We present a first attempt to derive the cosmological evolution of the kinetic luminosity function of AGN based on the joint evolution of the flat spectrum radio and hard X-ray selected AGN luminosity functions. An empirical correlation between jet power and radio core luminosity is found, which is consistent with the theoretical assumption that, below a certain Eddington ratio, SMBH accrete in a radiatively inefficient way, while most of the energy output is in the form of kinetic energy.

We show how the redshift evolution of the kinetic power density from such a low-ṁ mode of accretion makes it a good candidate to explain the so-called “radio mode” of AGN feedback as outlined in many galaxy formation schemes.

Low-frequency ($f \leq 10^{-6}$ Hz) quasi-periodic variability observed from radioloud, jet-emitting Active Galactic Nuclei (AGNs) can provide substantial inductive support for the presence of close ($d \lesssim 0.1$ pc) supermassive binary black holes in their centers. Such periodic variability may arise, for example, due to differential Doppler boosting along helical jet paths driven by the orbital motion or jet precession. If the jet path is non-ballistic, travel time effects can lead to a strong reduction of the observable period by up to a factor of $\gamma_b^{-2}$. Here we suggest a binary model where the optical periodicity with timescale of several years is related to accretion disk interactions, radio periodicity to Newtonian jet precession and periodicities in the high energy bands to the orbital motion of the jet. We analyze the explanatory potential of such a framework and comment on the possible origin of periodicities in AO 0235+16.