The delegation of decision authority is a routine part of engineering systems design. Examples include hiring an outside subcontractor and assigning specific design tasks to a subordinate engineer. In all cases, the delegator is responsible for describing the task to the delegate in a way that clarifies expected outcomes. Ideally, the delegate executes the task in a manner consistent with the preferences of the delegator. For systems design tasks, this means the delegate specifies an artifact with engineering attributes desired by the delegator. Whether such expectations are conveyed successfully can have a major impact on the success of a project. However, despite the central nature of such communications in engineering systems design and the potential consequences miscommunication, there has been little empirical investigation of decision delegation in design. This experimental study examines how the framing of decision task expectations impact task outcomes. Subjects participate in an engineering-themed computer game in which they make design choices. Subjects are assigned to conditions randomly, with some given instructions framed as traditional design requirements and others given instructions in a value-driven design framing (e.g., to maximize a figure of merit). A total of 472 engineering students participated in the study. Results indicate that task outcome expectation framing can impact task performance, with design requirement task framings exhibiting advantages over a value-driven framing. This suggests theoretical arguments favoring value-driven framings may not capture adequately the effect of human behavior.