Policy Wonk

What Should We Make of the UW Minimum Wage Study?

UW researcher Jacob Vigdor responds to criticisms of his team's analysis.

By Hayat Norimine June 26, 2017

Ed murray minimum wage r6mdxs

Rarely does a study put University of Washington researchers at odds with the city, but that's what happened today with new analysis released on the city's minimum wage. Seattle officials have in the past looked to the top-rated university's experts as a resource to develop policy—UW Law (Hugh Spitzer) for the city's income tax, UW School of Public Health (Jim Krieger) for the soda tax, UW Aquatic and Fishery Sciences (Jeff Cordell) for the seawall's salmon habitat. This time, though, a University of California Berkeley professor briefed the city on the effects of the minimum wage the same morning a group of UW researchers came out with their study on the same topic. 

On Monday mayor Ed Murray-championed minimum wage bill took a hit when UW researchers released conclusions that the legislation led to fewer hours worked (a 9 percent cut) and a total loss in wages ($125 a month). Berkeley professor Michael Reich pointed out that UW's results are three times more than what minimum wage critic David Neumark estimated. Robert Feldstein, director of the mayor's Office of Policy and Innovation, on Monday sent a letter to Jacob Vigdor—a professor from UW's Daniel J. Evans School of Public Policy—echoing criticisms from Reich. Before the UW study came out, the mayor's office asked him for comments, according to a letter from Reich dated Monday.

"Although we appreciate the time and effort invested by your team, professor Reich's concerns lead us to believe that your report may not, in fact, tell the full story of the effects of Seattle's minimum-wage policy," Feldstein wrote to Vigdor. "We would hope that future work will address these substantive and methodological concerns, and that your team will be open to looking at how to examine more precisely the characteristics of Seattle's labor market." 

The study had very different results to previous minimum wage reports that have been published—including a Berkeley study published just last week–and Reich at a city council briefing Monday said UW's findings "are not credible. They have a lot more work to do."

But, as Vigdor pointed out, both UW and Berkeley studies reported the same results among restaurant jobs. The minimum wage didn't hurt those jobs. UW's report included employees' number of hours worked and included seven faculty members, as well as some graduate students.

"If you do an apples-to-apples comparison, there's absolutely no contrast between our reports," Vigdor told PubliCola. "The catch is that restaurant employment is not the same thing as low-wage employment. It's when we look specifically at low-wage employment that we're picking up very different results." Vigdor said a lot of restaurant jobs, especially in Seattle, cater toward affluent Seattleites and hire employees who have a fair amount of experience in the service industry. That means higher wages.

Criticism: The UW report excluded multi-site businesses. 

Explanation: UW researchers used data from the Employment Security Department, to which employers send their data every three months. Let's say business owners have two locations, one in Seattle and one in Bellevue. Owners then have two options when they're reporting their numbers—they can either add to an existing ESD account or open up a second account to keep their locations separate. With a multi-site business, you can't tell which employees are based in Seattle and which are based in Bellevue, Vigdor said. Employees outside of the city wouldn't be affected by Seattle's minimum wage bill. 

So yes, it's possible some of those jobs "lost" actually just got transferred outside the city. But Vigdor said a survey showed those businesses—which tend to be larger corporations with more employees—are also more likely to report reducing employment as a consequence of minimum wage. 

Criticism: The UW report compared Seattle to other parts of the state, as opposed to similar cities across the country. 

Explanation: Data varies quite a bit depending on the state. For example, it would be tough to compare Seattle to San Francisco because there's a completely different state agency that collects employment data in California. The quality of data collected by Washington state's Employment Security Department is way better—it includes the number of hours worked. 

Criticism: The UW report went too low by capping low-wage jobs to $19 an hour. 

Reich said over the same period, jobs at all pay levels increased at single-site businesses.

Explanation: Vigdor said the researchers put a lot of thought into where to draw the line between low-wage and high-wage jobs. They repeated their analysis 13 different ways, he said, experimenting with wages anywhere between $13 to $25 an hour. "The results don't change." 

Criticism: The UW report didn't adequately account for Seattle's economic boom and how that factors into the results. 

This is the toughest factor to consider. With an economic boom, employers replace low-wage jobs with higher-wage jobs to compete with other employers in the area—so maybe jobs aren't disappearing, just paying better. Reich pointed out that it's difficult to separate out the boom's effects, especially in the UW report. The control cities UW used (other cities in the state) aren't booming like Seattle, and may not be adequate controls at all.

Explanation: "You have to ask yourself, 'Was Seattle in the economic doldrums until January 2016, when it suddenly grew explosively?' That's the only kind of boom that could explain away our results," Vigdor said, which showed those employment impacts appear beginning of 2016. That's exactly when the $13 minimum wage began. "Anybody living in Seattle back in 2015 will probably recall that it was booming pretty well back then too." 

It's fair to say UW did its due diligence—and though some may still dispute the researchers' claims, it certainly wasn't out of any sort of negligence. 

The big takeaway? Vigdor said it's too early for that. The UW study isn't perfect, and Vigdor said they're not claiming any sort of conclusion. The message is not, "Seattle's minimum wage is terrible for poor workers." (Though some headlines offer some derivative of that.) There's still a lot of work to do and a lot of unanswered questions. Researchers have 30 other papers in the works on the minimum wage, Vigdor told PubliCola, and they're not expected to finish that research until 2025. 

"We get this question asked repeatedly," Vigdor said. "What should Seattle do about this? What should another city that's contemplating another minimum wage think about this study? There's still more that we need to know before you make any kind of judgment call like that."

For one thing, the UW study doesn't elaborate which low-wage employees are being hurt the most. Is it teenagers who still live with their parents, or, say, single mothers who work well over 40 hours a week and barely cover their rent? Which workers are being impacted, and how? The data researchers received from the Employment Security Department include names, social security numbers, and information about the hours they work and their wages, but they know nothing about the people beyond that. 

The next step, Vigdor said, is to bring data in from the Department of Social and Health Services—to keep track of who's eligible for benefits like SNAP, Medicaid, and housing assistance—and Department of Licensing to get birthdates. Because of the sensitivity of that private information, Vigdor said, taking appropriate cautions to protect people's identity and finalizing a contract with DSHS will take another few months. 

Updated June 27, 2017, at 10:34am: This post provides more explanation on the economic boom and how it would affect the study. 

Show Comments