KOHEN AVOCATS · PARIS, FRANCE
90 Day Blackout: Are Employees Paid When a Workplace Loses Power?

Hassan Kohen is a Paris-based lawyer admitted to the Paris Bar (Avocat au Barreau de Paris). Initial strategic analysis: €80 incl. VAT, personal reply within 24 hours.

100% confidential · Legal professional privilege · No obligation

Article generated by artificial intelligence, through a process designed and supervised by the firm

Sources: U.S. court opinions and official U.S. government publications, read for this article

Maître Hassan KOHEN, avocat au Barreau de Paris
Hassan KOHEN
Attorney at the Paris Bar

90 Day Blackout: Are Employees Paid When a Workplace Loses Power?

Speak to a lawyer 06 89 11 34 45

If a workplace loses power, the first pay question is not how long the outage lasts. It is what the employee is actually asked to do while the business is unable to operate normally. A covered, nonexempt employee who keeps working, or who must wait under restrictions that make the time part of the job, may have compensable work time. A genuinely relieved employee may be treated differently. An employee who qualifies for a federal salary-basis exemption needs a separate analysis of the workweek and the reason for any salary deduction.

The phrase “90 day blackout” has prompted questions about extended interruptions, but it does not establish that a nationwide civilian blackout has been forecast. An October 29, 2026 Department of War report describes a proposed military installation designed to operate for 30 days apart from outside utilities. The report describes resilience planning, not a prediction that civilian workplaces will lose electricity for 90 days. The discussion below addresses a hypothetical outage of any length and uses federal wage law, with Tenth Circuit examples, to show which facts a worker should preserve and compare. The discussion is general in scope and describes checks for comparable workplaces; it makes no findings about the current circumstances of any named party.

I. What happens to your pay when the power goes out at work?

A. Separate work, waiting and time you can genuinely use for yourself

Start with a timeline. Record when the electricity failed, what work was underway, when instructions changed and when the employee was released. The answer may change several times during a single shift. A cashier might keep serving customers by hand, then wait inside for a system to restart, and later be told to go home. Those are different periods, and one label on a timesheet may conceal facts that matter.

Federal 29 C.F.R. § 785.11 explains the Department of Labor’s interpretation that work an employer knows or has reason to believe is being performed is work time, even if the employer did not request it. Preserving perishable stock, making a required safety check, answering customer requests or completing a paper record can therefore matter as actual work. The useful question is what the employee did and what a supervisor knew or had reason to know, not whether the computer system was functioning.

Instructions should be distinguished from assumptions. A worker told to secure inventory and report completion has an assigned task. A worker who chooses to reorganize a personal locker while completely relieved of duties presents a different question. If the employer has a way to report unexpected work, use the actual reporting procedure and retain a copy of what was submitted. When the ordinary clock is unavailable, a contemporaneous note, a saved message or a paper sign-in can help reconstruct the period. None of these records needs to be perfect to be useful; accuracy about what is known and what is estimated is more useful than an invented minute-by-minute account.

Waiting is not automatically unpaid just because no customer is being served. The Department of Labor’s federal interpretation in 29 C.F.R. § 785.14 looks at the working arrangement, the service performed and the surrounding circumstances. Under § 785.15, a factory worker waiting for machinery to be repaired is an example of on-duty waiting when the employee cannot effectively use that time for personal purposes. A comparable outage question is whether the employee must remain available for an unpredictable restart, rather than being free to make meaningful plans.

Consider a warehouse employee who is told to remain in the building because a generator may start at any moment. If the employee must periodically check a temperature display, respond to deliveries and resume handling goods whenever directed, write down each instruction and the periods between tasks. Those facts can support a conclusion that the employee was waiting as part of the job. By contrast, merely being inconvenienced by an outage does not settle whether a particular interval was work. The legal answer depends on the actual restrictions, the nature of the work and what the employee could do with the interval.

Federal 29 C.F.R. § 785.16 describes a different situation: an employee is completely relieved from duty for a period long enough to use effectively for personal purposes. Its general explanation includes advance notice that the worker may leave and will not have to start work until a specified hour. The duration must be assessed in light of the circumstances. A message saying that the site is closed until a stated reopening time can be useful evidence; a vague instruction to stay ready for an immediate call is materially different.

Suppose a supervisor first says to wait beside a loading dock, then announces that everyone is free to leave and that no one must resume work before a stated time. The earlier interval and the later interval should be assessed separately. Preserve the actual message, not just a later recollection that the shift was canceled. If the employee was also told to watch for urgent calls, note that instruction too. A fact such as being allowed to leave does not by itself answer every on-call question; the practical restrictions during the time away still matter.

Remote work can also change during an interruption. A worker whose own electricity stays on may be asked to call customers, coordinate deliveries or monitor a backup system while a main office is closed. Record the period actually spent on those tasks and the instruction that required or permitted them. If an employee is later told that no further response is expected until a specified time, preserve that change as a separate entry. A closure notice for one building does not describe the duties of every employee working elsewhere.

