Bond Outlook

If the first quarter of 2026 was defined by the eruption of the Middle East oil shock, the second quarter has been the story of its partial unwinding. On 17 June, the US and Iran signed a 14-point memorandum of understanding (MoU) to pause hostilities and restore commercial transit through the Strait of Hormuz. Brent crude saw its largest monthly decline since March 2020. It fell roughly 21% over June to the mid-to-upper-US$70s per barrel, beating a sharp retreat from the US$100+ levels of the crisis peak.

The relief has been real, but incomplete. A series of tit-for-tat flare-ups in late June and again in mid-July have renewed uncertainty. Combined with disputes over the terms of the MoU and lingering questions over who controls the Strait, these have kept Brent anchored in the upper-US$70s, rather than returning fully to the pre-war level in the low-US$70s. Vessel traffic through the Strait has recovered but remains below its pre-conflict norm, and a backlog of shipping is still working its way through. The durability of the ceasefire poses a key vulnerability for the global macro outlook, but the tail risk of a sustained US$100+ oil price has receded materially from where it stood in March.

South African (SA) assets, which had borne the brunt of the risk-off rotation in the first quarter, staged a notable recovery over the second quarter as the oil price retreated. Having ended March at its weakest, around R17.19/US$, the rand strengthened into a R16.00-R16.50/US$ range from early May and traded around R16.30-R16.40/US$ by quarter-end. The 10-year SA government bond (SAGB) yield richened[1] meaningfully from the c. 9.3% peak reached in March, as the second-round inflation tail was priced out. Importantly, the recovery was underpinned not only by the oil price retreat but also by a run of constructive domestic news: a stronger-than-expected first-quarter GDP print, Moody's shift to a positive outlook, and Fitch's first upgrade of SA in over two decades. The FTSE/JSE All Bond Index (ALBI) delivered a solidly positive return for the quarter of 7.87%, ahead of cash (1.67%) and 6.48% from inflation-linked bonds (ILBs). This was mostly driven by a flattening of the yield curve from the 10-year area and longer. This brings its 12-month return to 21.48%, which is still well ahead of cash (7.07%) and ILBs (19.61%). ILB returns lag nominal returns over the longer term; however, in the last year, they have demonstrated their defensive nature in fixed income portfolios, as their year-to-date returns have beaten nominal bonds and cash (5.28% from ILBs versus 4.25% from nominals and 3.35% from cash).

AN ACUTE SHOCK OR CHRONIC PAIN?

The first-round inflationary pain came through largely as anticipated as the fuel price spike fed directly into headline inflation. Crucially, however, the June retreat in the oil price has capped what looked in March like an open-ended second-round risk. The disinflationary path can reassert itself over the second half of the year, but faces upside risks: a renewed oil spike if there is a complete breakdown in the MoU and the possibility of an El Niño-driven drought lifting food prices, both of which the South African Reserve Bank (SARB) explicitly flagged in its May deliberations. Our revised inflation outlook has inflation peaking close to 5% imminently, averaging 4.2% in 2026, 4% in 2027, and 3.8% in 2028 (below the SARB’s May forecast but more persistent in 2027 and 2028). This is lower at a headline level but still shows the persistence in inflation. This is due to the second-round effects of the higher oil price, together with elevated food prices, which are driven by a higher risk of an El Niño-driven drought.

The SARB moved preemptively against second-round impacts of inflation at its May meeting by hiking rates 25 basis points (bps). Market expectations remain hawkish. As Figure 1 shows, the expectation is for two more 25bps hikes, taking the repo rate to 7.5% and for it to remain there over the forecast period. Coronation’s view remains that policy rates were restrictive coming into the Middle East conflict, hence providing a cushion against an aggressive rate hiking cycle. In addition, the external nature of the inflation shock, coupled with the poor local-demand backdrop, provides little reason for aggressive or prolonged restrictive policy from the SARB. We expect only one more rate hike of 25bps at the July meeting, bringing the repo to 7.25%, given the recent increase in the oil price and long-term inflation expectations, as reflected by the Bureau of Economic Research’s latest survey. The room to ease the policy rate will present itself in the second half of 2027, the magnitude of which will depend on the second-round effects of the oil price and food price developments. However, one could expect a reduction of the repo rate to 6.5% by the end of 2027 or in the first quarter of 2028. This is not priced by markets and could be a tailwind for front-end to belly bond yields.

