OVERVIEW
How coffee systems shape farmer outcomes
Coffee supports rural livelihoods across East Africa, but the way value moves through the sector is uneven. Project Roast compared Kenya and Uganda to understand how institutional design affects farmer outcomes, and what kinds of commercial changes could improve incomes.
325
farmers interviewed
23
cooperatives covered
2
coffee systems compared
While Kenyan coffee is globally recognised and typically achieves higher prices, farmers in Uganda often retain a larger share of the final export value. This raises a fundamental question: what determines how value is distributed across the chain. The findings show that market structure, payment systems, and information flows play a larger role than production practices alone.
What matters most is simple:
-
Training does not increase value capture.
-
Price and quality information does not reach farmers in a usable form.
-
Sales routes drive income.
-
Payment timing affects outcomes.
CONTEXT
Two coffee systems, different trade offs
Coffee sectors in Kenya and Uganda operate under very different systems. Kenya’s system is more structured, with cooperatives, processing facilities, and an auction based sales mechanism. Uganda operates a more liberalised system, with multiple buyers and faster transactions at farm level.
These differences shape how farmers access markets, how quickly they are paid, and how clearly they understand the relationship between quality and price.
The diagrams below illustrate how coffee moves through each system, from farmer to export.
Kenya
Uganda
Kenya’s coffee system is more centralised. Around 90% of coffee still moves through the Nairobi Coffee Exchange, and many smallholders sell through cooperatives, with multiple processing and marketing layers before reaching buyers.
This supports coordination and standardisation, but also creates distance from the final buyer. Payments are often delayed and feedback on quality is limited, weakening the link between farmer effort and income.
.
Uganda’s system is more liberalised. Farmers can sell to a wider range of buyers and are typically paid quickly, often at the point of sale.
This improves liquidity and provides clearer commercial signals, but the system is more fragmented, with weaker coordination and less consistent traceability. Farmers often operate with greater flexibility in how and when they sell, but with fewer structured support systems.
The study draws on fieldwork, surveys, and value chain analysis to assess how these systems function in practice and their implications for farmer income and value capture.
FINDINGS FROM THE STUDY
What the evidence shows
1. Training does not increase value capture
Across both countries, over 90 percent of farmers reported receiving training and applying good agronomic and quality practices.
Adoption rates were high even among farmers without recent training, indicating that core knowledge is already widespread.
Additional training tends to repeat existing knowledge and has limited impact on income. At best, it can lead to small yield improvements, but it does not change the price farmers receive or shift them into higher value markets.
This reflects a broader constraint. Farmer income is not determined by production practices alone, but by how coffee moves through the system after it leaves the farm. Pricing, processing, aggregation, and sales routes play a central role in determining outcomes. In Kenya, for example, coffee is pooled at the cooperative level and often again at later stages, which breaks the link between individual farmer quality and final prices.
This has important implications for current interventions. A large share of donor funding continues to focus on training and compliance, but the evidence suggests that these investments alone are unlikely to improve incomes unless structural constraints in the value chain are addressed.
2. Farmers in Kenya lack visibility on market needs
Farmers in Kenya often do not see how quality, grading, and sales routes translate into the prices they receive. While pricing structures, grading systems, and auction results exist within the system, this information does not flow back to farmers in a clear or timely way.
This is largely due to how the system is structured. Coffee passes through multiple layers, including cooperatives, processing stages, brokers, and the exchange, which creates distance between farmers and final buyers. As coffee is aggregated and pooled, feedback becomes generalised, delayed, or disconnected from individual farmer output.
As a result, farmers have limited visibility on how their coffee is assessed, how prices are determined, or how much value is created beyond the farm gate. Even where grading systems are defined, they are not communicated in a way that farmers can act on or link directly to their own production decisions.
This limits the ability of farmers to improve outcomes. Without clear and timely feedback, there is no reliable basis for adjusting practices, targeting higher value markets, or choosing between different sales routes. Improvements in quality may not be recognised or rewarded, weakening incentives over time.
The issue is therefore not the absence of information, but the absence of usable information. What matters is not whether data exists in the system, but whether it reaches farmers in a form that informs decisions.
In Uganda, faster transactions and more direct sales create clearer and more immediate price signals at farm level.
3. Sales routes drive income
How coffee is sold plays a central role in determining farmer income. In Kenya, cooperatives that sell through a mix of channels, including both the Nairobi Coffee Exchange and direct buyers, tend to achieve higher returns than those relying only on the exchange. Evidence from the study suggests that these mixed models can deliver around 20 percent higher value for farmers.
This difference is not primarily driven by large changes in price per unit, but by how the system enables value to be captured. Direct or hybrid routes create closer relationships with buyers, allow for better alignment on quality requirements, and increase the likelihood that higher quality coffee is recognised and rewarded. They also introduce a degree of flexibility that is missing in more rigid systems. Cooperatives and farmers are better able to choose when and where to sell, respond to market signals, and avoid being locked into a single pricing mechanism. This strengthens their position within the value chain.
