We are in a world of much worse liquidity and moving towards Fed lift-off. In such an environment some shifts in sentiment on domestic idiosyncratics can have a profound potential effect on market trends – but only if they are on the market’s agenda. Here we update our occasional series of risk outlooks for South Africa. We look at external risks (Fed, ECB, China – including linkages) and local ones.
Not a huge number of events with high probability of major impact are going on in the coming months, though there are important developing underlying themes. The potential for a market shift on the terms of trade shock and current more important even than some of the more subtly complex events like the xenophobic violence and the gold sector wage round. The inflation story and SARB will obviously be key. The battle for the 2017 ANC electoral conference has already begun – and will shape everything between now and then.
In one of our bi-annual risk updates we look at the coming six months or so and the key events, domestic and foreign, for South Africa markets – as well as the longer-run factors and stories we know investors are following (or will need to follow) that can have a later impact on markets.
Our broad strategy stance remains to be short ZAR into Fed lift-off (targeting around 13.0) and then we see a turnaround in ZAR back to around 12.0 through mid-next year as the SARB out-hikes the Fed in its move back to neutral and given its current framework and inflation fear “cycle”. We have a bias towards a steeper FRA curve than the SARB is currently pricing, but we expect a flatter bond curve as rotation from equities into bonds by local investors continues as a medium-run theme and given the hiking story and carry backstop by foreigners through Fed lift-off for a relatively steep curve.
Below we summarise and rank the risk factors we outline in this piece.
Fig. 1: H2 2015 (and beyond) risk outlook
We have recently pushed back our house Fed lift-off call to the December meeting given the sluggishness of the growth rebound even with decent labour data. That view is unchanged after yesterday’s FOMC meeting. However, as we still think the Fed will provide some advance warning of a hike through the previous month, we maintain our view that the next SARB hike will be a 25bp hike at the November MPC meeting as markets more definitely pre-position for a hike at that time with stress coming into EMFX and ZAR in particular with a sticky current account.
We currently put a 40% probability on an October hike vs 25% for September. The intervening risks will be that a hike is possible in September if there is a strong rebound in the high frequency growth data together with decent labour market prints. Inflation would also have to tick higher even if it still doesn’t show upside momentum.
For South Africa the problem is that it is easy to construct a scenario of stress from EM fund outflows, FX hedging by foreign investors in bonds and equities etc. However, the countervailing force will be that there is an increasing expectation by investors of a duration rally (carry trade) after the Fed hikes. The question then arises of the amount of front running of that view by investors, and hence the fact duration might be quite well supported into the event (or volatility might happen from a lower yield level / flatter curve).
With markets very much set on a rate hike this year, however, any delay in Fed take-off might have more limited support for EM and ZAR if this carry trade strategy gets pushed back.
As such we continue to describe the Fed lift-off as a major volatility event for South Africa – where USDZAR may well reach towards 13.0 and the front end steepen – but we do not characterise it as a “call the IMF” risk event or a major blow-up potential. However as South Africa’s current account deficit is now around 2pp wider than at the start of Fed QE, and the net FDI funding position is arguably weaker (because of greater outflow from South Africans investing in the rest of the continent and elsewhere) so portfolio flows must be watched very carefully.
Figure 2 shows that there has been a decent degree of stability in bond flows by foreigners. That said, with about 35% ownership of government debt by foreigners and net debt liabilities of around ZAR537bn or 14% of GDP (at end March 2015) in the international investment position, the risk becomes clear. However, more worrying in terms of short-term dynamic may be the strong build-up of positioning by equity investors. As Figure 2 shows there has been a significant build-up more recently in the equity market by foreigners, replacing we think selling by locals who are shifting into bonds. The risks here stem not just from Fed contagion but also the impact of rate hikes and China stock market contagion. The JSE is already down 6% from the start of May.
