Rural households in northern Ethiopia had no access to frankincense production and trading. However, following the recent devolution of forest use-rights, rural communities in norther Ethiopia are getting access through organizing frankincense cooperatives. This study examines the effects of membership in the rural frankincense firms on income and poverty by applying instrumental variable, propensity score matching and Rosenbaum bounds methods on survey data from five rural villages in frankincense woodland areas of Northern Ethiopia. Results indicate that both membership and amount of households' investment in shares in the frankincense cooperative firms have statistically significant positive welfare impacts. The average treatment effect on the treated from matching methods for the welfare indicator variables (log-transformed household per adult equivalent income, poverty head count, poverty gap, and poverty severity) are robust against hidden bias arising from unobserved confounding variables that simultaneously affect assignment into frankincense forest cooperative firm membership and each of the welfare indicator variables.