The $15 Question

A major new study says Seattle’s ambitious minimum wage hike is hurting workers. Should Democrats support it anyway?

A restaurant like this Chipotle Mexican Grill in Seattle might have cut back its low-wage jobs because of the city’s increased minimum wage.

AFP/Getty Images

Last Monday, a team of economists from the University of Washington delivered an unflattering early verdict on Seattle’s historic $15 minimum wage ordinance: The law, which has yet to fully phase in, was already hurting the very people it was meant to help. By making it more expensive to pay their employees, the researchers found, the city had led businesses to curb jobs and hours for low-wage workers, who as a result were earning an average of $125 less per month.

Labor activists and progressive politicians responded to the study with a combination of criticism and indifference. The union-backed Economic Policy Institute issued a lengthy rebuttal questioning the paper’s methods and pointing out some apparent inconsistencies in its findings. By Friday, meanwhile, the city of Minneapolis had become the latest progressive bastion to pass its own $15 pay floor, joining the likes of New York, D.C., the entire state of California, and, of course, the booming tech hub where researchers had just reported that the experiment was going badly wrong.

And that is the state of America’s minimum wage debate in a nutshell. Rather than let the early test cases play out for a bit, a growing number of left-leaning politicians have simply decided to embrace the idea of a $15 minimum wage as a central piece of the progressive agenda—and not just in wealthy cities. When Sens. Bernie Sanders and Patty Murray introduced a bill in April to raise the federal minimum to $15 an hour, they landed 22 co-sponsors, including Senate Minority Leader Chuck Schumer. Two years ago, a similar bill got just five co-sponsors. The idea has rapidly gone mainstream in Washington while economists are still figuring out what it might do to the workers it’s supposed to benefit.

Politically, that makes perfect sense. The movement for a radically higher minimum wage has energized grassroots activists while giving an ailing labor movement some rare concrete policy victories at an otherwise dark moment for progressive priorities. It’s also fairly popular: According to the Pew Research Center, 52 percent of all registered voters supported a $15 pay floor in 2016, including 82 percent of Clinton voters and 21 percent of Trump supporters. It polled especially well among lower-income Americans who would theoretically benefit from the policy, with 67 percent of those from families earning less than $30,000 a year favoring it. For Capitol Hill Democrats trying to revive the party’s electoral prospects, embracing the Fight for $15 is the obvious, and maybe even correct, move.

But if you like the idea of crafting public policy based on evidence, the fact that a $15 minimum wage is quickly becoming Democratic Party orthodoxy should make you feel at least a little bit ambivalent. Because the party may be sacrificing the best interests of low-wage workers, even if it’s giving them exactly what they want.

* * *

When Seattle became the first major American city to pass a $15 pay floor in 2014, it was hard for anyone to predict what would happen next. Economists had been arguing for decades over whether raising the minimum wage killed jobs for people on the bottom of the economic pyramid like teens and fast-food workers. One line of research, preferred by conservatives, suggested that, yes, making low-cost labor more expensive led to modest but noticeable employment losses for those vulnerable groups. Another line, preferred by liberals, suggested the effect on jobs was slim to nonexistent. But almost all of the studies had been based on relatively small minimum wage increases. Would a more dramatic hike have a more dramatic impact on the job market? It was unclear. As University of Massachusetts–Amherst economist Arindrajit Dube, who has done as much as anybody in recent years to show that minimum wages might have a negligible effect on jobs, told me at the time, “We are not going to be able to predict very well the likely employment impacts of this policy.” Researchers simply had never studied a hike at this scale before.

To its credit, Seattle didn’t simply enact its bold new policy in 2015. It also commissioned the University of Washington team to study its results. The working paper that group produced last week is interesting and alarming but also possibly flawed and by no means the last word anybody will have on this subject. It’s also still only an early draft, meaning the researchers will continue refining their results and addressing criticisms as time goes along.

One potential advantage their study has over many previous minimum wage studies is its data. Washington is one of only four states that collect information on both earnings and hours for individual workers each quarter, which allowed the researchers to examine the new pay floor’s effect on the full array of actual low-wage jobs instead of relying on rough proxies like restaurant or retail employment, which is what economists have tended to do in the past. That’s an important difference for a few reasons. First, it’s possible that business owners could react to a higher minimum by cutting entry-level jobs and hiring more experienced workers they don’t have to train or can trust with more responsibility in return for a bigger paycheck. Or, instead of paring their headcount, managers could just cut their employees’ hours to save money. The wage and hour data let the Washington team poke at those possibilities, which a more traditional study just looking at the number of people working at McDonald’s, Arby’s, and Chipotle might miss.

Here’s how the study worked: To determine how the policy had affected Seattle’s labor market, the authors created a “synthetic” version of the city to use as a comparison—essentially, a mathematical mash-up of other cities and suburbs in Washington that, once stitched together, closely tracked the actual Seattle economy before the ordinance went into effect, and could therefore depict an alternate reality where Seattle’s pay floor didn’t rise. Based on this counterfactual, they found that the law had reduced the total number of hours worked by low-wage employees by 9.6 percent. The total number of jobs paying less than $19 an hour—their threshold for “low-wage”—fell by 6.8 percent, compared with the counterfactual. Obviously, many workers received raises as a result of the new minimum. But thanks to all those shortened shifts and lost cashier gigs, the average low-wage worker ended up taking home $125 less each month. Which, for those workers, is quite a lot.