In contrast, exchange only routes rely on centralised sales through pooled systems. While this supports coordination and scale, it limits flexibility and reduces the ability of farmers and cooperatives to influence outcomes. Coffee is aggregated, quality is standardised, and the link between individual performance and price is weakened.
The implication is that income differences are driven less by production practices alone and more by how farmers are connected to markets. Access to multiple sales routes, and the ability to engage more directly with buyers, is therefore a critical driver of value capture.
4. Payment timing affects outcomes
Payment timing is a structural factor shaping farmer outcomes. In Kenya, farmers selling through cooperative and auction based routes are paid only after coffee has been processed, sold, and proceeds distributed. This can take several months, in some cases up to seven months, creating a long delay between harvest and income.
This delay creates a cash flow gap at the point when farmers need money most. During harvest, farmers must pay for labour and basic production costs, but income arrives much later, limiting their ability to manage farms effectively and invest within the same season.
The effect is not only on liquidity, but on production decisions. Increasing output often requires hiring additional labour during harvest. However, with payment delayed, farmers are less able to take on these costs, even where higher production could increase total income.
The delay also increases exposure to price uncertainty. Farmers have limited visibility on the final price realised between delivery and sale, weakening the link between effort, quality, and income.
In contrast, farmers in Uganda, are typically paid at or near the point of sale, improving liquidity and supporting more immediate decision making during harvest.
WHAT DRIVES THESE OUTCOMES
The constraints sit around the farmer, not only on the farm
These findings point to a consistent pattern across both countries. Outcomes are not primarily constrained by knowledge at farm level, but by how the system functions around the farmer. Limited market access, delayed payments, weak feedback loops, and low visibility on pricing reduce the ability of farmers to act on what they already know.
This means that improving practices alone is not sufficient. Farmers may adopt better agronomic or post-harvest techniques, but if these improvements are not recognised, rewarded, or linked to clear market signals, they do not translate into higher income.
The issue is therefore not only about production, but about how value is transmitted through the system. Where feedback is delayed or unclear, payments are uncertain, and sales options are limited, the link between effort, quality, and income weakens.
This helps explain why outcomes differ across systems. In more layered structures, additional coordination can come at the cost of transparency and responsiveness. In more direct systems, faster transactions can strengthen incentives, even where formal structures are less developed.
This has important implications for how the sector is currently supported. A large share of donor and development funding continues to focus on training and compliance. However, where the underlying constraints relate to market access, pricing visibility, and payment timing, these interventions have limited impact on income.
Addressing these outcomes therefore requires shifting focus from improving practices alone to improving how markets function around the farmer.
WHAT CAN BE DONE NOW
Three practical responses
Develop a simple, dynamic information channel embedded within existing farmer communication systems to ensure reach and accessibility.
This channel should answer common farmer facing questions, for example: how does quality affect the price I receive, how do different sales routes impact what I earn, and what do I need to do to meet traceability and EUDR requirements.
1. Build a farmer information channel
Connect farmers and cooperatives more directly to buyers through structured buying models. This can include direct partnerships and more transparent pricing agreements.
Reducing reliance on single sales channels helps improve price realisation, expand market access, and create clearer feedback loops that strengthen the link between quality and income.
2. Connect farmers and cooperatives directly to buyers
3. Create a forward finance instrument
Introduce a forward finance instrument at harvest to improve farmer liquidity. Provide cash to farmers at the moment it matters most.
By mobilising funds earlier, this reduces reliance on high cost credit, lowers exposure to price volatility, and enables better selling decisions and planning.
DEFORESTATION AND TRACEABILITY RISKS
Coffee driven deforestation is limited but concentrated in ecologically sensitive areas
Project Roast also examined deforestation risks linked to coffee, which are increasingly important given the current focus on traceability and EUDR requirements. While overall expansion may appear limited, there is a risk that renewed growth shifts into high altitude, ecologically sensitive areas.
The maps below illustrate where recent tree loss overlaps with protected areas and high value ecosystems across coffee producing regions. While not all deforestation can be attributed to coffee, these patterns highlight where ecological risks are most concentrated and where future expansion could have disproportionate impact.
Kenya shows lower land pressure than Uganda because the coffee sector is shrinking rather than expanding. Limited expansion may therefore reflect weak sector economics rather than strong environmental protection.
In Kenya, much of the recorded tree loss comes from urbanisation or other land use change, rather than farmers expanding coffee into forests.
In Uganda, the picture is different. Even small areas of expansion can have a large ecological impact, as coffee is often grown in high altitude, high biodiversity zones near forest edges. As temperatures rise, farmers are also moving upslope, increasing pressure in these sensitive areas.
Current risk maps do not fully capture this, as they do not track altitude shifts where warming temperatures are pushing farmers into forest edge zones.
Digital traceability systems therefore need to reflect this by:
-
Flagging ecologically sensitive zones such as high altitude areas and protected area edges
-
Identifying where expansion pressure may reappear if coffee profitability or suitability improves
-
Defining risk based on ecological value and actual expansion patterns, not just simple tree loss
