The Reserve Bank’s monetary policy committee (MPC) took the extraordinary step at its May meeting of publishing a 3% inflation target scenario alongside its usual baseline forecast and declaring it “more attractive” than the option of sticking with the current 4.5% inflation target.
This bold move has been widely interpreted as a deliberate signal to markets, policymakers and society to prepare for a pivotal recalibration of South Africa’s monetary policy framework.
It has been 25 years since South Africa introduced inflation targeting. Unlike most emerging markets, which have lowered their targets at least once since then, South Africa has lacked the courage to reduce the 3%-6% target band to 2%-4% as originally planned — something Bank governor Lesetja Kganyago believes was “a major monetary policy mistake”.
After 2017, the Bank moved from implicitly tolerating inflation at the top end of the target band (6%) — which had allowed higher inflation expectations and higher inflation to become entrenched — to explicitly targeting the 4.5% midpoint of the range.
At the time, the economy was on a long-term structural disinflationary path (later interrupted by the pandemic). As a result, the Bank managed to achieve the new target and anchor inflation expectations at a lower level without sacrificing any growth in the process.
Several research papers have suggested that another reason the transition was so painless was that the Bank’s hawkish commitment to 4.5% was fully believed. Or, to quote Kganyago: “Inflation was not forced down by a recession; it was managed lower by clear and credible communication.”
The chief lesson learnt was that the higher the inflation rate, the higher the interest rate a country must have. Conversely, if you want low interest rates, you must have lower inflation and then set about achieving it. Unfortunately, most of South Africa has yet to internalise this message.
Since 2021, the Bank has made no bones about the fact that even at 4.5%, South Africa’s inflation target remains an outlier relative to other emerging markets, most of which have point targets of around 3%. Our target is even higher relative to most of our developed-nation trading partners, which imposes various costs on the economy and makes it less competitive.
The Bank’s modelling shows that successfully shifting to an inflation target of 3% would allow the repo rate to be almost 50 basis points (bp) lower in 2025 and 150bp lower in 2027, falling below 6% in the outer year instead of remaining above 7% as currently forecast.
And all this would come at little cost to the economy, given that headline inflation has been below 3% for three months (2.8% in May). The Bank also has significant credibility, which would help reduce the transition costs.
Though real GDP growth would be slightly weaker in the short term, the Bank’s 3% scenario shows that by 2027 growth would be stronger, at 2% as opposed to 1.8% if the target remains unchanged.
If having the MPC publish a glowing 3% scenario wasn’t convincing enough, the Bank simultaneously released a working paper on the topic co-authored by its chief economist and MPC member, Chris Loewald.
It estimates that with the inflation target at 3%, the fiscus could save about R130bn in the first five years (or R600bn by the end of the decade) as new debt issuance would benefit from a stronger rand, reduced interest rates and lower inflation.
The paper concludes that “now is a compelling time” to move to an unambiguous point target of 3%. Kganyago appears to wholeheartedly agree. In his summation of the Bank’s biennial conference on inflation targeting, held in Cape Town in March, he argued that “the current target is too high and inconsistent with price stability”.
Lower inflation could bring substantial economic benefits, including permanent fiscal gains, he added, while the costs of transitioning to lower inflation are likely to be “minimal”, particularly if there is strong central bank credibility.
Kganyago concluded that establishing a concrete point target, such as 3%, to be met within a sufficiently long time horizon, “offers the strongest case” for ensuring flexibility while quickly anchoring inflation expectations at the new target and realising the gains from lower inflation
At the May MPC press conference, he declared: “A lower target will lock in lower interest rates … South Africa would be a lower-inflation, lower-interest rate economy.”
Such certainty suggests that if the Bank was the only institution making the decision, the target would have been lowered already. In fact, though a new target has yet to be officially announced, the Bank’s overt messaging has started to influence the rand, which strengthened immediately after the May MPC meeting.
So, though Kganyago has said he will wait for the finance minister to agree on the proposed new target, the Bank has clearly already started the process of managing inflation expectations lower.
Bureau for Economic Research (BER) chief economist Lisette IJssel de Schepper believes the time for the shift to a lower target has arrived.
“The macroeconomic environment, the current level and path of inflation, and market pricing all point to a unique window for change,” she says.
But clarity, credibility and co-ordination will be critical.
“Lowering the target carries risks, including political and fiscal, and the benefits are realised only if the move is clearly communicated and institutionally owned, and if the target is actually achieved,” she adds. “This is not just a modelling exercise; it’s a policy decision with reputational consequences.”
