In this work we present the results of an ab initio study aimed at the prediction of the chiroptical spectra of dissymmetric bidentate Co(III) complexes at modest levels of theory. The restricted Hartree-Fock/random-phase approximation is employed to obtain excitation energies and rotatory strengths for tris(ethylenediamine)cobalt(III), tris(1,3-diaminopropane)cobalt(III), and tris(oxalato)cobaltate(III). Numerically, the calculated optical properties compare quite favorably to experiment. A modified RHF/ligand field CI procedure is employed to account for some of the error which is encountered at the RPA level. The features of the SCF wave functions and the structure of the RPA transition density matrices are found to be strikingly different from what would be predicted by classical ligand field theory. Our results are analyzed in terms of similarities to and differences from related efforts.