The traditional approach to optimal economic operation of industrial processes has been the use of a hierarchically structured control system. This hierarchical structure comprises a steady-state economic optimization layer that computes optimal operating set points of the system and a dynamic control layer using model predictive control to track these set points. One limitation of this structure is that the economics of the process is not being dynamically optimized. To overcome this limitation, economic model predictive control, which optimizes the economics directly in the dynamic layer in order to give a superior economic performance, has been proposed to replace the hierarchical approach. This paper investigates this notion of superior economic performance and proposes a redesign of the traditional hierarchical structure that gives an economic performance that is as good as that of economic model predictive control. Specifically, a new approach to selecting the objective function in the model predictive control layer is presented.