Regional science and economics studies increasingly use Defense Meteorological Satellite Program (DMSP) night‐time lights data to measure spatial inequality. These DMSP data are a poor proxy in this context because of their spatially mean‐reverting errors, which yield significantly lower inequality estimates than what sub‐national GDP data show. Inequality estimates from DMSP are also lower than what newer, research‐focused and more accurate, satellites show. We demonstrate this bias using county‐level data from China and the United States. The errors in the DMSP data distort estimates of both the level of, and trend in, spatial inequality.