You price scrap in steel. The plant pays in hours.

Here is an uncomfortable observation from two decades of walking automotive and aerospace plant floors. A stamped bracket fails the go/no-go gauge, drops into the red bin, and finance books a €4.30 loss. The press hour that made it is gone. Nobody books that. Two trade outlets ran pieces within 48 hours of each other this week – Quality Magazine on machine time lost to scrap eating the bottom line, Quality Digest on the '20% tax' paid by anyone who tries to inspect quality in. Both are right. Both still under-count the damage, because the accounting underneath them is built to under-count it.

What the scrap report hides

Standard costing prices a rejected part at material plus direct labour. That is the number on the monthly scrap report, and it is fiction by omission. The press absorbed its hour at €380 whether the part was good or bad. The energy went in. The tooling wear went in. And in a sold-out plant – which most automotive plants are, otherwise you would not be running three shifts – a scrapped part is unsold product at full margin.

Run the arithmetic honestly. The part sells for €18, the steel in it costs €4.30, and standard cost books a €4.30 write-off plus a few minutes of operator time. Full cost says you lost the €13.70 contribution, the machine absorption, the energy – and the slot. Thursday's 14:00–15:00 press window does not reappear on Friday afternoon. Capacity is not inventory. It does not sit in a warehouse waiting for a better week.

You can re-melt the steel. You cannot re-melt the hour.

Scrap is a capacity problem disguised as a cost problem.

Costs get negotiated, deferred, trimmed. Hours simply vanish, one press cycle at a time, and the plant pays whether or not anyone signs the debit note. By the time the variance report lands, the hour is a month old and the press has moved on.

The inspection tax nobody budgets

The '20% tax' argument is just as real and just as invisible. Walk any plant with an unresolved quality problem and count the hidden factory: two sorters at end-of-line, a containment cage of suspect stock, a third-shift inspector re-checking parts the first shift already produced, an extra dimensional report because the customer stopped trusting your release. That is a factory you built, staffed and depreciated, whose entire output is the rediscovery of work you already paid for once. I have walked plants where it ran to double-digit headcount, and none of it showed up as a line item anywhere.

None of it makes a single part good. Every quality manual since Deming says you cannot inspect quality into a product; the budget process keeps pretending otherwise, because sorting labour hides in overhead the same way scrap hides in standard cost. Two line items, one shared delusion. The scrap bin, meanwhile, is the only department that never misses its output target.

Price it honestly and the projects fund themselves

At Witte Automotive I spent as much time on finance as on the shop floor, and it was time better spent than any slide deck. We ran A3 after A3 and stood up a Q-Wall, but the turning point was walking the controller through the full-cost maths of scrap: one month of rejects re-rated at machine-hour cost plus lost contribution, plus the wages of every pair of hands touching containment. When the denominator changed, the funding conversation changed. Projects that had been 'nice to have' cleared the hurdle rate in a single meeting.

Later, building the QA/QC department of a 900-plus-employee greenfield plant from zero, I front-loaded prevention deliberately: real PFMEA depth before launch, proper MSA on every gauge doing anything consequential, QRQC discipline on the floor. Where we needed containment, it carried a named end date on the QRQC action list. Defect costs came down 70%. QRQC's entire economic point is speed – process drift caught in hours instead of shifts, which means the machine hours come back. Drift caught in time is capacity you keep. Drift caught at final inspection is capacity you donated.

None of that required new theory. It required pricing the old theory correctly.

The cheapest capacity you will ever buy is the capacity you stop scrapping. The euro-per-part number lets finance nod and file the memo; the machine-hour number forces a decision, because hours are the one resource the plant cannot requisition more of by Friday. So print this week's scrap report, cross out the standard cost column, and take the machine-hour number to finance instead. In twenty years I have yet to see prevention lose an honest ROI comparison. The only thing it ever needed was an honest price tag on the alternative.