Oromia Coffee and Dairy Development Project (OCDDP): Final Report, End-line Evaluation Study

Overview

This end-line evaluation covers a five-year coffee value chain project, July 2016 to April 2021, funded by Comic Relief (UK). Solidaridad Eastern and Central Africa implemented it with the Oromia Coffee Farmers Cooperatives Union through 13 cooperatives in five weredas of Oromia Region: Abaya, Gelana, Bule Hora and Kercha (West Guji Zone) and Nensebo (West Arsi Zone). The project had two components, coffee production (GAP and climate-smart agriculture) and dairy development aimed mainly at women, plus diversification and women's micro-enterprises. The evaluation assessed achievements against the MEL framework, relevance, effectiveness, efficiency, value for money and sustainability, and drew lessons for scaling up.

Key findings

• Average household coffee land grew from 1.3 ha at baseline to 2.08 ha, an increase of 0.78 ha or 60%. It fell slightly in Kercha (-6%).
• Average annual cherry production per household rose from 4.49 quintals to 10.73 quintals (139%). Average trees per hectare rose from 1,790 to 3,411, and the share of households yielding over 10 qt/ha rose from 30% to 40%.
• 85% of beneficiary farmers were trained directly or indirectly on improved coffee practices. The project distributed 1,543,450 improved seedlings and 3,334 hand tools. Farmers' main complaint was that training sessions were too short.
• Households using organic fertiliser rose to 89% from 60.8% at baseline. Use of chemical fertiliser alone fell from 4% to 0%, and pesticide use on coffee fell from 7% to 0%.
• Households selling mainly to their primary cooperative rose from 40% to 78%. However, most cooperatives cannot buy dry coffee, so farmers sell it to traders at lower prices.
• Women's say in household decisions grew most in selling dairy products (from 3.6% to 44.1% women deciding), selling chicks and eggs (2.2% to 46.4%) and buying household utilities (2.8% to 36.3%). Men became more dominant in buying agricultural inputs (5.2% to 37.1%). Men decided on cooperative membership in 58.7% of households. 60.1% of respondents said women are involved in cooperative leadership.
• Value for money: £961,762 was spent against a £968,715 plan, reaching 5,397 of 7,000 planned beneficiaries (5,145 of 5,700 coffee; 252 of 1,300 dairy). Cost per beneficiary was £178.20 against £138.39 planned. The body of the report calls this a 29% inefficiency (£39.82 per beneficiary), while the executive summary states 13% (£18.12). Projected coffee NPV is £4,122,000 (IRR 24%) and projected dairy NPV £653,698 (IRR 48%), both based on stated assumptions.

In figures

Recommendations

• Help primary cooperatives produce improved seedlings, at least for their members, through intensive technical training on nursery management and financial support.
• Encourage and support model and lead farmers who have land to set up nurseries that produce seedlings for their own use and for sale to other farmers.
• Continue and expand training on modern coffee practices so it reaches households not trained directly, with longer and more frequent sessions, demonstration plots and farmer-to-farmer learning.
• To sustain GAP and climate-smart production, keep working on soil fertility, improved seeds and seedlings, improved tools and natural landscape and agro-ecosystem practices.

Lessons learned

• Ending the improved seedling supply is the main risk to sustainability. Farmers expect distribution problems once funding for nursery workers' wages stops, and cooperatives and wereda agriculture offices lack the money and skills to run nurseries.
• Higher yields came with higher production costs. About 65% of farmers spend over birr 2,500 per quintal, roughly double the baseline, so the project's benefit shows in yield rather than lower costs.
• Delays, especially in distributing heifers (received only a few months before the evaluation) and seedlings, reduced value for money and made performance hard to measure.
• Women deciding on a sale does not necessarily mean they benefit directly or equally from the income.
• Seed supplied by the project covered less than 10% of the seed beneficiary farmers planted, and very few women benefited from heifers. Scale of input supply was a limiting factor.

Methodology

End-line household survey of 203 beneficiary households (102 Abaya, 36 Bule Hora, 32 Gelana, 25 Kercha, 8 Nensebo) using multi-stage cluster and purposive sampling with probability-proportional-to-size allocation across the 13 cooperatives. It was complemented by 12 FGDs with 120 participants, KIIs at wereda, regional and federal level, document review and observation, with descriptive SPSS analysis compared against the baseline.

Limitations

• Coffee planted after the intervention had not yet been harvested, so impact on production, yield, cost and income could be assessed only for existing trees.
• Gelana wereda was not covered by the baseline, so baseline comparisons exclude it.
• The sample size was set by logistics, limited time and assumed low heterogeneity among respondents rather than full statistical sampling.
• Data limits restricted the value-for-money analysis. The dairy component was only 1–2 years old, and data on heifer prices, income and expenses and loan repayment were lacking, so financial projections depend on assumptions.
• The report advises caution in interpreting the milk production comparison because of data limitations.