Not everyone is on board
While the joint National Treasury/Bank macro standing committee has been discussing the new targeting framework for some time, and the technical work is far advanced, the Treasury recently asked its staff to develop three alternative scenarios of its own. This has disappointed economists who had been hoping the two institutions would issue a joint paper.
The Treasury’s apparent foot-dragging likely stems from concern over the stickiness of administered (or public sector-controlled) prices, which have a strong impact on the formation of inflation expectations.
The danger is that if much of government, state-owned enterprises (SOEs) and public sector unions fail to buy into the new 3% target, and moderate their tariff setting and wage demands accordingly, it would mean that demand has to be forced down in the private sector (through higher interest rates) at a time when South Africa is hardly growing.
RMB Morgan Stanley economist Andrea Masia estimates that assuming administered CPI (excluding fuel) stays at its long-run average rate of 6.5%, the remainder of the basket (essentially private sector inflation) needs to average no more than 2.6% for the overall blended inflation rate to equal 3%. As of April 2025, private inflation was 3.2%.
The Bank’s counterargument is that persistently high administered price inflation is no reason not to lower the target. This is because, while administered prices do bid up headline inflation, the reverse is also true — a decline in headline inflation implies a drop in a wide range of input costs that feed into the determination of administered prices.
Wages, for instance, are a significant input cost, constituting about 28% of total municipal operating expenditure. The inflation indexation of public sector wages drives up (or down) SOEs’ and municipalities’ operating costs; this in turn feeds into higher (or lower) tariffs.
In short, the Bank argues that a decline in headline inflation will also lower administrative price inflation.
But, of course, it’s not as simple as that.
First, in terms of the multiyear public service wage settlement, wages are set to rise by CPI inflation in 2026 and 2027 but with a floor of 4%. So, even if CPI is below 4% in these years, wages will rise by 4%, which will hinder the drive towards a 3% inflation target.
Second, SOEs and municipalities jack up their tariffs, rates and service charges far in excess of headline inflation each year, not just to cover salary and other input cost increases but also to cover their own inefficiencies.
The challenge, therefore, is not just to get the entire government and the trade unions to buy in on the desirability of a lower inflation target (which is hard enough), but to ensure that municipalities and SOEs become more cost-conscious and efficient — something the Treasury has tried and mostly failed to do.
That said, a recent wave of research has highlighted the substantial long-term economic benefits of lower inflation, such as reducing uncertainty and making planning easier. This would encourage investment, boost productivity and support stronger long-term growth.
Indeed, the Loewald paper shows, using the Bank’s models, that though the combination of a stronger exchange rate and tighter real interest rate reduces economic activity by 0.3 percentage points in the first year, once inflation and long-term borrowing costs start to fall, private consumption and private investment strengthen.
Over the longer term, a lower inflation rate will also result in less real appreciation of the exchange rate, increasing the competitiveness of local producers and resulting in better export performance.
All this underpins sustained long-run gains for economic growth, with GDP growth rising by 0.25% after five years and about 0.4% after a decade.
And then there are the fiscal benefits — lowering inflation would also help to lower public debt. This occurs through stronger growth and lower interest rate premiums and borrowing costs.
The Loewald paper argues that, with the government debt ratio having tripled over the past 15 years to more than 75% of GDP, the country’s fiscal position needs “a robust reset”. This cannot be achieved, it says, without better co-ordination between the Bank and the Treasury, grounded in lower inflation and a serious effort to reduce the cost of government borrowing.
According to the Bank’s modelling, lowering the inflation target to 3% sparks expectations that inflation will fall, resulting in the market accepting cheaper borrowing rates (because the country’s embedded inflation risk premium will be reduced).
The impact on borrowing costs will, however, still take time to be realised because reducing the inflation risk premium will not lower the service costs on existing fixed-rate debt, only on future debt. But as this legacy debt is refinanced, overall debt service costs will slowly grind down, making it easier for the Treasury to stabilise the debt-to-GDP ratio.
Though the debt ratio will briefly increase in the short term (as slower nominal GDP growth initially lowers the ratio’s denominator), this effect is likely to dissipate as borrowing costs improve.
Crucially, these prospective fiscal gains would come at a time when the government faces significant refinancing demands and high borrowing costs. Over the next decade, about half of the government’s long-term domestic debt (R2.2-trillion) and foreign currency debt (R340bn) is due to mature. This is set to impose sustained funding pressure on the state and keep debt service costs elevated.
