Debates about working time tend to focus on quantity. Are American workers having to put in too many hours, especially those with caring responsibilities? This is of course a vitally important question. But it is not the only one. It matters not only how many hours people work, but how much control they have over them. Irregular work hours lead to income volatility for hourly wage workers, increasing the difficulty of making ends meet. “Just-in-time” scheduling practices put workers in a vulnerable financial position—both by destabilizing earnings and by disrupting their access to safety net programs—and make it difficult for them to arrange childcare, attend school, or pick up a second job.
What is just-in-time scheduling and who is affected?
Just-in-time scheduling is used to match labor supply to consumer demand. Workers typically receive their schedules on short notice and may have their shifts changed or canceled at the last minute. Employers may put workers on call with no promise that there will be work available for them, or send workers home without pay if demand has slowed by the time they arrive for their shift. This frequently means that workers aren’t guaranteed a minimum number of hours per week.
Two in five wage and salary workers over the age of 15 know their schedules less than one month in advance, according to data from the American Time Use Survey Leave and Job Flexibilities Module. Nearly one in five know their schedules less than one week in advance. Men tend to have shorter scheduling notice than women, due at least in part to gender differences in employment in industries or occupations where short notice is common.
Less educated workers and Hispanic or Latino workers are particularly likely to experience short scheduling notice. Differences between racial and ethnic groups appear to be driven by more than differences in occupation. Other research has found that among workers in retail and food services, non-white workers are 10% to 20% more likely to experience several other markers of unpredictable scheduling, including on-call shifts, clopenings (back-to-back closing and opening shifts separated by fewer than 11 hours), and involuntary part-time work.
Industry matters a good deal for predicting whether a worker will be subject to just-in-time scheduling. Three in four workers in the leisure and hospitality industry receive their schedules less than one month in advance, and most have fewer than two weeks’ notice. Construction and agricultural workers are particularly likely to receive their schedules less than one week in advance.
Industries where just-in-time scheduling is less common (such as public administration) may nevertheless contain occupations subject to short scheduling notice. Some workers may even prefer a schedule that varies from week to week. But just-in-time scheduling has become the standard in industries such as hospitality and retail, which accounted for about 20% of all jobs before the coronavirus outbreak.
The SHIFT Project out of the University of California, Berkeley started surveying hourly retail and food-service employees at some of the nation’s largest firms about their work schedules and well-being in 2016. One-third of SHIFT respondents had less than one week’s notice of their work schedules, and their hours varied by an average of 32% from week to week. This means that a worker who worked an average of 25 hours per week in a given month probably worked as few as 20 hours in one week and as many as 30 hours in another.
Most workers in the SHIFT Project report having little control over when or how much they work. Half of SHIFT respondents said they had no input into their schedules, and another third said they had some input but that their employer ultimately decided. Additionally, more than one in ten respondents reported having a shift canceled on short notice in the previous month, and one in four reported working on-call shifts that may or may not result in work.
Just-in-time scheduling cuts labor costs in the short run, since businesses only have to pay for the workers they need at the moment. But they shift these costs onto workers by destabilizing work schedules and pay.
Material hardship from unpredictable hours
For workers paid by the hour, unpredictable hours mean unpredictable incomes. Three in four hourly wage workers between the ages of 26 and 32 report that their hours fluctuate from week to week, according to an analysis of the NLSY97. The average gap between a given worker’s shortest and longest workweek in a one-month period is 10 hours.
Similarly, an analysis of banking records by JPMorgan Chase found that Chase customers experienced a 20% change in labor earnings from month to month on average. A new working paper, also using Chase bank account records, finds that household consumption is highly sensitive to these month-to-month income fluctuations. This is particularly true for black and Hispanic households, who have fewer liquid assets (given the racial wealth gap) to buffer themselves against temporary income shocks.
Most people report that they would prefer to have a predictable income. More than three-quarters of participants in the U.S. Financial Diaries research study say that financial stability is more important than “moving up the income ladder.” This issue has become more salient as household income has become more volatile since the 1970s, in part because of greater variation in hours worked.
Income fluctuations can have significant consequences for families’ financial well-being. One in three workers with variable schedules say they are “just getting by” or “finding it difficult to get by” financially, compared to one in five of those who set their own schedules or have stable hours, according to the Federal Reserve’s 2018 Survey of Household Economics and Decisionmaking. The SHIFT Project finds that service-sector workers with a 50% swing in work hours are at a 13% higher risk of hunger hardship and an 11% higher risk of residential hardship than those with steady hours.
