On-Call and Standby Pay: When Waiting Has to Be Paid

Waiting is paid when the waiting is really work. That is the whole federal test, and almost every on-call dispute is an argument about which side of the line a particular rota falls on. If your people are stuck at the site, or on a short enough leash that they cannot live their evening, you owe them wages. If they carry a phone and can go to dinner, you usually do not.

The trouble is that "usually" hides a lot. Two employers can run what looks like the same rota and land on opposite sides of the line, and a state rule can flip the answer entirely. This guide covers the federal test, the factors courts weigh, the states that go further, and how to build a rota that stays where you meant it to.

What is the federal test for paid on-call time?

Federal law asks one question: is the employee engaged to wait, or waiting to be engaged? The Department of Labor puts it plainly in its hours worked fact sheet: the facts may show the employee "was engaged to wait (which is work time)" or "was waiting to be engaged (which is not work time)."

The phrase comes from two Supreme Court decisions handed down on the same day in 1944. In Armour & Co. v. Wantock, fire guards stayed overnight at a factory, free to sleep, eat and play cards, and the Court held their time was compensable. Its reasoning is still the cleanest statement of the principle: "Refraining from other activity often is a factor of instant readiness to serve, and idleness plays a part in all employments in a standby capacity."

In Skidmore v. Swift & Co., employees stayed in a company fire hall or "within hailing distance" several nights a week. The Court refused to write a formula. It said that whether such time counts "is a question of fact" depending on the agreement, the nature of the service, "and all of the surrounding circumstances." That is why on-call pay is so hard to settle with a one-line policy. Nobody can hand you a number of minutes or miles that makes a rota safe.

What the DOL does give you is a rough sorting rule. Its fact sheet says an employee required to remain on call on the employer's premises is working. An employee who is required to remain on call at home, or who may leave a message where they can be reached, is "not working (in most cases)." Then comes the sentence that matters most for operators: "Additional constraints on the employee's freedom could require this time to be compensated."

So the premises are an easy call. Everything else is about constraints, and how many you have stacked up.

Which factors decide whether on-call time is paid?

Courts and the DOL keep returning to the same handful of questions. A 2018 DOL opinion letter (reissuing a 2009 letter) is a good worked example. A county ambulance service required on-call staff to reach the garage within about five minutes, though nothing was in writing and late responses were not disciplined. Employees carried pagers, wore uniforms and were told to avoid alcohol. They were called back about three times a week, usually for around an hour. The DOL still concluded the on-call periods were not compensable, because the restrictions were "not so restrictive as to convert the on-call time into hours worked."

Look at what the agency weighed: response time, geographic limits, discipline for being late, uniform and alcohol rules, and how often the calls came. Those are the working factors. Put in operator terms:

• Response time. Thirty minutes to be on site is a very different leash from five. The shorter it is, the harder it is to do anything else.
• Geographic limits. "Stay within ten minutes of the clinic" costs someone their weekend plans. "Stay reachable" does not.
• Consequences for missing a call. Discipline for a late response signals real control. A shrug does not.
• Frequency of calls. An employee who is pulled in once a month has a very different shift from one pulled in twice a day.
• Practical limits on personal activity. Required uniforms, bans on drinking, a requirement to bring equipment home. Each is small. Together they add up.

Frequency is the factor operators underestimate. In Gilligan v. City of Emporia, the Tenth Circuit held that city water and sewer workers on mandatory pager duty (one hour to respond for one plaintiff, thirty minutes for the other) were not entitled to pay for the standby time, in part because call-backs averaged less than once per day. The court described the earlier firefighter case, Renfro v. City of Emporia, as one where employees "were called back an average of three to five times a day," and frequency there was "a pivotal factor." Same circuit, same city, different outcomes, and the difference was how often the phone rang.

One caution on all of this. Courts apply these factors to a whole arrangement, not one factor at a time, and the federal cases are a patchwork by circuit. A rota that passes in one part of the country is not guaranteed to pass in another.

What happens when on-call time interrupts sleep or overtime?

Two things change the picture even on a rota you thought was clean.

First, once a callback turns into actual work, that work is hours worked, full stop. It counts toward the 40-hour overtime line like any other time. Our guide to calculating overtime covers the arithmetic, and our overtime policy guide covers writing the rules down. A common surprise is an employee who works a full week, takes two evening callbacks of ninety minutes each, and crosses into overtime without anyone having scheduled it.

Second, sleep. The DOL fact sheet says an employee on duty for less than 24 hours is working even if allowed to sleep when not busy. For duty of 24 hours or more, the employer and employee may agree to exclude bona fide, regularly scheduled sleeping periods of not more than eight hours, provided adequate sleeping facilities are furnished, and no time can be excluded unless the employee gets at least five hours of sleep. Oregon's BOLI guidance spells out how interruptions work there: time spent on an interruption counts as hours worked, and if interruptions leave the employee without five continuous hours of sleep, the entire sleep period counts.

Rest between shifts matters too. A callback at 2 a.m. followed by a scheduled 7 a.m. shift is exactly the situation covered in our post on minimum rest between shifts, and a few jurisdictions regulate it directly.

Which states go further than the federal rule?

Quite a few. The federal test sets the floor, and states can add requirements on top. The table below covers three states. It is a sample, not a census. Rules change, and a missing state does not mean a missing rule.

