Time Clock Rounding: What the Law Actually Allows

Time clock rounding is legal under federal law, but only on one condition: over time, it can't shortchange the employees whose punches get rounded. That condition is where most rounding policies get into trouble. The popular "7-minute rule" describes the math of quarter-hour rounding. It isn't a grace period, and a policy that only ever rounds in the employer's favor fails the federal test no matter what it's called.

California is stricter. Its Supreme Court has banned rounding for meal periods, an appeals court has ruled that an employer whose system captures exact time has to pay all of it, and the Supreme Court is still deciding whether neutral rounding is allowed there at all. Further down, a worked example puts a dollar figure on the gap between rounding that's actually neutral and rounding that leans one way.

What does the federal rounding rule actually allow?

The federal rule on time clock rounding is one paragraph long. 29 CFR 785.48(b) says employers may record time to the nearest 5 minutes, or to the nearest tenth (6 minutes) or quarter (15 minutes) of an hour, "provided that it is used in such a manner that it will not result, over a period of time, in failure to compensate the employees properly for all the time they have actually worked." That's the entire test: does the rounding average out, or does it quietly shave time off every shift.

The Department of Labor still applies that test to modern timekeeping. In a May 28, 2026 opinion letter (FLSA2026-8), the Wage and Hour Division looked at a hospital that let staff clock in up to 7 minutes early, rounded those punches forward to the scheduled start, and didn't let anyone clock out before the shift ended. Its conclusion: if employees are doing compensable work in that early window, the policy isn't neutral under 785.48(b), because the rounding only ever benefits the employer. The letter said the setup would likely comply if employees who clock in up to 7 minutes late also got credit for their scheduled start time, so the rounding could run both ways.

One detail in the letter matters for managers. Under federal law, clocking in, and waiting between the punch and the first real task, generally doesn't have to be paid, so rounding only costs employees money when they're actually working during the rounded minutes. The rule from 1961 and the letter from this year are saying the same thing: rounding is a convenience, not a discount.

Why is the 7-minute rule so often misunderstood?

The "7-minute rule" isn't a rule at all. It's just the midpoint of quarter-hour rounding: a punch 1 to 7 minutes past the quarter rounds down, a punch 8 to 14 minutes past rounds up. Washington's Department of Labor & Industries spells out the mechanic plainly in its recordkeeping policy (ES.D.1, last revised September 2025): an employee who clocks in 1 to 7 minutes late must be paid for the entire quarter hour, and at 8 to 14 minutes late, pay may begin at the nearest quarter hour. The same math applies at clock-out, in the other direction.

The part managers get wrong is treating "7 minutes" as a grace period that runs one way. It doesn't. It's a boundary that has to cut both ways across the full punch record, arrivals and departures, early and late, for the policy to stay neutral. A policy that rounds every clock-in forward and every clock-out backward isn't using the 7-minute rule. It's using half of it.

What turns rounding into wage theft?

The math fails quietly, which is exactly the problem. In May 2025, the Department of Labor recovered $594,313 in back wages and damages for 419 workers at a Leesburg, Florida, construction contractor after investigators found its rounding methods improperly reduced hours, leaving overtime unpaid for hours worked over 40 in a workweek. It doesn't take intent. A rounding formula that rounds down more often than it rounds up will do it on its own, shift after shift, for as long as nobody checks the aggregate.

That's the detail worth sitting with: rounding doesn't just cost straight time, it can wipe out overtime. If a rounding rule shaves a few minutes off a shift in a week that would otherwise cross 40 hours, the employer hasn't just underpaid those minutes at the regular rate. It has skipped the time-and-a-half premium on them. Multiply that across a crew and a year, and a "neutral-looking" policy can carry a real liability nobody budgeted for.

A hypothetical shows how this happens without anyone choosing it. Say a store rounds every punch to the nearest quarter hour, which is neutral on paper. Its staff tend to clock in a few minutes early and start setting up, then clock out a few minutes after close while they finish the last tasks. A 6:54 clock-in rounds to 7:00, a 3:05 clock-out rounds to 3:00, and 11 minutes of work disappear from the shift. The rule is symmetric. The punches aren't, so the result isn't either.

How far has California moved against time rounding?

California didn't ban rounding outright. It backed away from it in three steps, and the direction of travel matters more than any one ruling.

See's Candy Shops, Inc. v. Superior Court (2012) is where a California appeals court adopted the federal standard: a rounding policy is lawful if it's fair and neutral on its face and doesn't, over time, fail to compensate employees for all the time they actually worked. For nearly a decade, that was the California rule, borrowed straight from 785.48.

Then the California Supreme Court carved out an exception. In Donohue v. AMN Services, LLC (2021), the court held that employers can't round time punches for meal periods at all. California's meal period rules are precise: a 30-minute meal period that begins no later than the end of the fifth hour of work. The court said those rules are "designed to prevent even minor infringements," and rounding is incompatible with that.

The opinion's own example makes the point. Picture an employee who gets a 21-minute lunch, from 12:04 to 12:25. AMN's system rounded punches to the nearest 10 minutes, so it would have recorded a compliant 30-minute lunch from 12:00 to 12:30, and the employee "would have lost nine of the 30 minutes" without the system ever flagging it. The court also held that time records showing noncompliant meal periods raise a rebuttable presumption of a violation, so an employer has to explain the short or late breaks its own records show.