Losing access to safety net programs due to varying hours
To make matters worse, many safety net programs come with work requirements that aren’t compatible with volatile work hours. For instance, some TANF and SNAP recipients are required to work a minimum number of hours per week. Workers with variable hours may drop below the work hours requirement in precisely the weeks when they have the greatest need for an income buffer.
Work requirements are intended to incentivize work among those who are able, but this assumes that individuals in need of assistance can control their own work hours. Employer-controlled unpredictable scheduling violates this assumption. Despite this, the USDA recently expanded SNAP work requirements, and several states have moved to add work requirements to Medicaid in the last few years.
Our colleagues at The Hamilton Project found that about a quarter of SNAP participants who are typically exposed to work hours requirements fall below the required threshold during at least one month over two years. According to their analysis, many SNAP and Medicaid recipients who could be newly exposed to work requirements would likely be disqualified due to work-related (that is, involuntary) employment gaps. Another study found that adults with more unpredictable schedules were at a higher risk of exiting child-care assistance programs after short enrollment spells.
(Many work requirements for public assistance have been waived during COVID-19.)
Work-family conflict worsened by variable hours
Unpredictable work schedules also make it difficult for employees to schedule everything else in their lives, from dentist appointments to second jobs. This is particularly challenging for parents who have to arrange childcare without advance notice of when they will have to be at work. Young children of U.S. food and retail workers subject to last-minute scheduling go an average of 15 days per year without childcare or supervision, compared to 9 days for those who are not subject to last-minute scheduling. Some research has found that unpredictable work schedules are associated with negative behavioral outcomes for children.
Many proposals to address work-family conflict focus on increasing workplace flexibility. Greater worker control over scheduling can reduce work-family conflict.
Experimental evidence suggests that most people would prefer to have an inflexible schedule than an unpredictable one. Researchers gave jobseekers at a U.S. call center the choice of either a standard on-site job or a randomly selected alternative, such as employer discretion in scheduling (by which employers assign shifts one week in advance) or employee-controlled flexible scheduling. Applicants were willing to take a much larger wage cut to avoid employer discretion in scheduling they were than to obtain any of the employee-friendly alternatives.
Pioneering cities, states, and companies lead the way on fair scheduling
Fair scheduling legislation has been enacted in the state of Oregon and cities such as San Francisco, New York, and Seattle. Many of these laws mandate that employers pay workers for last-minute scheduling changes (“predictability pay”) and for on-call shifts. Some place restrictions on “clopening” shifts by requiring a minimum rest period between shifts. Evaluations of some fair scheduling ordinances are currently underway.
The Schedules That Work Act, first introduced in 2015 by Senator Elizabeth Warren and Representative Rosa DeLauro, would bring many of these reforms to the federal level. In its current form, the bill covers workers in retail, food services, cleaning, hospitality, and warehousing. A key provision requires employers to pay workers for scheduling changes made with fewer than 14 days’ notice. The bill would also require employers to provide a “bona fide business reason” for denying an employee’s request for a schedule adjustment.
Some companies have moved toward greater worker control over scheduling in the absence of legislation. Gap implemented two-week advance scheduling notice at its retail stores and eliminated on-call shifts back in 2015. The company also partnered with a team of researchers to randomly assign some of their retail stores to implement additional stable scheduling practices, such as giving employees more consistent start and end times and limiting variation in hours. These measures increased median sales by 7% in treatment stores.
Walmart, the U.S.’s largest private employer, spent several years testing a system that provides more fixed shifts and worker-controlled scheduling options. Early results showed that the new system reduced absenteeism and turnover, and the company announced that it would implement these practices nationwide in 2018. All of this occurred in a tight labor market, when employers typically have to offer higher pay or better workplace practices to attract and retain workers. The situation now is very different, as weekly unemployment claims soar to unprecedented levels. There is some danger that fair scheduling laws could impede employment recovery later on by making it riskier for employers to hire workers during a period of uncertain consumer demand. But workers should not have to relinquish control over their time in order to earn a (volatile) paycheck. Reasonable constraints on just-in-time scheduling practices would improve workers’ well-being. They may also benefit companies in the long run by improving morale and reducing turnover.
As more families rely on more than one earner, the politics of time are becoming as important in some ways as the politics of money. Uncertainty is bad for most workers; it may be bad for business too.