Massachusetts is the instructive one, because it is stricter than federal practice in how it frames the default. Under 454 CMR 27.04, on-call time is compensable working time unless the employee is not required to be at the work site or another location and is effectively free to use the time personally. The same section carries the state's reporting pay rule: an employee scheduled for three or more hours who reports and is not given the expected hours must be paid for at least three hours at no less than the minimum wage. Massachusetts guidance on that rule, such as a Department of Labor Standards opinion letter, treats it as a reporting-pay provision rather than an on-call rule, so read it for the case where someone is called in and sent home.

Washington's Department of Labor & Industries takes the more employer-friendly line on location, but with limits. Its hours worked policy says employees who are merely required to leave word where they can be reached are not working while on call, and adds that restrictions on where and when the employee may travel "may change the character of that 'on call' status." Its overtime policy adds a point many operators miss: if you pay for on-call time that is not hours worked, that payment must still be included when calculating the employee's regular rate for overtime.

Oregon deserves a second look because of its scheduling law. The BOLI page notes that employers covered by predictive scheduling must typically pay one-half the regular rate for scheduled on-call hours where the employee is not asked to work. If you operate in a city or state with a fair workweek law, check whether on-call hours trigger a separate premium. Our predictive scheduling laws guide lists where those rules apply.

California has its own wage orders and reporting-time rules, and they differ by industry. If you employ people there, check with the state and read the Industrial Welfare Commission wage order that covers your occupation rather than relying on a general summary.

How much callback pay do you owe when someone is called in?

Here the federal and state pictures split. The FLSA question is whether time was work and whether it was paid. It sets no minimum for being called in: the Department of Labor's hours worked advisor says that when an employee reports as directed and is sent home before doing any work, the FLSA does not require the employer to count the time or give show-up pay. Minimum guarantees come from states, and they vary in both the number of hours and the trigger.

The Massachusetts three-hour rule above is one example. California has a rule aimed squarely at callbacks: an employee required to report a second time in one workday who gets less than two hours of work on the second reporting must be paid for two hours at the regular rate, according to the state's Division of Labor Standards Enforcement. The same page says reporting time pay does not apply to employees on paid standby. In New York, the wage order for miscellaneous industries requires at least four hours of call-in pay, or the regularly scheduled shift if that is shorter, at the basic minimum wage, for an employee who reports for work at the employer's request or with its permission.

The related distinction to hold onto is between on-call pay (compensation for the waiting) and call-back or reporting pay (a guaranteed minimum when someone actually comes in). Plenty of employers conflate them, and the result is a policy that pays one and forgets the other. Our guide to reporting time pay covers the second half in detail, including what happens when you send someone home early.

If you pay a flat stipend for being on call, spell out in writing whether the stipend covers the waiting only, and what rate and minimum apply when an employee is actually called in. A stipend does not erase the obligation to pay for hours actually worked, and under both the federal rules and Washington's policy it counts toward the regular rate.

A worked example. A clinic pays a nurse $25 a week to carry the after-hours phone. Over the week she works 38 scheduled hours, takes one callback that runs 90 minutes, and takes a second that runs 45. Her hours worked are 38 plus 2.25, or 40.25. She is over 40, so 0.25 hours are overtime. The $25 stipend also belongs in her regular rate, because federal rules say on-call pay is included in the regular rate even when the waiting itself is not hours worked. Spread over 40.25 hours, it adds about 62 cents an hour to her regular rate and about 8 cents to her overtime pay for the week. None of this is large. All of it is the kind of thing that surfaces in an audit.

How do you build an on-call rota that stays unpaid?

If your goal is a rota where the waiting is genuinely not work, the design choices are mostly about reducing constraints. In practice that means:

1. Lengthen the response window. Thirty to sixty minutes to respond remote, rather than fifteen to reach the site, leaves people room to live.
2. Drop geographic limits you do not need. If the job can be done remotely, say so. If it cannot, be honest about how far someone can drive and still be useful.
3. Do not discipline for the wrong things. A policy that penalizes a late response is a control. If you need reliability, build it with a backup person, not a threat.
4. Rotate fairly and cap frequency. A rota where one person takes every Saturday is a rota where that person is working Saturdays. Our shift differential pay post covers the premium side of that conversation.
5. Track actual callbacks. Keep a log of calls, response times and hours worked. If frequency creeps up, your rota is drifting toward compensable.
6. Pay for the waiting anyway, if the rota is tight. When your operation needs someone within ten minutes of the site, paying a modest hourly rate for standby is often cheaper than litigating it later.

Hours worked from callbacks should land in the same timekeeping system as everything else. Our guide to calculating work hours covers the basics of getting that right.

The honest summary is that there is a real trade-off. A rota that is cheap to run is one where staff are lightly constrained, and a rota that guarantees fast response is one you will probably end up paying for. Pick which you want, then write the policy to match what you actually do, not what you wish you did.

What should you do on Monday?

Pull your current rota and score it against the factors above. How long do people have to respond? Where do they have to be? What happens if they are late? How often does the phone ring? Then check the one or two states where you employ people, because a state rule can override a perfectly good federal answer.

If any answer points toward paid time, decide now whether to pay for the waiting, shorten the leash, or both. Put the decision in writing, record callbacks as they happen, and feed those hours into overtime. That is the whole discipline, and it is a lot cheaper than the alternative.

This article is general information, not legal advice. On-call and call-in rules vary by state, city and industry and can change, and the figures here come from the official pages linked above as of November 2026, so confirm the current rule with your state labor agency before you set up a rota.