The all-pass sections found in HRTFs were analyzed and compared to thresholds of audibility found in a parallel study. Furthermore, the audibility of all-pass components in HRTFs was tested in a listening experiment. The investigation showed that HRTFs can be represented by their minimum phase components as long as they have the correct interaural time difference. The thresholds were furthermore used to investigate the audibility of the all-pass components in headphone transfer functions. Specifically, the consequences of equalizing these transfer functions were considered. The results suggest that equalization can be done by means of minimum-phase filters. Discarding all-pass components in general, leads to substantial savings in computation when implementing binaural synthesis. Supported by STVF/FTP.