ARCH – Divestor

Inflation – boosts prices, but boosts costs as well

When inflation is mentioned, instinctively one’s financial reaction is to own hard assets – the asset value will rise in nominal terms. In real terms, the value stays static – if you go and buy a lawnmower, that lawnmower will keep its value as a great grass-clipping instrument minus the accumulated depreciation of usage. Non-depreciating hard assets are even better, but owning a stack of silver bars does nothing except sit there and look pretty unless if you’re planning on processing the material into some better industrial usage (like analog photography!).

Another logical place would be to invest in companies producing hard assets, such as mining companies. However, the cost to haul materials out of the ground and to refine them are also subject to inflationary pressures.

Looking at the metallurgical coal market, spot Australian met coal has been reaching lofty levels (about US$350/ton), so the few publicly traded companies out there have been able to make a fortune given that costs are typically much lower than this price level. The demand for steelmaking coal is still strong despite recession concerns (although headwinds are forming – take a look at STLC, for example).

The commodity, however, has to get itself out of the ground and loaded onto ships. There is a gigantic volume of material to process. This takes capital and expertise – capital was fairly easy to obtain since most of the coal companies have delevered and can access plenty of money, but it is clear the expertise (having people with brains and experience to do the job) is getting more expensive by the month.

On October 2, 2023, ARCH delivered the following guidance (an earnings warning):

… due primarily to ongoing challenges mining in the first longwall district at its Leer South mine [,] Arch is revising its full year 2023 guidance for coking coal sales volumes to 8.6 to 8.9 million tons and its average metallurgical cash cost guidance to $88 to $91 per ton.

Contrast this with their July 27, 2023 release which had guidance at 8.9 to 9.7 million tons and a $79 to $89/ton cash cost.

On October 12, 2023, AMR delivered the following guidance (another earnings warning):

On top of some weather-related problems that caused vessel delays in the quarter, we experienced mechanical issues at DTA that hampered the ability to load and ship our coal. … Along with lower-than-expected shipment volumes in the quarter, we sold some lower-priced tons from the development areas at new mines during the pricing trough early in the quarter, which negatively impacted our average realizations for the period.

In light of the logistics challenges we have experienced throughout the year, we lowered our overall shipment volume guidance and tightened the ranges to reflect our expectations for the balance of the year. Additionally, due to further investments in employee wages as well as the significant movement of the met coal indices, which directly impact sales-related costs, we are increasing our Met segment cost of coal sales guidance for the full year.

“due to further investments in employee wages” – i.e. you need to pay people a lot to show up to work for a very dirty job these days! This was a triple whammy – lower volumes, lower realization of pricing, and increased costs. On the cost side, it went from $106-112/ton to $110-113/ton.

These are not likely to be the only, nor last warnings on costs coming out of commodity companies going forward. We are seeing these costs increase on almost all commodity firms – the question is how well each individual firm can roll up their sleeves and retain talent. Eventually the demand-supply dynamic of the commodity product will normalize and resemble some function of cost and when this occurs, the low cost producers will survive, while the higher cost producers will face increasing financial pressure until some entities break and cease production. Given the lack of capital pipeline in the fossil fuel world, this might take longer than a traditional commodity cycle, but it will eventually occur.

There will be a day when I will be writing about things other than commodities, but valuation-wise, many of those commodity equities are still trading at valuations that are nowhere close to those of the broader markets, both for ESG-exclusion reasons and anticipation that the industry is somehow going to be imperiled by some phase-out. It is very ironic that this belief is one of the primary causes of the industry’s profitability at present – capital constraint restricting supply is creating a higher price environment than if capital were flowing freely.

Costs matter – a brief look at coal

There is a paradoxical rule in investing that when you anticipate the underlying price of whatever a company sells to rise, you want to be invested in a higher cost producer. The reason for this is embedded leverage. In a flat to declining price environment, you want to be invested in the low cost producer.

An example will suffice.

Say the market rate for widgets is $100. Company A (high cost producer) can make widgets for $90 a piece, leaving $10 of profit per widget. At a 10x multiple, the company would be worth $100 a widget. If the price of widgets goes up to $200, the company would be worth $1,100 a widget, 11x your money at the same multiple.

Company B (low cost producer) makes widgets for $50 a piece, leaving $50 of profit per widget. At the same multiple, it would be worth $500. If the price of widgets goes up to $200, Company B would be worth $1,500 or a mere 3x. Not bad, but nowhere close to the high cost producer.

The reverse is true – especially if the price of widgets goes below the costs of some producers. If the price of widgets goes to $70, Company A will suffer (they will have to dig into their balance sheet), while Company B will still make a living.

