This paper provides an account of a South African tax-benefit microsimulation
model—SAMOD—which has been developed for use by government over the past ten years.
The two datasets that underpin the current version of SAMOD are introduced, and the model’s
tax and benefit policies are described with discussion of the various data challenges and
assumptions that had to be made in order to simulate them, with particular emphasis on the
value-added tax policy. Simulations using the two different underpinning datasets are compared
with reported administrative data. The paper concludes by highlighting three future
developments for SAMOD.