Why are miners always trying to buy each other?
Mining capex collapsed with commodity prices nearly a decade ago and has never really recovered, with free cash used to fund share buybacks instead. Buybacks were a way to improve production per share without actually improving production.
The charts below, from Jefferies, are a little old but show the trend:
The strategy’s now exhausted but digging a new mine takes a long time, is very expensive, and annoys a lot of people. When labour and materials costs are rising, there’s not much sense in making ten-year bets on commodity prices doing the same. And when mining stocks are trading at below replacement cost, it’s much easier to buy than build. The pay-off is immediate.
Mining stocks are currently below replacement value?
Mostly, yes. It’s largely on concern about Chinese iron ore demand so copper specialists have been doing fine. However, even though copper is providing about 40 per cent of Anglo’s earnings this year, a profit warning in December deepened the perception that it’s a bit lost strategically.
Why does BHP want to own Anglo American?
It doesn’t, it wants to own some of Anglo American, specifically the bits that don’t need fixing. Anglo’s an old-fashioned conglomerate full of part ownerships and poison pills, most of which are South African. Jefferies has a structure table:
BHP wants Anglo to get rid of its majority stakes in two South Africa-listed miners, with its proposal (and headline price tag) conditional on Anglo demerging Kumba Iron Ore and Anglo Platinum. If that happens, Anglo would be recentred around Chilean and Peruvian copper with output of about 760,000 tonnes a year. Buying that would improve BHP’s copper exposure by about 40 per cent.
There’s also Australian coking coal and Brazilian iron ore, plus a not-too-significant diamonds subsidiary and a long-dated polyhalite project in the North York Moors National Park that defies rational explanation. Some of these assets are below world-class, such as the Brazilian nickel mines, but none is considered particularly high risk either politically or geologically. A formal separation would make South Africa’s wildcat strikes and rolling power outages someone else’s problem.
It’s also important that without South Africa, Anglo is approximately the shape of BHP’s existing operations. Sticking the companies together would make BHP usefully bigger but it wouldn’t really change the investment case, nor the corporate structure. Charts from JPMorgan:
So BHP’s the obvious bidder?
Good gracious, no. BHP completed the purchase of copper and nickel miner Oz Minerals less than a year ago, and its liabilities from the Samarco mine dam disaster are not yet known. It’s also no friend of South Africa, having created and demerged South32 in 2015 to get rid of all exposure.
Rio Tinto (which bought a South African business, Richards Bay Minerals, off BHP in 2012) is probably better placed to consider big M&A right now. Glencore is probably not, given everything else it has going on, but that has not always stopped it when an opportunity arises to add volume to its trading business.
There’s also the likelihood that bits of the conglomerate will be of interest to peers like Vale, already an Anglo partner in Brazil, which has former Anglo boss Mark Cutifani chairing its base metals business.
Given the risk of interlopers, has BHP offered a knockout price?
Again, no. Not even close. The offer is 0.7097 BHP shares and shares in Johannesburg-listed Kumba and Amplats. At the open, that valued Anglo at £31.1bn, or £25.08 per share, but with BHP shares down 2.4 per cent at pixel time it’s now slightly less.
Given Anglo’s conglomerate nature, finding a knockout price isn’t easy. It’s generally agreed that Anglo trades below its sum of the parts value, but that’s been true for at least 20 years, and the discount just reflects problems a new owner would inherit. If there was a quick and easy fix it would’ve happened by now.
Jefferies analyst Christopher LaFemina, a veteran in the field of mining megamergers, thinks a “price of at least £28/sh would be necessary for serious discussions to take place, and a takeout price of well above £30 per share would be the outcome if other bidders were to get involved”.
Here’s his working:
While significant diligence would be required to estimate the substantial synergies of this combination, we assume as an initial base case that synergies would equate to 5% of Anglo’s total operating costs (excluding Kumba and Amplats). This is not inconsistent with what we have seen with large mergers in mining historically. This implies an average annual EBITDA benefit of ~$750m, which compares to our 2025 EBITDA estimate of $7.6bn for Anglo (ex Kumba and Amplats). This synergy estimate is likely to be conservative. If we include our estimate of synergies on an after-tax present value basis ($4.1bn), we estimate Anglo fair value to be 2824p/sh, which equates to a US$42.6bn equity value. That is 28% above the most recent Anglo share price, and we believe it is a reasonable starting point in estimating what price might be enough to get a deal across the finish line.
Will this get through competition regulators?
According to the sellside, probably. There are a few potential sticking points in copper, given its national strategic importance, and individual approvals will take a while.
But the thing about mines is that they’re often natural monopolies in the areas they operate, selling into a global market. The Anglo-owned Quellaveco and BHP-owned Antamina mines are important to Peru, for example, while being a fraction of global copper production. If they were to be brought under the same ownership does it represent a market concentration problem or a political one? And if it’s seen as a political problem, could the antitrust remedies be sales of non-core assets while copper is retained?
Are things going to kick off in South Africa?
Maybe, not least because there are national elections on May 29. But the proposal to demerge Anglo’s listed subsidiaries isn’t an obviously hostile one domestically, and it wouldn’t be a clean break for BHP, since it would still have Anglo’s South African diamonds and a majority stake in the Samancor chrome operations. (Ironically, the Samancor stake was previously owned by BHP and found its way to Anglo via the South32 spin-off. Mining M&A often resembles musical chairs.)
Remind me, what’s the point of all this?
Cost cutting. Berenberg analyst Richard Hatch isn’t convinced that Anglo offers turnaround potential but there’s plenty of jobs that can go at head office:
In terms of synergies, there are limited clear operational synergies to us, maybe some incountry synergies in Chile, Peru, Australia and Brazil, but the main saving would be through corporate G&A.
And Anglo’s reputation is for paying higher than the industry average, particularly for management jobs. Its 2023 annual report has a useful map to indicate where the headcount cuts can happen:
Back to Berenberg:
Overall, we can see the sense in the deal for the copper assets, but BHP is potentially buying a group of assets that need some care and attention, which, in our view, offer limited upside at this point. Current valuation multiples would also imply a slightly dilutive deal for BHP. It would also have to assume a net debt position of USD10.6bn, which would increase post-closing of deal, we think, as both Kumba and Anglo Platinum, which are consolidated, are net cash. Overall, we expect Anglo to push for a higher premium.
What does this mean for North Yorkshire?
The crucial question, and it’s likely to be mixed news for the residents of Sneatonthorpe. Anglo rescued UK-quoted Sirius Minerals in 2020 — taking on Woodsmith, its hugely ambitious project to dig fertiliser out of the Yorkshire Moors. BHP also has a fertiliser project, the Jansen potash mine in Saskatchewan, Canada.
BHP’s project is much further along than the Yorkshire money pit, which has yet to receive a final investment decision and is unlikely to progress quickly under Anglo’s ownership without outside investment. Berenberg says BHP “can probably delay the Woodsmith project given it is already growing substantially in potash via Jansen in Canada.”