Exposure effects are policy relevant only under strong assumptions about the interference structure

Savje (2023) recommends misspecified exposure effects as a way to avoid strong assumptions about interference when analyzing the results of an experiment. In this discussion, we highlight a key limitation of Savje's recommendation. Exposure effects are not generally useful for evaluating social policies without the strong assumptions that Savje seeks to avoid. Our discussion is organized as follows. Section 2 summarizes our position, section 3 provides a concrete example, and section 4 concludes. Proof of claims are in an appendix.