PITMOD – a microsimulation model of Personal Income Tax

PITMOD is a microsimulation model of Personal Income Tax (PIT) undertaken in collaboration with the South African Revenue Service (SARS). The initiative is funded through the Southern Africa – Towards Inclusive Economic Development (SA-TIED) programme.

This programme is “intended to support policymaking in the Southern Africa region by working closely with researchers to close knowledge gaps crucial to the achievement of inclusive growth and economic transformation.” Click here for more details.

SASPRI is contributing to this programme in a number of ways including:

  • Introducing the tax-benefit microsimulation model SAMOD to the South African National Treasury (NT) and the South African Revenue Service (SARS).
  • Collaborating with SARS, the NT and UNU-WIDER in researching the impact of recent VAT changes on poverty and inequality.
  • Providing an online training course on SAMOD for SARS and the NT.
  • Investigating the impact of COVID-19 on poverty and inequality and analysing the measures put in place to ameliorate its effects.
  • Collaborating with SARS in the development of an administrative data based personal income tax model (PITMOD) which uses the EUROMOD platform. 

PITMOD uses anonymized individual level administrative records supplied by SARS from tax certificates and tax returns, including assessed tax returns, supplemented with third party information on medical aid scheme contributions. The administrative data contains the fine-grained details about incomes that are necessary to simulate South Africa’s PIT rules in detail. The administrative datasets are used to construct input datasets for PITMOD, where the detailed PIT rules are modelled. These rules are applied to the individuals in the datasets to calculate each individual’s tax liability. PITMOD has a separate set of rules and a separate underpinning dataset for each tax year. Currently, PITMOD simulates the rules for the 2018, 2019, and 2020 tax years, with later years in the process of being incorporated. PITMOD’s full datasets contain one record per individual and comprise approximately 15 million individuals for each tax year (10% and 1% samples have been created which enable PITMOD to run more quickly and still generate nationally representative results). The model’s output dataset(s) can then be analysed with the PITMOD Summary Statistics tool to produce summary statistics in table and chart formats.

PITMOD has been compared to SAMOD (the survey-based microsimulation model) and also used to explore options to deepen and broaden the PIT base.

SARS is the primary user of PITMOD and training for SARS and National Treasury staff took place in February 2023. There is also an online training course for users at the National Treasury Secure Data Facility (SDF). SARS plans to share the PITMOD model with various units including the legal division and regional offices, and there is a future plan to expand the use of PITMOD by housing a version of the model in the National Treasury SDF.