The New (Old) Minimum-Wage Debate

A major downside to raising the minimum wage is that it could decrease employment. If employers have to spend more for each person they hire and each hour they assign, they might do less hiring and cut back on hours. But don’t worry, say the policy’s defenders: While older research did tend to find negative employment impacts, newer, better research does not.

That line has taken hold in the public debate, but it’s not quite true. And with Joe Biden’s advocacy of a $15 nationwide minimum wage, it’s prompted some strong pushback from within the economics profession — including allegations that politics have played too much of a role in deciding what gets published and what gets discussed.

“Anyone who thinks there’s a consensus in economics on the effects of changing the minimum wage should talk to someone who’s recently tried to publish a paper on the topic,” tweeted Jennifer Doleac, an economics professor at Texas A&M University, earlier this month. “I study lots of controversial topics but would never go near that one, thanks. So political. A nightmare.” She added that “the stories are enough to make me discount the past 10 years of published research in this area.”

Tyler Cowen, the George Mason University economist and popular blogger, was even harsher in a recent post: “I am sorry to speak in such terms, but the reality is that an allied cabal of activists and left-wing economists have combined on social media to insist on a particular approach to minimum wage economics and to bully those who disagree.”

And this week, David Neumark and Peter Shirley released a paper reviewing the academic literature. On balance, they find, this literature still suggests that higher minimum wages damage employment.

Neumark and Shirley rounded up a total of 66 economics papers on the minimum wage, dating back to 1992, and tallied up what those papers found. (These studies tend to report a lot of numbers estimating the effects in different ways, so Neumark and Shirley asked the authors which estimates they felt best represented their findings; the vast majority responded, but Neumark and Shirley needed to pick themselves for the rest.) The upshot? About half of the papers found a statistically significant hit to employment from minimum-wage hikes, and another 30 percent found a negative effect that wasn’t statistically significant. As they put it, that’s “a clear preponderance of negative estimates.”

Interestingly, notwithstanding any political shenanigans in the publication process, more recent studies were not less likely to find negative effects. And studies were most likely to find negative effects when they focused on the groups of workers most likely to be affected by the minimum wage: The results were strong for teens, young adults, and specific individuals who made low wages before increases went into effect, and weaker for low-wage industries studied as a whole (because they include many workers who make well above the minimum).

This new roundup certainly doesn’t end the debate over employment and the minimum wage; indeed, it’s already facing the usual round of criticism. (See, for example, here.) But it ought to make minimum-wage fans a little less confident that their policy won’t have unintended consequences.

Meanwhile, there are several other aspects of this topic that deserve a lot more discussion and study before we more than double the minimum wage but that get left out when we focus so intently on employment effects — so, while I have your attention, a quick word on those.

First, we can’t forget that those on the left might support a higher minimum wage even if it reduces employment to some degree. If some workers lose their jobs or work fewer hours, while others get big raises, low-wage workers might still come out ahead on the whole. (Or they might not.)

Second, when workers make more, who pays for it? Well, some mix of business owners and customers, though the precise blend is disputed. Notably, many low-wage employers, such as discount retailers and fast-food restaurants, serve lots of low-income customers, and most minimum-wage workers do not live in poor households (because, for example, they’re teenagers or second earners). This policy, therefore, doesn’t just transfer money from well-to-do business owners to poor workers, but also from poor customers to middle-class workers.

Third, how much of a minimum-wage increase is the government going to take back in the form of taxes and benefit cuts? Berkeley’s Labor Center recently pointed out that hiking the minimum wage to $15 would affect lots of families that receive safety-net benefits ($107 billion in benefits total). The Labor Center sees this as a “public cost” of having a low minimum wage, but the finding also highlights a major limitation to raising the wage: The safety net can shrink quickly when people make more of their own money — especially if they have kids and incomes just above the poverty line, in which case each dollar earned kills about 50 cents via taxes and benefit cuts — so the folks who need the cash most might not get as much of it as you’d think. The federal government, which is funded disproportionately by the rich, will take a big cut.

And fourth, exactly how high can you push a minimum wage before it starts to do more harm than good? No one is saying a $50 minimum wage would be a good idea, and even some advocates of a higher minimum wage admit things could look worse once you go past, say, half or two-thirds of the median wage. Fifteen dollars is two-thirds of $22.50, which is above the U.S. median of $19.14 (as of 2019), to say nothing of poorer states and rural counties. In some places, $15 would be above the median.

The seemingly simple act of raising the minimum wage amounts to a rather elaborate income-redistribution scheme involving workers, employers, customers, and Uncle Sam, and we don’t know for sure how all of it will add up. The results may not impress; one attempt to simulate the overall effects found that, even if it doesn’t reduce employment, “an increase in the national minimum wage [increases consumer prices in a way] that is more regressive than a typical state sales tax and allocates benefits as higher earnings nearly evenly across the income distribution.”

With all that in mind, my own view is that if we think people should be paid more, we should subsidize their wages with tax dollars. At least that way we’ll know who’s paying and who’s benefiting before we set the process in motion, we won’t single out the customers and employers of low-wage workers for punishment, and we won’t risk throwing people out of their jobs.