The platinum miners are back at work. Commentators are at pains to point out that there were no winners in the strike. It’s true. The longest strike in this country’s history has been detrimental to all involved; devastating to mineworkers and those in surrounding and labour-sending communities; damaging to the producing companies involved; and harmful to the nation’s economic growth, trade balance, and ultimately, to its credit rating. The remuneration package finally agreed upon was a long way from the Association of Mineworkers and Construction Union’s (AMCU) original demand, and the cost to company is not significantly higher than what was originally on offer. By our calculations it will take the average worker about eight years to recover earnings lost as a result of the five-month stoppage.
The extremely unstable labour environment, combined with the deeply cyclical nature of extractive industries, raises the question: Are South African miners investible at all? The South African gold mining industry sets a disturbing precedent for its peers: increasing depth, declining grades and labour-intensive production, combined with poor capital allocation and lax cost control, have made for dismal returns for shareholders over the long term. Over the last ten years the gold index has delivered zero return, while the rand gold price is up more than fivefold.
Shareholders in the platinum miners would appear to think that there is an investment case to be made. The shares of the three companies affected by the strike have held up reasonably well during the five-month long strike, all things considered. Lonmin, the most dramatically impacted of the three, has lost just over 20% of its value, Implats is down 10% and Amplats is actually up 8%. It could have been much worse, considering that together they have lost over 1m ounces of platinum production, or R24bn in revenues, while still carrying 40% to 50% of their cost base at the affected operations.
We share the view, and own both platinum equities and metal in client portfolios. The platinum sector displays some attractive fundamentals not shared by, for example, the gold sector. In the October 2012 issue of Corospondent we wrote extensively on this topic, but a brief recap of the supply and demand fundamentals follows:
- South Africa and Zimbabwe dominate the supply of platinum group metals (PGMs), producing roughly 80% of global mine supply of platinum and rhodium, and 45% of palladium.
- Platinum production at South African mines peaked at 5.3m ounces in 2006. In 2014 the impact of the strikes will in all likelihood reduce production to close to 3m ounces.
- PGMs are used principally in vehicle autocatalysts to reduce exhaust gas emissions. Laws governing these emission standards continue to tighten, and emerging market standards tend to lag those of developed markets. This translates into a reasonably predictable increase in demand for PGMs over time.
- In the case of platinum, jewellery forms an additional component of demand, in this case dominated by China. While growth will be more sensitive to movements in price, over time we would expect a growing middle income consumer class to drive additional demand.
- The combination of the aforementioned points results in market deficits. Recycling of PGMs provides an additional source of lower-cost supply that goes some way to meeting the shortfall. The recycling market is somewhat opaque, but one can make some assumptions on how this is likely to grow over time based on historical vehicle loadings.
- In addition to all of this, investment has emerged as a meaningful component of demand, particularly with the launch of platinum and palladium exchange traded funds (ETFs) in South Africa. In the absence of wild swings in metals prices, we would consider this to be an additional, relatively stable source of demand growth over time.
Ultimately we see the PGM markets being in fundamental deficit over our investment horizon, even without taking investment demand into account. The world will need all the mine supply and recycled metal it can get. What this means is that, without significant above-ground stocks, metal prices need to incentivise continued investment in mine production if the supply gap is to close.
Currently, PGM prices don’t do that. As can be seen in the chart below, in the second half of 2013 (a period reasonably unaffected by strikes excepting a six-week stoppage at Northam), 43% of mine production incurred cash losses after all expenses and capital expenditure. The average free cash margin earned was 6.8%, which provided a negligible return on industry invested capital. This is clearly not a sustainable situation.
Despite this looming deficit, there was very little movement in metals prices during the course of a strike that impacted close to 50% of South African production. In the year to date, the price of the metals basket is only up 10% in US dollars, and most of this move happened before the strike commenced. This may surprise some. We think there are a few explanations for this:
- The PGM miners, as well as some of their customers, went into this strike well prepared for a lengthy toppage. Finished metal inventories were high and these stocks were run down to meet contractual delivery obligations, as were in-process pipelines.
- Demand for certain metals, particularly platinum, is still reasonably weak. Platinum is used principally in diesel autocatalysts. Western Europe is the dominant diesel passenger car market, and auto sales in that region are only starting to show signs of recovery for the first time in seven years.
- There is clearly some above-ground stock in excess of normal inventory holding requirements that has been built up during recent years of low global growth. The market is opaque and the precise quantity is unknown. But there was little evidence of tightness in the market as the strike became increasingly prolonged, suggesting that the market was at least in part being supplied from these stocks.
The strike has reignited some of the long-held concerns among customers and investors. The lack of a reliable source of supply may spur renewed efforts by autocatalyst fabricators to replace PGMs in their products. The basic wage of R12 500 touted by AMCU has not been attained, and this may prompt another round of protracted industrial action when the new wage agreement ends in two years’ time.
However, the recent turmoil may have significant positive consequences that may be less well appreciated. In the short term, pipelines and inventories will need to be replenished, and it is likely that the market will start to feel some tightness as this happens. More importantly, certain long-term developments are likely to emerge:
- The industry for once has the opportunity to restructure meaningfully. This may result in the disposal of marginal mines and the closure of others. Management teams were in the past hampered in their ability to reduce their labour complement, which resulted in them attempting to grow their way down the cost curve by producing more ounces while trying to keep the cost base stable. This time the approach should be quite the opposite. Unlike the previously dominant NUM, AMCU is not aligned to the ruling party, and the new mines minister’s involvement in resolving the strike has resulted in a greater appreciation of the economic reality faced by the mines. There is talk of redrafting certain pieces of draconian labour legislation, and despite AMCU’s demands there is no undertaking in the wage agreement that prevents restructuring or retrenchments.
- In addition, the parties have agreed to open up discussions on productivity-based remuneration. This has obvious benefits for both employees and shareholders, but was a no-go area for labour previously.
- There has been a level of co-operation among the CEOs of the big three producers during the strike that is almost unprecedented. All three are relatively new to their positions and come with little baggage. This can only be positive for the industry.
- Above-ground stock levels that have acted as an overhang on the market would have been drawn down somewhat during the course of the strike, and this is likely to continue as pipelines and inventories are rebuilt.
Ultimately we would expect a picture to emerge of an industry producing fewer but more profitable ounces. Mechanised production will grow as a proportion of total output, but this doesn’t mean that good quality, high-grade conventional (i.e. labour-intensive) production will disappear. A healthier, more efficient industry supplying a tighter market is likely to be the result.
If we feed our assessment of metal prices required to sustain the industry along with our expectations of longer term production levels into our models, the normal earnings that result suggest that the platinum miners are trading on single-digit price earnings multiples. Owning these stocks in client portfolios requires patience and some courage; they have underperformed the market for some time, yet certainly still don’t look inexpensive on near-term multiples. But we believe the fundamentals will win out, and that the events of the recent past are likely to expedite this.
There is of course a risk that at the expiry of the current wage agreement the industry will return to another period of instability, and that costs will continue to creep up as wage rates are renegotiated higher yet again. We believe it is unlikely that such a protracted strike will recur, but continued pressure on the cost base, which will impact on the profitability of the producers, cannot be ruled out. In this scenario, given our view on market balances, metal prices are likely to benefit. It is for this reason that, in addition to investment in the equities, we have exposure to both platinum and palladium ETFs in client portfolios.