Ready to destroy a longstanding argument against raising the minimum wage?

The US House of Representatives recently passed a bill raising the minimum wage—the first such increase since 2009. In the extremely unlikely event that the law is also passed in the Senate, the bill would incrementally raise the minimum wage over time until it hits $15 an hour in 2025.

This ambitious (if ill-fated) legislation is based in part on a changing consensus within economics. Opponents of minimum wage hikes have long argued that an increase would result in job losses. They claim that as the minimum wage rises, firms would lay off workers and automate their jobs with cheaper machines. But studies conducted over the last few decades have shown a more muted result, with minimum wage increases resulting in few jobs lost, and in some cases, jobs gained.

• Whatever happened to Six Sigma?

Economists have interpreted these findings as evidence that the labor market is a “monopsony.” Just as a monopoly is the only company selling a particular good or service in a given market, a monopsony is the only company buying one. Because a monopsony labor market is not competing with other firms to hire workers, it does not have match wage offers. Instead, employers are able to pay individuals less than they would in a competitive market (and pocket the difference). This is consistent with the finding that employers don’t typically lay people off after a minimum wage increase. They already were paying people less than they would in a competitive market; the minimum wage merely brought the wage closer to its true value.

But how is it possible that a given company is the only company hiring labor? There are many employers in the labor market, especially for the “low-skill labor” most likely to be affected by the minimum wage. As such, many believe the labor market is effectively competitive.

There are two flaws with this reasoning, though.

• Photos: The aftermath of Hurricane Dorian in the Bahamas

First, there’s no such thing as “low-skill labor,” just low-wage labor. Many people earning wages at or near the minimum are highly skilled. Their skills just do not happen to be particularly remunerative.

Second, evidence suggests there’s no such thing as a single “labor market” in which everyone—or even a group of people with similar skills—competes. Instead, the labor market is more like the dating market, where the goal is to find a good overall match—and everyone is looking for something different. Just because there are plenty of fish in the sea does not mean it is easy to find the type of fish you are looking for.

Pay isn’t the only factor when choosing a job

In fact, if we assume a single labor market exists, then we are effectively assuming that people don’t care about their work environment at all, just their wages. That’s obviously not true: People tend to have very strong preferences about the companies they work for, just as they have strong preferences about who they date and who they select as a partner. These preferences vary substantially from person to person.

As a result, two firms that are similar in most respects and offering jobs with equivalent wages could be operating in two entirely different labor markets.

A number of factors, not just wages, would impact your decisions about where to apply and ultimately work. One firm may offer a shorter or more pleasant commute. One could be more convenient to your child’s school. One might have health insurance coverage for a specific ailment you have, while the other doesn’t. One expects you to use an Apple computer, and the other a PC. One expects you to work some nights, while the other expects you to work some weekends.

While the firms could compete on salary, they cannot compete across all the dimensions that we actually care about. This means that employers have much more bargaining power than economists originally anticipated when they conceptualized perfect competition in the labor market.

Now imagine you have job offers from two companies. One is much closer to your home than the other, and you value a short commute. The first company therefore can pay you a lower salary because of that differential. But there are other people who live closer to the second company, so that company can also pay workers a lower salary. In the end, they will probably both offer a lower wage. Because there is always somebody with different preferences than you, it means both companies will ultimately pay workers less than what they would pay in a competitive market.

Even in job markets where we would expect the variety of preferences to be muted, we still see evidence of monopsony. For example, the Amazon task marketplace Mechanical Turk seems to be the ideal frictionless marketplace with hundreds of thousands of buyers. Mechanical Turk lets companies hire workers for small, discrete tasks (such as paying $0.25 for data entry or to respond to a survey question). But experimental studies examining this marketplace show that buyers in fact are able to set low wages due to a lack of competition—and that’s due to varying preferences regarding the type of work. People tend to choose surveys or data entry tasks that they think will be at least a little interesting. Even in online marketplaces with many buyers and sellers, people care enough about what kind of work they are doing that each person is effectively in their own labor market.

Each one of us is essentially our own labor market

Economists often assume that people share similar preferences. But the evidence suggests that people have wildly different preferences with respect to how and where they work, and what they work on. Each of us not only has a unique set of skills but a unique set of preferences. This puts us each in our own labor market, even if firms appear to have relatively uniform preferences about optimal characteristics for potential hires.

If companies were not monopsonies, we would expect them to raise wages without the minimum wage forcing them to, because they would have to compete. That’s one reason the minimum wage can be an effective economic and social equity tool for mitigating the bargaining advantage that firms possess.

How high the minimum wage should be set is an empirical question—but the fact that it hasn’t been increased in 10 years indicates that a lot of Americans are due for a raise. Monopsony employers aren’t going to increase low wages on their own.

Matt Darling is a vice president at the non-profit behavioral design lab ideas42, where he applies behavioral science to labor and workforce development to improve economic justice.

 

Sign up for the Quartz Daily Brief, our free daily newsletter with the world’s most important and interesting news.

More stories from Quartz:

• China says Taiwan was always part of it, but an old stamp shows a telling propaganda snafu

• Like Pakistan, India may soon face the catastrophic costs of steroid-driven nationalism