Flowing Visual Patterns
Market Intelligence · Fiscal policyStrategy Note
Market Intelligence Signal · 26 October 2026

The state's books are healing. Households are still waiting

The 2026 MTBPS shows debt stabilising at 78.4% of GDP and bonds rallying. But growth was cut to 1.2%, inflation is 4.5% and prime is 10.75%. The gap between the state's balance sheet and the household's is the real signal.

Reading time · 8 minTopic · MTBPS 2026Format · Strategy note

Finance Minister Enoch Godongwana's Medium-Term Budget Policy Statement on 21 October was the best fiscal news South Africa has had in some time. Revenue beat February's plan, the deficit came in below target and long bonds rallied. Yet the growth forecast was cut, and the only direct relief for households, a fuel-levy cut, expires on 5 January 2027.

Debt-to-GDP · 2025/26
78.4%

Stabilised, and set to dip to 78.2% in 2026/27. Interest takes 21c of every rand of revenue.

GDP growth · 2026 forecast
1.2%

Cut from 1.6% in February. Treasury and the SARB now agree, after a Q2 contraction.

Fuel levy relief
R10bn

R1.50/l off petrol and R3/l off diesel, 4 Nov to 1 Dec; a smaller cut to 5 Jan 2027.

01The fiscal story: repair, with fragile foundations

Gross tax revenue for 2026/27 is R17.1 billion above the February estimate, mainly from corporate tax, dividends tax and VAT. The consolidated deficit is 3.9% of GDP against 4.0% planned, the primary surplus is 1.7% and rising to nearly 3% by 2030, and weekly bond auctions are being cut, with buybacks planned. The 2048 bond yield fell by as much as 18 basis points, to 9.16%, and the rand firmed to R16.36/$ on the day (News24).

Treasury is candid about the weak points: much of the revenue gain stems from commodity prices it expects to fade, and growth was cut to 1.2% for 2026, rising to 1.6% in 2027 and about 2% by 2029.

Real GDP growth forecasts, % (National Treasury)
0.0%0.5%1.0%1.5%2.0%2.5%1.6%2026 (Feb Budget)1.2%2026 (MTBPS)1.6%2027 (MTBPS)2.0%2029 (MTBPS)
MTBPS speech, 21 Oct 2026. Grey bar: February Budget forecast for 2026. The speech gives no 2028 forecast.

02What helps, what hurts the outlook

The improvement is real, but part of it rests on factors outside Treasury's control.

Helping

  • Tax revenue R17.1bn above the February plan (corporate tax, dividends tax, VAT)
  • Deficit 3.9% of GDP vs 4.0% planned; primary surplus 1.7%, rising to nearly 3% by 2030
  • Weekly bond auctions cut; bond buybacks planned for 2026/27
  • Capital payments are the fastest-growing spending item (5.3% a year)
  • All three major rating agencies took positive actions earlier in 2026

Hurting

  • 2026 growth forecast cut from 1.6% to 1.2%
  • Revenue lift leans on commodity prices Treasury expects to fade
  • Interest takes 21 cents of every rand of revenue
  • Strait of Hormuz closure lifting oil and food prices; El Niño risk
  • R38.1bn of savings over the MTEF, incl. R13bn from grant fraud and R7.8bn from the housing grant
  • Fuel relief ends 5 January 2027
A healthier state balance sheet is not the same as a healthier household budget.

03The household squeeze: Treasury vs the data

September CPI rose to 4.5%, the second monthly increase, with fuel 27.9% dearer than a year ago and transport inflation at 10.8%. Core inflation is 4.0%, and the SARB expects to return to its 3% target only towards the end of 2027. Prime is 10.75% after two hikes since May.

Headline CPI inflation, % year on year, January–September 2026
2.0%3.0%4.0%5.0%6.0%Jan 26FebMarAprMayJunJulAugSepSARB target 3%3.0%5.0%4.5%
Stats SA (P0141), via Stats SA, SAnews and eNCA. The 3% target is the SARB's.
Treasury viewLatest data
Growth1.2% in 2026, 1.6% in 2027, 2% by 2029GDP −0.2% q/q in Q2; SARB also sees 1.2%
PricesLower inflation target helps contain borrowing costsCPI 4.5% (Sep); SARB sees 3% by end-2027
Debt78.4% of GDP, stabilised; 78.2% in 2026/272048 bond yield fell up to 18bps, to 9.16%
HouseholdsR10bn fuel relief; grants reach 18.1m peopleDebt 62.2% of disposable income; saving −1.3% (Q1)
JobsInfrastructure and capital spending leadUnemployment 33.6% (Q2)

Commentators: Standard Bank's Elna Moolman sees consumer relief and possible further rating action; CAM's Maarten Ackerman doubts growth reaches 1% this year; Momentum's Sanisha Packirisamy says temporary relief “cannot shield consumers indefinitely” from high oil prices.

