Copper-Indexed Pricing: How Wire and Cable Costs Move Between Price Books

Most categories have a price book that changes when the supplier decides to change it. Wire and cable has a price book that changes when a metals market moves, which is to say continuously, in both directions, without anyone sending you a letter. If you stock building wire, welding cable, battery cable, extension cords, speaker wire or anything else where a conductor is most of the mass, you are running a small commodity position whether or not you think of it that way.

The cost is two numbers, not one

A cable price is built as a base plus a metal adder. The base covers everything that is not conductor — insulation and jacket compound, machine time, packaging, freight, the mill's margin. The adder covers the copper in the product at whatever the mill's reference price is that week.

cost per 1,000 ft = base + (copper lb per 1,000 ft × copper price)

Mills express this in different ways. Some publish a price book quoted at a stated base copper price and issue adjustment tables when the market moves outside a band. Some republish the whole book on a fixed cycle — weekly, twice a month — with the metal already folded in. Some quote a firm price only for the life of a quotation, measured in days. The mechanics differ; the structure does not. What arrives in your inbox is a single number per line that is really two numbers stuck together, one of which is stable and one of which is not.

This matters because your instincts about supplier price files are calibrated on the other kind. A cost that rose 12% usually means a supplier decided to charge 12% more, which is permanent until they decide otherwise, and the correct response is to reprice. A cost that rose 12% on a copper-indexed line may mean nothing of the sort. It may reverse next month. It may also be a permanent increase wearing a metal move as a disguise — and telling those apart is the whole skill.

Splitting base from adder using files you already have

You do not need the mill's internal cost sheet. Two consecutive editions of the same price file, plus the reference copper price each one was built on, are enough to solve for the copper content per unit:

implied copper lb per 1,000 ft = (cost_new − cost_old) / (index_new − index_old)

Worked on three lines from one file, where the reference price moved from $3.60 to $4.20 per pound:

Once you have the implied content, the base falls out. For the 12 AWG line: 198.00 − (40 × 3.60) = $54.00 of base, confirmed by edition B at 54.00 + (40 × 4.20) = $222.00. Now you know that $54 of that price is a supplier decision and the rest is a market. At $3.60 copper the base is 27% of cost; at $5.20 it is under 21%. You are mostly buying metal.

The increase hiding inside a price cut

Here is why the test earns its keep. The next edition arrives with the reference price down from $4.20 to $4.05 — a fifteen-cent decline. Costs fall. Nobody reviews a file where costs fell.

The 10 AWG line should have come down $9.00 on 60 lb of content and came down $6.00. Three dollars per thousand feet of base increase went in while the headline number was falling. It is under one percent, it is invisible to a diff that only reports lines that went up, and it is the part that does not reverse. The next time the index climbs a dollar, you pay the new base on top of it.

This is the same structural blindness that makes a price-file process worth building: a diff that only surfaces increases answers the wrong question on an indexed category. What you want is the residual after the index is accounted for.

What the index does to margin at a fixed retail price

Take the 12 AWG line as a 250-foot coil, so cost per coil is a quarter of the per-thousand price. Retail sits at $79.00 and nobody has touched it in a year.

Margin is (price − cost) / price and the hold price is cost / (1 − target margin). From $3.60 to $4.40 — a move of eighty cents a pound, unremarkable by the standards of an industrial metal — the line goes from 37.3% to 27.2%. Ten margin points, no email, no announcement, nothing in the file that says "increase". If your floor is 30%, you crossed it somewhere around $4.15 and have been selling under it since.

Repricing something that moves both ways

The awkward part is that the obvious response is also a trap. Suppose you catch the move at $4.20, do the arithmetic, and raise the coil to $88.58. Two months later the index sits at $4.05, cost per coil is $54.00, and your margin on that line is 39.0% — above where you started and above the rest of the category. On a competitive commodity item where trade buyers price-shop by the coil, being 12% above the market is not a comfortable place to sit, and you will not notice, because nothing alerts on excess margin.

Chasing every edition in both directions is not the answer either. It burns hours, it confuses customers, and on a category where quotes go out and get accepted days later it creates a mess. What works is a policy with two triggers, written down once:

• Reprice down when the index-implied cost has fallen more than a set percentage — five percent is a reasonable starting band — and has stayed there across two consecutive editions. Two editions is what stops you from chasing a spike back down and up again.
• Reprice up immediately on a floor breach, regardless of band. A line under your floor is losing money now, and symmetry is not a virtue here: the cost of being briefly overpriced is a lost sale, the cost of being under the floor is a sale you would rather not have made.

Between those triggers, do nothing on purpose. That is a decision, and it is a better one than the accidental version where the line drifts for a year because nobody was watching. The general mechanics of the recalculation — and why markup percentages mislead here — are in recalculating margin and price after a cost increase, and the case for a per-family rather than per-catalog threshold is in choosing a margin floor. Indexed lines are a strong argument for a family-level floor: they need a wider band than a stable durable does, because they will spend part of every year moving through it.

What to record, and why the usual field is not enough

Shopify gives you one cost per variant, with no history and no context. On an indexed line that field is nearly meaningless on its own — a cost of $55.50 tells you nothing about whether that was a normal week or a peak. What makes the history interpretable is three things stored together:

1. The cost, with the file's effective date. Not the date you got round to importing it.
2. The reference index value the file was built on, if the mill states it, or the market price on the effective date if it does not. Without this, next year's you cannot tell a base increase from a metal move.
3. The unit basis. Per foot, per hundred feet, per thousand feet, per coil, per reel — and reel lengths are not always round. Mixing these produces cost errors of one or two orders of magnitude, and the direction that flatters you is the one you never catch. Normalising basis before the diff runs is the same discipline described for per-M pricing in industrial price books.

The rest of the copper catalogue

The mechanism is not confined to cable. Any line where metal is most of the mass moves the same way and deserves the same treatment: brass fittings and valves, transformers and motors, alternators and starters, welding leads, busbar, grounding hardware, some heat exchangers. For remanufactured automotive parts there is a second connection — scrap value is part of what makes a core worth returning, which is why core charges are not a fixed fact of nature either.

Aluminium conductor is the standing alternative in some applications and moves on its own index, usually with less amplitude in absolute dollars. Where both are genuinely usable, tracking the delivered cost of each is a real second-sourcing decision rather than a preference — the framework for costing that split is in second-sourcing suppliers. Where they are not interchangeable, ampacity and code decide, and the price comparison is beside the point.

A short routine

Four steps, run whenever a new edition lands:

1. Record the effective date and the reference index value alongside the file.
2. Compute implied metal content on three or four benchmark lines. Stable content means a pure pass-through; a shift means look closer.
3. Compute the residual — actual change minus index-explained change — per line, and sort by residual rather than by headline change. This is the list of things your supplier actually decided.
4. Re-run the floor check on every indexed line at the new cost, including the lines that went down, and act on breaches only.

None of that requires software. It requires that you keep the old files and write down the index, which is the part almost nobody does, because at the time it feels like admin rather than information. It becomes information about eighteen months later, the first time somebody asks why margin on the wire category is four points below where it was and there is a folder that can answer.