NextFin News - Bank of America is making the most hawkish call on South African interest rates among major forecasters, expecting two rate increases in 2026 when the economist consensus remains split and the central bank has guided rates as stable through year-end. The first of those hikes arrived in May. Bank of America is betting the second comes on 23 September, a move that would lift the repo rate to 7.25% and leave South Africa's borrowing costs higher for longer than the Reserve Bank itself has signalled.
The divergence is a clean fault line in one of emerging markets' most closely watched policy decisions this year. The Reserve Bank of South Africa meets this week with its policy rate at 7.00%, where it has sat since a 4-2 hold on 23 July surprised the dominant analyst view for an immediate follow-through increase. Bank of America's two-hike forecast for 2026 — May delivered, September expected — puts it firmly in the tightening camp against a divided economist community and a central bank that, as recently as July, projected rates would remain broadly unchanged through December.
The Outlier Forecast and the Consensus It Defies
Bank of America's call is an outlier among forecasters, even if the derivatives market has partially come around. Forward-rate agreements — contracts that lock in future interest rates — price a rate increase in 2026 as a near certainty and attach more than an 85% probability to a 25-basis-point move at this week's meeting. A separate market gauge of the 23 September decision put the implied probability of a quarter-point increase to 7.25% at roughly 90%. Traders, in other words, are no longer arguing with the hawks.
Forecasters are a different matter. Investec chief economist Annabel Bishop, who had previously put the odds of a September hike at roughly even, now sees better than a 50% chance of a 25-basis-point move that would lift the policy rate to 7.25% and the prime lending rate to 10.75%. KPMG lead economist Frank Blackmore is more hawkish still, expecting increases in both September and November. But a consensus survey of economists is almost evenly divided between a hike and a hold for the remainder of 2026, and PSG economist Johann Els expects the central bank to hold, arguing that the May increase and cooler inflation expectations limit the need for further tightening. Into that split steps Bank of America with one of the firmest hike calls on the table.
The stable-rate guidance Bank of America is betting against is explicit. The SARB said in July that it expected rates to remain broadly stable through year-end, and its own forecast curve points to policy holding while inflation drifts back toward target. After six rate cuts across 2024 and 2025, the policy rate sits a full percentage point below its 2023 peak of 8.25%. The 28 May increase to 7.00% was the first rate rise since 2023, ending a two-year easing cycle, and the 4-2 July hold that followed was the committee's signal that one hike was enough for now. Bank of America is arguing that signal has already been overtaken by events.
There is a reason the outlier label sticks. The SARB's July hold was a genuine surprise that looked defensible at the time: headline inflation had just peaked at 5.0% in June and was showing its first signs of cooling. But the external environment moved faster than the committee's forecast horizon. The Federal Reserve's September rate increase — its first in more than three years — reset the emerging-market rate calculus overnight, and Bank of America is betting the SARB will not be allowed the luxury of waiting for domestic data to confirm what the currency is already signalling.
Why the Fed's Move Changed the SARB's Calculus
The transmission channel is the rand, and it works through the carry trade that has funded South Africa's deficits for years. The country has run positive real interest rates for most of the past decade, and that differential is what attracts foreign capital into local bonds. When the Federal Reserve raised its own benchmark rate by 25 basis points in September, to a 3.75%-4.00% range, it narrowed that differential at the margin. The rand weakened from about 16.26 per dollar to 16.40 in the immediate aftermath of the Fed decision. By 21 September the currency was trading at 16.24 per dollar, down 1.38% over the past month — a modest move by South African standards, but pointing in the wrong direction for inflation.
A weaker rand is not an abstract financial-market event in an economy where most fuel is imported and priced in dollars. Governor Lesetja Kganyago supplied the mechanism in the July policy statement himself, warning that renewed conflict in the Middle East, which has driven up oil and fertiliser prices, "could warrant further tightening if higher fuel costs spill over into food prices and core inflation." That sentence is Bank of America's thesis rendered in the central bank's own words: Fed tightening narrows the interest-rate differential, the rand softens, imported inflation rises, and the SARB follows — even against its own preference for stability.
The inflation data hand the hawks their opening. Headline consumer inflation accelerated to 5.0% year-on-year in June, the highest level in five months, before cooling to 4.3% in July — the first decline in that run. That is still above the midpoint of the SARB's 3% target, which carries a tolerance band of plus or minus one percentage point, putting the upper edge at 4%. The central bank expects inflation to average around 4% for 2026 and to remain above 4% for the rest of the year. On the SARB's own forecast, in other words, inflation spends the better part of two years hovering at the top of, or above, the target band — not the picture of a central bank with room to cut.
There is also the question of credibility, and it cuts both ways. The SARB said in July that its policy stance was restrictive enough to return inflation to target within two years. South Africa's inflation-targeting framework transmits policy over roughly 12 to 24 months, which means a delay today is a problem the committee will still be managing well into 2027. If the Fed is tightening and the rand is softening at the same time, holding rates while a major anchor-economy central bank moves in the opposite direction risks a disorderly currency move that would do more damage to inflation than a 25-basis-point hike would do to growth.
