Ramaco's Gallium MOU Makes News. It Doesn't Make Cash Flow.

Ramaco announced this morning that it had signed a non-binding memorandum of understanding with Bedrock Semiconductor to negotiate supplying gallium feedstock from its Brook Mine project in Wyoming. Read as a headline, it is an easy story to like: gallium is a critical mineral with a genuine U.S. supply-security narrative, more than 95% of global production is Chinese, and Ramaco is recasting itself as a domestic supplier to the semiconductor age. The shares' reaction was the more honest news. They were down roughly 8% on the morning the announcement crossed the tape, a market verdict I would not argue with.

Read the document the way a cash-flow analyst has to. The MOU outlines plans to negotiate a supply arrangement, and beyond that it commits little. Any actual supply or offtake deal is expressly subject to due diligence, the negotiation and execution of definitive agreements, and, most importantly, "successful development of the Brook Mine project." That last clause is doing all the work, because the Brook Mine is an exploration-stage property that currently produces nothing. This is also not a new genre of news for Ramaco. In late May the company signed a similar non-binding arrangement with REalloys that envisions Ramaco supplying mixed rare earth carbonate from the same project, and management has hinted at further MOUs before year-end. Understood for what it is — a handshake and a negotiation, not a contract and certainly not a dollar of revenue — the Bedrock announcement is a rounding error for valuation.

The numbers that actually matter arrived two weeks ago in the second-quarter report. Revenue was $144.8 million, down 5% year over year. The company posted a consolidated net loss of $15.4 million, or $0.26 per diluted Class A share, roughly in line with the year-ago loss. Adjusted EBITDA — earnings before interest, taxes, depreciation, and amortization, a rough cash-earnings proxy — collapsed to $5.7 million from $9.0 million. Realized met-coal prices averaged $116 per ton, down 6%, and cash margin fell to $17 per ton from $20. Management cut its full-year production and sales guidance. Widen the window and the picture is worse: over the trailing twelve months, operating cash flow is negative, free cash flow is deeply negative at roughly minus $135 million, and the EBITDA margin is effectively zero while return on invested capital sits near negative 9%.

To be fair to the operation, this is a low-cost, well-run coal business. Cash costs of $99 per ton marked a fourth consecutive quarter under $100, and Ramaco entered the third quarter with roughly 97% of the top end of its revised production guidance already booked, including 2.5 million tons fixed at an average near $121 per ton. But low cost is not the same thing as profitability when prices fall, and right now the coal engine that is supposed to fund the transformation absorbs more cash than it generates. A business that sells first-quartile-cost coal and loses money anyway tells you exactly where prices sit, and exactly how much of the current enterprise value depends on a project that does not exist yet.

That project remains a very large, very distant option. The latest Hatch conceptual study — internal modeling, not a reserve statement — sketches a Brook Mine worth $3.4 billion to $8 billion in net present value and $600 million to $1.3 billion in average annual EBITDA, with 75% of projected revenue tied to semiconductor-adjacent critical minerals such as gallium, germanium, and scandium rather than the rare earths that get the headlines. The same study prices construction at $3.2 billion plus another $0.8 billion in contingency, targets first production around 2031, and will not deliver a pre-feasibility study until spring 2027. The pilot plant that has to validate this process at commercial scale does not run until 2027, and no mineral resource has yet been converted to reserves. None of this makes the Bedrock news wrong; it makes it early. Gallium from coal is still a hypothesis until pilot data says otherwise.

That leaves the question I care about most, behind the headline and the hype alike: who pays for it, and can today's cash flow survive meanwhile? The balance sheet passes the survival test, and I want to give it full credit. Liquidity at June 30 was more than $400 million, including $282.5 million of cash and roughly $118 million of unused revolver capacity, and the company raised on the order of a billion dollars over the past year to fund the Brook Mine. But the ledger also shows roughly $650 million of total debt against a stock market capitalization near $716 million, with net debt around $170 million and equity of only about $376 million. And the equity cushion that raise created has been spent aggressively: $66 million of repurchases in the first half at an average price near $14.44, for shares that now trade around $11.30, while the Class A cash dividend sits suspended. I can see the logic of buying back stock below the highs. I would not mistake a leveraged bet on optionality for a cash-flow statement.

The valuation obliges the same honesty. At about $11.30, Ramaco trades near 1.9 times book and 1.39 times sales with negative earnings, no usable P/E, and no meaningful EBITDA multiple, because on a trailing basis there is no EBITDA to multiply. For a cash-flow buyer the standard toolkit does not even apply, and that absence is itself the finding. Cheap is not the right word for a stock whose downside is bounded not by earnings but by whether a $4 billion, 2031-dated gallium-and-rare-earth bet works. Contrast it with an income name in the same sector that still trades on real numbers: Alliance Resource Partners, per market data, sits around 6 times trailing EV/EBITDA with a 9.4% dividend yield and a positive earnings multiple. That is a trade a cash-flow discipline can underwrite today. Ramaco is a different species entirely.

The market, to its credit, has already done much of the deflating. Ramaco shares are down roughly 37% year to date and more than 50% over the past twelve months, against a 52-week range of $8.56 to $57.80; the crowd paid $57.80 for the rare-earth story, has given most of it back, and greeted this latest piece of the same story with a shrug and a sale. That haircut means much of the bearishness is in the price, and the option is not worth zero. But a fallen speculation is still a speculation until the cash flows materialize, and none of the four events that would turn this from story into investment have occurred: conversion of resources into reserves, pilot-scale proof of gallium extraction, committed financing for a multi-billion-dollar construction bill, and a return to positive consolidated free cash flow from the coal business that is supposed to carry the journey. A non-binding memorandum of understanding with a semiconductor company is a negotiation, not earnings. It restores no dividend, and it does not reduce the $4 billion the mining bet will ultimately need. The margin of safety that deep-value investing demands is precisely what this stock lacks right now, so I am watching from the sidelines and will revisit Ramaco when the story starts producing numbers instead of press releases — either after the facts have earned a higher price or, more likely, at a better price before they do.