On Election Day, voters in four right-leaning states, Alaska, Arkansas, Nebraska, and South Dakota, will weigh in on whether to raise their state-level minimum wages. The expectation is that these measures will succeed by wide margins. (There will also be a local referendum in monolithically liberal San Francisco to raise the minimum wage to $15 an hour, an effort that, per Steven Greenhouse of the New York Times, a reporter widely considered friendly to organized labor, has attracted little opposition from local business groups.) All four of the states intend to raise their minimum wages above the current federal minimum wage of $7.25, yet they notably intend to keep their minimum wages below the higher $10.10 federal wage floor proposed by President Obama and most of his Democratic allies.
In February, the Congressional Budget Office concluded that a higher wage floor would have significant disemployment effects while also raising household incomes for those low-wage workers who manage to remain employed. Real incomes would decline for other households, including those obligated to pay higher prices due to rising compensation costs, but the CBO finds that the net impact on real income would be mildly positive, and more positive for low-income households than for households middle- and upper-income households well above it. The CBO was careful to acknowledge that because many low-wage workers are not members of low-income households, a substantial share of the income gains would flow to lower-middle- and middle-income households.
One issue that merits closer attention, and that ought to have had some bearing on the debates in Alaska, Arkansas, Nebraska, and South Dakota, are the impacts on an increased wage floor on the composition of the low-wage workforce:
Some studies have found large elasticities for particular groups of adults, such as high school dropouts or African Americans in their 20s, but most of the adults who would be affected by the $9.00 and $10.10 options would not fall into those categories. A study that tracked directly affected adults regardless of their education, age, or race suggests that their employment is less sensitive to increases in the minimum wage than that of directly affected teenagers. One explanation for that lower degree of responsiveness is that employers facing an excess of workers or of job applicants tend to favor adults over teenagers. Supporting that explanation is research suggesting that encouraging employment among low-wage parents reduces employment among younger, childless adults.
This discussion strikes me as decidedly incomplete. Yes, it may well be true that elasticities for adults are lower than for teenagers regardless of education, age, or race. Yet when we are assessing the impact of an increased wage floor on the adult population, it is important to determine if adults with lower levels of education are more likely to find themselves locked out of the workforce then those with higher levels of education. Is the claim that the disemployment effect will be identical for those who haven’t completed high school as it will be for those who have? That is, if we assume that an increased wage floor will induce an increase in labor force participation by some workers and a decrease in labor force participation by others, what will be the impact on the less-skilled? The headline number from the CBO analysis was that employment levels would fall by 500,000 on a net basis – it seems reasonable to expect that employment levels would fall further among high school dropouts.
Granted, these assessments are based on a federal wage floor of $10.10, which would of course be higher than the levels at issue in Alaska, Arkansas, Nebraska, and South Dakota. It is also true, however, that these states depart from the national average in several respects. Alaska has a per capita personal income substantially higher ($46,778 in 2012) than the U.S. as a whole ($42,693); Nebraska ($43,143) and South Dakota are slightly higher ($43,659) than the U.S. as a whole; and Arkansas ($34,723) is one of the poorest states in the country. The share of American adults (25 and older) with at least a high school diploma was 85.3 percent as of 2010. Among the states with minimum wage referendums tomorrow, that share ranges from 91.5 percent in Alaska to 82.5 percent in Arkansas. Nebraska (90.1) and South Dakota (89.9) are clustered closely together on this metric. It’s also true that the demographic composition of the population varies across these states; Alaska has a relatively large indigenous population (14.7 percent) and a very small black population (3.9 percent); Nebraska and South Dakota are among the whitest states in the U.S.; and Arkansas has an above-average (15.6 percent) black population.
So while the discussion of the minimum wage referendums have largely focused on what these states have in common — they’re relatively politically conservative — it hasn’t focused on the fact that among them, Arkansas is unusually poor and that its adult population has an unusually low average skill level. The consequences of a substantial increase in the local wage floor will likely have different consequences in Arkansas, in light of its history of deprivation and isolation, and where higher consumer prices associated with rising compensation costs will have more bite due to its low income levels, than in Alaska, which is considerably more affluent. And while both Nebraska and South Dakota had unemployment rates of 3.6 percent as of August, unemployment in both Alaska (6.8 percent) and Arkansas (6.3 percent) is fairly high. I can’t help but think that Arkansas is making a mistake. But better that Arkansas is making a state-level decision that, as inflation and productivity growth proceeds apace, will be less binding than new federal legislation, which will be less responsive to its particular conditions.
(As Charles Hughes of the Cato Institute observes, only one of these states, Nebraska, has a state-level earned-income tax credit.)