LESS COMPELLING, STILL ATTRACTIVE

Fiscal concerns in SA have taken a back seat to geopolitical events. Many of these concerns we had coming into the oil price spike have remained. Thankfully, though, nothing has occurred to derail the gradual fiscal recovery path that SA has been on. Developments over the shorter term continue to be supportive of the fiscus. These include lower funding costs due to the lower bond yields; increased use of non-traditional funding instruments (floating rate notes, infrastructure-specific bonds, and offshore concessional financing); higher revenue expectations due to higher metals prices; and a harder line on expenditure wastage (withholding equitable share from financially irresponsible municipalities). The effect of the short-term developments is self-reinforcing: a better fiscal trajectory leads to lower bond yields which in turn leads to government funding at less of a discount. In Figure 2, one can clearly see that as yields have come lower, government has managed to get more for their issuance, which has helped them reduce nominal issuance and might help alleviate pressures in the short term.

These short-term positives treat many of the symptoms but not the disease itself. Over the longer term, 3%-4% growth is required in order to ensure that debt accumulation slows to a more manageable pace. This would need to happen in an environment where local government infrastructure in many of the key cities is crumbling due to poor management, increasing the risk of further investment required to stabilise rather than improve current conditions. There have been steps taken in the right direction but nothing substantive enough to shift the needle of local government service delivery as yet. These fundamental concerns combined with overall policy direction post the local government elections (November 2026), the ANC elective conference (December 2027), and National elections (2029) will continue to obscure the longer-term outlook for SA’s finances.

SA bond yields have benefitted from the risk reversal post the signing of the MoU and have compressed to improved levels (8.5% on 10-year SAGB). This could be thought of as expensive, albeit not at the levels reached towards the end of 2025 (8% on 10-year SAGB). We would argue that bond yields are sitting at fair levels, given that the upside for inflation has been capped in the near term and, together with cash levels, are more likely to decrease than increase over the next three years. Figure 3 illustrates that the implied real yield (the yield earned over and above the inflation rate) of the 10-year SAGB is still well above its pre-Covid level, but lower than its post-Covid average. However, if we incorporate our forecasts for inflation, we can see that it returns to its post-Covid average (>5%) quite quickly, and remains well above the pre-Covid average (c. 3%). Both the absolute and relative levels remain attractive and in line with the long-term average, which suggests yields are probably at fair value.

The value point on the nominal bond curve has shifted slightly longer than previously indicated given the compression in bond yields and the relative steepness of the curve. In Figure 4, we look at breakeven across the bond curve relative to three target cash levels (7%, 8%, 9%) over the next three years. One can think about this as the amount of bond yield widening one can tolerate before the total return of the bond equates to the targeted cash level. Taking the 15-year bond as an example, its current yield is 9%. Over three years, its yield could widen to:

• 9.9% (0.9% widening) before the total return matches a 7% cash return
• 9.5% (0.5% widening) before the total return matches an 8% cash return
• 9.1% (0.1% widening) before the total return matches a 9% cash return

Bonds trading at a lower yield relative to the targeted cash return (e.g. all the bonds with a maturity of less than 12 years trade at a yield below 9%) will need to see their yields compress to match the higher return and, therefore, have no breakeven protection (represented in the graph as a negative yield movement). Based on this analysis, the value point has definitely shifted into the longer-dated bonds with a maturity of longer than 12 years, given the yield curve is still quite steep in this area and the higher yields they offer relative to cash.

