A significant discrepancy of up to 0.6 Myr exists between radio-isotopically calibrated and astronomically tuned time scales of the late Eocene-Oligocene. We explore the possible causes of this discrepancy through the acquisition of “high-precision” 206Pb/238U dating of zircons from 11 volcanic ash beds from the Umbria-Marche sedimentary succession, which hosts the Global Stratotype Section and Point for the base of the Oligocene. Our results indicate that the four 40Ar/39Ar dates from the Umbria-Marche succession, which underpin the late Eocene-Oligocene portion of the Paleogene geomagnetic polarity time scale in the 2012 edition of the Geological Time Scale, are anomalously old by up to 0.5 Myr. Conversely, when integrated with the established magnetic polarity record of the Umbria-Marche succession, 206Pb/238U (zircon) data from this study result in Oligocene magnetic reversal ages that are generally equivalent to those obtained through the tuning of Ocean Drilling Program (ODP) Site 1218 (equatorial Pacific). Furthermore, our results indicate that the late Eocene tuning of ODP Site 1218, and International Ocean Discovery Program (IODP) Sites U1333–1334 (equatorial Pacific), to the 405 kyr eccentricity signal is accurate, at least back to 36 Ma. Propagating the full uncertainty of our radio-isotopic data set and, where appropriate, taking into account locally derived astronomical time scales, we arrive at an age of 34.09 ± 0.08 Ma for the Eocene-Oligocene boundary and 28.11 ± 0.17 Ma for the base of the Chattian.