Fig. 2: Foreigner holdings - cumulative Fig. 3: Terms of trade
The SARB views the ECB action as growth supportive for the eurozone. At present we are going through the peak time of summer front-loading of ECB QE meaning the largest support for easy monetary policy into EM curves and markets. However, we are experiencing only very limited support for even neighbouring countries in CEE when adding in the Greece crisis as well, and for instance steeper curves there have not yet fallen back to previous levels seen through the start of ECB QE.
As the ECB QE pace normalises back through H2, we think there is unlikely to be any real effect on EM and South Africa in particular.
What could be an additional supportive step would be an extra monetary policy loosening by the ECB through H2 in response to low growth or sticky inflation expectations. However, with market 5y5y inflation expectations having recovered to closer to the ECB target, such a move is unlikely in our view. Any such step would be supportive for ZAR if it ever did happen. Through the SARB’s eyes it would be a marginal aid to further hiking.
Commodity prices have overall not being supportive of South Africa through the past year. While South Africa initially benefited from lower oil prices, a slide in metals prices offset this, as well as the volume impact from much greater import volumes of oil by Eskom. As we can see in Figure 3 terms of trade have continued to drag further down – they are actually now at the weakest point since mid-2004. As such while there is marginal support from a narrowing of the volume balance of trade, it is being offset totally by terms of trade. This is part of the core of our view of a lasting sticky current account deficit.
Viewing commodity prices through this lens of vulnerability in the current account has been confirmed by the MPC at the last meeting. It is not looking at it as a growth negative in the first instance, but into its “fear cycle” which ultimately leads into inflation and so tighter monetary policy.
The commodity price story, however, is important from several angles. Firstly it encapsulates both the China slowdown and hard landing risk story but also the Fed lift-off (and global demand) story. As such, a better outcome for China would ease some of the commodity price pressures and vice versa.
We think while there is some volatility to come on oil prices as the Iran deal moves to implementation, and the Fed move might be supportive for demand-sensitive metals at the margins, overall the environment of terms of trade will remain very unsupportive for H2.
We have mentioned a few factors above tying in China – the commodities story and shocks into foreigner holdings (especially equities).
However we also need to consider the direct impacts of a China slowdown on South Africa. We delve into this in Figures 4 and 5.
Fig. 4: Partner shares of exports Fig. 5: Export growth by country
The peak share of China trade was in Q4 2013 at 13.7% of total exports. Since then it has fallen back significantly to a low of 7.0% in Q3 of last year before recovering slightly to 9.9% in Q1 of this year. Part of the China slowdown risk is therefore already incorporated in in our view while limiting the impact of further falls. The steady growth of Africa trade which now accounts for around 30% of exports has taken up the slack together with a rise in EU export share (both from the UK and eurozone) to 22% from a base of 16.5% in Q4 2012.
Looking at export growth by region is difficult given the impact of loadshedding for the past nine months, and the differing impact of that on different export commodities going to different countries. Overall exports have fallen by 9.6% y-o-y into March. Meanwhile, EU outperformed this by 3pp (including UK by 16pp), while US outperformed by 13pp. Africa outperformed slightly whilst China underperformed by 17pp, shrinking by 26.4% y-o-y into Q1.
China exports get particularly hit by loadshedding as a large part of them are raw and processed steel products which is particularly electricity intensive, and a smaller component from aluminium smelting which is probably the most energy intensive activity in the South African economy.
Our house view remains of a quite sharp slowdown that will become evident in Q2 GDP – falling from 7.0% to 6.6%, before a slight tick up to average around 6.7-6.8% in H2. Long run we see China growth at 6.5% next year and 6.2% in 2017 after 7.4% last year and 6.8% this year.
We think the market is only partly accounting for some of the contagion risk from China into South Africa though these are more likely to be around reinforcing a current account and terms of trade story as well as impacting the pace of growth recovery and sentiment than (as we have seen) a major direct shock. This would be particularly the case if we entered a more hard landing scenario for China. As such, we must continue to watch the impact on equity flows for South Africa, China PMI and growth data as well as the sustainability of Chinese policy responses to their equity market and growth issues.
