Short reads on commodity prices, tariffs and cash. We take a budget, a business case or a guidance statement, rebuild it as a few drivers, and stress test it with 10,000 scenarios. Then we show what the one number leaves out: the odds, the downside, and the two or three drivers that decide it. Each piece says what the markets set, what management sets, and what a hedge is worth. Every number from a run is a real run.
// public-company figures from annual reports and guidance, rounded · the private manufacturer is an anonymised composite · hedges are in the models where the company has them · the models and ranges are our illustrative assumptions · not investment advice
Eight drivers priced from the US commodity sheet. Six prices nobody in the building sets carry 61% of the swing. Free cash flow goes negative one year in three. A pricing clause beats a steel hedge.
Read the article → 02Tariffs · USAChina, Mexico, EU. The duty bill runs $9M to $17M and ties up $3M of cash before it hits the P&L. There is no hedge for a tariff, yet management controls more than half of the swing: pass-through and USMCA paperwork.
Read the article → 03Chemicals · EuropeOn our numbers the guided €8.0–8.4bn band is a one-in-four outcome. 84% of the swing is set by markets. One hedging decision moves the odds from 26% to 65%.
Read the article → 04Aerospace · EuropeThe market prices Airbus on one number. Even after the hedge book, the dollar moves EBIT about as much as the factory does, 45% versus 40% of the swing. A 3–8% delivery miss takes the guide off the table.
Read the article → 05Steel · GlobalEBITDA swung between $5bn and $19bn on the same plants. 81% of the swing is set in Shanghai and Rotterdam, so the lever is the balance sheet. Coverage of capex plus returns is three years in four.
Read the article →Each piece rebuilds a company as a simple driver model, the kind an FP&A team sketches on one whiteboard. Plan values are tuned to the public accounts, or for the private cases to a typical cost structure. Every driver gets a range, a shape and, where it matters, a correlation. Commodity and tariff ranges come from public forward curves and bank forecasts. Then 10,000 Monte Carlo scenarios per run through our engine. Sensitivities move one driver at a time between its P10 and P90. What-ifs re-run on the same seed, so the change in the answer comes from the assumption, not from sampling noise.
Hedging is taken into account. Where a company discloses a hedge, the range is the exposure after it: Airbus's currency book, ArcelorMittal's own mines. Where it does not, the base case is unhedged and a what-if shows the hedged book, so the hedge has a price instead of a footnote. Each article also splits the swing into what the markets set, what suppliers or the cycle set, and what management sets. The models and ranges are ours. None of it is investment advice.