- within Law Department Performance, Corporate/Commercial Law and International Law topic(s)
It is becoming increasingly unlikely that SARS will meet the collection targets required to achieve the additional R35 billion reflected in National Treasury's monthly publications. Data up to December show that only R248 million has been collected over and above the initial projections. By our estimation, SARS would need to increase revenue collections by 1.29 in January 2026, 1.47 in February and by 2.08 in March to achieve this target, based on historical revenue trends. Does this imply a potential VAT increase in this afternoon's budget speech? Or a reduction in planned expenditure to manage current debt levels?
At SNG Grant Thornton, we have modelled the revenue trajectory to assess the feasibility of achieving this additional target within the remaining fiscal period. Follow us for timely budget news updates. We will summarise how any announced changes may affect you as an individual and or your business.
How the above estimates were determined:
We applied a Seasonal ARIMA (SARIMA) model to forecast each revenue component using historical monthly data from 2021–2025. The model specification captures short-term persistence, underlying growth trends, and strong annual seasonality, which is particularly important for tax revenues that exhibit recurring intra-year patterns. Notably, revenue spikes — particularly in March — are driven by structural features of the tax calendar, including corporate income tax provisional payments, year-end settlements, VAT reconciliation cycles, and compliance-related adjustments. These institutional payment patterns create predictable seasonal peaks that the SARIMA framework is designed to learn from historical data.
The raw model projections for January to March 2026 produced total revenue forecasts of R8.75billion, R8.62billion, and R7.92billion respectively. However, fiscal planning requires that monthly totals align with predetermined revenue targets of R11.33billion (January), R12.69billionn (February), and R16.44billion (March). To reconcile the statistical forecasts with these policy targets, we applied a proportional scaling factor (λ), calculated as Target ÷ Revenue Forecast. This resulted in scaling factors of 1.29, 1.47, and 2.08, respectively. Each revenue component was multiplied by the corresponding λ, producing adjusted forecasts that exactly match the target amounts.
Economically, these scaling factors imply highly aggressive collection assumptions relative to historical dynamics: approximately 29% above model-based projections in January, 47% above projections in February, and more than double (approximately 108% above baseline, 208% of the forecast level) in March. The particularly large March adjustment reflects both the inherent volatility of end-of-fiscal-year collections and the elevated fiscal target embedded within the current revenue framework.
In summary, the SARIMA model provides a historically grounded structural projection, while the proportional adjustment imposes the policy-determined revenue requirement, highlighting the magnitude of additional collection effort implied for early 2026.
These projections underscore the fiscal pressure facing policymakers and frame the potential revenue or expenditure measures that may be announced in the 2026 Budget Speech.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
