<p>This paper uses a sample of lone mothers (and former lone mothers who are now repartnered) drawn from the 1997 Family Resources Survey to analyse the potential effects of reforming the UK system of child support. The main deficiency of the data is that non-resident fathers cannot be </p><p>matched to the mothers in the data, and this is overcome by exploiting information from another </p><p>dataset which gives the joint distribution of the characteristics of separated parents. The effects of reforming the child support system are simulated for the amount of maintenance liabilities, the amount paid and the net incomes of households containing mothers-with-care and of households containing non-resident fathers. The likely effects of the reform are simulated at various levels of compliance. The analysis highlights the need for further research into the incentive effects of child support on individual behaviour.</p>