Anglo American Plc is listed on the London Stock Exchange (AAL.LN) and on the Johannesburg Stock Exchange (AGL SJ). It is also traded in the USA through an unsponsored ADR on the OTC pink sheets (OTCPK:AAUKY) where one ADR consists of two shares. For the purpose of this article and calculations I use a closing price of $9.51 on 10 July 2013 for the ADR and rely on the JSE closing price on 10 July 2013, which was R190.61 for AGL at a usd/zar exchange rate of 10.04.
Anglo American has exposure to a portfolio of commodities listed below but does not have any exposure to oil and gas. The two differentiators in the portfolio are the platinum and diamond interests, which makes Anglo unique compared to the other commodity companies; I cannot however ascertain whether this is unique in a good or a bad way.
Net asset valuation basis
Never used as a valuation method to buy a commodity asset but useful as a starting point, Anglo American trades at .70 of net asset value. Deducting intangibles the price to book ratio goes up to .80. The reason a company trades below a price to book ratio is that one of the fundamentals of the company is so poor that the return on equity (RoE) will be inadequate to compensate the shareholders. Shareholders sell the share until the share price drops to a price that provides an adequate RoE or a reasonably good risk-adjusted chance of an adequate RoE. This capital will then be allocated elsewhere, as capital goes where it is treated best. Thus a discount to NAV in a commodity company is interesting but not compelling.
A sort of a "Sum of the parts" calculation
Anglo American owns 69.7% of Kumba Iron Ore (KIO SJ) which is listed on the JSE. That translates to 374 116 749 shares at a closing price of R459.03 on 10 July 2013, which equates to a valuation of R118 906 274 568. This represents 45% of Anglo's share price. KIO contributed 48% of Anglo's operating profit, so a sensible relationship exists here.
Anglo American also owns 82% of Anglo American Platinum (AMS SJ) which is also listed on the JSE. It comprises 263 391 521 shares at a closing price of R280 on 10 July 2013, which equates to a valuation of R60 042 731 127. This represents 23% of Anglo's share price. AMS produced an operating loss of USD120 million, but it still represents 23% of the share price. This relationship appears out of kilter.
The platinum industry in South Africa is in turmoil with labour issues, ineffective government and cost escalations that are above inflation. However, as the majority of the world's platinum deposits (around 75%) are in South Africa, the platinum industry should self-balance. Production is being cut by AMS, which is the big daddy in the platinum industry, having the most reserves by a multiple over its nearest competitor. But being this big comes with the responsibility of being the swing producer. AMS have tried to exercise this responsibility by attempting to reduce ounces mined, shutting down high cost shafts. As a consequence they got blasted by the ANC government and had to rework some of their numbers. In due course, should Adam Smith's invisible hand work, the smaller mines should get squeezed out of production and long term (5 years) the industry should be profitable again. Note that there are still some mines (like Northam) undergoing expansion phases that will introduce more supply, based on decisions made years ago.
These two listed companies together represent 68% of Anglo's share price but only 46% of its operating income. The residue of the share price is R61 (R190.61 x (1-.68) = R61). This residue represents 54% of Anglo's operating profit. Calculating the P/E of this residue is done as follows:
Total Net operating income per AFS 6 164 million
Deduct KIO and AMS operating income 2 860 million
Residue of Anglos operating profit 3 304 million
Less Interest expense (Per AFS 31Dec 2012) 377 million
Less Tax at 30% assumed 878 million
Net Profit After tax 2 049 million
No of shares(Source Bloomberg) 1 393 710 000
Earnings per share USD 1.47
Earnings per share rand @ 10.04 14.76
P/E 61/14.76 = 4
Thus for the rest of Anglo's business you are paying a princely P/E of 4.
But wait, there's more.
The rational CEO
Assuming that the CEO of a commodities company is a rational allocator of capital, which s/he should be, and uses return on operating asset as a metric, s/he would look at the table below and ask some hard questions.
We know that the commodities cycle is soft and is probably going to get softer, which gives Thermal Coal a get out of jail free card for having a return on net operating assets of 8%. In due course this too should pass, in spite of the USA's changing the dynamics of the energy industry by being able to export coal in the foreseeable future as a result of huge gas discoveries.