“Lowering the inflation target presents an opportunity to relieve some of this pressure and realise significant fiscal savings as new debt issuance benefits from a stronger rand, reduced interest rates and lower inflation in the medium term,” concludes the Loewald paper.
It estimates that about R130bn of fiscal savings could be realised in the first five years, rising to R600bn by the end of the decade.
If done in tandem with a shift in the Treasury’s borrowing strategy to emphasise short-term T-bills and inflation-linked borrowing (because rates on these forms of borrowing decrease as inflation falls), the state could save almost R870bn on debt service costs, while the debt stock would be R150bn smaller over 10 years.
Of course, while shifting to a lower inflation target should lower interest payments on debt over time, critics warn that tax revenues could take an immediate negative hit. If this happens and the Treasury doesn’t respond by ratcheting down spending growth, we could be in even bigger trouble fiscally.
Not so rosy
While Codera Analytics CEO Daan Steenkamp believes a lower inflation target is desirable, he is not fully convinced by the Bank’s modelling.
He sees the Bank’s published 3% scenario as a “best case” view, since it assumes that inflation expectations and price setting will align rapidly to the new target, and that fiscal policy will become supportive — assumptions he considers unrealistic.
IJssel de Schepper also considers the Bank’s 3% scenario to be “very optimistic”, noting that the model assumes away many of the frictions that make South Africa’s inflation dynamic so persistent — administered prices, labour market rigidity and structurally high inflation expectations.
Instead, it assumes that backward-looking inflation expectations will seamlessly translate into forward-looking outcomes and that the economy will experience no shocks over the medium term.
“This is extremely unlikely to transpire,” she says. “To be sure, should the Bank move to a lower inflation target, we don’t think its new baseline forecast would be as rosy as its scenario.”
The problem is that for as long as expectations settle on 4.5%, and are not reset to 3%, prices and wages will tend to increase at above 3% levels. The Bank may then have to keep interest rates higher in the near term to choke off demand to push inflation lower.
That lower demand will translate into a positive “sacrifice ratio” — how much near-term growth must be sacrificed to reduce inflation by a percentage point.
However, if companies, unions and consumers reset their expectations to 3%, price- and wage-setting behaviour will moderate accordingly, and the Bank won’t need to choke off demand by raising rates. In that event, the sacrifice ratio will be zero.
How fast inflation expectations fall depends largely on the extent to which everyone believes that the Bank will remain resolute. This depends on the Bank’s credibility and the effectiveness of its communications strategy. The higher the Bank’s credibility, the stronger the effect on expectations and the less work the nominal policy rate will have to do (see box: The Maradona theory of interest rates).
The Bank estimates that any growth sacrifice would be small, ranging between 0% and 0.5% of GDP. But sacrifice ratios are notoriously difficult to quantify. Indeed, there are nearly as many estimates as there are economists.
In the South African literature, some notable estimates range between 2% and 4%, though four estimates presented at the Bank’s biennial conference suggest that South Africa’s sacrifice ratio is low, zero or even negative.
The importance of communication and credibility
Given this uncertainty, Prof Philippe Burger, dean of the faculty of economic and management sciences at the University of the Free State, feels it would be prudent to assume the sacrifice ratio is going to be positive (that some growth will be sacrificed) and then take steps to reduce it.
“The sacrifice ratio is not something to be ignored, but it is also not a fixed, immutable parameter, unresponsive to the way policy is implemented,” he explains. “It can be reduced through clear policy targets, pursued unambiguously and resolutely.”
This calls for a continuous communication strategy — and thorough engagement in structures such as Nedlac — that stresses the need for everyone to adjust their expectations and, therefore, their price- and wage-setting behaviour, he adds.
Given the uncertainty over the potential sacrifice ratio, there appears to be a consensus that a shift to a 3% target should be approached gradually.
The Loewald paper recommends that while a 3% target should be announced now, to take advantage of the ongoing deceleration in global and domestic inflation and the dampening effect this will have on inflation expectations, the Bank’s goal should be to achieve the target only by the end of 2027 (see box: How not to do it).
“That will give companies, unions and consumers time to glide, rather than dive, into the new target range,” says Burger. “It has the added advantage that as the Bank demonstrates success in gliding inflation [down], its policy credibility will strengthen, which means that expectations will be reset with less need to push demand lower.”
But, warns IJssel de Schepper, if expectations don’t fall, the Bank could find itself “chasing credibility rather than building it”.
She says: “There is no question that pursuing a lower inflation target could come at a short-term cost to economic growth (which is dismal already) and initially weaken fiscal metrics. However, successfully achieving the target would come with long-term benefits.
