Higher minimum wage have losers, US studies find
By New York Times, Jan. 12, 2017
CHICAGO—A growing number of economists have found that many cities and states have considerable room to raise the minimum wage before employers meaningfully cut back on hiring.
But that conclusion may gloss over some significant responses to minimum-wage increases by individual employers, according to two new studies. And those reactions may, in turn, raise questions about the effectiveness of the minimum wage in helping certain workers.
The findings, presented over the weekend at the annual meeting of the American Economic Association, the nation’s premier gathering of academic economists, come as many cities and states are raising their minimum wages.
California and New York last year approved gradual increases to $15 per hour. Proponents argue that raising the minimum is one of the most practical ways of improving living standards for the working poor and reducing inequality.
To test that proposition, John Horton of New York University conducted an experiment on an online platform where employers post discrete jobs—including customer-service support, data entry and graphic design—and workers submit a proposed hourly wage for completing them.
Horton, working with the platform, was able to impose a minimum wage at random on one-quarter of about 160,000 jobs posted over roughly a month and a half in 2013. If a worker proposed an hourly wage that was below the minimum, the platform’s software asked him or her to raise the bid until it cleared the threshold. In some cases the minimum wage was $2 per hour, in some cases $3 and in some cases $4.
At first glance, the findings were consistent with the growing body of work on the minimum wage: While the workers saw their wages rise, there was little decline in hiring.
But other results suggested that the minimum wage was having large effects. Most important, the hours a given worker spent on a given job fell substantially for jobs that typically pay a low wage—say, answering customer e-mails.
Horton concluded that, when forced to pay more in wages, many employers were hiring more productive workers, so that the overall amount they spent on each job changed far less than the minimum-wage increase would have suggested. The more productive workers appeared to finish similar work more quickly.
The traditional way most studies determine if employees are hiring a different kind of worker after the minimum wage rises is to consider certain characteristics, like race, age and education level.
The problem is that one worker can be much more productive than another with the same demographic profile, potentially masking the productivity upgrade that Horton documents. He was able to overcome this problem because the platform gave him access to precise data reflecting productivity, like past wages. When the minimum wage increased, employers tended to hire workers who had earned higher wages in the past, suggesting that they were looking for a more productive work force.
If the pattern Horton identified were to apply across the economy, it would raise questions about whether increasing the minimum wage is as helpful to those near the bottom of the income spectrum as some proponents assume.
The higher minimum wage could cost low-skilled workers their jobs, as employers rush to replace them with somewhat more skilled workers.
“There’s nothing about my paper that says raising the minimum wage is a bad idea—it may be that the trade-off is worth making,” Horton said. But there is “this consideration we probably haven’t considered.”
Some economists are skeptical that employers would respond the way Horton describes if an entire city or state increased its minimum wage, as opposed to just a single employer or subset of employers.
“I think the way to relate this result to the real world labor market is to consider what would happen if Wal-Mart had to raise wages,” said Arindrajit Dube, an economist at the University of Massachusetts at Amherst, who has studied minimum wage laws.
Wal-Mart has said that after it increased wages in 2015, it was able to attract more productive workers.
“This experiment is probably telling us much the same thing,” Dube added in an e-mail. “But if we want to know what would happen if NY or CA raised its minimum wage to 15/hr, I doubt that this online experiment—neat as it is—will shed much light.”
When the minimum wage goes up for everyone, it is not so easy for employers to substitute better-skilled workers because the new minimum would not offer a more attractive wage. In many cases, more highly skilled workers see their wages rise after minimum-wage increases to keep them above the new minimum, making it all the more difficult to lure them away.
Zane Tankel, chief executive and equity partner in a group that owns and operates several dozen Applebee’s restaurants in the New York City area, said replacing low-skilled workers with higher-skilled ones after the state’s recent minimum-wage increases is “not something that we try to do.”
Tankel argued that differences in the productivity of low-level workers in his industry are not very big.
“It’s just a lot more money for the exact same job description,” he said. He is accelerating automation in his restaurants, including tablet devices for ordering and payment, to offset the costs of the higher minimum.
Horton is quick to acknowledge that there are many reasons his experiment might not capture employer behavior in the wider economy. But he says a higher minimum wage could attract more highly skilled workers who were not previously in the labor market—say, college students. At the same time, less-skilled workers might lose their jobs and drop out of the labor force.
More broadly, he said, the contribution of his paper is to show that one impulse of many employers in the face of a minimum-wage increase will be to find more productive workers, even if there are limits on how much they can follow through on this desire.
“There are lots of reasons to think this is probably happening and we haven’t detected it because we don’t have the data,” he said. “I know in my career, people have fine-grained opinions about who’s better than who, and we talk about them endlessly. We’re constantly ranking. But when we talk about other labor markets, we pretend the same distinctions don’t exist.”
A second study presented at the conference suggests another way that employers may respond to a rising minimum wage: simply going out of business.
The husband-and-wife research team of Michael Luca of Harvard Business School and Dara Lee Luca of Mathematica Policy Research identified the ratings of tens of thousands of restaurants in the San Francisco area on the web site Yelp and found that many poorly rated restaurants tend to go out of business after a minimum-wage increase takes effect.
By contrast, highly rated restaurants appear to be largely unaffected by minimum-wage increases, and overall, there is no substantial decline in the number of restaurants after a minimum-wage increase.
Though the couple’s study is still being refined and they have yet to explore the reason, one possibility is that workers at poorly rated restaurants tend to be less productive than workers at highly rated restaurants, making it difficult for those restaurants to survive when paying workers more. (It’s also possible that highly rated restaurants are better able to pass rising costs on to their customers.)
The results are broadly consistent with a 2013 study by the economists Daniel Aaronson, Eric French and Isaac Sorkin, showing that a sizable minimum-wage increase in New Jersey resulted in many lost jobs as numerous businesses closed, but an almost offsetting number of new jobs as other businesses opened. New York Times News Service