Data

In collaboration with the National Treasury (NT), the South African Revenue Service (SARS) and the United Nations University World Institute for Development Economics Research (UNU-WIDER), anonymized administrative tax data is made available for research purposes. South Africa is one of only a few countries globally to grant access to this type of data for research purposes. This is a unique opportunity as access to national administrative data for researchers, outside of senior officials, is rarely granted. The administrative data plays a central part in the SA-TIED programme and aims to feed into policy formulation. 

Accessing tax administrative data at the NT

3 work streams use these data for research, namely:

These work streams will periodically call for research proposals during the duration of the programme (2018-2021) and can be found under the Opportunities tab.  Apart from responding to the request for proposals, proposals (max 5 pages) can be sent to ntsdf@treasury.gov.za outlining the research question, the relevant literature, the proposed methodology, exact data requirements and the anticipated time needed in the lab. Please ensure your proposal showcases that you have read papers written on the tax data relevant to your proposal. This will ensure that you understand the data limitations and can place your proposed contribution into the tax data literature.

Resources

Click here for guidelines on compiling a data appendix and citing the tax data. 

Data description

These are micro data, which means the units of a population are individuals or businesses. Data records are linked by using the variables they have in common. Personal information is removed so individuals and firms cannot be identified.

Currently, six datasets are available for research:

  1. Company Income Tax (CIT) (2008-18)
  2. Pay-As-Your-Earn (PAYE, payroll or IRP5) (2008-2018)
  3. Value-Added Tax (VAT) (2009-2018)
  4. Customs data (2009-2018)
  5. Individual Panel (which merges IRP5 and ITR12 data) (2011-18)
  6. SARS-NT/CIT-IRP5 panel merges the CIT, IRP5, customs and VAT data (2008-18) 

Two matched panels, the Individual Panel and the SARS-NT Panel CIT-IRP5 have been developed. The Individual Panel matches payroll (IRP5) and self-assessed (ITR12) taxpayer information to create a more complete overview of formal employment in South Africa. You can find a detailed description of the Individual panel in this SA-TIED Working Paper.

The SARS-NT panel contains linked data from the CIT, VAT, Customs and IRP5 and is on the firm-level. You can find a detailed description of the SARS-NT panel in this WIDER Working Paper.

Data security

The data is anonymized to comply with global best practices on data security and can only be accessed at a secured data facility located in Pretoria, South Africa, where UNU-WIDER researchers work with NT and SARS staff to maintain the database and share knowledge and capabilities with anyone given access.

Research

If you are planning on responding to a SA-TIED call for research proposals, please check the papers at the bottom of this page for completed, current, and ongoing research to avoid the duplication of existing research.  Final papers resulting from the analysis of these data will be made available on our research page alongside all other SA-TIED final papers.   

previous research collaboration between NT and UNU-WIDER has already produced a series of studies utilizing tax administrative data.

Some independent research studies have also been undertaken by staff at the National Treasury. An overview of this research can be accessed here.

Please feel free to send an email to sa.datateam@wider.unu.edu with additional questions regarding the data.

 

 

Working paper
Estian Calitz, Eva Muwanga-Zake, Alexius Sithole, and Wynnona Steyn
Nowadays, tax depreciation allowances are used less as instruments of macroeconomic stabilization and more as long-term measures to stimulate investment. This paper tabulates the types of accelerated depreciation allowances in South Africa and calculates the...
April 2020
Public revenue
Working paper
Aroop Chatterjee, Léo Czajka, and Amory Gethin
This paper estimates the distribution of personal wealth in South Africa by combining tax microdata, household surveys, and macroeconomic balance sheet statistics. We systematically compare estimates of the wealth distribution obtained by direct measurement of...
April 2020
Inequality
Working paper
Ada Jansen, Winile Ngobeni, Alexius Sithole, and Wynnon Steyn
A key objective of many governments is to improve tax revenue mobilization. One way to achieve this is by improving tax compliance. This requires accurate knowledge of the tax gap, i.e. the difference between what...
April 2020
Public revenue
Working paper
Francesco Amodio, Michele Di Maio, Yifan Li, and Patrizio Piraino
We study the relationship between product market competition and labour market outcomes in South Africa. We combine firm-level data from tax records with individual-level data from the labour force survey. We estimate markups across sectors...
March 2020
Inequality
Working paper
Lawrence Edwards and Ayanda Hlatshwayo
This paper uses detailed firm transaction data on manufactured exports to analyse the dilution of the real exchange rate-export relationship in South Africa over the period 2010 to 2014. Our empirical results show that firms...
February 2020
Enterprise development
Working paper
Ayanda Hlatshwayo, Friedrich Kreuser, Carol Newman, and John Rand
This paper uses matched employer-employee data from South Africa to examine the extent to which technology transfers between firms through the hiring of workers. Allowing for differential spillovers based on observable technology differences between sending...
February 2020
Enterprise development