Initial Measurement Period: Hours, Stability, and Penalties

The initial measurement period is a tracking window of 3 to 12 months that an applicable large employer uses when a newly hired variable-hour or seasonal employee’s schedule makes it impossible to know at the start date whether that person will average 30 hours per week.¹ At the end of the window, the employer calculates the employee’s average weekly hours and, if the number lands at 30 or above, must offer coverage for a locked-in stability period that follows. Getting the mechanics wrong can trigger penalties above $3,300 per full-time employee for 2026, so the rules reward careful setup.

Which New Hires the Window Applies To

Two categories of new employees can be placed in an initial measurement period. A variable-hour employee is one whose hours are unpredictable enough that the employer cannot reasonably determine at hire whether the person will average at least 30 hours per week.¹ A retail associate hired without a guaranteed weekly schedule or a per-diem nurse whose shifts float with demand fits this description.

Seasonal employees also qualify. The IRS accepts a reasonable, good-faith interpretation of “seasonal,” and the common benchmark is a position where annual employment runs six months or less and is tied to a recurring time of year.² Part-time hires expected to work fewer than 30 hours can be placed in the window as well when there is any realistic chance their hours could climb.

Employees who are plainly full-time on day one do not belong in this process. If the position clearly requires 35 or 40 hours every week, coverage has to be offered within the normal waiting period. The initial measurement period exists for genuinely ambiguous schedules, not for delaying coverage where the answer is already obvious.

How Long the Window Runs and When It Starts

The employer picks any duration from 3 to 12 months.¹ Most employers choose 12 months because it captures a full cycle of busy and slow stretches and produces the most reliable picture of actual hours. Whatever length gets picked has to be applied consistently within each category of employees. You cannot give one variable-hour cashier a 6-month window and another a 12-month window when they share the same job classification.

Two start-date options exist. The employer can begin tracking on the employee’s actual first day of work, or on the first day of the calendar month following that start date.¹ Starting on the hire date keeps the timeline clean because each employee’s clock runs from their own start. Starting on the first of the following month aligns with payroll cycles but creates a partial-month gap at the front end that eats into the administrative window later on.

Counting Hours of Service

Measurement runs on hours of service, which is broader than time actually worked. Every hour for which the employee is paid or entitled to payment counts, including vacation, holidays, and paid sick leave. Unpaid leave generally does not count, but any compensated hour does.¹

For hourly workers, the employer counts the actual hours recorded in payroll. For salaried and other non-hourly employees, the regulations offer three equivalency methods:

Daily equivalency: 8 hours credited for each day the employee works or is entitled to pay.
Weekly equivalency: 40 hours credited for each week the employee works or is entitled to pay.
Pay-period equivalency: a set number of hours matching the pay cycle, such as 80 hours for a semi-monthly period.

An equivalency method cannot be used to undercount hours for an employee who actually works more, but it provides a workable shortcut when tracking exact hours for salaried staff would be unreasonable.³

At the end of the window, the employer totals hours of service and averages them across the measurement period. The federal threshold is at least 30 hours per week on average, which translates to 130 hours per month.⁴

The Administrative Period After the Window Closes

Once measurement ends, the employer needs time to run the numbers, notify eligible employees, and coordinate enrollment with the insurance carrier. The administrative period is the buffer built in for that purpose. Federal regulations cap the combined length of the initial measurement period plus the administrative period: together, they cannot extend past the last day of the first calendar month beginning on or after the first anniversary of the employee’s start date.¹

In practice, that works out to roughly 90 days plus the remainder of any partial month at the front. If measurement began on the first of the month after the hire date, the partial month between the hire date and that first-of-month start consumes part of the administrative window. Starting measurement on the actual hire date generally leaves a cleaner 90 days. Blowing through the cap risks violating the waiting period rules and triggering separate penalty exposure.

Stability Period When the Employee Qualifies

When the math confirms an average of 30 or more hours per week, the employer locks in full-time status for a stability period lasting at least six months, and it cannot be shorter than the measurement period itself.¹ A 12-month measurement period produces a 12-month stability period. A 6-month measurement period produces at least a 6-month stability period.