We think the macro data theme will broadly be between questioning the pace of any future rebound in growth as base effects from the end of the platinum and manufacturing strike catch up as we move through H2, and the level of vulnerability on the current account front.
Q2 current account data is shaping up to come in pretty well before the full terms of trade shock hits in H2. There may even be a print of closer to -4% of GDP as a one-off before we see wider services, income and trade balances all dragging the current account back to -5%. There should be some support on the back of this for ZAR then (looking through any Fed risk premia shock). On top of that, this kind of profile should be supportive at the margin for growth though obviously offset in part by weak domestic demand.
At the same time we expect manufacturing to continue to be suppressed by ongoing loadshedding though still to exhibit some signs of adaptation. Base effects will be manufacturing’s enemy, however, as the bounce back from the mining strike last year (on upstream and downstream industries) catches up. As such we see average manufacturing growth of -1.5% in H2 compared with -0.2% in H1. Retail sales should continue to show more robustness averaging (albeit volatile) around 2.5% y-o-y through H2.
While it is not until the end August that we get GDP (and then mid-September for BoP) we expect growth to drop back to around 1.9% y-o-y for Q2 and then 1.6-1.7% for H2.
The other big macro data theme is inflation – though after the last print for June it is likely to be less dramatic through H2 than we had originally thought. A very delayed turn in food inflation and core high but not shifting much higher yet has moved back the target breach points for headline and core we see to the December print which is not out until January of course. Hence, we see the H2 inflation theme to be slowly grinding higher core and leaping higher (but within target) headline. The risks are tilted to the upside, however, if the food inflation cycle turns higher much quicker than expected and indeed as a weaker ZAR feeds in fast via petrol prices especially (ie risks to an earlier headline breach), though we think the risk are more balanced around core.
Eskom will be ever present through H2, with continual loadshedding (ie a number of days every week seeing at least evening peak outages) though the number of full day and loadshedding events is expected to be still more limited to maybe one a fortnight or so. The situation will be highly weather dependent, however, as Eskom pushes a much more aggressive maintenance schedule than ever before through the coming months. No new supply is expected from coal in H2 with the next unit of Medupi not expected until H2 2017. The Eskom story will still be very heavily dependent on unforecastable shocks like unplanned outages and accidents. Broadly, however, the electricity supply and Eskom build story is unlikely to shift or have much delta in the next six months we think.
One key point to watch will be the application by Eskom to NERSA for a full MYPD-III tariff reopener for the next two fiscal years. Both we and the SARB expect baseline 13% tariff increases for the two years based on underlying MYPD-III awards of 8% (already agreed) plus RCA cost recovery for previous years’ costs (including OCGT usage). Environmental levy legislation adds upside risks. However, the core risk around waiting for this application is what other, additional tariff increase Eskom asks for to recoup investment and general working costs – that could bring tariff increases into the 20-25% range. This application was originally meant to be submitted in July but with the reject selected reopener by NERSA, Eskom appears to be taking much more time over this next round and so we expect it in August or September. This event should garner quite some market focus.
On fiscal, the MTBPS will most likely be in the second half of October. The document will likely struggle to overcome “peak credibility” of the budget and how it handles the strong revenue collection performance of the fiscal year so far. By this we mean the political pressure to spend the “surprise” extra money, as well as the political pressure not to go into primary surplus next year. With the macro forecasts of the National Treasury already looking quite sound, there should be some inbuilt credibility to the document. Progress on procurement redress and tax collecting efficiencies will be important to monitor, as well as reaching targets through underspend. That said given no new policy or tax announcements possible (though could be implied) there is unlikely to be much to shock.
