Standard hours are the time a job is supposed to take; earned hours are that standard time multiplied by the good units you actually produced. Divide earned hours by the actual clock hours worked and you get labor efficiency, the labor-reporting shop's cousin of OEE's Performance factor. The two terms are related, not interchangeable.
On a labor-reporting shop, these two numbers run the whole performance conversation, and they get muddled constantly. Standard hours are a rate you set once per part. Earned hours are a running total you rack up as good parts come off the line. Confuse them and your efficiency reports stop meaning anything. This post pins down each term, shows how earned hours drive the efficiency and utilization ratios, and flags the traps that quietly corrupt the numbers.
What are standard hours?
A standard hour is the amount of work that should be completed in one hour of normal operation, the engineered time for a task, set from time study, historical data, or a predetermined system. Expressed per part, it is your standard time per unit: this bracket takes 0.10 standard hours to assemble, that casting takes 0.25. It is a rate, fixed until someone re-studies the job, and it is the yardstick every other labor number is measured against.
Everything downstream depends on that yardstick being honest. A standard set loose, padded so operators always beat it, makes every shift look efficient and hides real loss. A standard set to an unreachable ideal makes good crews look like they are failing and teaches them to ignore the report. Standard time is to labor reporting what ideal cycle time is to OEE: get it wrong and every ratio built on it is wrong in the same direction.
What are earned hours, and how do they differ?
Earned hours, also called standard hours earned or standard hours allowed, are the credit a shift banks for what it actually produced: standard time per unit multiplied by good units. Make 400 brackets at 0.10 standard hours each and you earned 40 hours, no matter whether it took the crew 35 hours or 50. Standard hours are the rate; earned hours are the rate times output. That is the whole distinction, and it is the one people blur.
The word that carries the most weight is "good." Earned hours are credited on good units only, parts that pass. Rework and scrap earn nothing, because the standard is the time to make a sellable part, not any part. Leave the quality gate off and a crew can "earn" hours making defects, which is exactly how a labor report ends up celebrating a shift that shipped a pile of rejects. Tie earned hours to first pass yield and the number stays honest.
How do earned hours drive efficiency and utilization?
They feed two different ratios that answer two different questions, and a third that combines them. Work the numbers in this order:
- Set the standard time per unit. From time study or history, fix how long each part should take under normal conditions. This is the rate everything else is measured against.
- Count good units. Tally only the parts that passed. Scrap and rework are excluded, they cost time but earn none.
- Compute earned hours. Standard time per unit times good units. This is the output side, expressed in hours of standard work banked.
- Record actual hours. The real clock time the crew spent running those jobs. This is the input side.
- Divide for efficiency. Earned hours divided by actual hours worked, times 100. Above 100 percent means the crew beat standard; below means it ran behind. This answers "how fast did they work while they were working?"
- Divide for utilization. Actual hours worked on jobs divided by the hours the crew was present and available. This answers "how much of their paid time was on productive work at all?" Multiply efficiency by utilization, and by a quality factor, and you get overall labor effectiveness the labor mirror of OEE.
Where these numbers sit in the wider picture:
- Labor productivity is tracked nationally the same way, at scale. U.S. manufacturing sector labor productivity, output per hour, rose 2.0 percent in 2025 the largest annual gain since 2010, according to the U.S. Bureau of Labor Statistics Productivity and Costs program. Earned-versus-actual hours is the same output-per-hour idea measured at one machine or cell.
- The accounting side formalizes the gap. The direct labor efficiency variance equals standard hours allowed minus actual hours worked, multiplied by the standard labor rate, the dollar value of running ahead of or behind standard. The definition and worked examples are laid out by AccountingTools' labor efficiency variance reference.
What does a worked example look like?
Take two operators on the same job with a standard of 0.10 standard hours per good part, both present for an 8-hour shift. The efficiency figure only means something once you also know utilization and quality, the table shows why one number alone misleads.
| Measure | Operator A | Operator B |
|---|---|---|
| Good units produced | 420 | 360 |
| Earned hours (0.10 × good) | 42.0 | 36.0 |
| Actual hours on the job | 8.0 | 6.0 |
| Efficiency (earned / actual) | 105% | 150% |
| Utilization (on-job / attended) | 100% | 75% |
| Effective output (earned / attended) | 84% | 75% |
Read the bottom row, not the efficiency row. Operator B posts a dazzling 150 percent efficiency, but only because two of the eight attended hours never got clocked onto the job, a setup, a material wait, a problem nobody logged. Measured against the time the plant actually paid for, B delivered 75 percent while A delivered 84 percent. This is why a supervisor who rewards efficiency alone trains the floor to hide idle time rather than eliminate it: the fastest way to a big efficiency number is to keep the clock off during every problem. Earned hours against attended time is the number that cannot be gamed the same way, because idle hours count against it whether they were logged or not. The habit worth building is to publish both columns side by side and coach to the effective-output figure, so a shift that ran flat out for six hours and then waited on parts is credited for the work and flagged for the wait, instead of hiding one inside the other.
How does this relate to OEE?
The two frameworks are the same idea aimed at different resources. OEE measures how well the equipment ran; earned-versus-actual hours measures how well the labor ran. Efficiency is the labor twin of OEE's Performance factor, earned hours to actual hours mirrors ideal output to real output. Utilization mirrors Availability. Add a quality gate and you have overall labor effectiveness, built exactly like OEE but with people as the asset. On a labor-intensive line, both stories matter, and they should be read together as part of one set of manufacturing KPIs not in separate reports that never reconcile.
The link is also practical. A crew's earned-hours efficiency can crater not because anyone worked slowly but because the equipment kept stopping, every machine breakdown or minor stop steals actual hours with no earned hours to show for it. That is why blaming a low labor-efficiency number on the people is so often wrong: the loss frequently sits in the six big losses on the equipment, not in the crew. Read the two together and the real cause shows up.
What are the traps that corrupt these numbers?
Three, and each is common enough to assume it is happening until you check. First, loose or stale standards: if the standard time was set generously or never updated after a process change, efficiency runs above 100 percent all day and measures nothing. Re-study the standards on the jobs that matter, the same way you would re-baseline cycle time. Second, no quality gate: crediting earned hours on all units instead of good units rewards making scrap. Third, gaming through utilization: an operator can post spectacular efficiency by only clocking onto easy jobs and leaving the clock off during setups and problems, which is why efficiency without utilization is a number to distrust.
The deepest trap is manual reporting itself. When earned and actual hours are keyed from paper travelers at the end of a shift, the numbers arrive late, rounded, and quietly optimistic, nobody logs the ten minutes lost hunting for a fixture. Plants that capture good-unit counts and on-job time straight from the line the way Harmony does (see the platform) get earned-versus-actual hours that reconcile to reality, and you can price what a point of efficiency is worth alongside your equipment losses with the OEE calculator. Honest inputs are the whole game; the ratios take care of themselves.