04Triangulation: does the data back the MTBPS?

We set Treasury's outlook against twelve signals covering inflation and rates, growth and jobs, household finances, confidence and markets.

SignalLatestVerdictWhy it matters
InflationCPI 4.5% (Sep), up from 4.4%. Fuel +27.9% y/y; core 4.0%ChallengesAbove the 3% target; caps real income gains and room to cut rates.
Interest ratesRepo 7.25%, prime 10.75% after +50bps since May. Next MPC 19 NovMixedSARB baseline sees rates stable, then cuts; fuel relief may ease hike risk.
Producer pricesPPI 5.0% (Aug), down from 5.7%MixedPipeline pressure easing, but still above CPI.
GDP−0.2% q/q in Q2 (Q1: +0.4%)ChallengesTreasury's 1.2% needs a second-half rebound.
JobsUnemployment 33.6% in Q2 (Q1: 32.7%)ChallengesFiscal repair is not yet reaching hiring.
Household debt62.2% of disposable income; debt service 8.4%; saving −1.3% (Q1)ChallengesHouseholds are dissaving; little buffer for higher costs.
Retail sales+3.3% y/y in Aug (volumes); −0.3% m/mSupportsSpending is holding up, led by clothing.
Vehicle sales61,645 new vehicles in Sep, +12.7% y/ySupportsBig-ticket demand intact despite dearer credit (Naamsa).
Consumer confidenceFNB/BER CCI −13 in Q3 (Q2: −19)MixedImproving, but still negative.
Business confidenceSACCI 134.3 (Sep); RMB/BER 38 (Q3)MixedThe data is strong; managers stay cautious.
ManufacturingAbsa PMI 50.7 in Sep (Aug: 45.8)SupportsFirst reading above 50 since May.
MarketsRand R16.36/$ (from R16.43); 2048 bond yield down up to 18bpsSupportsInvestors back the debt path.
4

support the MTBPS outlook

4

mixed

4

challenge it

FVP interpretation

The signals split evenly. Markets, retail, vehicles and the PMI back the MTBPS story; inflation, GDP, jobs and household debt do not yet. The state's balance sheet is improving faster than the household's. Whether that gap closes depends on the 19 November MPC and on oil. The classification is FVP's judgment, and the data are from different dates (household debt Q1, GDP and jobs Q2, CPI Sep, markets 21 Oct).

05What it means in rands

A household using 200 litres of petrol a month saves about R300 in November and R200 in December. October's R3.33/l increase cost the same household R666 a month, and a R1m, 20-year home loan costs about R336 a month more than before May's hike. For a business running 1,000 litres of diesel a month, November's saving is about R3,000 (FVP calculations).

Illustrative monthly rand effects for a household (FVP calculations)
R0R200R400R600R800−R300Nov levy saving (200 l)−R200Dec levy saving (200 l)+R666October petrol rise (200 l)+R336Home loan R1m, +50bps
FVP calculations. Fuel use is an assumption. Home loan: R1m over 20 years, prime 10.25% to 10.75%.

06The fuel relief: a bridge, not a fix

The R10bn levy cut is the MTBPS's only direct relief for households, and it is weighted to diesel. That helps transport, logistics and farming most. But it arrives on top of existing upward pressure on the pump price, and it ends on 5 January 2027.

What it does

  • Cuts the fuel levy by R1.50/l (petrol) and R3/l (diesel) from 4 Nov to 1 Dec, then R1/l and R1.50/l to 5 Jan 2027
  • Costs R10bn, funded by returning R10bn from the Central Energy Fund to the National Revenue Fund and the Contingency Reserve
  • Favours diesel: a firm using 1,000 l a month saves about R3,000 in November; a 10,000-litre fleet about R30,000 (FVP calc)
  • A household using 200 l of petrol saves about R300 in November and R200 in December (FVP calc)

What it doesn't do

  • Offset existing pressure: pre-relief under-recovery was about R1.52/l on petrol and R1.25/l on diesel (CEF, to 20 Oct)
  • Guarantee a lower price: the petrol cut roughly cancels that under-recovery, so pumps may barely fall on 4 Nov (FVP arithmetic)
  • Change what sets the price: oil and the rand still decide it, and the relief ends on 5 January
  • Settle the rate debate: Cosatu hopes it averts a November hike; Anchor Capital expects a hold at 7.25%
FVP interpretation

The relief protects November's price more than it lowers it. For households the saving is real but small; for diesel-heavy businesses it is meaningful for about two months. Plan on the assumption that oil and the rand, not the levy, set the price from January. The comparison with under-recoveries is arithmetic, not a forecast.