Ratings updates are not expected from Fitch and S&P until 4 December though Moody’s is expected to announce in September. The key rating pillars for each are growth, current account, fiscal/debt, labour markets and political backdrop. With no major change in the medium-run outlook expected on any front here, nor shorter-run shock, we expect no ratings changes in H2. Our baseline remains that given excessive benefit of the doubt by the agencies that reform must eventually come, downgrades will only occur after the 2017 elective conference in start 2018. A major disruption on the gold sector wage front through strike and violent labour conflict may tip Moody’s into negative watch – but this is not our baseline.
The gold sector wage round is garnering quite a large volume of investor interest. This is justified in our view as the outcomes of this wage round will be an important benchmark for the much bigger wage rounds next year in the wider mining sector (including coal and platinum) as well as manufacturing. Importantly, the gold sector wage round is one of the first “normal/cyclical” rounds to be exhibiting what we call “sweep up wage demands”. This is very large 50-100% increases in the wages of the lowest paid to basically establish a much higher minimum wage in a sector than set by government. Such moves are important for considering South Africa’s competitiveness particularly at a time when so many jobs in the mining sector are at stake (and the drag on manufacturing FDI or job creation).
An industry that wants to push the case of higher costs as a reason for automation, restructuring and slimming down is, in our view, not totally averse to higher demands and hence unions are giving them some leeway before a dispute is declared and formal arbitration is started. General wage demands have now fallen from 15% to 9.5% as unions focus in more on the lowest paid. We think resolution is possible without strikes around 8.0% if miners agree to sweep up wages. An enfeebled NUM (and COSATU behind it) and AMCU seemingly off the boil as well have also helped this wage round avoid strikes so far, together with unions understanding the job loss dynamic in other mineral areas.
The nuclear build program issue is probably one of the strangest areas of interest for investors. Not that it is unjustified, with 25% of GDP potential contingent liabilities on the horizon. Our view remains firstly that South Africa does not need 9.6GW of nuclear energy at around ZAR1/kwh (according to CSIR) more than any other fuel form (current base load coal is ZAR0.3/kwh; new coal is expected to be around ZAR0.8/kwh – though this is more a signal of its gross inefficiency in build costs than true cost). H2 will be important for this story as the Department for Energy is expected to launch formal procurement by end September with a decision of strategic partner by end Q1 2016. Our baseline remains that Rosatom is most likely to be selected given the funding overlay it can provide; though it may not be cheapest cost, it should be one of the lower bids. Ultimately we believe the decision remains (geo)political. Markets should watch closely the Treasury guarantees (contingent liabilities) and feed in prices offered up as part of the tender process, and then the structure of the final outcome.
Politics and legislation
For us the defining ark of narrative that will define the next two and half years will be the 2017 elective conference of the ANC – most likely held in December of that year.
There is currently a lot of debate about whether the ANC is shifting left or right. We continue to believe that both are happening at once though still within a context of tenderpreneurship, cadre deployment and vested interests. Basically, we believe that in the face of serious underperformance in job creation or crises in parastatals a need to balance vested interests still wins through. Nevertheless, in H2 some of the key themes in this area we will be watching are:
Reshuffle. A cabinet reshuffle and restructure is overdue, though its timing is uncertain. The movement of responsibility for Eskom into the Department of Energy is possible and shifts to the leadership of the Department of Public Enterprise. We do not believe any change will be made to Minister of Finance Nene and changes to other economic roles like DTI’s Rob Davies, while possible, still seem unlikely to us.
There is a long-run, currently largely dormant, debate over the SARB’s mandate that we suspect may start to rear its head through H2 as the SARB hikes rates again and intensify as we reach 7.50% neutral rates in a year’s time whilst job creation remains unsatisfactory and growth surprise risks skew to the downside (and with tight fiscal providing little room to move there). This issue is just starting to appear on investors’ radar. We have continually believed that the current mandate would change in the long run to put an accent on job creation (even if without specific labour market numerical targets). We think the National Treasury and the SARB would push back strongly against such a move in the short to medium run so we see a change as unlikely before the ANC policy (and then elective) conference of mid-2017 meaning a change may be only possible from the Budget in February 2018. An early change could be possible if much greater popular dissatisfaction with growth (placing the blame on the SARB) were to emerge.
