I don't think this point is settled, especially regarding the details of band bending, but migs only seem to occur in the case of fully evaporated interfaces. In the case of spin-coated organic layers, the interface between metal and organic isn't atomically clean, and so the metal wavefunction does not physically overlap with the electronic states in the organic semiconductor. The phenomenon of band bending in conjugated polymers has been reported (see references), and it results from charge transfer between the HOMO or LUMO density of states in the polymer and an effectively Ohmic electrode. 

http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.106.216402

http://advances.sciencemag.org/content/1/10/e1501127.full

 thank you very much for the clarification.

I'll try to look into it.

Dont be surprise that Fermi-pinning might not be valid between organic metal- organic semiconductor interface

Article Investigation into the Gaussian density of states widths of ...