The Counter-Case: Why the SARB May Still Hold
The strongest argument against Bank of America's call is that the SARB has already done the heavy lifting. PSG's Johann Els expects the central bank to hold, reasoning that the May increase and cooler inflation expectations limit the need for further tightening. The July inflation print of 4.3% was a genuine cooling, and if August's data — due ahead of the meeting — confirms the trend, the Monetary Policy Committee can point to its own restrictive stance working rather than to a fresh inflation impulse. This is not a marginal point: the SARB's mandate is forward-looking, and policy is set for where inflation will be in 18 months, not where it was last month.
The growth side of the mandate also argues for caution. The SARB raised its 2026 growth forecast to 1.4% from 1.2% in July, but that remains below the economy's longer-run average. South Africa's economy grew by 1.1% in 2025, better than the 2023 and 2024 outcomes but still anaemic by the standard of a country with an unemployment rate above 30%. A rate hike into fragile growth carries a political as well as an economic cost, with local elections on the horizon making the timing awkward. Adam Phillips of Umkhulu Treasury put the dilemma plainly:
"The SARB is in a predicament, as to whether to increase rates or not. A hike would attract yield investors, but would mean a curtailing of domestic activity."
Finally, forward guidance carries a price. After the 23 July hold already surprised the dominant analyst view, a second surprise in the form of a rate increase would require the committee to walk back its communicated stability — something governors generally avoid unless the data force their hand. Governor Kganyago has built a reputation for data dependence over a long tenure, and that reputation is itself a policy instrument: a governor known for moving only when forced is harder to second-guess, but also slower to pivot when the pressure arrives. The SARB's own forecast curve points to stable rates through year-end, and central banks pay a market credibility premium for reversing course twice in quick succession.
What a Second Hike Would Mean for the Rand, Bonds, and Borrowers
If Bank of America is right and the SARB delivers a 25-basis-point increase on 23 September, the immediate beneficiaries are yield-seeking investors in rand-denominated assets and the currency itself, which would find support in a preserved interest-rate differential. The benchmark 2035 government bond was already firm ahead of the meeting, with its yield edging down 1.5 basis points to 8.615% on 18 September — a market braced for higher rates but not yet fully committed. That positioning is what makes the decision asymmetric: because the market already prices the hike at roughly 90%, a September increase alone should move assets little. The repricing risk sits on the other side — a surprise hold would likely weaken the rand and push bond yields higher on a credibility premium.
The exposed are households and businesses carrying debt referenced to prime, which stood at 10.50% after the July decision and would climb to 10.75% if the September 25-basis-point increase passes through mechanically. For an economy still working through the aftermath of six cuts over two years, that is a meaningful tightening of financial conditions into a recovery the SARB itself describes as fragile. Household debt levels remain elevated from the pandemic era, and the pass-through from repo rate to prime is essentially automatic — there is no discretion in the link, which is why borrowers feel every 25 basis points within weeks of an MPC decision.
History offers a sobering anchor. South Africa's benchmark interest rate has averaged 11.68% from 1998 through 2026, reaching an all-time high of 23.99% in June 1998 and a record low of 3.50% in July 2020. Even after the May and a potential September increase, the policy rate at 7.00%-7.25% remains well below that long-run average and a full percentage point below the 2023 peak. The hawks are not trying to return to 2023; they are trying to stop inflation from settling permanently above the target midpoint. The doves are arguing that getting there does not require another hike on top of one that has not yet fully worked through the economy.
Verdict: A Cyclical Squeeze, Not a Regime Shift
The right read is that Bank of America has correctly identified a cyclical squeeze but is likely right more on timing than on the ultimate destination. The pressure on the SARB is real and mechanical: Fed tightening, a softer rand, and imported fuel inflation are not arguments that disappear. A 25-basis-point hike on 23 September is the base case, and the market's 85-90% pricing reflects that. This is a cyclical, mean-reverting impulse — a reaction to an external rate shock and a commodity-price spike — not a structural shift in the SARB's inflation-targeting framework. The framework itself has not changed: the target remains 3% with a tolerance band of plus or minus one percentage point, and the transmission horizon remains 12 to 24 months.
But the hawkish case has limits, and they are visible in the committee's own split. The 4-2 July vote showed two members already wanted a second consecutive increase; the fact that four did not is evidence that the committee does not see itself in a runaway-inflation scenario. The SARB's structural preference is for a restrictive-but-stable path that returns inflation to 3% over two years without breaking the recovery. That preference does not vanish because one global bank revises its forecast, nor because the Fed moves once.
The forward look splits cleanly by time horizon. Short term, a hike this week is the base case, with the rand likely to strengthen if the SARB delivers it and the guidance stays measured; a surprise hold would probably do the opposite. Medium term, the November meeting is the real decision point, and two specific signals would validate Bank of America's hawkish stance beyond September: core inflation printing at or above 0.3% month-on-month for two consecutive months, or the rand breaking decisively through 16.50 per dollar on a sustained basis. Either would suggest imported inflation is feeding through faster than the SARB's models assume. Long term, the structural direction is still down — the policy rate remains a full percentage point below its 2023 high, and the SARB's own forecast curve points to easing once inflation is anchored near 3%.
The market has priced one hike. Bank of America is the outlier calling for the year's second. The difference between them is not really a forecast — it is a bet on whether the Fed has taken South Africa's interest-rate policy out of Pretoria's hands. On the evidence so far, the SARB still holds the pen, but the next two inflation prints will decide whether it keeps it.
Explore more exclusive insights at nextfin.ai.
