Quality
Manufacturing Scrap Rates by Industry: What the Benchmarks Actually Support (and What They Don't)
Most published scrap-rate benchmarks trace back to nobody. Here are the ranges that are actually defensible by process type, why a single number per industry is a fiction, and the arithmetic that turns your own rate into dollars.
Search for manufacturing scrap rate benchmarks and you will find the same suspiciously round numbers repeated across a dozen sites, none citing a source. The honest situation is this: there is no government series for scrap rates, no audited cross-industry survey published on a schedule, and the definitions vary so much between plants (is rework scrap? is startup scrap counted? is it measured in pieces, weight, or dollars?) that a single industry number would be meaningless even if someone measured it. What can be stated honestly is a range by process physics, because the process sets the floor: how much material a process must remove, how it starts up, and how it fails.
Ranges the process physics will defend
Typical total material loss by process, stated as ranges with the driver named
- Machining from billet or bar: 20% to 60% of input material leaves as chips by design. This is engineered yield loss, not quality scrap; quality rejects typically add another 1% to 5%.
- Injection molding: runners, sprues, and startup purges commonly consume 2% to 10% of resin, less with hot runners and regrind reuse. Reject rates on a stable process typically run under 2%.
- Stamping and blanking: skeleton and edge trim commonly take 10% to 30% depending on nesting, which is why blank-nesting software pays for itself. Piece rejects are usually low single digits.
- Casting and foundry work: gates, risers, and remelt returns can cycle 30% to 50% of poured metal, most of it recovered in-house. True scrap (rejected castings) commonly runs 2% to 8%, higher on complex geometries.
- Electronics assembly: board-level scrap is typically under 1% at mature volume, but a scrapped assembly late in the line carries the full accumulated component value, so the dollars concentrate even when the percentage looks small.
A scrap percentage without a definition attached is not a benchmark. It is a rumor with a decimal point.
Why the dollars matter more than the percentage
Two plants with identical 3% scrap rates can be losing very different money, because scrap cost equals the rate times the value accumulated at the point of failure. A part scrapped at final inspection carries material plus every operation performed on it; the same defect caught at the first operation costs a fraction. Material prices set the base of that pyramid, and they move: steel mill products currently index at 361.44 index (1982=100) (Jun 2026), up about 16.9% from a year ago, while plastic resins read 310.75 index (1982=100), up about 17.2% from a year ago. When input prices climb, every scrap point is worth more recovery effort at the same headcount, which is why scrap programs that were marginal at last year's material prices can clear the hurdle at this year's.
How to benchmark yourself without fooling yourself
Write the definition first: which losses count, measured in what unit, over what period, at which operations. Then compute the rate two ways, in pieces and in dollars weighted by accumulated value, because the piece rate tells you where defects happen and the dollar rate tells you where money dies. Track the trend against your own history rather than against an industry number nobody can audit. A plant that cuts scrap from 4% to 3% has a real, provable win; a plant celebrating that it beats an unsourced benchmark by a point has learned nothing. The comparison worth making is internal, longitudinal, and priced in current material dollars.
Why scrap costs more than it used to
- 1990: 112.10 (Archive begins 1990; selected years shown)
- 1996: 115.60
- 2001: 99.10
- 2002: 110.10
- 2008: 189.30
- 2014: 198.60
- 2020: 197.20 (When scrap was cheap to waste)
- 2021: 449.71
- 2026 (latest): 361.44 (The cost of every rejected part)
The 36-year record shows steel making a full round trip, which is why point-in-time comparisons mislead so badly here. Its high came at the close of 2021 around 449.71, gave way over the following years to 266.51 by the end of 2024, and has climbed since to 361.44. That leaves it 20% below the peak and well off the floor, so whether today looks high or low depends entirely on which year you anchored to.
Use the scrap accounting cost calculator to turn your rate and accumulated value per stage into an annual dollar figure worth acting on. Price your scrap
Published 2026-08-05.