BASF guides EBITDA in a band of €8.0–8.4 billion. The band says where management aims. It does not say how likely that is. We built a simple driver model and ran 10,000 scenarios. On our numbers the band is a one-in-four outcome. Most of that comes down to one line: gas.
// figures from public reports, rounded · estimates and ranges are ours, not the company's · not investment advice
BASF is a €65 billion revenue business built around integrated sites. Crackers at the front, hundreds of downstream plants behind them. Its biggest cost is feedstock and energy, a bill in the high teens of billions. That bill is set by the gas market, not by BASF.
So a single-point forecast makes little sense here. Management knows this and guides a band instead. That is the honest choice. The band only leaves one question open: how likely is it?
Rounded: revenue around €65bn. EBITDA before special items around €8bn. Guidance in a band about half a billion wide. The call adds an oil price, a euro-dollar rate and a GDP number. It does not add a probability for the band, or a sensitivity you can rebuild.
That is the gap we filled. We rebuilt the P&L as drivers, gave each driver a range, and ran it.
Six drivers. EBITDA = volume × price − feedstock & energy − other variable costs − fixed costs. Plan values tuned to the public accounts give €8.2bn, the middle of the guide. So the model aims where management aims. It just adds a spread around that aim.
| Driver | Plan | Range | Shape | Why |
|---|---|---|---|---|
| Gas & feedstock | €18bn bill | −20% … +45% | triangular | gas falls slowly and spikes fast, so the range is skewed up |
| Price level | €65bn rev | −6% … +5% | triangular | chemical prices follow energy with a lag; ρ +0.3 with gas |
| Volumes | index 1.00 | −4% … +3% | triangular | higher prices cost some volume; ρ −0.2 |
| Fixed costs | €13.8bn | −2% … +2% | uniform | the one line management fully controls |
You may set the gas range differently. That is fine. The gas range is the real debate, and it is a better use of an hour than arguing about the mid-point.
Hedging is in the model. BASF buys part of its gas on longer contracts and passes part of the cost through to customers with a lag. The pass-through is the +0.3 correlation between gas and price. The contracts are not disclosed as a hedge ratio for the year ahead, so the base case runs the full market range on the bill, and the what-if further down shows the hedged book with gas capped at +10%. The distance between the two is what the hedge is worth. If you know the real hedge ratio, the answer sits between them.
Guidance band versus the simulated range
EBITDA before special items, €bn. Blue bar: P10 to P90. Dot: P50. Light band: the guided €8.0–8.4bn.
The band sits in the top quarter of the distribution, not in the middle.
On our ranges the band is not the central case. It is the optimistic quarter. Landing inside the half-billion band is a one-in-twenty result. That is not because management is wrong. It is because the honest spread on European gas is much wider than the band. The band was never meant to carry that information.
What moves the year
Share of the total EBITDA swing, one driver at a time between its P10 and P90.
Gas drives half the swing. Fixed costs, the line most cost programmes work on, drive 4%.
Who sets the driver
The same swing as above, grouped by who sets the driver. Bars: share of the total swing.
84% of the swing sits in prices set at the TTF hub and in the chemical market.
Management cannot control 84% of this year. It can control how much of that reaches the P&L. The levers on the part it does not set are the hedge ratio on the gas book, indexed pricing so that price follows gas faster than a quarter, and the balance sheet it carries into a bad gas year. None of those is a cost programme. The what-if further down prices the first one.
Bear (P10, about €3.0bn): gas spends the year near the top of its range. Prices recover only part of it, with a lag. Volumes slip. Nothing unusual happened. The drivers just moved the same way at the same time. This is the case the balance sheet has to be sized for.
Base (P50, about €6.4bn): an ordinary year that still lands €1.6bn under the bottom of the band. The cost side is skewed against the plan, and the plan does not say so.
Bull (P90, about €9.4bn): gas falls and prices hold. Note that the upside clears the top of the band by a billion. The band is too narrow in both directions.
Clip the gas range to −20% … +10%, the effect of a hedged book. Everything else unchanged, same seed.
Chance of reaching €8.0bn
Same model, same seed. Only the gas range changes.
The P10 rises from €3.0bn to €6.4bn in the hedged case.
One risk decision moves the guide from a long shot to better than even. It also lifts the bad-year floor by three billion euros. Whatever the hedge costs, you can now compare it with a number.
The gas bill is paid every month. The price recovery comes a quarter or more later, if it comes. In the bear case that gap is a working-capital draw before EBITDA shows any of it. A CFO running this distribution would size the revolver on the P10 gas year, not on the guided band.
Three questions, none of them "which end of the band": What probability do you attach to the band? What share of next year's gas is hedged, and at what cap? At what gas price does the band stop holding? A CFO who runs a distribution internally can answer all three quickly. Those answers are worth more than the band itself.
A guidance band is a distribution without the probabilities. The ranges behind it can be rebuilt from public accounts in an afternoon. Analysts will do this anyway. It is better to know the odds before they do.
// method: simplified driver model, plan values tuned to public accounts; 10,000 Monte Carlo scenarios per run through the ScenarioCast engine, correlations and hedges as stated in the model section; sensitivities one-driver-at-a-time at marginal P10/P90; what-ifs re-run on the same seed. Figures rounded. Our estimates, not the company's. Not investment advice.