Clients of female sex workers (CFSW) are at high risk of HIV transmission. Circumcision reduces the risk of heterosexual transmission of HIV. A total of 353 CFSW were interviewed using a combined interviewer-computer-assisted method. Amongst the participants, 28.0% were circumcised, whilst 63.8% of the uncircumcised participants were willing to take up circumcision (conditioned on being briefed about a 50% potential risk reduction effect). In a stepwise logistic regression model, frequency of patronizing female sex workers and factors related to the Health Belief Model were significantly associated with conditional willingness. Amongst uncircumcised participants, 20.9% anticipated risk compensation (i.e. would use condoms less frequently after being circumcised). Adjusting for background variables, inconsistent condom use during commercial sex and self-reported STD history in the last 6 months were significantly associated with anticipated risk compensation. It is feasible to promote circumcision among CFSW but such programs also need to promote condom use.