Markets can anticipate these leverage effects and compensate valuations accordingly – in particular price to earnings multiples decrease as prices increase. But over market cycles, costs matter.

I’m looking at earnings of coal companies, and the contrast between ARCH and BTU is quite striking.

In Q1-2023, ARCH produces its metallurgical coal at a cash cost of US$82.66 per short ton, while BTU is $151.13. In Q4-2022, HCC was $123.40, while AMR was $112.97. Teck reported US$103 per metric ton, which is about US$94 per short ton. (In the case of Teck, there is a bit of an accounting fudge factor as some of this cost is the amortization of “capitalized stripping”, which creates unevenness in cash flows, a technical matter well beyond the point of this discussion).

As met coal prices come back down to earth (they were as high as US$450 per short ton last year and are roughly US$260 or so presently), low cost producers should start to feel the pinch on their cash flows.

It leaves the question why one would want to invest in a company producing a commodity in a lowering cost environment, and that is where some market skill comes into place – there is an anticipation of cyclicality in these companies. You can also play expectations against each company by engaging in pair trading – long one, short another (and pray that your short doesn’t get bought out).

However, there is one raw number that really counts – cash dividends. If you’re going to get paid a reasonable return on equity, it still might be good enough.

In this respect, ARCH’s 50/50 plan (which is giving 50% of free cash flow directly off as special dividends and the remaining 50% for debt/capital/remediation/buybacks) has a certain elegance to it. As more shares get repurchased, the amount of the dividend that gets distributed will rise over time. It is like a very strange version of dollar cost averaging except the company is deciding to do it for you.

In 2022, ARCH gave out about $25/share in dividends. I do not anticipate this level of distribution will continue. For one, they will start paying significant cash income taxes which will reduce the dividend stream. However, there is a reasonable chance that the cash payouts will continue being in the double digit percentages, coupled with share appreciation through buybacks. Another paradox about having high amounts of cash flows is that you want to see the stock price lower, not higher – the reason is because reinvestment (in the stock) can compound at higher rates when done at lower prices.

It would not shock me in the least to see some more consolidation in the sector. We’re already seeing Teck trying to avoid one.

Also, for reference, read my December 2019 post on Arch. Even after Covid-19, this write-up is aging pretty well.

Revisiting ARCH

ARCH has been up and down like a yo-yo for the past half-year, ranging from roughly its current lows of 115 to a high of about 170 per share.

They have been able to cash in significantly in the post-Covid metallurgical coal boom, which is also instigated by the lack of capital invested in the industry.

I’ve been revisiting the math with this company.

I made some significant projection errors with my previous April 26, 2022 post. I improperly accounted for the shares outstanding (16 million vs. 19.8 million actual) and also underestimated the cash collection cycle when it come to the Q2 dividend. I was off by a mile, estimating an $11.60 dividend when it was actually $6.00! In fairness to my projection, the company did earn about $25/share on my mistaken input of 16 million shares, but they allocated some excess dividend cash to asset retirement.

I’ve sharpened my pencils since then and hopefully will be a little more precise. While at times I can be a spreadsheet warrior and try to calculate numbers to the nearest decimal point, investment analysis is a really strange business where in most cases it is intellectually wasteful to try to be exactly correct, but optimally be mostly correct with your assumptions and directions. We will apply the same standards here.

Balance sheet-wise, ARCH ended Q2 with approximately $191 million net cash, not including the $100 million they stashed away for asset retirement obligations. This assumes the capped call transaction is cashed in, and the convertible debt is converted. This positive net cash value represents about 10% of the market cap of the company, although for the purposes of this analysis we will make a conservative assumption and ignore the net cash on the balance sheet.

We will use 19.8 million shares as the denominator, although it is quite possible ARCH did perform share buybacks in Q3.

The key statistic in Q2 was the average met coal sales price of US$286 per ton. They already have committed sales in North America for US$216, and seaborne for US$284. The rest is spot sales, which for most part should be at comparative prices.

I see that Australian coking coal futures are trading around US$264 spot and US$310 for Q1-2023 coal (quite the contango).

The point is that Q3 met coal sales pricing should be around the ballpark as Q2, or about $400 million in net income.

This time, however, the company will have fully funded the reclamation funds and paid down the debt, so they can fully utilize the free cash flow for the 50/50 capital allocation model (half to dividends and the other half to either buybacks, capital preservation or the like). In Q2 the dividend was reduced by $40 million (~$2/share) than it otherwise should have been due to the $80 million contributed to asset retirement.

