This paper tries to justify the observation of different return patterns in the upstream and downstream sectors of US beef production. It builds a dynamic rational expectation model separating the cow-calf and feeding sector with the former sector being the residual claimer. The model shows that the cow-calf operation has positively autocorrelated return pattern while the feeding operation return only reflects random shock. Empirical study shows that 85.4% of the Ricardian rent is passed through to the upstream sector, and the downstream sector can only claim the unexpected return resulting from random shocks.