Using an original dataset from the Vietnamese catfish sector, I analyze the impact of vertical coordination options, namely contract farming and vertical integration on farm performance. Farm performance is measured in terms of yield and revenue per hectare. The effects of vertical coordination are estimated using a maximum simulated likelihood estimator and a two-stage least square regression with instrumental variables to account for exogenous farm and household characteristics and the sample selection bias. The results show that vertically integrated farms have substantially higher yields and revenue per hectare than non-integrated farms. The levels of gains, which can be attributed to integration, are large and consistent under different estimation procedures. There is no difference between non-integrated and contract farms in terms of farm performance.