The time-resolved anisotropy produced in polarized fluorescence photobleaching recovery experiments has been successfully used to measure rotational correlation times in a variety of biological systems, however the magnitudes of the reported initial anisotropies have been much lower than the theoretically predicted maximum values. This small time-zero anisotropy has been attributed to fluorophore motion, wobble and rotation, during the photobleaching pulse. We demonstrate that inclusion of the possibility of saturation of the fluorophore's transition from its ground state to its excited state during the photobleaching pulse leads to the prediction of reduced time-zero anisotropy. This eliminates the need to rely solely on the assumption of fluorophore motion during the photobleaching pulse as the cause of the reduced initial anisotropy. We present theoretical and experimental results which show that the initial anisotropy decreases as both the bleach pulse intensity is increased and bleach pulse duration is decreased so as to keep the total integrated bleach pulse constant. We also show theoretical and experimental results demonstrating that at high excitation intensity the effects of saturation cause the steady state fluorescence polarization to decrease. We estimate that saturation may occur using common photobleaching conditions.