Inventories are quantities of a physical commodity held within the supply chain for future use, processing, sale or delivery rather than immediate consumption. In storable commodity markets, they buffer flows that rarely match perfectly in time: when inflows exceed outflows, inventories can build; when outflows exceed inflows, they can draw down. In commodity-market usage, stocks is often used as a synonym for inventories, especially in official reports, but a published stock figure inherits the source’s own coverage and classification rules.
The reported number has a boundary
Published inventory data are therefore not necessarily a count of every unit of the commodity that physically exists. Petroleum statistics may measure primary stocks while excluding retail or end-use holdings; grain statistics may distinguish on-farm from off-farm stocks; exchange warehouse reports may divide metal by warrant or reporting status. Two series both described as stocks or inventories can consequently represent materially different pools of supply, so comparisons require consistent geography, product definition, storage coverage and observation date.
Physical existence is also different from economic availability. Material can be far from the relevant consumption or delivery point, constrained by quality specifications or logistics, or required for the operation of a storage system. Underground natural-gas storage illustrates the distinction particularly clearly: base or cushion gas is maintained to support reservoir pressure and deliverability, while working gas is the portion available for withdrawal. The headline quantity in storage can therefore be larger than the volume that is readily available to the market.
A build or draw is an observation, not a price signal
An inventory build means the measured stock increased over the observation interval; a draw means it decreased. That fact alone does not identify which flow caused the change or what prices should do next. Production, consumption, imports, exports, processing schedules, transport disruptions, seasonality, measurement adjustments and revisions can all affect the observed balance. Low inventories may accompany physical tightness and stronger demand for prompt supply, while high inventories may coincide with greater storage pressure, but the price implication depends on expectations, location, accessibility and prevailing market structure.
For research, the useful question is not simply “How much inventory is there?” but “What is being counted, where is it, how available is it, and against what seasonal or operational benchmark?” That discipline prevents a source-defined stock statistic from being mistaken for the market’s entire usable supply.