A growing area of research into rebound effects from increased energy efficiency involves the application of demand-driven input-output models to consider indirect rebound associated with re-spending decisions by households with reduced energy spending requirements. However, there is often a lack of clarity in applied studies as to how indirect rebound effects involving energy use embodied in supply chains have been calculated. We focus on a theoretical debate regarding the treatment of reduced energy requirements by energy producers and their up-stream supply chains as energy spending decreases with improved efficiency. We show that both the magnitude and direction of embodied energy rebound effects are highly sensitive to what is assumed to be part of potential energy savings, which we argue should be considered in terms of energy savings anticipated by decision makers. We also extend the focus of most studies of rebound effects via embodied energy impacts to consider impacts on energy use and CO2 emissions embedded in international supply chains and consider how these are reflected in alternative definitions of rebound.