After two years of building stock, the chemicals industry spent 2023 unwinding it, hard. The swing was not bad luck; it was the bullwhip, amplified by batch runs and volatile raw materials.
Source: AlixPartners, The great destocking (2023).
When supply was tight in 2021 and 2022, buyers all along the chain ordered ahead and padded safety stock to protect themselves. When demand softened, that padding reversed all at once. AlixPartners estimated specialty chemical players had to unwind more than $140 billion of excess inventory, with demand forecasts cut 10 to 30 percent against plan. That is the bullwhip effect written large: small changes in end demand amplified into violent swings upstream, paid for in cash and idle capacity.
Chemicals do not flow one unit at a time. You run a batch, which means minimum campaign sizes, changeover costs and a strong pull to 'make a full run while the reactor is set up.' That economic logic is sound until demand turns, at which point a full campaign of a now-slowing grade becomes a tank of working capital you cannot easily move. Planning has to weigh the run economics against the real forward demand, not just the cost per kilo of a big batch.
Many formulations, coatings, adhesives, specialty blends, have shelf-life and stability limits. So the inventory is not just slow; some of it expires. A grade overbuilt in a strong quarter can pass its stability window before the market recovers, turning excess into write-off. The buffer has to respect the clock, which means sizing it on real demand and real stability, not on reactor convenience.
Feedstock prices move with energy, petrochemicals and tariffs. Buy raw materials on a reactive, top-up basis and you ride every spike; buy them against a reproducible demand plan and you can time and size the purchase deliberately. In a sector where raw materials are a huge share of cost, the quality of the demand signal behind procurement is a direct lever on margin.
The producers who came through the destocking in better shape forecast demand reproducibly, so campaign sizing is weighed against real forward demand rather than reactor economics alone. They size buffers on real variability and respect stability windows, so overbuilt grades are flagged before they expire. And they plan raw-material purchasing on that same demand signal rather than reacting to every price move. It is the deterministic, auditable approach DemandIntel brings to chemicals and coatings makers, running on your own data, on top of the ERP.
Buyers padded orders and safety stock during the 2021-22 shortages, then reversed it all at once when demand softened. That bullwhip left specialty chemicals unwinding more than $140 billion of excess, with forecasts cut 10 to 30 percent versus plan.
Minimum campaign sizes and changeover costs pull toward full runs, which is efficient until demand turns and a full batch of a slowing grade becomes trapped working capital. Planning weighs run economics against real forward demand, not cost per kilo alone.
Size buffers on real demand and respect each grade's stability window, and flag overbuilt grades before they pass it, so excess is sold or redeployed rather than written off as expired.
Yes. Buying feedstock against a reproducible demand plan lets you time and size purchases deliberately instead of riding every price spike, which matters because raw materials are such a large share of cost.
The destocking was not a surprise. It was a buffer everyone built in good faith, unwinding all at once, because nobody's number could be trusted.
Picture your next S&OP meeting: one number on screen that nobody argues with. See it free, on your own data, in 45 minutes; and if it does not hold up, we part as friends.