By Ethan Bernstein If the past is any guide, the Schedules that Work Act (S 1772/HR 3071), fair work scheduling legislation introduced in Congress last week by Sen. Elizabeth Warren, D-Mass., and Rep. Rosa DeLauro, D-Conn., will ignite debate between employer and employee advocates. Yet solutions favoring employers and those that meet employees’ needs may be closer than you think. The debate masks how much smart labor scheduling can benefit both sides.
The bill would provide retail, restaurant and building cleaning employees with at least two weeks’ notice of schedules and compensate employees for last minute schedule changes, split shifts and on-call work. It would give employees in all occupations the right to request scheduling accommodations without fear of retaliation, and require employers to accommodate requests from certain employees, unless they are unable to do so for bona fide business reasons.
Recent articles on retailers such as Starbucks, Victoria’s Secret and Wal-Mart have highlighted workers facing unpredictable, erratic schedules with too few hours and too little stability. In the press, the issue often plays out as a struggle between workers, whose lives are turned upside down by volatile schedules, and businesses, who need flexibility to respond to fluctuating demand in industries with tight margins.
But research suggests that there is actually considerable common ground between the parties.
At its core, labor scheduling software is actually quite simple. It does two things: 1) processes lots of data — more than a human being could — to predict future staffing needs; and 2) uses those predictions to automatically create a schedule without human effort.
There is nothing inherently evil or miraculous about that. What matters is how it is implemented. The power of that software could be used to schedule individuals into 15-minute shifts that push them perilously close to unsustainability and wreak havoc on their lives. Or it could be used in conjunction with worker preferences to create schedules that provide stability to individual workers, within the constraints of predicted demand. In short, the increasingly common plight of workers is not due to systems themselves, but rather how they are used. It is a managerial issue, not a technological one.
And that’s why I, as a management professor, believe there’s hope.
The retailers who have been highlighted in the press — and many who haven’t — have learned from their initial mistakes with these scheduling tools. Implementation v2.0 looks very different than Implementation v1.0.
Implementation v1.0, for example, relied on temporary or part-time workers to fill the seemingly random allocation of shifts that some scheduling systems churn out, even though research shows a tipping point after which adding more temps and part-timers leads to decreased motivation, training, and profits. Implementation v2.0 is using better predictability to provide regular employees with more stability.
Implementation v1.0 also ignored the potential unintended consequences of employees feeling increasingly monitored. Implementation v2.0 is giving the data to employees, not just collecting it, so they can contribute to the scheduling process, using their mobile devices to swap shifts and provide availability information.
Implementation v1.0 was all about cost. Under pressure to reduce labor costs, managers were incentivized to make scheduling decisions that were bad for business: one study of a women’s apparel retailer’s 41 stores found them all to be understaffed at peak periods, leading to significant losses. Implementation v2.0 is about a smarter approach to management, which a team of us is studying at a major US retailer consistently rated among the “best places to work”.
Here’s the key point: data-driven, automated scheduling software is a huge step forward, but implementing it as an advancement for retailers at the expense of their employees misses the point. It actually undermines the true value of the systems. When was the last time a retailer aimed to employ a worker who is too sleep-deprived and unhappy to serve a customer well? Early results of our research indicate that, when the right mentality and tools are employed, employers can use scheduling software to engage employees, incorporate their preferences into the scheduling process, and maximize productivity and sales. Implementation v2.0 is real progress.
The unfortunate reality, however, is that most retailers can’t seem to skip right over v1.0 to v2.0. It’s like needing to load Windows 95 on your computer before upgrading to Windows 10.
Rather than debating who should “win and lose” in the scheduling fight, we should be discussing how to get more organizations to Implementation v2.0. Legislative efforts such as the Schedules that Work Act can improve scheduling for workers and retailers by nudging managers towards Implementation v2.0. Ironically, all of the bill’s requirements are more practical (not less) because of the technology we have available– and they can make sense for employers and employees alike.
Ethan Bernstein is an assistant professor of business administration in the organizational behavior unit at Harvard Business School.