When an employer proposes a corrected timecard, compare each interval with the notes and source messages before accepting the total as an accurate description. It may be helpful to identify a start, a release, a return and a second release on separate lines. The point is to make the payroll question concrete: which minutes involved an assigned duty, which were restricted waiting, and which could be used for personal purposes? The answer can be different for coworkers even when the same power interruption affected both.

These federal provisions are Department of Labor interpretations of hours worked, and the case outcomes below turn on evidence. They do not establish a general federal guarantee that a nonexempt worker receives the entire scheduled shift whenever a business closes. State reporting-time rules, a contract, a collective bargaining agreement or an employer policy may create a different question. Identify the state and the applicable document before treating an expected shift as the measure of pay.

B. Check whether the employee is nonexempt or paid on a qualifying salary basis

Pay frequency and the label on a paycheck are starting points, not conclusive classifications. Ask whether the federal wage rules cover the job, whether an exemption is claimed, and what the employee actually does. A person can receive a salary and still need a nonexempt analysis. Likewise, an employee described as a manager should not assume that a title alone resolves the exemption question. The documents worth collecting include the offer or pay agreement, a recent pay statement, the assigned schedule, the job duties and the policy used for the outage.

For covered employees to whom the federal overtime requirement applies, 29 U.S.C. § 207(a)(1) generally requires at least one and one-half times the regular rate for work beyond 40 hours in a workweek, subject to statutory exceptions. The linked Government Publishing Office edition is the 2024 Code. The federal Department of Labor’s 29 C.F.R. § 778.101, displayed as current through October 29, 2026, describes the same general 40-hour standard; the employee’s applicable exemption and the current hours-worked materials remain relevant to the calculation. The legal workweek, rather than a calendar month or an assumed 90-day period, is the practical unit to inspect.

An outage can interrupt one shift and produce extra work in another. An employee might be sent home on Monday, return for a lengthy cleanup on Tuesday and answer dispatch requests later in the week. Add the actual work and any waiting time that qualifies as work to the correct workweek before comparing that total with the payroll record. Do not count the same period twice simply because a call and an on-call assignment overlap. At the same time, do not omit a short work task solely because the normal electronic time clock was down.

The federal salary-basis rule is different. Under 29 C.F.R. § 541.602, an employee who qualifies for the relevant exemption generally must receive the full salary for a workweek in which the employee performs any work, subject to specified exceptions. The regulation also says an exempt employee need not be paid for a workweek in which no work is performed. For an employee ready, willing and able to work, it says deductions may not be made for time when work is unavailable. A week with some work is therefore a different question from a complete workweek with none.

That distinction is not a promise that every salaried worker is exempt or that every deduction is prohibited. The same regulation lists exceptions, and the facts behind an absence matter. A closure ordered by the employer is not necessarily the same as an employee’s separate personal absence. If a pay statement shows a reduction, compare the workweek, the duties performed and the written reason for the reduction with the actual exemption and the listed exceptions. Ask for an explanation of the payroll entry without changing or reconstructing the employer’s records.

A paid-leave balance raises its own practical question: what does the applicable policy say, and what happened to the salary payment? A bank of leave and the federal salary-basis rule are not interchangeable concepts. Review the leave policy, the recorded balance before and after the closure, and the paycheck. The federal text cited here addresses salary basis; the rules governing a particular leave deduction can depend on other law and documents and should not be inferred from the fact of a blackout alone.

II. How can you tell whether an outage makes on-call time compensable?

A. Compare real restrictions with what the courts actually decided

Being reachable does not always mean that every minute away from work counts as work. Federal 29 C.F.R. § 785.17 distinguishes an employee required to stay on the employer’s premises or so close that the time cannot be used effectively for personal purposes from an employee who merely leaves word about where they can be reached. Real assignments can fall between those examples. The practical inquiry includes response time, frequency of calls, location restrictions, consequences of missing a call and the freedom actually available.

Three published federal decisions of the United States Court of Appeals, Tenth Circuit, illustrate that comparison. In Pabst v. Oklahoma Gas & Electric Co., 228 F.3d 1128 (10th Cir. 2000), decided September 20, 2000, the court affirmed compensation for technicians whose frequent alarms and short response requirements substantially restricted their on-call time. In Boehm v. Kansas City Power And Light Company, 868 F.2d 1182 (10th Cir. 1989), decided February 28, 1989, the court held that the particular storm-response policy did not make all of the linemen’s claimed off-duty time compensable. In Norton v. Worthen Van Service, Inc., 839 F.2d 653 (10th Cir. 1988), decided February 17, 1988, the court affirmed the finding that the drivers could use the particular waiting periods for personal activities. In Barnes v. Omnicell, No. 23-1336 (10th Cir. 2024), decided May 28, 2024, the United States Court of Appeals, Tenth Circuit, cited all three opinions while affirming summary judgment on the on-call facts before it; Barnes is a nonprecedential order and judgment. Together, the decisions show why a worker should compare actual calls and practical freedom instead of relying on the word on-call alone.

The federal Tenth Circuit decisions are examples from that court, not a rule that every federal circuit must resolve every factual dispute identically. They do not decide the pay of a reader’s employer. The later Barnes order is useful evidence that the panel discussed the earlier decisions in 2024, but its nonprecedential status and the limits of a targeted citation search matter. It is not a complete check of every subsequent treatment.

