Equinox Gold (EQX)

My first DD went up 30x. Four years later, this is my second.

I'm not a gold bug, I'm just a value investor who really likes EQX's story:

Some veteran commodity investors believed gold would enter a secular bull market. To prepare, they merged with or acquired rather cheap mines (due to high production costs, uneconomical at prior price levels) producing 500k oz/year. (For example, los filos costs $2600/oz right now, so they're running it with a skeleton crew. They would like to add infrastructure too, but permitting's an issue.) Why? To simulate a highly leveraged bet on gold prices.

They then bought up the the Greenstone mine, under construction. Almost all (150-250 million) per year OCF went into construction keeping earnings negative to flat earnings for years. Construction just finished, so this quarter will see ~30m in FCF (and earnings!) freed by completed capex (perhaps 50m with current gold prices.) Upon completion, management bought the other 40% of the mine from a partner. Management also purchased shares on the open market (Ross Beaty added a million shares!) Greenstone will produce 400,000 oz around $1000 each, over half a billion in FCF. At 50% capacity already, Greenstone may give an extra 60m in FCF. Peer mines are valued at 4x the company's current market cap.

Last quarter saw 120k oz sold at $2000 f0r 240m in revenue with 50m EBITDA. With Greenstone and current gold prices, we may expect 170k oz sold at $2300, for 391m in revenue, with ~220m EBITDA (remember, Greenstone production is cheap.) The quarter after might then see 220k oz ~300m EBITDA.

I expect ~1.4B in debt, minus whatever they pay off. Specifically, 500m drawn + 500m drawn from a revolver (with 100m still available) and 450m in convertible notes of which 140m's in the money. Even if gold drops 20%, that'll be cut in half in a year. They have about 250m in cash.

For the future management reiterates they will just operate Greenstone to deleverage and add infrastructure to increase capacity and reduce cost at other mines (however e.g. Castle Mountain permitting is expected to take until 2028, so capex won't be an issue short term.)

Summary: 2.2B market cap, 1.4B debt, forward P/E under 2.


Notes:

  • gold financing window is opening with higher prices (some recent announcements didn't even involve warrants)
  • gold driven cashflow trade higher (e.g. senior gold miners trade at 3x net present value) (which.. is crazy!)
  • gold enjoys the lowest cost of capital in commodities
  • central banks and gold ETFs buy gold, not gold mines. The move in boullion's already happened, but equities are lagging, because stocks are often a Giffen gold, investors start caring only after prices move.

Since a deposit declines when mined, the implied cost of equity capital is negative! (I.e. shareholders' carry cost is negative.)

A few quarters of steady operation should see EQX establish itself as a near senior (Greenstone is the only such mine outside of a senior, there is no uncertainty due to exploration etc.) Currently EQX trades at a low p/nav in its peer group, yet has the highest reserve basis and growth profile.

With Greenstone, it'll move up that value curve, and our implied holding cost is negative. Concretely, with a 20% decrease in gold price (my personal models use $2000/oz) that means a 10B market cap by 2026. (3x NPV would be 38 billion dollars.)

Concerns:

  • sensitivity to gold prices : obviously
  • acquiring the other 40% of Greenstone, management diluted shareholders : but management also bought shares on the open market, and issuance for an accretive acquisition isn't dilutive (higher EPS after)
  • inflation drives gold costs
  • many talking heads in commodities (Rick Rule, Ross Beaty et al.) mention it off hand, which made me worry I was caught in the spell of social media charlatans : but people in industry confirm they are respectable and their past performance is amazing
  • the other mines' AISCs are abysmal, which could be indicative of poor management etc. : they explicitly bought cheap mines, uneconomical assets stranded until a new price cycle, which has come!
  • (precious metal) mining stocks often do poorly. Rick Rule says 80% are uninvestable : The coordination problem between management and shareholder incentives is quite bad in the space, where managers e.g. just like looking for gold, not operating mines etc. There's so much money pouring into junior miners and !#%@ exploration companies. Much of this EQX position just me thinking "well they are on the cusp of 1 million oz! Management has experience (successful mines e.g. PAAS) behind them. They shouldn't be compared to juniors." then waiting for a rerating like we saw with AMR.

Risk factors I have no response to:

  • gold people are weird, always dooming about the world etc. : I worry about this
  • Taiwan war, since China's the main driver : $1400/oz is where investing here'd be a bad idea for me (compared to holding WAL more). Getting cut off from Chinese demand could easily do that, but the whole market'd crash, so… I worry about this too. This is the main risk factors for my other plays HCC, GLNCY.)
  • vague accident at Greenstone e.g. something exploding, stopping production for some weeks or freak weather

Materials:

Actually there's little analysis not from the company itself, which is why I wanted to write this. Unfortunately, I primarily program valuation models and and the numbers mostly just scream "BUY ME".


I repeat, I don't actually like gold. There's little industrial consumption, so it's value is purely fiat (whereas USD at least has the government requiring tax payments in USD, requiring organizations to accumulate it.) But…

Here are the drivers of gold's price today:

  • sanctions evasion via gold (note Brazil, Mongolia, Kazakhstan, China etc. who do business with Russia are accumulating to settle USD denominated trade instead of treasuries, Basel III gave physical gold a high RSF.) China's central bank holds 70 million oz, India's 27 million.
  • regions with cultural affinity to gold (China, India, Middle East) are booming (n.b. poorer Indians tend to save in silver) who often hold considerable assets in gold. Last quarter, Chinese consumers bought 9,945,000 oz. and Indian consumers > 5 million oz. (Indian imports projected at ~7-800 tons/year, which doesn't count domestic production)
  • while gold's done well on some time frames (e.g. 10x since 2000) it has barely compared to other asset classes or inflation (when you don't cherry pick the data). Personally, I ascribe this to huge growth from tech, while populations with gold affinity didn't see much wealth accumulation. Nowadays however, the Indian Ocean's the nexus of world growth (see Gavekal.)
  • Currently precious metals are ~.5% of total assets, with a 1.5-2% historical average. Mean reversion occurs over longer periods (let's hope we haven't had a regime shift!) You can have a lot of fun comparing corporate earnings or energy costs to gold etc. and everywhere it's extremely low.
  • US treasuries destroyed a lot of wealth recently. Even after losing 40%, they were offering yields under half the inflation rate. This challenged the 60/40 equity/bond portfolio because a 40% allocation guarantees a loss substantial enough to not meet fiduciary obligations in 10-15 years, with institutions exploring other options (e.g. 60/40 equity/trend or even 36/64 equity/trend) driving further disintermediation. (N.b. currently, that liquidity is going into big tech, small caps etc. have seen huge outflows.) And lower rates help gold in the same way they help bonds.



View Reddit by VeqqView Source

Leave a Reply

Your email address will not be published. Required fields are marked *