This paper uses both parametric and non-parametric approaches to estimate technical, allocative, and economic efficiencies for the agriculture production in sixty provinces of Vietnam in the period 1990-2005. Under different technology specifications, both approaches show that the average technical, allocative, and economic efficiency estimates were not high, and there would be a large room for the studied provinces to improve their agricultural production efficiency. To examine consistency of the estimates from two approaches under different specifications of returns to scale, we use Spearman rank test, and the results indicate that parametric and non-parametric approaches provide different estimates.