Something unusual has happened on the share markets over the past 16 months. The JSE has outperformed the S&P 500 index by a large margin, improving its rand value by 30.5% (including dividends reinvested) while the US S&P 500 (also valued in rand) has gained just 17.7% since January 2024.
Yet over the longer term, the past 15 years, the JSE has trailed the S&P by a large margin. The rand value of the S&P 500 is up by as much 16.5 times since 2010, compared with a 5.8 times gain for the JSE.
Fundamentals — that is increased earnings and improved economic performance discounted by generally lower interest rates — account for much of the rising value of the S&P index. S&P earnings per share in dollars have grown 4.1 times since 2010, while the index has increased 5.2 times, with much of the rerating realised after 2023.
By contrast, JSE earnings have grown faster than share values since 2010 — by 4.2 times, compared with the index which has risen 3.4 times. A derating, rather than a rerating. The temporary earnings boom of 2022-23, linked to metal prices and global supply side constraints, did little to boost the JSE.
The 10-year US Treasury bond has provided an average annual return of 3.9% over 2010-25 — thus providing equity investors with almost an average 10% per annum extra return over bonds. Past performance might have suggested that an extra 4% per annum from a well-diversified basket of equities would have been enough to justify the exposure to US equities.
The S&P 500 has delivered an average 18.7% per annum rand return since 2010 and 13.6% per annum on average in dollar terms. It surely can’t get much better than it has been for holders of US equities over the past 15 years.
By comparison, the equity risk premium for SA investors has averaged only 2.8% per annum since 2010. The yield on an RSA bond since has averaged 9.1% per annum (the inflation rate was 5.1% on average, providing an impressive real return of 4% per annum for a low risk asset) while the total annual return on the JSE averaged 11.9%.
Such exceptional returns over an extended period required a combination of unexpectedly good news — about the operating profits of the listed firm and about the risks attached to these profit flows, used to discount these expected flows. SA mostly disappointed investors on these metrics. An annual equity risk premium of 2.8% does not compensate for the extra risk. At least an extra 4% per annum would have been expected.
Yet recently, over the past 15 months — thanks largely to the increases in the price of gold and the strong growth in the earnings reported by gold mining companies — JSE earnings have outpaced S&P earnings (both in rand) by about 25%.
The risks attached to RSA bond yields also declined thanks to the formation of the government of national unity (GNU) in mid-2024. That is, risks as measured by the spread between the RSA dollar denominated bonds and their US equivalents. This reduced the cost of capital for SA businesses, the returns they must exceed to add value for shareholders, regarded here as RSA long bond yields plus an equity risk premium of 4% per annum. The establishment of the GNU reduced SA risk, while the budget dissonance has raised it.
Since 2010 JSE earnings have badly lagged S&P earnings, while the differences in the cost of capital for SA and US businesses have widened. The sovereign risk premium attached to RSA bonds have mostly risen, driving up long bond yields and the cost of capital. Slow GDP growth and the threat this results in for the tax base and fiscal stability have meant higher required returns from SA bonds and equities, and lower valuations.
These required returns, or hurdle rates, for investors and firms rose from about 11% per annum in 2013 to about 14.5% now. That is about 6.5 percentage points per annum higher than the cost of capital for the average US company, using the same 4% equity risk premium. The lack of expenditure on extra productive capacity by SA business, a GDP growth depressor, is easily explained by these baleful facts.
The JSE desperately needs the stimulus of unexpected growth in listed company earnings and in GDP, as does the SA economy, with less SA risk attached to them. Sustaining the GNU would reduce risks and interest rates. Better — much better — the GNU could help deliver surprisingly faster growth. That is, do more for the economy than merely limit the mistakes the government would otherwise make, as is the presumption implicit in the slow growth forecasts.
• Kantor is head of the research institute at Investec Wealth & Investment. He writes in his personal capacity.
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.