Seattle is at the center of a giant pissing match about minimum wage

Last week, we told you that a new academic paper showed Seattle’s experiment with a $13 minimum wage is working. Then one day later, another paper came out. It said Seattle’s experiment with a $13 minimum wage isn’t working.

And we’re not sure who’s right.

Most economists think minimum wage increases can help workers. But they also believe there’s a point at which a minimum wage can be so high it destroys jobs and hurts the people it’s intended to help. Problem is, nobody knows exactly where that point is.

Enter into this debate Seattle. Motivated in part by grassroots agitation from groups like Fight for $15, the affluent Pacific Northwest metropolis — along with a number of other liberal-leaning cities — passed laws raising the minimum wage much higher than the federal mark of $7.25 an hour. That in turn has given economists an opportunity to peck at the data in an attempt to figure out whether the cities are pushing minimum wages too high.

And that’s exactly what Seattle has done according to the paper written by economists at the University of Washington. The authors report that after the city started to phase in its $13 minimum wage, low-wage employees, defined as those making $19 an hour or less, fared worse than they should have: Their overall number of working hours fell 9 percent and an estimated 5,000 jobs were lost. If accurate, those results mean low-wage workers’ earnings fell by an average of $125 a month last year.

That, of course, is the exact opposite of what a minimum wage increase is supposed to do. “These results,” the paper said, “suggest a fundamental rethinking of the nature of low-wage work.”

However, the more positive paper, from University of California, Berkeley economists, disagrees. It studies in particular the fast-food industry and its workers, on which economists who study the minimum wage often focus because the industry relies so heavily on minimum-wage labor. The Berkeley study found the wage increase boosted weekly wages and, more importantly, didn’t kill jobs.

“These findings of no significant disemployment effect of minimum wages up to $13 significantly extend the minimum wage range studied in the previous literature,” the authors wrote.

So who’s right?

Economists aren’t scientists — they’re social scientists. Instead of conducting their own experiments, they use natural experiments. And as a result, they run into a big problem: It’s impossible to know with absolute certainty what would have happened if the thing that happened hadn’t happened.

In other words, what if Seattle hadn’t phased in a $13 minimum wage in 2016? Would low-wage workers now be better or worse off? Economists use a bunch of statistical techniques to try to answer questions like that, but all of the methods involve myriad small choices that, if handled incorrectly, can skew results and introduce bias.

For example, both papers attempted to use wage and employment data from areas outside Seattle — areas unaffected by the minimum wage hike — to approximate what would have happened in Seattle without the hike. But they did it in different ways.

The University of Washington group looked at areas within Washington state — but that doesn’t necessarily mean those areas are all that similar to Seattle. And if the group you’re comparing to Seattle isn’t much like Seattle, it’s possible that the comparison is biasing the results. And some, like one of the economists involved in the Berkeley paper, argue that since Seattle is the only big, booming, techy, affluent city in Washington state, it doesn’t make much sense to compare it to the state as a whole.

Berkeley’s economists made different choices about how they looked at the city, examining mostly counties outside Washington state that are somewhat similar to Seattle when it comes to metrics like population. They then used a computer to construct a sort of weighted average of counties that mirrors the historical experience of Seattle, and compared what the effect of the minimum wage change in Seattle was to what happened to fast-food workers in this “Synthetic Seattle” over the same period of time.

It’s worth noting that one section of the University of Washington paper focused on the food-and-drink service industry, finding that within that sector, the minimum wage increase did, in fact, raise wages without hurting jobs — which you could certainly argue dovetails with the Berkeley paper’s findings about fast-food workers.

And if an industry that’s so reliant on minimum wage workers wasn’t hurt by a minimum wage increase, how could the city as a whole have been hurt by higher minimum wages? One weakness with the University of Washington study is that it doesn’t include “multisite employers” in its analysis, meaning it doesn’t capture businesses that have several locations — like, say, McDonald’s and Starbucks. That could mean the U-Dub paper undercounts job and wage growth.

I have more faith in the Berkeley study than the University of Washington study, but I’m not an economist, and my faith may be a case of confirmation bias because I support policies that seek to raise wages and help counteract inequality. Liberal groups like the Economic Policy Institute, along with some left-leaning economists, have also been critical of the University of Washington paper. In a fairly extensive review of the research, EPI economists wrote that they found “the authors’ very large estimated job losses suspect.”

On the other hand, the University of Washington study was able to get hold of more granular data from the state on actual earnings of employees, something that most studies of the minimum wage don’t do. And the usage of that detailed data could explain why the findings differ from typical findings on the minimum wage. Furthermore, the $13 minimum wage is fairly high by historical standards, and it’s entirely possible that Seattle might indeed be hitting that harmful minimum wage ceiling.

But nobody, not even all of these economists, really knows for sure. So the arguing over Seattle — and no doubt over the next minimum wage hike — will continue.