Coverage stays in place throughout that window regardless of what the employee’s hours actually look like. Someone who averaged 35 hours during measurement but drops to 20 hours three months into the stability period keeps their coverage as long as they remain employed. This is the protection the look-back method offers workers with unpredictable schedules.

When the Employee Falls Short of 30 Hours

If the measurement result comes in below 30 hours per week on average, the employer is not required to offer that worker health coverage. The non-full-time stability period cannot exceed the initial measurement period by more than one month, and it cannot extend beyond the remainder of the standard measurement period in which the initial measurement period falls, plus any associated administrative period.¹

The employee’s hours then get picked up by the employer’s standard measurement period used for all ongoing staff. If hours rise during that standard window, the employee can qualify as full-time for the next stability period even after missing the initial one.

Overlap With the Standard Measurement Period

Every new hire eventually moves from the individualized initial timeline into the employer’s standard measurement period. That transition almost always produces a stretch where the employee is being tracked under two overlapping periods at the same time. The rule is strict: if the employee qualifies as full-time under either period, coverage must be provided for the corresponding stability window.¹

Two situations come up regularly. In the first, the employee qualifies during the initial measurement period but not during the overlapping standard measurement period. Coverage must continue through the entire initial stability period before the employer can treat the worker as non-full-time. In the second, the employee does not qualify during the initial measurement period but does qualify during the standard one. Full-time treatment then applies for the stability period tied to the standard measurement, even if that stability period starts before the initial one ends.

If a gap opens between the end of the initial stability period and the start of the standard stability period, the employee’s status from the initial period carries through until the standard period begins.² From that point on, the employee’s status is determined by the standard measurement results, just like any other continuing employee.

Rehires and Breaks in Service

When a former employee returns, the employer has to decide whether to treat the person as a brand-new hire, which allows a fresh initial measurement period, or as a continuing employee whose prior timeline resumes. Under the general rule, an employee with a break of at least 13 consecutive weeks and no credited hours of service can be treated as a new hire.² Educational organizations use a 26-week threshold instead, reflecting longer academic-calendar breaks.

A separate rule of parity covers short-tenured workers. If the break in service is at least four consecutive weeks and exceeds the number of weeks worked before leaving, the employer can treat the returning worker as a new hire even when the break was shorter than 13 weeks.² Someone who worked eight weeks, left for ten, and returned would meet the parity test because ten exceeds eight.

If neither test is met, the returning worker is a continuing employee. Prior service must be credited, and whatever measurement or stability period was running when the person left resumes as though the gap never occurred. For workers who were already in a stability period with full-time status, coverage has to be offered immediately upon return. This is one of the more common mistakes with seasonal workforces.

Penalty Exposure for Getting It Wrong

Applicable large employers who fail to offer coverage to workers determined full-time under the measurement method face penalties under two provisions of Section 4980H. The 4980H(a) penalty applies when the employer does not offer minimum essential coverage to at least 95 percent of its full-time workforce. For 2026, the amount is $3,340 per full-time employee annually, calculated monthly at one-twelfth of that figure. The first 30 full-time employees are excluded from the count, but every worker beyond that adds to the assessment.⁵

The 4980H(b) penalty is narrower but still expensive. It applies when the employer offers coverage that is either unaffordable or fails to provide minimum value, causing one or more full-time employees to receive a premium tax credit on the marketplace. For 2026, the amount is $5,010 per affected employee annually, again calculated monthly. Total 4980H(b) liability is capped at what the 4980H(a) penalty would have been, so an employer never pays more under (b) than they would have under (a).⁵

Mishandling the initial measurement period is one of the fastest routes to these assessments. An employer that miscounts hours and classifies a full-time worker as part-time has effectively denied coverage to someone legally entitled to it. If that worker then enrolls in a marketplace plan with subsidies, the 4980H(b) penalty attaches for every month the coverage gap existed.

1
Internal Revenue Service. Notice 2012-58 – Determining Full-Time Employees for Purposes of Shared Responsibility for Employers Regarding Health Coverage
2
eCFR. 26 CFR 54.4980H-3 – Determining Full-Time Employees
3
GovInfo. 26 CFR 54.4980H-1 – Definitions
4
Office of the Law Revision Counsel. 26 USC 4980H – Shared Responsibility for Employers Regarding Health Coverage
5
Internal Revenue Service. Rev. Proc. 2025-26