We present evidence that the failure to account for proper motion is responsible for most of the regional biases seen in the astrometry of asteroids produced by the Pan-STARRS sky survey, as well as the 0.02 arcsec northward bias seen in our own Apophis astrometry.