The analysis of pulsar time-of-arrival data is intimately bound up with planetary ephemerides. Highly accurate ephemerides are required for Earth and Moon and, to a lesser degree, for the other planets, in order to make full use of the timing data for millisecond-class pulsars. These data, in turn, present an opportunity for improving planetary ephemerides in a variety of ways. Fitting the Earth and Moon orbital parameters to the timing data is the obvious first step, though it is less valuable in the short term for many applications than using the current accumulation of spacecraft-tracking and lunar laser ranging data. By themselves, the pulsar timing data convey no information on the orientation of Earth's orbit, since each pulsar's position on the sky must be determined from those same data. However, independent pulsar position measurements by VLBI, in combination with the timing-derived positions, can serve to fix the orientation of Earth's orbit with respect to the radio reference frame and thereby link the planetary and radio frames. In the long run, the acquisition of timing data over increasing time spans and with improving precision should prove to be an important factor in determining the shape, as well as the orientation, of Earth's orbit. In addition, pulsar timing over a sufficiently long span can directly measure a planet mass through the reaction of the rest of the solar system. The effect must be observed for a major fraction of the orbital period of the planet in question so that the signature can be separated from that of the ordinary spin-down of each pulsar. Finally, pulsar timing can be used to probe gravitational physics, a field with far-reaching consequences and a basic part of the framework for constructing the ephemerides.