Cleveland-Cliffs Q3: profit improves, cash conversion still matters

A profitable quarter leaves a cash conversion question

Cleveland-Cliffs reported a return to quarterly profit on October 19. For the three months ended September 30, 2026, consolidated net income was $165 million and adjusted EBITDA, a non-GAAP measure, was $612 million. The useful follow-up for CLF investors is how much of that improvement can become cash available to reduce debt.

The release reports $224 million of operating cash flow and $152 million of purchases of property, plant and equipment. Subtracting those two reported amounts gives $72 million. This is a simple operating-cash-less-capital-spending calculation, not adjusted EBITDA or a claim that every dollar was available for shareholders.

The quarter supports an improved operating assessment. It also shows why a much larger forward earnings number needs a separate cash test. An investor who uses adjusted EBITDA as a proxy for immediate debt repayment would skip both the cash costs below that measure and the timing of customer collections and inventory investment.

Measure the improvement against the right baseline

The company's July 23 release forecast approximately $575 million of third-quarter adjusted EBITDA, following $286 million in the second quarter. The actual result therefore exceeded management's prior guide by $37 million, or about 6.4%. That comparison is against the company's own forecast; it is not evidence of a beat against analyst consensus.

The July report also provides an operating baseline: second-quarter steel shipments were 4.025 million net tons and average selling price was $1,124 per net ton. In the latest quarter, those measures reached 4.150 million and $1,177. The changes point to both volume and realization contributing to the improvement.

For modeling, preserve the difference between net tons and metric tons, and between steelmaking segment figures and consolidated results. Multiplying an average steel price by shipments will not automatically reproduce every component of consolidated revenue. A model should explain the scope of its inputs before using the resulting margin as a valuation anchor.

Receivables and inventory explain why earnings are not collections

The third-quarter cash-flow statement shows $116 million used by receivables and $170 million used by inventory changes. Payables and related accrual changes provided a partial offset. Those operating movements help explain the gap between the earnings improvement and cash retained after investment.

A working-capital outflow is not, by itself, evidence of poor receivables quality or unsalable inventory. It can accompany an expanding business. The next question is whether collections and inventory turnover subsequently release cash, or whether continued growth keeps requiring funding. Neither outcome should be assumed from a single quarter.

A practical follow-up is to compare subsequent operating cash flow with sales, customer balances and inventory together. If profits rise while those balances repeatedly absorb the improvement, the debt-reduction schedule needs more scrutiny. If cash conversion strengthens without an unsustainable reduction in required investment, the earnings recovery becomes more useful to creditors and equity holders.

The 2027 illustration contains different kinds of assumptions

Page 10 of the earnings presentation starts with roughly $2.45 billion of annualized third-quarter adjusted EBITDA. It adds $350 million from index-contract discount and extras changes, $850 million from fixed-price resets and $1.1 billion from prevailing price forecasts, then subtracts $250 million of cost impacts. The illustrated endpoint is approximately $4.5 billion.

Management identifies the price-forecast component as the current futures curve and Canadian order book. That is a materially different basis from a fixed-price contract reset. The chart is an illustrative bridge, and its starting point annualizes one quarter; it does not establish a fully contracted annual earnings floor.

Using the chart's rounded figures, removing only the $1.1 billion price contribution reduces the endpoint to about $3.4 billion. This is a mechanical sensitivity, not a prediction: costs, demand and contract realizations could move together. Its purpose is to expose how much of the illustration depends on the pricing environment rather than to invent a second company forecast.

When evaluating the outlook, keep contract terms and market-price assumptions on separate lines. Even an improved contractual price formula still needs actual customer purchases and operating execution before it produces earnings. Revisit each line when new disclosures arrive instead of treating the entire bridge as one confirmed milestone.

What would strengthen or weaken the investment case

The release guides to about $715 million of fourth-quarter adjusted EBITDA. That provides a nearer checkpoint than the full-year 2027 illustration. Investors can compare the eventual result with the announced expectation while separately assessing cash generation and changes in borrowing.

Stronger evidence would combine sustained operating improvement with better cash conversion and documented debt reduction. Weaker evidence would include rising funding needs despite improving reported earnings, or a deterioration in the price assumptions supporting the longer-term outlook. These are conditions to monitor, not outcomes established by today's release.

This analysis uses the earnings release, its presentation and the prior-quarter release. All three originate with Cleveland-Cliffs; they provide primary disclosure rather than independent confirmation of management's forecasts. It makes no claim about an intraday share-price reaction or fair value. The research task now is to reconcile the operating recovery with cash delivery, using dated results rather than assuming the forecast has already been earned.