“So, it’s a case of short-term pain for long-term gain. The question the Treasury is probably asking is whether now is the right time to go through that pain.”
Mixed messages
The actual mechanics of changing the inflation target requires nothing more than an exchange of letters between the finance minister and the Bank governor. This means there is no need to wait for an official occasion such as the medium-term budget in October (currently the market’s best bet) or the national budget in February.
Asked at the May MPC press conference whether the Bank would go ahead without the support of the Treasury, Kganyago replied: “Of course we can go alone. Will we do it? We don’t think it’d be desirable”.
He explained that this was because inflation targets work when the whole of government is behind them.
This is precisely why the misalignment between the Bank and the Treasury is so unfortunate. Steenkamp’s view is that promising a lower target without first achieving co-ordination could backfire.
“We need a Team South Africa approach for a low-cost transition to a lower target — a social compact between the government, the Bank, SOEs, the relevant regulators and labour unions, combined with a government commitment to structural reform and supportive fiscal policy,” he says.
IJssel de Schepper also worries that a target change that is not fully owned by both the fiscal and monetary authorities could create confusion, warning: “If the fiscal authorities continue to signal doubt and the Bank moves alone, the credibility gains from a lower target could be diluted or even reversed.”
Recent market pricing (a flatter 10-year bond yield curve and stronger rand), though probably aided by global factors, is also likely indicative of growing investor confidence that South Africa could be heading for lower inflation outcomes on the back of a lower target.
The next few months will be critical. If the shift is announced soon, before expectations start trending up alongside actual inflation (which is forecast to pick up in the second half of the year), and there is clear buy-in across institutions, it could lower the cost of capital, support the fiscal position and stabilise inflation expectations over the medium term.
If not, the Bank risks undermining the gains it has taken decades to achieve.
“Low and steady inflation is worth striving for,” says Masia. “The Bank has built an immense amount of credibility over the years and has largely been successful in its inflation targeting strategy, so to settle for a targeting framework [3%-6%] that was crafted when these two outcomes were not yet in place seems like a wasted opportunity.”
In short, it is a time for boldness and ambition. South Africa has an opportunity to achieve permanently lower inflation and, therefore, permanently lower interest rates at very little cost. We should seize it.
THE MARADONA THEORY OF INTEREST RATES
This term was coined by former Bank of England governor Mervyn King. He used the two goals Argentine football player Diego Maradona scored against England in the 1986 World Cup to explain how central bank credibility, and forward-looking inflation expectations, can deliver lower inflation outcomes without necessitating steep hikes in policy rates.
“Maradona’s first ‘hand of God’ goal,” explained King in a speech in 2005, “was an exercise of the old ‘mystery and mystique’ approach to central banking. His action was unexpected, time-inconsistent and against the rules. He was lucky to get away with it. His second goal, however, was an example of the power of expectations in the modern theory of interest rates.”
Maradona ran from inside his own half, beating five players before scoring, all by running virtually in a straight line. Because the English defenders reacted to what they expected Maradona to do — move left or right — he was able to go straight on.
Monetary policy works in a similar way, assuming that price setters will adjust their behaviour as long as they believe the central bank will not deviate in its pursuit of its goal. In that event, official interest rates might not need to rise on a vastly different path than would otherwise have been the case.
“Companies, unions and consumers will need to believe that if they do not adjust prices and wages in line with a 3% target, the Reserve Bank will implement policy to reduce inflation to 3% even if it results in lower growth,” explains Prof Philippe Burger of the University of the Free State.
“It is that belief … which will ensure an adjustment in expectations and obviate the need for the Bank to actually implement such limiting policy.”
HOW NOT TO DO IT
Brazil’s experience constitutes a warning about how not to shift to a lower inflation target and why co-ordination with the fiscal authorities and the rest of the government is so important.
In 2018, Brazil elected to reduce its target in increments of 0.25 percentage points over six years from 4.5% in 2019 to 3%, initially set for 2024 and 2025. However, it ultimately abandoned this approach amid tension between the fiscal and monetary authorities.
“The Brazilian experience illustrates that moving very slowly can be costly as fiscal policy decisions change, the economy experiences shocks, and markets do not see the move as credible,” says the Loewald paper.
It concludes that the lesson from Brazil’s experience is that if the implementation phase is too drawn out, the expected costs will rise and the risk of a policy reversal will increase.
Bisseker is an economics writer and researcher at the BER
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.