Platinum also gets a reprieve as it self-balances: the closing of high cost shafts across the industry will result in a price increase. This may take time as the current ANC government will do its best to protect jobs.
With regard to diamonds, Anglo American has just increased its stake in De Beers to 85%, with the Botswana government holding the residue minority stake. De Beers controls 35% of the rough diamond market and although this commodity is worthless unless you are a prisoner of war, clever marketing and aspirational women should ensure that the industry does well over the long term. Anglo hopefully would have done their homework on the fundamentals when purchasing De Beers's 40% interest. Over the long term the diamond industry should produce a satisfactory return on operating assets.
The problem child is nickel, which no one really seems worried about. It produces a 1% return on operating assets of 2 509 million dollars. There appears to be no big project coming on line that can produce an adequate return for this business. This is where the new CEO should make some hard decisions, as in my opinion it is a problem that does not seem solvable and will continue to reduce Anglo's return on equity over the long term. I suspect there should be some asset write-downs in this area of business.
If you adjust the return on net operating assets to "normalised" (whatever that may mean), assume no return for the platinum and nickel businesses, reduce iron ore return from 35% to 25% as the supply coming on line over the next 3 years is terrifying, reduce copper to 18% and assume diamonds will earn 8%, then the income statement model looks as follows:
You are buying Anglos at a forward P/E multiple of 7 which is very compelling, as platinum earns zero in this model.
The blue blue sky
Minas Rio for which Anglo took a USD4 billion write-down in the December 2012 results allegedly comes into production in late 2014 - let's say 2015. The only problem is that it will come into production when there is sure to be a supply glut of iron ore. However Minas Rio will be in the lowest quartile, so it should be the last man standing even though it has not yet climbed into the ring.
The low road selling price per ton was provided by Anglo in their Minas Rio presentation and the high road selling price was the average selling price realised in 2012. At a P/E multiple of 10 this project adds R40 to the share price on the low road and R80 on the high road. Obviously there are plans to increase production above 26.5 million tons, but we will ignore this on the basis of the old adage about crawling before walking.
The market is never wrong, so why are Anglo shares so cheap? The reasons may include the following:
- Extreme negativity on commodity prices. I recall an article in Forbes during the dot com bubble that said you can choose any basket of commodities and in five years' time their prices will be halved as a result of improved technology in accessing the commodity and improved technology in utilising the commodity. This is clearly incorrect, but it may nevertheless be today's mantra in the investing world.
- Commodity companies are over owned by the investing world as a result of the China story, and this has to reverse itself. This has to happen before a bottom is reached.
- A large number of Anglo's assets are in South Africa, which is proving itself to be investor unfriendly and very labour friendly: large multinationals are astonished by the protection that South African employees enjoy.
- Write-down on the Minas Rio project and write-down of some platinum assets in the December 2012 results.
Other factors that may count against Anglo American is that they are perceived to be arrogant. The anecdotal evidence is that when dealing with suppliers, their negotiating position is "we are Anglos, we will tell you what price we will pay." When sinking a shaft for a platinum mine in a JV, the JV partner raises concerns about the possibility of a pot hole in that area and the response is " we are Anglos, we understand this geology" - and sure enough Anglos hit a pot hole, costing billions.
For every single South African no portfolio is complete with Anglo American. The company has world-class assets and has been operating in the commodities business for a very long time. They have a long-term view of the world and that is how they position their assets. There appears to be an asymmetric pay off profile with the share price at R191, with loads of free optionality existing within the company. There also seems to be a huge margin of safety with this share and the investor is getting a lot of projects for free but it will take time to be realised and time, in this case, is measured in years. The conservative long-term investor who can face an mtm loss in the first year of ownership should be richly rewarded. Take EPS of R27, add R4 for Minas Rio and you get EPS of R31; at a P/E of 10 this gets you to a share price of R310, but it won't happen tomorrow.
Source: All numbers from Anglo American Plc 31 December 2012