Is 0716+714 a superluminal blazar?

We present an analysis of new and old high frequency VLBI data collected during the last 10 years at 5--22 GHz. For the jet components in the mas-VLBI jet, two component identifications are possible. One of them with quasi-stationary components oscillating about their mean positions. Another identification scheme, which formally gives the better expansion fit, yields motion with $\sim 9$ $c$ for $H_0=65$ km s$^{-1}$ Mpc$^{-1}$ and $q_0=0.5$. This model would be in better agreement with the observed rapid IDV and the expected high Lorentz-factor, deduced from IDV.