Manufacturing Hedging: Six Pain Points Across Both Sides

Manufacturing hedging differs from trading in one fundamental way: a manufacturer carries two risks pointing in opposite directions at the same time, and they never offset each other.

Upstream is raw material cost. Downstream is finished-goods price. When raw material rises while your selling price is locked, margin is squeezed. When the selling price falls after the material is already in the warehouse, margin is squeezed again.

Many manufacturers believe that having exposure on both sides makes them safe. That is a dangerous misunderstanding. The two exposures rarely line up on timing, quantity or product, so they do not cancel out. They erode profit in both directions at the same time.

Pain point 1: upstream exposure — procurement always runs half a step behind

The exposure sequence in manufacturing usually looks like this: sales takes the order in January, production schedules in February, procurement runs in March, delivery happens in May. The futures hedge, however, has to be in place in February. That means you are using a February judgement to hedge a cost that crystallises in May.

If the hedge volume is based on how much raw material you used last month, it lags sales by construction. If it is based on the sales plan, and the sales plan itself carries a 20%–30% adjustment range, then the hedge volume starts out with error already built in.

The most common mistake here is letting management set the hedge ratio by decree. "Raw material is a large share of our cost this year, so we hedge 70% across the board." The number sounds reasoned, but it does not map to any specific product, order or delivery window. It rescues you when prices rise and creates fresh mismatch when prices fall.

Pain point 2: downstream exposure — inventory wears two hats

Manufacturing inventory has two identities: it secures continuity of production, and it carries price risk. The financial statements never separate them, but operationally you have to manage them separately.

When prices fall, manufacturers go through a painful phase: high-cost stock takes an impairment at the same time as products already sold at a fixed price have their margin compressed. If the hedge is not in place either, that is a triple hit.

The difficulty in practice is that you rarely know how much you can lock. A sell hedge presupposes a future batch of goods to sell, but capacity moves: orders get cancelled, lines go down for maintenance, peak season brings output cuts, and raw material supply breaks. Too much locked is speculation, too little is naked exposure, and in manufacturing that boundary is far blurrier than it is for a trading house.

That is why the manufacturing hedge ratio should not be a static target but a set of tiered rules: signed orders locked at their confirmed quantity, pipeline orders locked at a probability discount, work in progress locked by completion progress, inventory locked by days of turnover. Each tier has its own coefficient. Write the rules down clearly and the hedge volume follows by itself.

Pain point 3: capacity swings turn locked positions into mismatch

This risk is specific to manufacturing. Trading houses largely do not have it.

Typical scenario: the company locks 70% of raw material hedging based on annual order expectations. Halfway through the year an important customer switches its order away and order volume falls 25%. The raw material hedge is still there, but the physical demand behind it is gone — what was a hedge has become a speculative position.

You cannot handle this with judgement in the moment. You need disposal rules preset in policy: what size of order change triggers a position adjustment, how much to adjust, who approves it, and within what deadline. Without that rule, every order swing leaves you choosing between two bad options: do not close, and the exposure turns speculative; close, and you take a real loss at exactly the volatile point.

Pain point 4: hedging sits apart from performance assessment, so the policy idles

This is an organisational problem, and the hardest one to solve.

If a sales person's performance is measured on current-period gross margin alone, then a floating loss on the hedge is pure negative. The rational response is not to hedge: no hedge means no floating loss, and the numbers look good at least for the assessment period.

A subtler version is hedging by halves. In slow months you do not hedge, because the floating loss would drag down this month's number; in peak season you hedge hard, because deliveries are due. The result is a hedge ratio completely disconnected from risk — an arbitrary operating habit.

The fix is not to criticise the salesperson. It is to change what the assessment looks at: show hedging contribution separately from gross margin, and bring exposure coverage quality into the assessment. Once the value of hedging can be seen on its own, hedging stops being self-punishment against performance.

Pain point 5: most manufacturers hedge only half the book

In manufacturing diagnostics we keep meeting the same pattern: a large share of companies hedge the upstream side and never hedge the downstream side.

The reason is understandable: raw material inflation is the issue every finance person raises, while "our product may sell for less" is never mentioned in a good year. But in the year demand weakens, an upstream-only hedge means cost is locked while price falls — hit on both sides.

Another common asymmetry: futures are used upstream while basis is never touched. Many manufacturers handle upstream exposure through long-term contracts or pricing agreements, and use futures on only a small number of products. That is not wrong, but you need to know explicitly which products you have left exposed.

Pain point 6: multi-site, multi-entity groups hide exposure behind internal trading

A group manufacturer has one distinctive feature: a single plant's margin swings wildly while the group total looks stable. That is because internal trading moves risk between entities.

The buying centre purchases heavily when prices are low; the selling centre sells heavily when margin is high. Both numbers look good, yet at group level the company is net exposed across a price band. What these companies need is not plant-level hedging but a net exposure figure after internal transfer pricing.

Five common manufacturing profiles and what each one needs

Manufacturing is not a homogeneous population. The table below sorts companies by operating model; each profile has its pain in a different place and its own priority for the fix.

Three questions you can ask yourself now

1. If raw material prices rise 15% next year, what share of your procurement does your current hedge position actually cover? Is that number calculated, or estimated?
2. If order volume falls 20%, what happens to the hedge positions you have already locked? Who decides, and when does it execute?
3. Is your inventory write-down risk covered by a hedge, or merely hidden behind an accounting provision for inventory decline?

If you cannot answer any of the three, or the answer is "I don't know", your risk management is still running on instinct.

One caution about a judgement that is easy to get wrong: hedging is not a universal remedy for manufacturers. If your main profit comes from capacity utilisation, yield or channel, hedging can only help so much. Hedging addresses one variable, price. When price is not your main variable, fixing the operating business first is the right order of work.

Sources and notes

The framework used in this article — the structure of manufacturing hedging, the two-sided nature of the exposure, and the analysis of hedge ratios, inventory tiering and assessment misalignment — comes from RMC's own observations and generalisations in corporate commodity risk management diagnostics. The industry practice data cited here — night-session pricing at roughly 35% of requests, 5–10 minutes of manual pricing per contract, 8–12 minutes of manual attribution per trade, and error rates of 1%–3% — comes from RMC's diagnostic records for manufacturing clients. This article has been reorganised and rewritten from RMC's diagnostic perspective. No specific client names or data are used, and every judgement here can be verified by the company itself. Nothing in this article constitutes a recommendation or evaluation of any specific software product.