Question NW1380 to the Minister of Finance
06 June 2017 - NW1380
Alberts, Mr ADW to ask the Minister of Finance
(1)With reference to liability for income tax, what number of members of (a) each opposition party and (b) the governing party were audited by the SA Revenue Service (SARS) in each of the past five tax years; (2) whether SARS makes use of a randomised agorithm to determine which persons should be audited; if not, what are the full relevant details in respect of how SARS in fact determines which persons should be audited; if so, (a) how does the algorithm function and (b) what are the statistical chances and probability of a person being audited two or three years in a row, based on the functioning of the algorithm; (3) whether any statistical correlation exists between persons who are audited and those to whom SARS owes refunds; if not, what is his standpoint in respect of the perception on the part of taxpayers that they are audited as soon as SARS owes them funds; if so, (a) what has he found to be the reason for the alleged perception and (b) what are the full relevant details in this connection?
Reply:
(1)(a)(b) Due to the secrecy provisions contained in Section 69 of the Tax Administration Act No. 28 of 2011, SARS is prohibited from disclosing any taxpayer information (Including whether or not a taxpayer is subject to an audit) to any person other than a SARS official.
The South African Revenue Service (SARS) makes no differentiation according to taxpayers’ political affiliation. SARS treats the tax affairs of all members of political parties in exactly the same manner as all other taxpayers in accordance with the compliance model. This model is premised on three components to encourage tax compliance: Education, Service and Enforcement.
(2)(a) The selection of taxpayers for audits is generated through an automated risk Engine and manual profiling which is designed based on principles such as:
- Random, empirical and statistical rule sets are deployed
- The comparison of declared information to third party information e.g. Banks, Employers, Insurance Companies, etc.;
- The verification of claims;
- Year on year comparisons on income, deductions and movements in the aforementioned to identify exceptions;
- The verification of incomplete and inconsistent submissions of tax returns.
- The compliance programme which informs profiling activity
(b) Any case triggered by the risk engine and above a stated threshold would be routed for an Audit. This might apply to one taxpayer for multiple years.
(3) (a)(b) SARS audits debit, credit and nil assessments. The risk is obviously higher with credit assessments and thus the percentage of credit assessments audited will be higher. This question must be viewed in light of the R30bn credit revised assessments SARS was able to prevent from flowing last year.
SARS cannot comment on taxpayer perception, however all audit alerts are based on a sound selection process which all taxpayers are subject to.