Skip to main content
Research

Gig Workers Value their Flexibility... a Lot

The promise of the gig economy is of working on a schedule of your own choosing. Does controlling their own time actually make gig workers better off? Using extensive data on Uber drivers, Yale SOM’s Judith Chevalier and her co-authors examined their driving patterns to understand the economic value of flexible scheduling. They found that rideshare drivers would have to earn as much as double to accept less-flexible arrangements.

Illustration of an Uber driver enjoying flexibility

Sean David Williams

By Jyoti Madhusoodanan

While visiting Chicago a few years ago, economist Judith Chevalier got into a conversation with her Uber driver. She learned that the driver’s primary job was running a small, traditional funeral home. But that didn’t cover the bills, so when there were lulls in her responsibilities at the funeral home, she drove for Uber. Driving for the rideshare company was one of the few jobs that allowed her to take days off every time there was a funeral to arrange, the driver said.

Chevalier was struck by her driver’s description of the modern gig economy. “Not many jobs will allow a worker to just show up when they feel like it, and work only for as long as they choose to,” says Chevalier, the William S. Beinecke Professor of Finance and Economics.

While many surveys have found that being able to choose when and where they work is important to employees, precisely how workers make the most of flexible scheduling for their personal benefit wasn’t clear. In a new study, Chevalier and three co-authors calculate the value of flexible work by looking at the minimum wage a person requires to make a job worthwhile at various times.


Read the study: “The Value Of Flexible Work: Evidence From Uber Drivers”

“Most Uber drivers would not participate at all if you forced them to work long hours. The magnitude of this need for flexibility is surprising.”

To measure the value of flexibility, the researchers gathered anonymized data on driving behaviors from Uber. The data covered all drivers on the UberX platform in the period from September 2015 through April 2016. After applying some filters, the researchers worked with data on 197,517 drivers who worked a total of 102,280,904 hours. The data was coded by time, so the team could track whether or not a particular driver was working for each hour of a week. Typically, a driver worked about 12 hours in a week. But there was a wide variation across drivers in how much and when those hours were. Even for a given driver, the hours they drove in one week were often very different from those they worked in the next week. Some drivers were morning workers, some worked evenings and weekends, and many drivers changed their patterns from week to week. “The patterns speak to a demand for this kind of flexibility,” Chevalier says.

Having confirmed that drivers did indeed make use of the flexibility offered by Uber, the researchers used the data to identify how an individual’s wage requirements—their reservation wage—varied over time. If they saw, for example, that a particular driver worked at a time when the expected wage was $15 an hour, but not when the wage was $25 an hour, they could infer that there were additional constraints on the person’s availability. They compared the variation in an individual’s reservation wage over the course of a week with the wage available at different times to determine how much value drivers put on being able to choose their own hours.

“Once we learned about different drivers’ willingness to work certain hours, we imposed what we call ‘counterfactuals’,” Chevalier explains. “We asked, okay what if you had to work eight hours in a row, or four, instead of this variable schedule?”

The researchers also determined how much more Uber drivers would need to be paid in order to accept less flexible work arrangements—for example, one in which drivers control their work hours but can’t adjust them on the fly, and another in which they select eight-hour shifts like a traditional taxi driver. They found that drivers would require almost twice as much pay to accept the inflexibility of the taxi schedule.

Flexibility, Chevalier concludes, is central to Uber’s appeal to its drivers. “Most Uber drivers would not participate at all if you forced them to work long hours,” she says. “The magnitude of this need for flexibility is surprising. Even drivers who work a lot do so pretty unpredictably, and might not drive if they had to work more conventional blocks of time.”

The finding likely extends to Lyft, TaskRabbit, and other companies that rely heavily on independent workers. But it’s not necessarily reflective of a typical worker’s preferences, Chevalier emphasizes. People who take on such employment may be doing so in addition to their primary job, or because their primary job is inflexible and they need more money, or for other reasons.

“The people we are sampling here are those for whom this uncommitted version of work is particularly attractive,” she explains. These people may be actors earning money between gigs, funeral home directors, or parents with a few hours to spare before their child gets out of school. “The gig economy is often the only choice that works around these other constraints.”

But the results could help employers of all kinds understand the monetary value of flexibility when trying to design or alter schedules or contracts, Chevalier says. “Once we understand a person’s work patterns and willingness to work—or not—for certain hours, we can use that to think about whether that individual might be willing to work a four-hour shift or an eight-hour shift,” she says.

The results also shed light on the debate over the shift toward a gig economy. As the authors point out, up to 30% of U.S. workers participate in independent work in some capacity. Critics have pointed to what they see as downsides of this trend, including a lack of benefits like overtime and health insurance. This study shows that the non-financial benefits offered by Uber and other employers—the flexibility to take care of family members, work alternate jobs, or just take a few hours off to watch the big game—have their own kind of economic value.

Department: Research