ARCH should be able to give off about $9/share in their Q3 dividend, based off of approximately $360 million in distributable cash. I am guessing that their accounts receivable balance will not bloat further during the quarter.

This will make the three-quarter average for dividends $7.75/share, or $31/share annualized.

Recall this is half of the company’s distributable cash flows, which annualized is about 27% of the current share price (US$115/share).

The company will probably dump the majority of the other half of free cash flow into share buybacks. Needless to say, at a price of a 27% implied yield and in a net cash situation, I do not disagree with using capital for buybacks. Even if they are the worst market timers on the planet, they would have bought back a million shares this quarter, which would take out 5% of the shares outstanding and they would be able to jack up the dividend even further – to about $9.50/share.

At US$280/ton for met coal prices, ARCH is a cash generating machine. The margin of safety is quite high.

However, many dead bodies are littered on the road of purchasing commodity stocks after cycle highs. If the world is heading into an interest rate induced global economic recession, it does not bode well for steel production, which in turn would sap demand for metallurgical coal production. Current indications suggest a mixed environment, which bodes well for future returns.

The only real threat, other than raw commodity pricing, is their tax shield. At the end of 2021, ARCH had reserved $500 million for a valuation allowance with respect to their income taxes. In the first half of 2022, they went through $120 million of this, which will result in their tax shield expiring around Q4-2023 at the current pace of their earnings. The blended tax rate for ARCH would be approximately 28% when this kicks in – reducing the returns significantly for 2024 and beyond, but still a very healthy amount.

Birchcliff Energy – hiding in plain sight

Sometimes an investment stares at you in the face and it is so obvious that it makes you wonder why others do not see it this way.

This is the case with Birchcliff Energy (TSX: BIR). Now that it has appreciated well beyond its Covid lows, I’ll write a little more about it in detail. I’ve been long shares of this (both common and preferred) for quite some time.

In 2022 it will produce about 79,000 boe/d equivalent (exit 2022 at approx. 82,000 boe/d), of which 80% of it is in the form of natural gas. All of this production is in the northwestern Alberta area, right up to the BC border.

Thus, the primary driver for this company is the state of the natural gas market. It has exposure to Dawn, Henry Hub and AECO.

Birchcliff is an unusual company in that they do not host quarterly conference calls. Instead, they issue information through large press releases and make it very easy to look at the assumptions. Although I have no problem sharpening my pencil and doing the leg works to do a proper pro-forma projection given various commodity price environments, Birchcliff expedites this process considerably.

There is some fine print to wade through, but the point is that BIR will generate $910 million in “excess free funds flow” (effectively cash flows after capex and projected dividend payments) with the average commodity prices as displayed in the release.

Notably, spot WTI and the spot Henry Hub price is well above their assumptions (US$114 and US$9.2 as I write this). Dawn typically tracks Henry Hub. Let’s ignore that spot is higher than modeled rates in the press release.

$910M of “excessive free funds” translates into $3.43/share.

At Wednesday’s closing price of $11.56, that is 3.4x or a yield of about 30%.

Normally companies are constrained with leverage and debt servicing. At the end of 2021, Birchcliff had $539 million in net debt (which includes BIR.PR.C) and another $50 million for the redemption of BIR.PR.A. The redemption of the preferred shares will result in a $6.8 million annualized savings on dividends (3 pennies a share, every bit counts!).

This will leave the company with a positive net cash amount of $270 million at the end of the year (the “Surplus”), unless they decide to blow some money on acquisitions and the like. Importantly, the math does not have to be adjusted for a leveraged return (indeed, it has to be corrected in the opposite direction).

The company will also be making enough money to eat through most of its tax shield ($1.9 billion at the end of 2021) and start paying income taxes in 2023, if the current price environment continues. Still, at US$88 oil, and US$5.50 Henry Hub for 2023 assumptions, the projection is for $535 million or about $2/share in free cash flow.

The stated policy on what to do with the cash surplus is to dividend it out beyond that which is to be used for strategic purposes. Management does not appear to be big on share repurchases other than to offset dilution that which has been issued from option plans (which is a real cash cost and will drag cash flows accordingly).

They will increase the dividend to $0.80/year in 2023, which is a $212 million outflow. This dividend can be maintained at price levels that are unlikely to be seen barring a great depression.

If they dividend the rest of their cash flows, when plugging in current commodity prices, they can give out far more than $0.80/year in dividends. It would be closer to around $2.80, or about $0.70 per quarter. Needless to say, if this is what they did, the market would find the yield (24%) tough to resist.