The ANC has been increasingly asserting that the poor in South Africa have come to rely too much on the state and that the government has done too much to encourage this. Such a stance has profound implications for fiscal and if it holds true means future consolidation to primary fiscal surplus would be possible. However, strong voices remain in cabinet to boost benefit coverage and generosity. Parallel to this is a discussion within the ANC about the role of government in providing large wage increases and increased numbers of jobs – something the right in the ANC (and National Treasury) is increasingly pushing back on as unaffordable. These themes will be important to watch into the MTBPS and beyond.
The slew of left-wing legislation remains broadly ever present either on the President’s desk waiting to be signed or in parliament. The climb down of DTI to offer greater arbitration opportunities on the Investment Promotion and Protect bill which looks set to be signed into law in H2, was very welcome though arguably late and an unnecessary distraction. We also watch land reform legislation, moves to accelerate control and power of traditional leaders and the fate of legislation restricting foreigner ownership in the security industry. All these factors might move in H2 and are important for the South Africa narrative.
Structural issues in the mining sector and government’s reactions to them remain key for H2. This story is set to develop strongly through H2 given firstly the gold wage round and terms of trade shock mentioned above, but also the developing plans by miners to lay off significant numbers of workers and the divestment potential of a number of majors. While in the past we have stated that all this provides a key opportunity for the creation of a black-owned local (supposedly private sector) mining champion, the job loss and divestment story is clearly moving a lot faster than is comfortable for the DMR to move forwards with plans in this area. Hence we are in this difficult place of the DMR understanding, we believe, the difficulties and cost competitiveness issues miners are facing, but at the same baulking at the political consequences of a slide in the mining industry. The mining chamber has recently confirmed our long-standing analysis that around half of gold miners are currently underwater (set to rise significantly with larger wage settlements), while some two-thirds of platinum shafts are probably still underwater. We think issues are likely to come to a head between the DMR and mining companies over jobs through H2 with a high level jobs indaba likely and the possibility of further threats to remove mining rights. Ultimately, as after Marikana, jobs losses will happen anyway but may again be done below the surface through natural attrition etc than in a smooth restructuring. The end point looks likely to be the same.
There are significant tensions within the ANC and government over Eskom. H2 is set to see the quiet dissolving of the war room (after it became ineffective and a talking shop) and pressure has eased off National Treasury to provide further sales of non-strategic assets in the short term after the sale of Vodacom to PIC. Despite headlines through H1 stating the ANC had warmed towards some restructuring and privatisation of parts of Eskom, we believe there is no conclusion to that debate and no majority for it within the ANC. H2 may bring some moves, however, one way or the other. We think the baseline of all future build being outside Eskom while Eskom is fixed through equity injections and increased tariffs over the medium run (ie to bumble along) remains the ANC’s operating mode on Eskom for the foreseeable future. Basically waiting out for more capacity to come online.
Issues like Nkandla and xenophobic violence have largely died down and are not currently on the market’s radar. However, the unpredictability of xenophobic violence and the difficult debate that is rumbling on locally around black empowerment mean it remains an “unknown known” that can reappear for markets.
All the time campaigning for the 2017 elective conference is we believe occurring in the background. This is unlikely to break out into the public sphere really until closer to 2017. However, we continue to watch Nkosazana Dlamini-Zuma, Zweli Mkhize, Cyril Ramaphosa and Gwede Mantashe through H2. Our baseline remains that a (policy) status quo, Zulu candidate is more likely to succeed Zuma.
Related to this we watch through H2 the role of the Gauteng ANC and its leadership who are trying increasingly interesting methods of turning round provincial and municipal institutions and structures to improve service delivery and are increasingly differentiating themselves from the ANC centrally. The backlash really of the ANC leadership and other provinces like KZN is of interest. This should play into 2017 and if it is at all possible for Ramaphosa to win.