Extinction risk assessments of marine invertebrate species remain scarce, which hinders effective management of marine biodiversity in the face of anthropogenic impacts. In order to close this information gap, we developed a metric of relative extinction risk that combines paleontological data, in the form of extinction rates calculated from the fossil record, with two known correlates of risk in the modern day: geographic range size and realized thermal niche. We test the performance of this metric - PERIL (Paleontological Extinction Risk In Lineages) - using survivorship analyses of Pliocene bivalve faunas from California and New Zealand, and then use it to identify present-day hotspots of extinction vulnerability for extant shallow-marine Bivalvia. Areas of the ocean where concentrations of bivalve species with higher PERIL scores overlap with high levels of climatic or anthropogenic stressors should be considered of most immediate concern for conservation and for management.