee buyers, researchers tested the impact of offering subsidized labor, loans, and targeted digital communication on small-scale farmers’ adoption of coffee rejuvenation practices. Although farmers are not contractually obligated to sell to this coffee buyer, over 90 percent of farmers in the study reported selling all or most of their coffee to them and many had attended their past trainings. Researchers randomly assigned 5,237 households across treatment arms and 767 households to a comparison group, for a total of 6,000 households assigned to one of the following groups:

1. Bundled labor subsidy and loan: In addition to the standard set of services, the coffee buyer offered farmers in this group subsidized workers to carry out the tasks associated with coffee rejuvenation, like stumping and uprooting trees. The coffee buyer paid for workers to complete these tasks on up to twenty percent of a farmers’ trees per year of the study. The coffee buyer also offered farmers a loan intended to replace the temporary loss in income that arises from rejuvenation practices. The loan was tied to the amount of trees included in rejuvenation and was disbursed to farmers during harvest season, after the buyer verified how many trees had been stumped or uprooted.
2. Bundled labor subsidy and loan, plus targeted digital communication: Smallholders in this group received access to the buyer’s standard set of services, the subsidized labor and loans, and also received personalized monthly phone calls and weekly SMS messages from an agronomist assigned by the coffee buyer. The calls initially included reminders to farmers that they were eligible for the loans and labor subsidy and provided information on how to access them. After loans and subsidies were provided, smallholders received weekly SMS messages with reminders on how to care for their new seedlings.
3. Comparison group: Smallholders in this group received access to the coffee buyer’s standard set of services, which include training, fertilizer, and the option to reach out to a call center for information about these services. Some farmers can also apply for loans, but these are not tied to adopting rejuvenation practices.

Researchers will work with the coffee buyer to track farmers’ engagement with the digital communication and their take-up of the labor subsidy and loans. Researchers will also measure how many newly planted seedlings survive and the share of stumped trees that begin bearing fruit. Administrative data from the coffee buyer will allow the researchers to track how much coffee farmers deliver to them during and following the intervention to measure how farmers’ yields and income from coffee change over time.

Results and Policy Lessons:
Results forthcoming; study ongoing.