This study investigates the factors that influence participation of households in devolved system of forest management by joining community forest associations (CFA). It further employs Propensity Score Matching (PSM) to measure the impact of household’s participation in CFA on farm forestry decisions. The analysis uses cross-sectional data from a survey of Kakamega forest communities in Kenya in 2010. Generally, our findings reveal that participation in CFA by households is influenced broadly by socio-economic and institutional factors, and that participation in CFA has a positive impact on farm forestry development. Policy makers and development practitioners, therefore, need to devise, implement and sufficient fund interventions that would promote development of community forest associations with the ultimate goal of increasing forest cover in the country.