SA Gold:outlook for 2018,Recovery in cash flows into 2018


Upgrade AngloGold, Gold Fields to Buy, Harmony to Hold on higher Rand Gold

Industry burning cash in well-supplied market unless there is a supply response

    We update our Rand forecasts in this note by an average of 13% over 2018E-23Eby changing our inputs to a blended 30- and 90-day trailing average. For 2018E weuse R14.22/USD. This leads to an average 27% increase in our Rand-based pricetargets and we upgrade AngloGold and Gold Fields to Buy recommendations;

    We have lowered our price targets by an average of 10% for the platinum minersafter marking to spot and slightly lowering both our 2017E and 2018E forecastRand-Basket price by 1%. We upgrade Northam to a Hold (R39ps PT), despite a4% cut to our price target, on recent share price movement. We have unchangedrecommendations on Amplats (Buy, R315ps PT), Impala (Hold, R36ps PT), Lonmin(Sell, R8ps/GBp48ps PT) and RBPlat (Hold, R28ps PT). We are negative overall onthe sector, as at spot further financial forecast and valuation downgrades wouldoccur - more than half of 2017E production is cash flow negative after SIB capex .We maintain our view that the Rand-price is likely to remain at or around themarginal cost of production until 2020E. A risk to our well-supplied market viewis proactive supply cuts by industry (not currently expected), which have thus farbeen insufficient to improve the price environment.Commodity price forecasts: Rand to remain strong, USD-prices to improve

    and Harmony is upgraded from a Sell to a Hold. We forecast the gold sector tosee a recovery in free cash flow in 2018E from a depressed level in 2017E, drivenby both improved margins and reduced project capex (2017was a particularlycapex-heavy year). Harmony is slightly behind the curve in this regard, withthe US$300m acquisition of Moab Khotsong still pending. We forecast Sibanye-Stillwater to deleverage to below 1x net debt to EBITDA (from current c.2.3x) bythe end of 2019E on both higher EBITDA from Stillwater and cash generated;

    We forecast the Rand-basket price (R/4E) to remain below R12,000(spot,R11,944/4E oz, 2016A R12,536/4E oz) for the remainder of 2017E and to averageR12,004/4E oz in 2018. On our relatively strong Rand profile (R12,23/USD andR12,22/USD in 2018E and 2019E) we do, however, expect higher USD-platinumprices of US$1,088/oz and US$1,250/oz in 2018E and 2019E respectively. So farthis year, the strong Rand has not resulted in higher USD-platinum prices as wouldhave been expected, pushing the Rand-basket price well into the cost-curve atcurrent levels.Company specific changes: 17E&18E earnings downgrades on lower prices

    however, we have them on a Hold recommendation (R20ps PT) as we do notforecast the palladium price staying above platinum over the longer term. At spotwe derive a R30ps PT for Sibanye-Stillwater and palladium bulls may well beattracted to this deleveraging story as the Blitz project ramps up at Stillwater inthe US. Key events to watch for in 2018include progress on projects like Kibali,Obuasi, Salares Norte, Stillwater Blitz, Hidden Valley and Gruyere; while Harmonyhas guided that it will seek US$200m of equity funding to complete the acquisitionof Moab Khotsong.

    The lower forecast 2017E and 21018E Rand-basket prices have led to earningsdowngrades for the platinum producers. We have reduced EBITDA expectationsfor FY1by 21% (Amplats), 62% (Impala), 8.5% (Northam) and GP for RBPlatby 35%. For Lonmin, our forecast EBITDA loss is actually 39% better as weupgrade Q3CY17and Lonmin has a September year end. Margins remain low andforecasts are highly sensitive to input assumptions as a result.DCF-based valuations, 1x NPV, Sector risks include commodity, FX forecasts

    Changes to price targets, forecasts and recommendations are summarised inFigure 3. Company-specific risks and detailed changes to estimates are availableper company within the report.