Haroon Bhorat, Chair: Investment Committee at Sygnia
One of the worst kept Budget secrets in recent times – perhaps a sign of the challenges of an undisciplined government of national unity (GNU) – was the proposed VAT increase from 15 to 17%. This was viewed as necessary to raise an additional R58 billion of revenue to finance expenditures not accounted for in the previous budget – mainly to cover the extension of the social relief of distress grant (SRD) and an additional 1 percentage point increase in the public sector wage bill from 4.5% to 5.5%.
VAT is the second most important source of revenue in South Africa. As the graphic below shows, a quarter of total tax revenue in South Africa comes from VAT payments – sandwiched between personal income tax (40%) and company tax (16%). The outcry around an increase has often hinged on the assumption that VAT is regressive, but careful assessment of the data – based on a World Bank study in 2016 and an updated study in 2018 – shows the opposite. When zero-rated goods are included, VAT is in fact mildly progressive, meaning the tax hike would be pro-poor in nature, albeit by a very small margin. While it cannot be regarded as a redistributive tool, the VAT increase, based on this study, is at worst distributional-neutral. Furthermore, the proposed increase was accompanied by a widening of the basket of zero-rated goods, which would have further supported the progressive nature of the increase.
In a 2018 report, the Davis Tax Committee considered the impact of a hypothetical VAT increase of 3 percentage points on economic growth, revenue and household expenditure. The increase was compared to revenue-target equivalent personal income tax (PIT) and company income tax (CIT) increases. The report found that the highest revenue generation (based on leakage estimates) would come from VAT and that GDP would contract the least with the VAT change (CIT yielded the greatest GDP reduction, of about 2.5%). In turn, the impact on household welfare – measured by the expenditure effect – was most negative for PIT and CIT. In other words, households’ net income was modelled as least negatively affected by a VAT increase. This offers a clear insight into why National Treasury preferred a VAT increase over PIT or CIT increases: VAT reduces growth and household expenditure the least and generates the highest gross revenue. I raised another caution about raising PIT in the October 2024 Sygnals, because revenue from the last income tax hike for top earners led to a decline in revenue from this tax bracket. Read slowly: VAT is not regressive and is the least of the three tax hike evils that Treasury has available to it!
Ultimately, however, the onus is on the Minister of Finance to cover the R58 billion additional expenditure and resolve this budgetary impasse. To do this, the above evidence could be carefully and deliberately circulated amongst Cabinet colleagues, which could lead to horse trading around a new – lower – VAT increase. National Treasury could also consider deeper expenditure cuts in politically less sensitive areas. Alternatively, the government could go back to the credit markets to raise more debt. The first two are probably the only realistic options available to National Treasury, as Minister Godongwana is highly fiscally responsible and is unlikely to deviate from the consolidation path. Indications are already that we are very close to moving off the Godongwana era debt forecasts – albeit by very small margins.
This would then leave a politically viable solution to negotiate a lower VAT increase and expenditure cuts elsewhere or a pure R58 billion cut in expenditure commitments. Treasury officials are no doubt re-examining where excessive spending is occurring in the state apparatus – with easier cuts being to dominant government institutions where a percentage deduction here and there could be tolerated. A larger, long-term job is to identify and shut down moribund and expensive government-linked bodies, allied with a much more concerted effort to reduce fruitless and wasteful expenditure. The 2024 Auditor General’s report found that such expenditure amounted to about R30 billion in 2024 alone – more than half the needed funds in the 2025/2026 budget.
As we approach the new Budget date, we will see either a complete withdrawal of the VAT increase – and thus significant expenditure cuts amounting to R58 billion – or a smaller VAT increase and more modest expenditure cuts. The harsh lesson from one long Budget year to the next is this, however: we are walking a very long, narrow fiscal tightrope to lower debt-to-GDP ratios – with very little room for error.
ENDS
