Last week, Thomas Firey blogged here on how proponents of higher minimum wages tend to be selective when reviewing the academic literature. In particular, they mischaracterize it as implying that papers tend to show minimum wage hikes do not affect employment.

Right on cue, an important new NBER paper from academics at the University of Washington examining Seattle’s two minimum wage hikes since 2014 suggests significant job and hour losses as the pay floor rose.

Seattle raised its minimum wage from $9.47 to $11 per hour in 2015 and then again to $13 per hour in 2016. The report concludes:

Using a variety of methods to analyze employment in all sectors paying below a specified real hourly rate, we conclude that the second wage increase to $13 reduced hours worked in low-wage jobs by around 9 percent, while hourly wages in such jobs increased by around 3 percent. Consequently, total payroll fell for such jobs, implying that the minimum wage ordinance lowered low-wage employees’ earnings by an average of $125 per month in 2016. Evidence attributes more modest effects to the first wage increase.

The paper seems significant both as a result in its own right and in addressing why sometimes results can be so different depending on methodology.

It suggests that the disemployment effects caused by reductions in demand for low-wage labor rise disproportionately as the wage floor increases. This is logical, but often needs to be articulated given many appeals to the effects of past increases in wage floors as evidence for new much higher minimum wages (see the “Fight for 15,” for example).

Perhaps more importantly, whereas much of the previous literature examines low-wage industries (usually restaurants or retail) or teenagers as proxies for those impacted by minimum wage changes, this paper uses a comprehensive data set allowing assessment on employment for all categories of low-wage employees across industries and demographics. This suggests that previous research examining the impact on individual industries such as restaurants may have significantly underestimated the negative effects on hours worked for low-wage employees.

Of course, as with all studies, there will be disputes about methodology. The paper itself notes that it does not include companies with multiple locations, such as fast food chains, which theory suggests could bias the results in either direction. They also acknowledge that some of the “jobs lost” may have been replaced by jobs in the broader area, making their results an overestimate. Overall though, this is further evidence, adding to an already large literature, suggesting raising the minimum wage to high levels has a very high cost indeed and that the competitive model of the labor market holds.