Supply Chains
Lean or Buffered: How Firms Decide How Much Stock to Hold
Inventory is money sitting still, and stockouts are sales that never happen. The balance between them rests on assumptions that have shifted.
This is less a set of instructions about the trade-off behind inventory levels than an argument, and it is worth saying so at the start.
The argument in brief
- Inventory consumes working capital, storage and risk of obsolescence.
- Just-in-time works best where lead times are short and reliable.
- Buffer placement matters more than total buffer quantity.
What inventory actually costs
Stock ties up cash that could be deployed elsewhere, and the cost of that capital is the largest component of holding cost for most firms. Storage, handling, insurance and shrinkage add to it, and for technology or fashion goods obsolescence dwarfs all of them.
Because these costs are continuous and stockout costs are episodic, the pressure inside a firm usually runs towards holding less. Finance functions can measure inventory precisely and can measure lost sales only by inference, which reinforces that pressure. The asymmetry in measurability, not a difference in economic importance, is what tilts many inventory decisions.
The case for running lean
Holding less stock frees capital, exposes quality problems quickly and forces the process discipline that keeps a line running smoothly. Defects hidden inside a large buffer surface immediately when the buffer is removed, which is a feature of the method rather than a flaw.
The approach depends on short, reliable replenishment, which in turn depends on nearby suppliers or dependable transport. Where those conditions hold, lean operation delivers lower cost and higher quality simultaneously rather than trading one for the other. Where they do not, the same practice transfers volatility onto the shop floor and produces frequent interruptions.
What changed the calculation
When transport is cheap, punctual and predictable, holding inventory looks like a waste of capital and lean logic prevails. When transit times become variable, the value of a buffer rises because it absorbs uncertainty the transport system no longer absorbs. The relevant variable is not average lead time but its variance, since safety stock is sized against variability rather than the mean.
Over a shipping cycle, a route that is slow but consistent can require less buffer than a fast route with an unpredictable tail. Firms that measure only average transit time therefore systematically under-size their buffers on volatile lanes.
Where to put the buffer
Holding finished goods buffers demand but commits you to a specific configuration that may not be what customers order. Holding components buffers supply while preserving flexibility, since the same parts can become several different finished products.
The decoupling point is where the chain switches from forecast-driven to order-driven, and moving it changes both cost and responsiveness. Pushing the decoupling point later usually reduces total inventory for the same service level, which is why late configuration is popular.
Buffer placement often matters more than buffer quantity, and it is the cheaper of the two levers to adjust.
Sizing a buffer honestly
Safety stock is a function of demand variability, supply variability and the service level the firm has chosen to offer. Raising the target service level towards completeness raises required stock steeply, because the last few percentage points cover rare events. That curve is why universal high availability is expensive and why firms differentiate service levels by product importance.
Segmenting the catalogue and holding deep stock only on critical or fast-moving lines is usually better than a uniform policy. A single company-wide service target is almost always the wrong answer applied to thousands of different products.
Tariff schedules are technical documents, and classification disputes turn on wording rather than intent.
The strategic layer
Beyond operational safety stock, some firms hold strategic reserves of inputs whose supply is concentrated or politically exposed. That decision is closer to insurance than to inventory management, and it should be evaluated as a premium against a scenario.
Holding a year of a cheap critical component costs little and buys time to qualify an alternative if the source fails. Holding the same duration of an expensive, perishable or rapidly obsolescing item is rarely defensible on the same reasoning. Distinguishing the two cases prevents a sensible principle from being applied where it destroys value.
The takeaway
Size buffers against variability, not against average lead time, and place them where flexibility is highest. This is general information, not financial advice.
Capacity takes a decade to build and one quarter to look like a mistake.
Questions readers ask
Is just-in-time discredited?
No, but its preconditions are better understood. It performs well with short reliable lead times and less well where supply variability is high, which is a statement about context rather than a verdict.
How much buffer is the right amount?
It depends on demand variability, supply variability and the service level chosen. There is no universal figure, and applying one target across a whole catalogue usually wastes money and misses sales at once.
