The anionic complex [NBu4][Pd(DMSO)Cl-3], as a "ligandless" system, was shown to be an active catalyst in the Mizoroki-Heck coupling of aryl chlorides in the absence of strongly sigma-donating ligands. To investigate the experimentally observed influence of halides and the amount of water on the catalytic activity, we employed a combination of experiments and theoretical calculations. The presence of water was shown to be critical for the formation of the active palladium(0) species by oxidation of in situ generated tributylamine. Oxidative addition to an anionic palladium(0) species was found to be the rate determining step of the catalytic cycle. For the ensuing steps, both neutral and anionic pathways were considered. It was shown that, in the absence of strongly sigma-donating neutral ligands, chloride ions stabilize the catalytic intermediates. Therefore, an anionic pathway is preferred, which explains the need for tetrabutylammonium chloride as an additive. The study of the influence of bromide ions on the catalytic activity revealed that the strongly exergonic displacement of the neutral substrates by bromide ions lowers the catalytic activity.