So many people do really detailed models on the trees and simply miss the forest. I am going to do a 3-4 hour job on Kumba.
Anglo has stranded Kumba, because it only owns 65% of it and not all of it. When Anglo did the deal with MMX in Brazil it was done directly, without Kumba being involved. This was very bad for Kumba, and also might be bad for the minority shareholders of Kumba. Any juicy non-SA iron assets Anglo gets will not go to Kumba. This just caps the business for Kumba, which is less than ideal. Anglo has to clean up its structure and it is actually doing a good job of it; the question is whether it is too little, too late.
If BHP gets Anglo, they will manage the MMX deal with Samarco as a neighbor, and they will manage Kumba as a stand-alone, but tie the marketing into their global marketing arm. The question that I ask is how to value Anglo’s iron ore business.
I am going to focus on Kumba in this post.
Kumba has roughly 31 mt of production, of which roughly 21 mt is at market rates, and the rest is sold to Mittal on a cost-plus basis. It has at least 25 mt/y of expansion in the North Cape, and it clearly has a lot of value.
I want to do this valuation differently, as I view Kumba as a perpetual asset (there is enough ore for a long time). Kumba has to have production costs in the $15-20 range, including rail. It is looking at a sales price next year of $60 on average with its lump fine. So each tonne of iron ore is worth $40-45 of margin. Assuming a 30% tax rate, they have $28-32 of earnings. The basic business is worth probably a 7% discount rate on earnings. If it was in Australia I would say 5%, but I am going to give 2% for country risk. So each tonne of production is worth $400-450 in market capitalization. There are 20 mt/y here, so there is a $8-9b in value of the current business. These estimates are very rough, not precisionwork.
On the expansion plans I am going to use the same model. They have 25 mt/y of feasible expansion, and in comparison to most junior projects these are very feasible projects. I am going to assume $100 per tonne to add this capacity, so $2.5b in capital would required. Assuming the projects are worth 70% of their value in development stage this expansion is worth $5.25b. The way I did this was to take $400 per tonne, subtract $100 for CAPEX (very high but safe), and multiply it by 70% to get $210 per tonne; and with 25 million tonnes, I saw $5b of value here.
So the whole company is worth, let’s say, $14b, but there are issues and the issues are what make this less attractive. Kumba only owns 74% of Sishen which means Kumba should have a market cap in the $10b range. I did not look up the market cap when cooking these numbers as that would be cheating. Anglo owns 65% so they have $6.5b in value here.
There is an extra $1-2b in value for BHP for this position. Getting Kumba will get BHP two things that they really need: 65% of Kumba’s ore goes into Europe, and getting that into their product mix will reduce their risk. The second issue is that BHP will get a significant step up on their position with Rio Tinto. Kumba is a quality asset of the first order, and frankly properly managed it is worth a lot. If BHP got Anglo, with their excellent track record of managing minor shareholders like the IDC in private companies, I could see them pushing for a delisting of the company and a take out the minority shareholders.
In my mind Kumba plus the MMX ownership has to be worth $10b in value to BHP in a takeover bid. In fact I would rather own Kumba and their share of the MMX project than, say, FMG in Australia.
Back to real work.
Benjamin