PROTECTION AND PRESERVATION

ILBs protect investors from the erosive effects of inflation, ensuring the real purchasing power of their capital is preserved over the life of the investment. They have a lower beta to nominal bonds and, therefore, protect investors during periods of high risk and inflation spikes. Over the last six months, ILBs have done a fantastic job of protecting investors, as they have outperformed nominal bonds (5.28% versus 4.25%). This performance has been driven by shorter-dated ILBs (maturity < five years), as real yields have compressed. This reflects the compression in the real policy rate and higher short-term inflation expectations, which have widened implied breakeven inflation (Figure 5). Breakeven inflation is the rate of inflation at which an investor would earn the exact same return on a standard nominal bond as they would on an inflation-linked bond of the same maturity.

Current five- and 10-year breakeven inflation figures are around the 4% level, which is in line with our expected inflation average over the next three years. In addition, real yields have seen significant compression: ILBs with maturities of less than five years now yield c. 3.5%, closer to our longer-term real policy yield of 2.5%-3%. If the five-year ILB yield compresses to 3% (a 50bps compression from current levels), it would generate a total return over the next three years of:

• 7.3% at 3% average inflation
• 8.2% at 4% average inflation
• 9.2% at 5% average inflation

As it currently stands, a five-year nominal bond will provide a total return of 8% over the next three years. So, not only does inflation need to stay above 4%, but real yields need to compress in the front end of the curve to support a position in the five-year ILB. The valuation underpin is therefore waning, but the diversification and insurance offered still holds some merit.

In Figures 6-8, we have run scenarios comparing the total return of ILBs to nominal bonds over the next three years under different inflation assumptions. In the 5% scenario (Figure 6), we assume a 1% sell-off in nominal bonds, a 0.25% sell-off in ILBs, and inflation averaging 5% over the period. As one would expect in this scenario, given that current breakeven inflation sits around 4%, ILBs provide a superior return.

In Figure 7, we assume 4% inflation average and no movement in real or nominal yields. It is therefore only the front-end ILBs that outperform the nominal bonds, given that their breakeven inflation is slightly lower than 4%. This, combined with the total return expectations for the shorter-dated ILBs, where yield compression is moving closer to the real policy rate, boosts the case for holding shorter-dated ILBs. However, the valuation underpin still rests on the real policy rate.

In the final scenario, we assume a 3% inflation average, a 50bps rally in nominal bonds, and a 25bps rally in ILBs. As can be expected, nominal returns far outstrip ILBs, given the lower-than-current breakeven inflation and the rally in nominal yields.

The conclusion from the above is that the case for holding significant amounts of ILBs in a bond portfolio has weakened on the back of inflation expectations and the total return of ILBs relative to nominal bonds under these new conditions. However, given the inherent protection that ILBs offer, they still warrant some allocation, which should be focused in the sub-five-year area, albeit in a reduced allocation than under previous assumptions.

While geopolitical risks and inflation uncertainty remain elevated relative to the benign conditions that prevailed prior to the Middle East conflict, the balance of risks has improved materially since the first quarter. SA bond valuations have adjusted to reflect this improved backdrop, leaving outright value less compelling than earlier in the year, but still attractive on a medium-term horizon given elevated real yields and the prospect of lower policy rates from 2027. We favour duration in the belly-to-longer end of the nominal bond curve. This is where the steepness of the yield curve provides the greatest compensation for risk, while maintaining a more selective allocation to shorter-dated inflation-linked bonds as portfolio insurance against renewed inflation shocks. Overall, we believe bond portfolios remain well positioned to deliver attractive real returns over the coming three years. This is supported by the combination of high starting yields; an improving near-term funding backdrop; a more credible, but still vulnerable, debt-stabilisation path; and a gradually easing monetary policy cycle providing a supportive backdrop for fixed income investors. However, we acknowledge that the recovery leaves less margin for error, and that the primary threats to the outlook remain a breakdown in the ceasefire and a renewed supply-side inflation impulse.

[1] Richened: Bond prices rose while yields fell

Insights Disclaimer