This is a very similar situation to Arch Resources (NYSE: ARCH), where the company will be giving out half of its free cash flow as a dividend and the other half to buy back shares. Considering its Q2 dividend will likely be around US$11/share, the obvious value of a share buyback is apparent. I wish Birchcliff would more actively consider it, at some cut-off threshold. For example, they can buy back shares until the price gets to a point where it is at 15% projected long-term free cash flows, a very conservative metric for a beneficial buyback. Right now that would imply that buying back below $15/share will clear that hurdle. At 12%, that number is about $19/share. There’s quite a way to go from current market prices.

None of this is a huge secret. It’s all in plain sight. It all relies on elevated commodity prices.

In a rising rate environment, cash flow talks – ARCH’s payout phase

“How much should I pay for the equity” is an easy question to ask, but computing all of the variables to come up with a range of prices is not so easy.

One of the components is the rate of interest. As interest rates rise, you want your cash flows today instead of tomorrow.

For instance, if your high tech company will give out a billion dollars in cash 10 years from today, the rate of interest has a significant effect on today’s capitalized value.

At 1% your billion dollars of cash 10 years out translates into $905 million.

However, at 3%, that same billion dollars turns into $744 million, or about an 18% difference from the above.

This is one reason why long term government debt has been decimated as of late. For instance, the TLT ETF (with an average term to maturity of 26 years) from the local peak at the beginning of December to present has rendered investors a 21% loss. The higher the long-term yield goes, the more damage that gets priced into the capital value. This is only mildly tempered by the 2.47% coupon and 3.01% yield to maturity – in other words, all things being equal, you would have to wait 7 years to recover the loss.

The converse is true for entities that will give out cash today instead of tomorrow.

Arch Resources reported their first quarter results today. There is a lot of details to digest, but the obvious headline is the declaration of a $8.11/share dividend, which is the result of their capital allocation policy to give out half of their free cash flow in dividends, while retaining the other half for other general purposes.

With the price of coal, both metallurgical and thermal, being sky-high as a result of a huge confluence of events (chronic under-investment, ESG, Russia/Ukraine, Australia/China, and overall demand for steel production), Arch and other coal producers that are still able to produce (this is the key – they need to have functional mines and not promises to build them) are making a fortune in free cash flow.

This quarter alone, operating cash flow minus capex was $271 million. This number was smaller than it otherwise would have been due to logistical constraints on the railway that serves the company.

In Q2, this number will be increasing because they will be able to further restore rail capacity, coupled with the average realized price to likely be higher.

My rough estimate for Q2’s dividend will be about $11.60/share, but this depends on certain variables being achieved. Barring a complete and total collapse of the commodity market, it will be around that figure.

However, to be “conservative”, let’s use the previous number – $8.11/share. Multiple this by four times and you get $32.44/share annualized. At Monday’s closing price of $131.32, that represents a yield of 24.7%.

Realize that the dividend represents half of the free cash flow available to the company before a reclamation (asset retirement obligation) reserve.

Obviously the stock market is going to find a 24.7% annualized yield to be very difficult to resist, no matter how vilified the sector of the company is. Pension funds, with both broad exposure in an equity market that is down about 12% year to date and a long bond market that is down 16%, have gotten murdered this year. They need an avenue for returns. The temptation is going to be too much for them to avoid.

Not surprisingly the stock market has decided it was too much to resist and Arch is up about 20% as I write this post. The yield at the implied $32.44/year dividend is now down to 20%.

The question is how much temptation there will continue to be going forward. Will yields compress to 18%? 15%? 10%? This really depends on how desperate the market is for a return, coupled with their impression on how durable the coal market is.

At the current pricing of coal, however, Arch will pay back its entire enterprise value to investors (either through dividends or share buybacks) in a couple years. Needless to say, this beats Microsoft equity which has a total return of 4%, based off of analyst estimates on their upcoming fiscal year ended June 2023.

Going back to my original topic of interest rates and cash flows – in a rising rate environment, present cash flows talk bigger than the promise of cash in the future. Arch (and other fossil fuel companies) is going to be a demonstration why.

A happy problem is to decide how to re-invest the cash flows. Internally for Arch, after building up a sufficient cash reserve, they will likely engage in some sort of equity buyback which will further juice up the stock price. So they are already making part of the decision for you with their cash flow, although there is a diminishing returns aspect to this decision (inevitably they will buy back too many shares at too high a price, like they did in 2019). It doesn’t mean that you have to re-invest your dividend into them – indeed, it would probably be a better time than to be stockpiling the cash for stormier days.