The Washington PostDemocracy Dies in Darkness

The brutal psychological toll of erratic work schedules

Unpredictable hours and variable pay may cause even more distress than low wages.

A young kitchen staff worker preparing flat-bread pizza in a fast-food restaurant. (iStock)

Stacey, a single mother, was getting 30 hours a week when she first started working at a big-box store in the Bay Area, in early 2015. But her hours had grown erratic by the time we interviewed her, a few months later: 20 hours one week, 12 the next, then 12 again, and then only eight. The following week was even worse: just four hours, and on a Saturday — meaning she’d need to pay for child care for her 8-year-old son.

The one bright spot in that short week was that “now the pain is going away,” she said — the discomfort she got in her feet and her legs from standing all day. But even that wasn’t worth the chaotic schedule and resulting financial pressure.

When we talked with Stacey (a pseudonym, as required by the research-ethics rules we worked under), the single mother was making a little better than California’s minimum wage of $9, and she relied on ultra-high-interest payday loans to get through the slow periods.

The rallying cry for millions of workers is a $15 an hour minimum wage. But in addition to low pay, erratic schedules are another bane of American workers, particularly in food service and retail: They interfere mightily with family life and are associated, our research finds, with poor sleep, psychological distress and lower levels of happiness.

There’s a strong case for raising the minimum wage. But our research suggests that regulations that impose some semblance of order on workers’ hours could have an even bigger impact on workers’ well-being than a raise.

That’s obviously not to say that we should choose between these two reforms, but the finding demonstrates just how disruptive modern just-in-time schedules are to workers’ lives — and the problem isn’t getting nearly as much attention from policymakers as low wages.

It makes sense that unpredictable hours cause unhappiness, but until now there’s been no data to explore the question. Since 2016, through a study called The Shift Project, we’ve been exploring the contours and consequences of just-in-time scheduling across the country (zeroing in at times on certain cities including Seattle, New York, and Philadelphia). We’ve surveyed 84,000 people, focusing on workers in 80 of the largest food-service companies (usually fast food) and retail chains, because those sectors are notorious for their use of erratic schedules. We asked workers detailed questions about their schedules, economic security, health, and general well-being. (Nationally, 14 million people are employed in food service, and 9 million in retail positions.)

Unpredictable schedules make sense from the employer’s point of view, from a purely economic perspective: The point is to precisely align staffing with demand, and thereby transfer risk from company payrolls to employees’ household balance sheets. But this approach has big costs in terms of employee welfare.

How to put the long-term unemployed back to work

Our survey affirmed the scope of the problem. Only about one in five of the people we surveyed work a regular daytime shift. About two-thirds of workers receive their weekly work schedule with less than two weeks’ notice, and one third get less than one week’s notice. Sixteen percent get less than 72 hours’ notice — a scenario that makes it basically impossible to plan child care, family meals or homework time.

A particular onerous task demanded of retail and food-service workers is to work a closing shift and then, immediately afterward, the opening shift (say, closing the store at 11 p.m. then returning to open a few hours later). That’s called a “clopening,” and half of our respondents said they’d worked one.

More than a quarter reported they’d been asked to be on call — meaning they set aside a block of time for the company but might not end up working or getting paid.

This kind of work exerts a toll. Forty-six percent of the people in our sample at least some psychological distress (defined in the survey as nervousness, feelings of hopelessness and worthlessness, and a sense of being overwhelmed). That appears to be significantly higher than for the typical low-income worker, though precise comparisons are difficult. And our analysis of the data showed that the distress increased as notices of scheduled hours grew shorter, when workers had shifts canceled with little notice, and when they worked “clopenings.” Sixty-four percent of workers who had had shifts canceled reported psychological distress, for example, compared with 43 percent of those who did not.

The picture was similar with sleep. Seventy-four percent of our respondents reported “poor” or “fair” sleep — and the more irregular hours, the worse the reported sleep.

Because of a lack of data, we don’t have a clear picture of change over time, although it seems clear that companies have gotten much more aggressive on this front. One exception to the lack of information involves work-hour variation. In the Great Recession, week-to-week work hour variation spiked, especially for low-wage and less educated workers, and remains high. In our data, workers report a 32 percent variation, in hours worked, month to month, even as their bills come with ruthless regularity.

Some cities are waking up to this problem. Since 2014, San Francisco, Seattle, New York City, and Philadelphia have passed laws regulating scheduling practices, to varying degrees, in retail and food service — as has Oregon. Seattle’s legislation requires such employers to provide at least two weeks’ notice of work schedules. If they make a change within that window, they must pay extra for added hours, and give half-pay for subtracted hours. Employers must also give workers 10 hours’ rest between shifts. If they ask workers to do a “clopening” shift tighter than that, they pay time-and-a-half for hours worked during the rest period. New York’s law requires that fast-food workers get 72 hours’ notice of shifts, bans last-minute shift cancellations, and forbids on-call shifts. (New York has separate rules for retail employees.)

Poverty is moving to the suburbs. The war on poverty hasn’t followed.

These provisions don’t require truly stable schedules, but they do meaningfully increase the predictability of working hours. Similar laws are being considered in Los Angeles, Chicago, Washington state and Connecticut. At the federal level, a “Schedules that Work Act” was last introduced in Congress, in 2017, but it languished.

Based on the connection we identified between schedule uncertainty and psychological distress, we estimated what would happen if certain policies changed in a city or state currently lacking any schedule regulation. If companies were required to give at least 72 hours’ notice of a shift change, up from less than that, average rates of psychological distress would drop by 4.5 percentage points for the affected workers. Eliminating on-call work would reduce distress by an impressive 15 percentage points. Meanwhile, raising the minimum wage by $4 would cause distress to drop by only 2 percentage points.

Again, the point is not to diminish the importance of a minimum wage, but to provide a fuller picture of the challenges low-income workers face.

We often hear calls to “make work pay,” but for workers’ lives to be manageable, employers — possibly with a strong push from lawmakers — also need to make work predictable.