Monarch butterflies (Danaus plexippus) are parasitized by the protozoan Ophryocystis elektroscirrha throughout their geographical range. Monarchs inhabiting seasonally fluctuating environments migrate annually, and parasite prevalence is lower among migratory relative to non-migratory populations. One explanation for this pattern is that long-distance migration weeds out infected animals, thus reducing parasite prevalence and transmission between generations. In this study we experimentally infected monarchs from a migratory population and recorded their long-distance flight performance using a tethered flight mill. Results showed that parasitized butterflies exhibited shorter flight distances, slower flight speeds, and lost proportionately more body mass per km flown. Differences between parasitized and unparasitized monarchs were generally not explained by individual variation in wing size, shape, or wing loading, suggesting that poorer flight performance among parasitized hosts was not directly caused by morphological constraints. Effects of parasite infection on powered flight support a role for long-distance migration in dramatically reducing parasite prevalence in this and other host–pathogen systems.