Sources: MTBPS speech (21 Oct); Central Energy Fund data and Anchor Capital via IOL (24 Oct); Cosatu via Mail & Guardian (22 Oct). The 4 November outcome depends on oil and the rand; the comparison is arithmetic, not a forecast.

07Strategic signals for consumers and business

01

Relief is temporary; the squeeze is not

The levy cut saves a household using 200 litres of petrol a month about R300 in November, less than half of October's R3.33/l rise (FVP calc). Plan for prices staying high after 5 January.

02

Infrastructure is where the money moves

Capital payments grow fastest (5.3% a year), IFISA supports R217bn of projects and 68 PPPs are in the pipeline. Construction, engineering and project finance benefit first.

03

Credit risk sits with households

Debt is 62.2% of disposable income and saving is negative. Lenders and retailers should watch arrears into the 19 November MPC.

04

Municipal and provincial money is rising

The local government equitable share rises R21.8bn and metros' fuel levy share is restored to 23%; R3.5bn more goes to smart meters. Service-delivery suppliers should track the flows.

08The honest read

Hypotheses, not findings

Fiscal credibility: strong. Revenue beat, deficit below plan, bonds rallied. Growth outlook: weak. 1.2% needs a second-half rebound after a Q2 contraction. Household relief: low. R10bn over about eight weeks, against fuel up 27.9% a year. Behaviours worth monitoring:

  1. Fuel-levy relief shows in November CPI and eases pressure on the 19 Nov MPC
  2. Households use relief to service debt rather than spend; retail growth slows after December
  3. Infrastructure and PPP awards pick up in 2027, lifting construction and engineering activity
  4. Rating agencies respond to the new debt path and the principles-based fiscal anchor.
Conclusion

The MTBPS repairs the state's balance sheet first. The household's is left to the Reserve Bank and the oil price. For executives that means:

  • Plan for fuel and energy costs staying high after the relief ends on 5 January 2027
  • Position for infrastructure, PPP and municipal-service spending as capital outlays lead
  • Watch household credit quality into the 19 November MPC and the February 2027 Budget

Fiscal credibility is a reason for confidence, not a reason to expect customers to spend more.

Next checkpoints: Fuel price adjustment and local government elections (4 Nov); Stats SA Q3 labour force and October CPI (November); SARB MPC (19 Nov); Q3 GDP (December); fuel relief ends (5 Jan 2027); Budget (February 2027).

Sources: Minister Enoch Godongwana, Medium-Term Budget Policy Statement speech (21 Oct 2026, gov.za); News24 “Bonds rally, rand gains” (21 Oct 2026); Stats SA CPI September 2026 (21 Oct, via SAnews, Stats SA, Rio Times), earlier 2026 CPI releases (Stats SA, eNCA), PPI August 2026 and retail trade August 2026 (via Trading Economics, Rio Times), GDP Q2 and QLFS Q2 2026; SARB MPC statement (23 Sep 2026, via SAnews); SARB Quarterly Bulletin household ratios (via Daily News, Ecofin); FNB/BER CCI Q3 2026; RMB/BER BCI Q3 2026; SACCI BCI (19 Oct 2026); Absa PMI September 2026; Naamsa September 2026; economist comment via The National, IOL and Mail & Guardian (22–24 Oct 2026). Sections labelled FVP interpretation or FVP calc are not from the sources.

Data note: The speech gives debt-service costs only as a share (21c per rand of revenue, per News24) and no debt peak; Treasury's full MTBPS tables could not be reached, so rand amounts are to be confirmed. Household debt ratios are for Q1 2026; Q2 figures from the September Quarterly Bulletin were not confirmed. NCR credit data for Q2 2026 are to be confirmed. Some union commentary cites 43.8% unemployment (expanded definition); this note uses Stats SA's official rate.