ADVERTISEMENT

A Sign to Go Slow on the $15 Minimum Wage

A Sign to Go Slow on the $15 Minimum Wage

(Bloomberg View) -- One thing almost all economic studies agree on these days is that higher minimum wages don’t throw many people out of work. A recent study of Seattle’s high-profile plan to raise minimum pay to $15 an hour by three University of California Berkeley economists found no drop in employment in the food-services industry as a result of the new higher wage. And a parallel study by the Seattle Minimum Wage Study Team at the University of Washington also estimated that the effect of the higher minimum wage on restaurant employment was zero. Other recent papers continue to find the same at the national level.

But that doesn’t mean it’s safe to raise the minimum wage to the levels that activists demand. Even if they don’t put people out of a job, higher minimum wages might force employers to cut workers’ hours.

Up until recently, that wasn’t happening in Seattle either. Seattle’s new policy mandated that the city’s minimum wage for large employers that don’t offer health insurance be raised in a series of steps from $9.32 in 2014 to $15 in 2017:

A Sign to Go Slow on the $15 Minimum Wage

Last year, the University of Washington team charged with evaluating the policy found that the 2015 rise to $11 hadn’t affected working hours very much. Last year's increase to $13, however, may be a different story. But the team’s 2017 follow-up found something more worrying -- although employment in Seattle’s restaurant sector has held up under $13, the hours worked by low-wage restaurant workers fell. The authors estimate that the reduction in working hours was so large that it more than canceled out the rise in hourly pay, leading to income losses for low-wage workers relative to what they would otherwise have taken home.

Now, there are some important caveats to this result. First of all, it’s just one paper -- most minimum-wage studies don’t look at the number of hours that employees work, since this data tends to be hard to come by. As a result, there are lots of studies about how minimum wage affects job numbers, but relatively few on how it affects hours.

Because the new Seattle Minimum Wage Study paper is just one paper, many people will inevitably question whether it used the right measurement techniques. Ben Zipperer and John Schmitt of the Economic Policy Institute have already written about a number of potential issues with the University of Washington researchers’ choice of methods. It’s possible that the study’s authors chose the wrong set of cities to compare to, or underestimated the number of workers who moved into higher-wage positions as a result of the policy change. They also didn’t measure chain restaurants, which constitute a large chunk of the industry. Until many more such studies are done, there will be a lot of uncertainty about whether minimum wage causes employers to cut working hours.

But despite these caveats and uncertainties, the University of Washington paper should give pause to the activists and politicians who want a nationwide $15 minimum wage.

For one thing, it reveals a new and disturbing channel by which minimum wages could hurt the very workers they’re designed to help -- reduced working hours. In the past, if an American worker had a job, that generally meant a guaranteed number of working hours per week, even if the job wasn’t full-time. But in recent years that has no longer been the case. A 2015 report by the Economic Policy Institute found that 10 percent of the workforce now has irregular or on-call work shifts. This irregularity is concentrated among low earners in service industries like the restaurant industry.

Non-guaranteed working hours give employers a new way to defray the higher costs imposed by a minimum wage hike. Instead of laying off workers, bosses can just call lots of their existing workforce in for fewer hours each week. Nobody will show up on the unemployment rolls, but workers’ incomes will drop all the same.

The University of Washington study raises the possibility that this has been happening a lot more than minimum-wage researchers have realized. Lots of the papers that find zero employment effects might be missing a drop in working hours. Because detailed data on hours is relatively rare, it will take some time for other research teams to follow up on this new angle.

Meanwhile, policy is moving fast. Even before the results of recent studies are fully digested, Seattle will raise its minimum wage again. The entire state of California will soon follow. And “Fight for 15” advocates want to extend the $15 minimum wage to the entire country, even to rural and poor regions where employers will have a much harder time paying that amount.

The University of Washington paper shows why this headlong rush might be unwise. It’s crucial to remember that even if small minimum wage hikes don’t hurt the economy, there is some level of minimum wage that just won’t work. If Seattle raised its minimum wage to $10,000 an hour, for example, all legitimate economic activity would become impossible, and businesses could only employ workers under the table.

Where does minimum wage policy start to hurt poor workers instead of help? There’s no general answer -- it’s different for each city, based on the strength of the local economy. The new University of Washington paper shows that at $13, Seattle may already have entered the danger zone. It’s a reminder that cities, states and especially the country as a whole should go slow, pay attention to the data and always be prepared to pull back if they go too far.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Noah Smith is a Bloomberg View columnist. He was an assistant professor of finance at Stony Brook University, and he blogs at Noahpinion.

To contact the author of this story: Noah Smith at nsmith150@bloomberg.net.

To contact the editor responsible for this story: James Greiff at jgreiff@bloomberg.net.

For more columns from Bloomberg View, visit http://www.bloomberg.com/view.