Then the Sixth District Court of Appeal went further in Camp v. Home Depot U.S.A., Inc. (2022). Home Depot's Kronos system recorded every punch to the minute, then rounded each employee's total shift time to the nearest quarter hour for pay. Over about four and a half years, that left the plaintiff with a net loss of 470 minutes, roughly 7.83 hours. The court held that when an employer "can capture and has captured the exact amount of time an employee has worked during a shift," it has to pay for all of that time.

The reasoning is worth knowing. The federal rule only asks that rounding average out across employees as a whole, but the court found "a complete absence of language" in California's Labor Code or wage order authorizing rounding that underpays an individual employee. The majority limited its holding to those facts. Justice Wilson's concurrence would have gone further, concluding that California law doesn't permit rounding "if it results in the underpayment of wages to any employee when due."

The California Supreme Court granted review in February 2023, and the question it took is broader than the one Camp answered: "Under California law, are employers permitted to use neutral time-rounding practices to calculate employees' work time for payroll purposes?" As of the court's list of pending civil issues, current as of October 9, 2026, Camp is still waiting on a decision, more than three and a half years after review was granted. Every California employer still rounding time is betting on how that question comes out. If you operate in California, that uncertainty alone is a reason to model what exact-time payroll would cost you before a court decides it for you.

Does rounding still matter if you also allow de minimis time?

Rounding and de minimis are two different doctrines that get treated as one. Rounding changes how a recorded punch converts to paid time. De minimis lets an employer disregard trivial bits of time that can't practically be recorded. Under federal law, 29 CFR 785.47 limits that to uncertain, indefinite periods of a few seconds or minutes, and says an employer can't arbitrarily ignore any part, however small, of the time an employee is regularly required to spend on assigned duties.

California's wage laws never adopted the federal version. In Troester v. Starbucks Corp. (2018), the California Supreme Court held that state law hasn't incorporated the FLSA's de minimis doctrine, and that it doesn't apply where an employer requires off-the-clock work on a regular basis. Starbucks had the plaintiff, a shift supervisor, clock out on every closing shift before finishing his closing tasks, which took four to 10 more minutes each day. Over 17 months, that came to about 12 hours and 50 minutes, worth $102.67 at the $8 minimum wage then in effect. Small, individually. Not small in the aggregate, and not random: it happened on a predictable schedule tied to the job.

On the argument that small amounts of time are too hard to record, the court said "employers are in a better position than employees to devise alternatives," such as restructuring the work, adapting their time tracking tools, or reasonably estimating the time. It left open whether time that's truly irregular or brief could still be disregarded.

The takeaway for a manager: even if your rounding policy is airtight, a separate habit of small unpaid tasks before or after the shift (opening a register, finishing a closing checklist after clocking out) is its own exposure. Fixing rounding doesn't fix that.

What does neutral rounding cost against employer-favoring rounding?

Here's the arithmetic that makes this concrete, using a hypothetical. Take a warehouse employee working five 8-hour shifts a week, 50 weeks a year, at $19.50 an hour.

Under a genuinely neutral quarter-hour policy, the rounding on any given punch is as likely to help the employee as the employer. Over a year, the gains and losses converge toward zero, which is the entire premise 785.48 relies on.

Now compare a policy that looks neutral on paper but rounds forward at clock-in and backward at clock-out (the "half the rule" pattern described above). Assume the employee is working during those minutes, which is the condition the 2026 opinion letter turned on, and that each punch loses anywhere from 0 to 7 minutes, spread evenly. That's an average of 3.5 minutes a punch, or 7 minutes a shift across the two punches. The same 7 minutes that was supposed to be a threshold becomes a daily loss.

That's one employee. A 50-person warehouse running the same skewed policy is looking at roughly $28,438 a year in wages the timekeeping system quietly kept, before counting a single hour of missed overtime premium. None of that requires anyone to have decided to underpay staff. It requires a rounding rule nobody has audited since it was installed.

What should a scheduling manager actually do about it?

Start by finding out which direction your rounding actually leans, because "we use the 7-minute rule" tells you nothing about that. Pull three months of raw punches against the paid totals your system generated and add up the difference, per employee as well as for the whole crew. A policy can come out close to even across a workforce and still leave individual people short, and Camp turned on one employee's net loss. If the difference isn't close to zero, you have your answer.

If you're in California, the safer move right now is not to round at all. Meal periods can't be rounded under Donohue, no matter what the Supreme Court decides about shift time. And if your time clock records punches to the minute, you're in the same position Home Depot was in under Camp. Paying the exact captured minute removes the argument entirely rather than betting on which way an unresolved case comes out.

Outside California, if you keep rounding, make sure it can actually run in both directions. If early punches round forward to the shift start, late punches inside the same window should round back to it too. That's the version the DOL's 2026 letter said would likely comply. Then treat the punch-versus-paid audit as a recurring task, not a one-time check, and pay particular attention to weeks near the 40-hour line. That's where a few minutes of rounding stops being a rounding error and starts being an unpaid overtime premium.

Whatever system you use, keep the raw punches, not just the rounded totals. Rounding is a policy choice you make on top of the data. It shouldn't be a decision your time clock makes for you before you ever see the number.

Getting overtime math right in the first place matters here too: a rounding policy that trims minutes near the 40-hour threshold interacts directly with how you calculate overtime, and a policy that rounds meal periods interacts directly with your state's break requirements. Neither problem shows up until someone adds up the minutes.

This article is general information, not legal advice. Rounding rules vary by state and California's are still being decided, so check current rules or talk to an employment attorney before changing your policy.