Anglo American's (AAL) platinum operations--conducted through its 80% ownership of Johannesburg-listed Anglo Platinum--are a source of great potential and great frustration. While it is the world's largest producer of platinum group metals, accounting for 40% of annual global output, Anglo Platinum is far from a reliable profit machine. In 2009, while $4.5 billion in sales made Anglo Platinum the largest segment by revenue, the business earned a meagre 1% operating margin, making it a laggard in Anglo's portfolio. While a weak platinum price had something to do with this, prices were high enough in 2009 for fellow South African platinum miner Impala (IMPUY) (the industry's second-largest producer) to turn in a 21% operating margin. The difference between Anglo and Impala, as for all commodity industries, comes down to cost: It's more expensive for Anglo more to dig the stuff out of the ground.
Anglo's leadership has made cost management a priority for the business and has followed through with some dramatic moves, slashing head count 20%, splitting large mine complexes into smaller, more nimble operations, and placing three higher-cost shafts on care and maintenance. Thus far, the results have been largely positive: Mine labour productivity in the first half of 2010 was up 11% from the year-ago period. Labour productivity is crucial for Anglo's platinum operations because labour (including contractors) accounts for roughly half of the business' total cash costs. Labor and contractor expenses typically account for one sixth to one third of cash costs at an open-pit copper mine, depending on extraction and processing methods.
Better labour productivity yielded an improvement in Anglo's cost profile. In local currency terms, first-half 2010 cash operating costs per platinum equivalent ounce were ZAR 11,493, up 7% from ZAR 10,775 in the year-ago period--not bad, considering significant inflationary forces (wages, electricity) and a decline in platinum head grade. Management has targeted a cash operating cost of ZAR 11,000 per ounce in nominal terms for both 2010 and 2011, with the long-term objective of moving the cost position of its platinum operations to the first and second quartiles of the industry cost curve. If achieved, this would represent a significant coup for management and major valuation driver for Anglo as a whole.
Despite recent signs of success, we have some doubts about how achievable this might be. Inflationary pressures are significant. Power costs are likely to double in the next three years, as state electrical utility Eskom has received approval for 25%-26% annual rate hikes through 2013. The pay rate for miners, far from stellar at $300 a month, is also likely to escalate at a high pace: Double-digit annual wage increases are the norm in the Bushveld. And despite Anglo's recent success in boosting labour productivity, the rapidly diminishing returns on those efforts will limit further gains. Labor constitutes 50% of Anglo's cash costs not because of inefficient or ill-deployed human capital, but because platinum mining in the Bushveld is extremely labour-intensive. Despite many efforts to mechanise the undertaking, the most economically efficient way to extract platinum-bearing rock from deep underground continues to be a small army of hand-drill-wielding workers.
Despite management's best efforts to improve Anglo's position on the industry cost curve, the company's relative cost profile is largely left to the circumstances of geology. In this regard, Anglo isn't particularly advantaged. In the Bushveld, home to the vast majority of Anglo's reserves, there are two main types of platinum-bearing rock, or reefs in geological parlance: Merensky and UG2. Generally, Merensky is more economical to mine for a couple of reasons. First, it yields more revenue per ton because of a higher platinum/palladium ratio (platinum being the more valuable). Second, Merensky tends to be less costly to process per ton because UG2 has higher chrome content, which makes it more difficult to process. At Anglo's underground mine operations, Merensky's share of production has accounted for 31%-36% over the past few years. The company expects this share to decline to 27% in 2011 and 24% in 2012, which is closer to Merensky's share of Anglo's proven and probable underground reserves (24% as of year-end 2009). By contrast, Merensky accounts for 45% of Impala's Bushveld reserves, which partially explains Impala's superior cost profile. All else equal, this trend toward more UG2 and less Merensky does not bode well for Anglo's cost curve position.
Even if Anglo succeeds in holding the line on rand-denominated costs, it's the dollar-denominated costs that ultimately matter, since platinum is a globally traded commodity priced in dollars. Assuming, for example, the prevailing rand/dollar rate of 6.95 holds through 2011, even if Anglo achieves its goal of holding cash costs at ZAR 11,000 per ounce (a 2% cumulative decrease from 2009's ZAR 11,236), dollar cash costs will have risen 19%.
See our analysis of Anglo American’s copper, iron and coal operations.
