Blog
Taxing the poor or fixing the system? A better alternative to VAT increases

With just days to go before the rescheduled 2025 Budget Speech, South Africa faces a ZAR 60 billion revenue shortfall amid slowing economic growth and diminishing international aid. The initial proposal to increase VAT from 15% to 17% met sharp opposition, prompting an unprecedented postponement, the first in the country’s democratic history. But shelving the VAT hike hasn’t resolved the broader question: how should South Africa raise revenue without deepening inequality?

Following the budget’s postponement, reports suggest that discussions have shifted toward a smaller VAT increase rather than the initially proposed two percentage points, although a final decision is still pending. Some Government of National Unity coalition members have opposed any increase, arguing that it would disproportionately affect lower-income households. With the fiscal gap still looming, uncertainty remains over whether VAT will be adjusted or if an alternative revenue measure will take its place.
 

Regardless of the outcome, the debate highlights a deeper issue: if not VAT, then what?

A tax on everything—but not on everyone equally

The case for VAT is straightforward: it applies broadly, is relatively simple to administer, and is harder to evade than other forms of taxation. However, VAT is not neutral. Unlike personal income tax, which increases with earnings, VAT is flat—meaning lower-income households contribute a greater share of their income toward it. New estimates from the SAMOD microsimulation model illustrate the impact. For the poorest households, the effective VAT rate on disposable income would have increased from 15.4% to 17.9%—a rise of 2.6 percentage points, resulting in a 17.5% increase in the amount of VAT paid per household. Meanwhile, the highest income group (the top 20%) would see a much smaller increase of 0.5 percentage points, from 3.9% to 4.4%—or 12.3% more VAT paid. 

Meanwhile, the VAT increase is estimated to raise the share of the population living in poverty by 0.12 percentage points, pushing an additional 75,600 people below the poverty line and bringing the total closer to 30% of the population. The most affected groups would be households headed by the elderly, where poverty rates would rise by 0.2 percentage points against an average increase of 0.12, and female-headed households, whose poverty gap would widen by 0.22 percentage points.

Why zero-rated goods don’t fully offset the burden

Supporters of VAT argue that its regressivity is overstated since essential food items like maize meal and fresh vegetables are zero-rated. While this provides some relief, it does not fully offset the burden on lower-income households. For lower-income households, basic food is only one part of their essential spending. Other necessities—electricity, transport, school supplies, and hygiene products—remain subject to VAT. In contrast, higher-income households face less VAT exposure, as they allocate a greater share of spending toward tax-exempt services such as private education, healthcare, and financial products. The result is a tax that applies broadly but weighs more heavily on those who spend most of their income on taxable goods and services.

Would a VAT hike actually raise revenue?

For a VAT increase to meet its revenue targets, it depends on whether consumer spending remains stable. If households change their spending patterns, the projected revenue may not materialize, requiring policymakers to seek alternative sources of income. The SAMOD model assumes constant consumption, meaning it does not account for shifts in spending when VAT increases. However, real-world evidence suggests that higher VAT rates can lead to behavioural changes that undermine revenue collection. In Malaysia, the introduction of a 6% Goods and Services Tax (GST) in 2015—a form of VAT—led to reductions in discretionary spending, particularly among low- and middle-income earners. The resulting economic strain led to the repeal of the GST in 2018 and its replacement with a lower Sales and Services Tax. Similarly, in Nigeria, research shows that VAT increases led lower-income households to adjust spending patterns, either reducing consumption or shifting to informal markets to avoid higher costs, undermining revenue gains.

While VAT remains an important source of government income, these cases highlight the risks of relying too heavily on consumption-based taxes, particularly in fragile economic environments. Instead of increasing a tax that weighs most heavily on lower-income groups, South Africa should ask: would the VAT increase generate as much revenue as projected, or should South Africa consider more equitable options?

A fairer alternative: reforming retirement tax deductions

If VAT exacerbates inequality and places additional pressure on lower-income households, what alternatives exist? Our SA-TIED research points to a fairer approach: reforming how retirement contributions are taxed.Currently, pension fund contributions are tax-deductible, meaning high-income earners, who contribute more, receive the largest tax breaks. Meanwhile, lower-income earners—who save little or nothing for retirement—receive no comparable benefit.The numbers are significant. In 2019/20, total retirement fund deductions amounted to ZAR 275 billion, representing 76% of all tax deductions and 10.1% of taxable income. The tax expenditure cost was ZAR 91.7 billion, or 16.9% of total final tax liability. While the average deduction was ZAR 39,121 per taxpayer, those earning above ZAR 1.5 million per year claimed an average of ZAR 175,193—far beyond what middle- and lower-income taxpayers could claim.

Replacing these tax deductions with a flat tax credit, similar to South Africa’s 2012 medical aid tax reform, would enhance equity while generating revenue. If deductions were converted to a 26% tax credit, the reform could raise ZAR 22.7 billion, increasing tax revenue by 4.2%. A 31% conversion rate would still be revenue-positive, generating ZAR 9.5 billion while better distributing benefits across income groups.

Unlike VAT hikes, which place a disproportionate burden on lower-income earners, a retirement tax credit approach would shift the revenue burden toward those better able to absorb it. Reforming the current system would not only raise revenue but also strengthen retirement security and create a fairer tax framework. With a widening fiscal gap and limited revenue options, South Africa faces difficult choices. Instead of relying on VAT increases, which hit the most vulnerable hardest, reforming tax breaks on retirement contributions could offer a more equitable and effective alternative. As policymakers consider their next steps, the debate is not just about revenue collection, it is about what kind of tax system best supports economic fairness and long-term stability.

Acknowledgement for technical assistance with the SAMOD model by Michael Noble (SASPRI)

Ada Jansen is Professor in Economics at Stellenbosch University, South Africa. She is also the Vice-Dean (Learning & Teaching) in the Faculty of Economic and Management Sciences at Stellenbosch University.

Wynnona Steyn retired from the South African Revenue Service, where she worked as an economist in the Macroeconomic Research Unit at the South African Revenue Service since 2009.

Abena Larbi-Odam is an Associate Communications Officer for the SA-TIED programme at UNU-WIDER.

The views expressed in this piece are those of the author(s), and do not necessarily reflect the views of the Institute or the United Nations University, nor the programme/project donors.