Cleveland-Cliffs’ adjusted earnings margin strengthened

In brief

• The adjusted earnings ratio calculated from the new report strengthened.
• The company metric and an investor’s share return are separate questions.
• Our currency model shows the condition required to break even.

Calculated from Cleveland-Cliffs’ report of 19 October, adjusted EBITDA as a share of revenue increased from 5.47% in the second quarter to 10.91% in the third. A Hungarian retail investor needs to consider both the metric’s accounting limitations and their own currency exposure to interpret the improvement. We examine the US steel company’s new results, not a share quote or a pre-opening market report.

Both the denominator and the adjustments matter

We use group revenue and the adjusted earnings metric from the original company statement. Both rows are in million USD; the periods are consecutive quarters of the same year.

• 2026 Q2: revenue 5,226; adjusted EBITDA 286.
• 2026 Q3: revenue 5,612; adjusted EBITDA 612.

Our formula divides adjusted EBITDA by revenue and multiplies by one hundred. The difference between the unrounded ratios is 5.43 percentage points. The change is therefore neither the share’s percentage return nor the growth rate of the company’s total sales. We keep the same denominator and the same adjusted metric throughout the comparison.

The chart is not copied from the company: it uses our own division of the source data, with rounding for display. Adjusted EBITDA means earnings before interest, taxes, depreciation and amortisation with further company adjustments. It should not be described as cash or earnings attributable to shareholders. The release’s reconciliation table helps trace the items separating it from the accounting result.

Chairman and Chief Executive Lourenco Goncalves used the phrase “strongest quarter in three years” in the release. This is a verified management assessment, not independent evidence of the next period’s outcome. Our narrower conclusion is that more adjusted earnings accompanied the revenue reported; the ratio does not produce a complete breakdown of the causes.

Whether the improvement lasts can also depend on subsequent pricing, sales mix and expenditure. An aggregate ratio does not separately reveal the margin on a new order. For a Hungarian industrial supplier, that is a useful methodological distinction: the US issuer’s figures are neither a domestic steel price list nor a guaranteed demand forecast.

The currency condition for breaking even

Our analysis of J.B. Hunt’s earnings ratio and report on US foreign-trade prices show other aspects of denominators and costs. We do not rank the companies against one another. The comparable idea is that operating performance, an asset’s price and an investor’s cash flows are distinct layers.

The Investor.gov international investing guide treats currency as a separate risk. Our example therefore asks what USD total return would be required to leave the ending value measured in HUF unchanged. The model does not estimate Cleveland-Cliffs’ expected performance and uses no actual exchange rate.

We assume a one-year holding period and a single cost of 1% deducted from the ending value. The USD/HUF change is minus 5%, zero or plus 5%. The USD return required to break even is then 6.33%, 1.01% or minus 3.80%. An increase in USD/HUF means the domestic currency weakens under this notation.

The second chart shows conditional thresholds, not a promised return. A negative threshold indicates that the model’s currency gain can offset some asset loss and the cost. That does not make a loss-making asset a good choice: the investor’s objective, horizon and tolerable risk still require separate consideration.

A reproducible method and an open outlook

The formula is one divided by the product of the currency factor and 0.99, minus one; multiply by one hundred for a percentage. The currency factors are 0.95, 1 and 1.05. Every other assumption remains unchanged, so the specified currency path alone causes the difference. Rounding should come only at the end of the calculation, otherwise checking it can produce an unnecessary discrepancy.

The cost here is a single deduction proportional to ending value. In a real account, purchasing, selling, currency conversion, account maintenance and custody charges can arise at different times. The Investor.gov fees guide emphasises understanding total costs. Our example is not a provider’s fee quotation and contains no adjustment for taxation or inflation.

In a favourable operating scenario, the higher adjusted earnings ratio could persist while revenue’s ability to generate cash also improves. In the risk scenario, less favourable product mix or costs could compress the ratio. These are analytical possibilities, not established future events. We attach no invented probability, price target or instruction to buy or sell.

Valuing the share would also require an entry price and subsequent cash flows. Without a verified current price, we therefore publish no valuation multiple. Past improvement does not automatically translate into future investment returns; market expectations could already have anticipated part of the change.

Timing matters when reading the ratio as well. Seasonal demand, maintenance or contractual repricing can affect comparisons between consecutive quarters. Our division does not separate those influences, so we do not mechanically annualise the result. Establishing lasting improvement would require examining more periods and consistently defined metrics, taking changes in the company’s scope into account.

An investor’s checking log can usefully separate the company source, their own calculation and the provider’s account statement. The first shows reported data, the second the inference, and the third the fee actually deducted. If they differ, units, reporting periods and cash-flow dates should be reconciled first. Clear documentation can reduce misunderstanding but does not eliminate the risk of investment losses.

Currency hedging would require separate product knowledge and cost analysis. We assume unhedged exposure, so no hedging transaction is included in the threshold. Contributions or sales in several instalments would likewise require a different calculation: instead of a single initial investment, each cash flow has its own date and conversion. The model answers one precisely defined question, not every possible account history. That limitation matters particularly when reconciling it with a provider’s statement.

The next material question is how adjusted earnings turn into cash flows and how the items explaining the difference evolve. Further company disclosures can be used to examine that. The source data and formula can be recalculated now, while the next period’s revenue, the future currency path and actual share returns remain open.