2014 Minimum Wage Hikes: Looking Back
In January 2014, 13 states hiked the minimum wage. This held out the promise of a very simple natural experiment: As the data came in, we could look to see how employment fared in these states relative to the rest of the country. Early reports indicated that, if anything, the states that hiked their wages saw particularly strong job growth. In fact, the states that didn't raise wages saw an abrupt drop in employment at the exact moment that other states boosted pay.
Alas, there were some problems with the simple approach, as Ben Gitis of the American Action Forum pointed out at the time. The overwhelming majority of workers counted in total-employment statistics aren't near the minimum wage; 9 of the 13 wage-hike states didn't raise wages very much, simply adjusting for inflation; and the abrupt drop in employment in January could be explained by rough winter temperatures, which were less pronounced in the East Coast states that had hiked their wages. Focusing on low-wage sectors and using some statistical wizardry to remove the effects of confounding variables, Gitis argued that the wage hikes had indeed been a blow to employment.
But now that the Bureau of Labor Statistics has numbers for all of 2014 -- and states have had time to recover from the terrible winter -- it's worth looking back at the basic data. I grouped wage-hike, inflation-adjustment, and no-change states together, and then tallied up the number of people employed there in four categories: total nonfarm, retail, accommodations and food service, and age 16-19. Then I graphed the numbers as a percentage of the people who were employed just before the hikes went into effect.
Here are the results. The wage-hike states are Connecticut, New Jersey, New York, and Rhode Island; the inflation-adjustment states are Arizona, Colorado, Florida, Missouri, Montana, Ohio, Oregon, Vermont, and Washington.
Some interesting trends here. The total-employment picture doesn't show much -- aside from the January weather problem, the same trends happened after the hikes as before. The strongest case that minimum-wage hikes hurt employment can be see in the accommodations/food service data, with wage-hike states growing slowly after, but not before, their laws changed. Retail's early-2014 problems were heaviest in wage-hike and no-hike states, but wage-hike states recovered better. And the 16-19 numbers, which are available only annually, are just weird.
The idea that modest minimum-wage hikes decrease employment doesn't pass the "trendline test" -- you can't just look at the numbers and see an obvious effect. Whatever effect there is, is likely pretty small.
Some technical notes: Some of the recent numbers are "preliminary" and may be adjusted later. New Mexico and South Dakota are missing from the BLS's accommodations/food service data, so they are not included in that chart. I excluded the District of Columbia from all the graphs because it hiked its minimum wage in the middle of the year. The very dramatic rise for inflation-adjustment states in the 16-19 graph comes almost entirely from Florida. I've contacted BLS about the New Mexico, South Dakota, and Florida anomalies, and I'll update this post when I hear back.
[Update: Heard back. They say the 16-19 data can be volatile even on the annual level (they're not even reported month-by-month), and that New Mexico and South Dakota indeed don't have accommodations/food-service numbers.]
Robert VerBruggen is editor of RealClearPolicy. Twitter: @RAVerBruggen