The caesarean section (CS) rate among white women aged 20-35 years and having their first baby was examined, comparing the private fee-for-service medical aid sector with Johannesburg Hospital. The chance of having a CS in the private sector was 50% greater than in the public sector (28.7% v. 19.5%). Twice as many CSs were done on weekdays as over weekends, and it is argued that only a quarter of these are accounted for by elective procedures (planned before labour begins). We also found that in the private sector the daily frequency of non-caesarean deliveries was 56% higher during the week than on Saturdays or Sundays. Considering non-caesarean deliveries separately, it is inferred that the rate of induction of such deliveries was 28.7% in the private sector compared with 2.8% in Johannesburg Hospital. The evidence strongly confirms the international experience that the CS rate in a given population is not objectively determined by medical factors and is strongly influenced by individual doctors' decisions. Moreover, fee-for-service reimbursement of doctors leads to increased intervention in delivery, in the form of more frequent induction of labour and more CSs.