In Pabst, the record included a short response requirement, frequent alarms and monitoring that substantially interfered with ordinary personal activities. A worker facing a power interruption should therefore preserve both the written instruction and what happened in practice: the number of alerts, their spacing, the time spent answering them, and whether a response required access to a particular place or device. A rule demanding a fast response can matter even when an employee is physically at home.

Boehm involved a different arrangement. The workers were expected to respond to storm emergencies, but the court examined the actual restrictions and concluded that the claimed off-duty time was not all spent predominantly for the employer’s benefit. Do not read the case as a rule that storm duty is never work. Its value is the need to distinguish the time spent performing a callout from the wider period in which an employee may have freedom to pursue personal activities. A personal log can identify when the obligation became active and what real choices remained.

Norton also turned on what the drivers could do between assignments. The evidence described personal activities during waiting periods and options that affected availability. A worker considering a similar issue should record whether it was possible to decline a call, exchange a shift, travel away from the workplace or set a period of unavailability. A written rule and what supervisors actually permitted may differ; keep evidence of both rather than assuming that one completely describes the other.

In the later Barnes decision, the court considered relatively infrequent service requests, an ability to use time for personal activities and response arrangements in the record. A single response requirement is therefore not the entire analysis. An employee whose device receives one optional message during a long release may have a different situation from an employee ordered to monitor a failing system continuously. A meaningful comparison identifies the full pattern, including periods with no calls, and does not remove inconvenient facts from the record.

B. Build a reliable record and compare it with payroll

Make a simple table for each affected workweek. One column can show the interval, another the instruction in force, and another the work or waiting actually performed. Add the source of each entry: a schedule, a message, a call log, a paper note or an honest estimate. Include the time the employee was completely relieved, the stated return time if there was one, and any later instruction that changed the arrangement. This method makes an ambiguous closure announcement easier to compare with what actually occurred.

Separate at least four kinds of time: active work before the outage, assigned work during the outage, waiting or on-call periods, and time when all duties were lifted. For waiting or on-call time, describe whether the employee had to remain at the site, how quickly a response was expected, how often the phone or radio sounded, and whether personal plans were interrupted. Preserve the original instructions even if a later explanation uses a different description. If a supervisor changes an instruction orally, a short contemporaneous note identifying the speaker and the substance can help.

Then compare the table with the employee’s timecard and pay statement. Check whether the payroll period contains all of the affected workweek and whether work reported on a paper form was later entered into the system. A discrepancy may come from a missing punch, a coding error, a difference over waiting time or a disagreement about exemption. Naming the specific interval and the evidence supporting it is more useful than asserting that an entire outage must be paid without distinguishing work from release.

For example, an employee may have worked two hours, remained at the workplace for an unpredictable restart, then been told at noon to leave until a specified return time the following day. Another employee may have gone home immediately but been required to respond to repeated messages within minutes. Those workers should not assume their situations are identical. In the first example, the on-site interval deserves careful comparison with the federal on-duty waiting interpretation. In the second, the real frequency and effect of the calls are central to the federal on-call analysis.

A third employee may qualify for a federal salary-basis exemption and perform a small amount of work during a week in which the office otherwise closes. That record should identify the workweek, the work performed, the normal salary and any deduction. A fourth may perform no work at all during a complete workweek. The salary-basis regulation expressly treats that situation differently, although an agreement or other applicable rule can still matter. Neither employee should infer exemption merely from a fixed paycheck.

When records are incomplete, describe their limits. A phone battery may have failed, a paper sheet may have been collected by a supervisor, or the outage may have prevented an electronic punch. Reconstruct what can be checked from messages, schedules, receipts, coworkers’ notes or later payroll entries, and distinguish an estimate from a recorded time. Avoid altering a message or presenting a reconstruction as an original log. A clear chronology helps a payroll reviewer understand the issue and preserves the factual distinctions on which the federal decisions depend.

Finally, identify the state where the work was performed and read the agreement or policy that governed the shift. State law and a collective agreement may address matters beyond the federal framework discussed here, including the effect of reporting to a closed workplace. The federal sources above do not establish a nationwide reporting-time guarantee or resolve the separate state-law question. If the employee works in more than one state, record where each period of work occurred so that the appropriate local rules can be identified.

The length of the outage alone is not a substitute for this analysis. A short interruption can include compensable assigned waiting, while a longer closure can include a period when a worker is completely relieved. The question changes again when an employee begins cleanup, handles customer communications or is placed under genuine on-call restrictions. Check the instructions and the payroll result for each phase instead of applying one answer to the entire event.

Conclusion

For a covered, nonexempt employee, distinguish work performed, employer-controlled waiting and a genuine release from duty before assessing an outage paycheck. For an employee who qualifies for a federal salary-basis exemption, examine the workweek and the reason for any salary deduction. Save the instructions, a truthful timeline and the pay record, then compare the specific disputed period with the federal sources and any applicable state or contractual rule. A military resilience plan is not proof that a nationwide civilian blackout is imminent.

📄 Official circular