Coral reefs in shallow-water environments (<30 m) are in decline due to local and global anthropogenic stresses. This has led to renewed interest in the ‘deep reef refugia’ hypothesis (DRRH), which stipulates that deep reef areas (1) are protected or dampened from disturbances that affect shallow reef areas and (2) can provide a viable reproductive source for shallow reef areas following disturbance. Using the Caribbean as an example, the assumptions of this hypothesis were explored by reviewing the literature for scleractinian corals—the reef framework builders on tropical reefs. Although there is evidence to support that deep reefs (>30 m) can escape the direct effects of storm-induced waves and thermal bleaching events, deep reefs are certainly not immune to disturbance. Additionally, the potential of deep reefs to provide propagules for shallow reef areas seems limited to ‘depth-generalist’ coral species, which constitute only ~25% of the total coral biodiversity. Larval connectivity between shallow and deep populations of these species may be further limited due to specific life history traits (e.g., brooding reproductive strategy and vertical symbiont acquisition mode). This review exposes how little is known about deep reefs and coral reproduction over depth. Hence, a series of urgent research priorities are proposed to determine the extent to which deep reefs may act as a refuge in the face of global reef decline.