This is pretty much the worst-case scenario for a minimum wage policy. Many people find the idea of losing a few jobs acceptable as long as low-wage workers make more money overall. But raising wages to the point where they force businesses to cut back so significantly that low-pay employees make less as a group is almost certainly not an outcome anyone wants.

Beyond its worrisome headline findings, the study also suggests that previous researchers may have missed the minimum wage’s job-killing effects in industries like fast food because they simply didn’t have detailed enough data. When the team looked at total employment in Seattle restaurants, it found the rising minimum wage had no apparent effect. But on closer inspection, they found a more complicated situation. Low-wage jobs did decline, but employers added higher-paid positions. This, they argued, suggested that low-skill workers were being nudged out of the job market in a way that the previous literature wouldn’t have detected.

The result is frightening for at least one other reason: Seattle’s minimum wage hadn’t even hit $15 during the period the researchers studied. By the end of 2016, it had maxed out at $13 for some large employers and $12 for smaller employers. And yet the ill effects were already kicking in.

Or were they? Critics have suggested that there may be another, more obvious reason why low-wage jobs are seemingly disappearing from Seattle. It’s a tech boomtown with an unemployment rate around 3 percent. With a red-hot labor market, people might simply be nabbing raises and moving to better-paid industries. The paper tries to control for Seattle’s zippy economy by constructing that “synthetic” version of the city out of other parts of the state. But in Washington, there’s really no other place remotely like Seattle, and it’s not really clear that an analogue made up from weighted bits of Spokane and Snohomish County (among other locales) can really stand in for it. If the researchers failed to account properly for Seattle’s boom, then that may be what’s actually driving their results, not the minimum wage.

Is there any actual sign that’s the case? Possibly. The Economic Policy Institute’s Ben Zipperer has pointed out that, taken at face value, the Washington team’s results suggest that the new minimum wage actually caused an increase in the number of jobs paying $19 an hour in the restaurant industry as well as a rise in the number of hours worked by those sorts of high-wage employees in all sectors. This is a bit puzzling, since the Washington team’s paper is premised on the idea that Seattle’s minimum wage shouldn’t affect jobs that pay more than $19 an hour.

Zipperer told me that was a sign the study was confusing a booming economy that was “moving from low-wage jobs to high-wage jobs” with job losses from the minimum wage. “That to me is a very big red flag,” he said.

Progressive think tankers aren’t the only ones who have cast doubts on the results. Salim Furth, an economist with the archconservative Heritage Foundation, also had questions about the research group’s methods. “If I was really sure that they got their estimates right, this would be libertarian spike-the-football time,” he told me, a bit wistfully. For the time being, he’s more interested in their raw data, which show that the absolute number of low-wage jobs in Seattle declined as the minimum increased—not exactly what you’d expect in a boomtown—as well as surveys they conducted in which employers said they had cut staff due to the hike.

As of now, the study’s authors aren’t ready to answer all of their critics’ questions. “We’ve decided we’re going to go through each of the points in detail and incorporate them into future iterations” of the paper, Hilary Wething, the paper’s principal data analyst, told me after I brought up some of Zipperer’s points.* ”I feel uncomfortable brushing off any of those comments as illegitimate.”

Which is fair enough. Good research takes time, and economists will spend a while duking it out over the results of this paper, large portions of which amount to a critique of the entire previous minimum wage literature. (Notably, the mayor of Seattle requested a dueling study from UC–Berkeley’s Institute for Research on Labor and Employment, which was released the week before the University of Washington paper and suggested the wage hike was going just fine. This was unsurprising, given that the authors have been the go-to source for papers espousing the benefits of a $15 minimum. Moreover, they used an odd-seeming set of suburbanish locales to create a synthetic Seattle, so their paper is vulnerable to its own methodological critiques.)

But politics moves quickly. Assuming that there’s at least a possibility the Washington study will turn out to be correct that a $15 minimum wage is too high even in a city like Seattle, what’s the right thing for progressives to do?

* * *

If your first rule in politics and policy is “do no harm,” the obvious answer to that question is to be cautious and embrace a more modest goal like an $11 minimum wage—which would still be far higher than the current federal floor of $7.25 an hour and (according to the Seattle study) on the edge of what’s economically responsible. After all, we’re talking about an idea that may end up backfiring and hurting precisely the people it’s supposed to help. Meanwhile, there are all sorts of other policies that could help low-income families, from boring, standard-issue Washington fare (like raising the earned income tax credit) to the more ambitious (like a universal child allowance).

But to state the obvious, measured liberal compromises aren’t exactly in high demand these days. And for mainstream Democrats who have struggled to articulate a clear, penetrating policy message, walking back their support for a $15 minimum wage—one of the few crystal-clear, plainly compelling ideas that the party leadership has embraced—isn’t really tenable, since it would make them look even flimsier in their beliefs. (Voters don’t give politicians points for keeping up with the fluctuating state of the economics literature.) It also doesn’t make much sense to try to squelch the only successful progressive economic movement of the past several years when you are trying to rally progressives to help oust from power a Republican Party pushing policy plans that would be far, far worse for the working class than a somewhat-too-high wage floor.

When it comes to the minimum wage, we’re still operating in a zone of empirical uncertainty at a time when voters are crying out for conviction. Embracing the Fight for $15 might not make for great policy. But it’s probably the only option Democrats have.

*Correction, July 6, 2017: This article originally misspelled Hilary Wething’s first name. (Return.)