Making sense of a JSE moving sideways and the conditions necessary to send the trajectory upwards
The recent performance of the JSE will have been disappointing to many South African shareholders. Since 1995, the JSE All Share Index has had its severe drawdowns. But these were more than compensated for when it came to the buy and hold investor. Since 1995 average annual returns, calculated monthly, were 12.9% compared to average headline inflation of 6.2% over the same extended period.
The worst months were when the JSE All Share Index was down by more than 30% in August 1998, 34% in 2003 and were as much as 43% down in February 2003. The best months came in the aftermath of these severe declines. Total 12 month returns were 40% in January 2001, 41% in July 2005 and 37% in March 2010.
By sharp contrast, between January 2016 and May 2017, the share market has moved very little in both directions, with comparatively little movement about this low average. The worst month over this recent period was February 2016, with negative annual returns of 4.4% and the best the close to 10% that were realised at January month end, 2017. (See Figure 1 below)
The JSE, since early 2016 however, presents very differently when the All Share Index is converted into US dollars. In dollars the Index itself (excluding dividends) is up by 24% since January 2016, a gain that compares very well with that of the emerging market benchmark (up 28%) over the same period and the 19% gain achieved by the S&P 500. (See figure 2 below)
In US dollar terms the JSE has, over the past two years, sustained its very close correlation with the EM Index of which it is a small part, perhaps 8%. Or in other words, the EM indices on average have recovered ground lost vs New York between 2011 and mid-2016, but have realised much more in US dollars than in local currencies, including the rand, which has strengthened materially against the US dollar since mid-2016. The JSE in 2017 to date (8 June) has gained 3% in rands; the EM Index is up 10% in rands, while the S&P 500, at record levels in US dollars, has gained but a mere 2% this year, when converted to the rand (now R12.84).
The rand has gained 6.3% vs the US dollar this year and is 20% stronger than its worst levels of R16.85 of early January 2016. The rand has also gained ground against most EM currencies over the same period. The rand blew out against the EM peers in December 2015 on the initial Zuma intervention in the Treasury. But it then recovered consistently against its peers as well as the US dollar after mid-2016 and is almost back to the level of 2012. The second Zuma intervention in Treasury affairs in late March 2017, when he sacked Minister of Finance Pravin Gordhan, has had very little effect on the USD/ZAR and on the value of the rand compared to a basket of eleven other EM currencies that have gained Vs the USD. (See figure 4 below)
Judged by this performance of the rand compared to other EM currencies shown in figure 3, the risks of doing business in SA rose significantly in December 2015 – but have receded markedly since. Moreover the spread between the yields offered in US dollars by the South African government, compared to the yield offered by the US Treasury, an explicit measure of country risk, is now no higher than it was before December 2015. The RSA, five year Yankee (US dollar-denominated) bond currently offers a yield of 3.7%. The safe-haven five year Treasury currently offers 1.75%. This risk spread rose from about 1.9% in mid-2015 to 3.7% in January 2016. It is currently about 2% or back to where it was before the Zuma threat to SA’s fiscal stability first emerged in December 2015 (see figure 5 below).
The Zuma interventions in South Africa’s fiscal affairs have clearly influenced the rating agencies. They have downgraded SA debt. The market place, it may be concluded, has largely got over the threat. The market must have concluded, rightly or wrongly, that the influence of President Zuma and the threat he represents to SA’s ability to service its debts is in decline. As it is often said, time (and perhaps leaked e-mails) will tell.
It should be appreciated that rand strength, considered on its own, is not directly helpful to about half the stocks listed on the JSE. These are the global companies with a listing on the JSE whose major sources of revenues and profits are outside the country. As such they are rand hedges and perhaps more important, hedges against SA specific risks. A combination of rand strength, especially for SA specific reasons, is not likely to be helpful to their rand values. A given dollar value for their shares, largely established by the global investor outside South Africa, will then automatically translate into lower rand values. Another way of making this point is to recognise that when the rand gains 20% against the dollar, it would take a more than 20% annual appreciation in the US dollar value of a stock to translate into an increase in the rand value of a dual listed company. This 20% or so is a very demanding US dollar rate of return.
This is a rate of return that Naspers, but few others of the global plays listed on the JSE, have been able to meet. Indeed, some of the important global plays listed on the JSE have suffered from weaknesses very specific to their own operations. For example MTN (Nigeria exposure), Richemont (luxury goods) and the London property counters (sterling) as well as AB-Inbev (the beer market) and Mediclinic (regulations in Dubai) have all had earnings problems of their own. Enough to drag back the JSE All Share Index in US dollars and even more so in the stronger rand.
But while rand strength is a headwind for much of the market, it can be a tailwind for the other half of the JSE much more dependent on the fortunes of the SA economy. But to help them, the strong rand and less inflation that follows need to be accompanied by lower interest rates. Lower interest rates stimulate extra household spending and borrowing that then becomes helpful for the earnings of retailers and banks. This move in interest rates has been delayed by the Reserve Bank, despite the recessionary forces that higher interest rates since January 2014 have helped to produce. But the recession coupled with the strong rand and the outlook for less inflation would make lower interest rates irresistible.
These recessionary forces have also been revealed by the earnings reported by industrial and financial companies listed on the JSE. Trailing Findi earnings per share are not yet back at 2015 levels – though they are now growing. In US dollars, Findi earnings per share are yet to recover to levels realised further back in 2011. (See figure 6 below)
These reported earnings moreover do not suggest that the Findi is undervalued – given current interest rates. They suggest the opposite, a degree of overvaluation that will need to be overcome by sustained growth in reported earnings. Naspers, with a weight of 27.3% in this index will have to play a full further part in this earnings growth. A regression model of the Findi, using reported earnings and short term interest rates as explanations run with data captured from 1990, explains the level of the index very well. The model suggests that fair value or model predicted value for the Findi was only 59 964 compared to the actual level of 72 732 at May 2017. That is, the model suggests that the Findi is now some 20% overvalued and has remained so for an extended period of time since 2013. This theoretical valuation gap has grown despite the stagnant level of the market as shown in figure 7 below.
The market has implicitly remained optimistic about a recovery of earnings, a recovery now under way but, that has taken an extended period of time to materialise given recession-inducing higher interest rates in SA. Lower interest rates – perhaps significantly lower rates – would be essential to justify current market levels. They will help by discounting current earnings at lower rates, but help more by stimulating currently very depressed levels of household spending and borrowing.
A stronger global economy will also help improve the US dollar value of the global plays listed on the JSE and, depending on the USD/ZAR, perhaps also their rand values. A weaker rand may help the rand values of the offshore dependent part of the JSE. But rand weakness might (wrongly) delay lower interest rates. The best hope for the JSE is a strong global economy, combined with a strong rand and a recession-sensitive Reserve Bank. Is this all too much to hope for?