When You’re Stuck on “Help Wanted”

The National Federation of Independent Business, for example, found in its January 2026 report that 88 percent of small businesses looking to hire workers reported there were few or no qualified applicants to fill open jobs.  

Economic theory—and common sense—suggests an easy solution for these employers: offer higher wages. But in many cases, companies aren’t biting. “There are people looking for work,” says Benjamin Friedrich, an associate professor of strategy at Kellogg. “If firms say they can’t find workers, there’s a price that should be able to clear this market.” 

To find out why many firms aren’t taking this approach, Friedrich—along with his collaborators Alison Zhao, also of Kellogg, and Michal Zator of the University of Notre Dame—created a mathematical model to identify the possible forces that drive these labor constraints. 

The researchers found that the most likely reason for firms’ hiring difficulties is their own inaccurate understanding of what the open roles are worth. Companies set wages too low to attract workers and then are too slow to correct the error, which ultimately creates hiring gaps. Friedrich and his coauthors found that this explanation matched real-world patterns in hiring data. 

“This inefficiency could really hurt productivity—it might mean that the wrong firms struggle [to hire],” Friedrich says. In other words, if poorly run companies can’t fill vacancies, that’s one thing—but if even healthy firms are shooting themselves in the foot, “that’s a loss to the overall economy.” 

Potential explanations 

Friedrich and his collaborators began by creating a mathematical model centered on a well-understood phenomenon that causes hiring difficulties: “search frictions.”  

This term refers to the fact that a firm looking to fill a vacancy “can’t see all possible candidates [for the job] at the same time,” Friedrich explains. Because not every suitable applicant puts themselves forward for every opening, firms need to work with what they get. Sometimes, that isn’t a good fit, which causes jobs to go unfilled. 

However, search frictions didn’t provide a good enough explanation for the hiring difficulties that the researchers observed in the real world. According to this version of the model, slow-growing firms with lower wage offers should experience the most hiring difficulties. But when the researchers analyzed data from the German Federal Employment Agency, which surveyed roughly 40,000 businesses between 1993 and 2019, “it was exactly the opposite,” Friedrich says. The data showed that faster-growing companies had the most trouble—and longer searches were associated with higher wages, not lower. 

This mismatch told the researchers that something was missing in their model, so they added three new factors to see if they could better explain why firms often don’t raise wages to fill open jobs. 

The first factor suggests that a firm might not advertise higher wages for an opening to avoid having current employees demand raises, too.  

The second captures that it’s very difficult to lower wages once they have been set at a certain level. That might make firms reluctant to offer high wages to attract job applicants. 

The third factor reflects that companies might lack information about the other job options applicants may have. If a company does not know that potential candidates may be fielding other offers for better pay, it won’t feel the need to raise wages for its own vacancies. 

Too low, too slow 

When the researchers plugged these three factors into the model, its predictions fit the real-world data much better. 

For one thing, the model correctly predicted that firms would be slower to raise wages in order to fill “peripheral” roles that are less related to their core business. An automotive manufacturer, for instance, may know much less about how to hire an accountant successfully than it does about hiring technicians, says Friedrich. “They [might have hired an accountant] five years ago, and now who knows what the going rates are?” 

The model also accurately predicted that the less concentrated a firm’s potential workforce is, the more difficulty it would have in setting an attractive wage to fill openings. For instance, a software company that tends to mainly hire software engineers has a highly concentrated workforce. And the company might have an easier time realizing it needs to raise wages when it can’t find the specialized workers it needs.  

But a restaurant looking to hire waitstaff would likely also be competing with other types of businesses, like warehouses and retail stores, to hire less-specialized job seekers. This lower workforce concentration means that any of those companies is less likely to be informed about how to set wages at a level that will fill openings quickly. They’re also slower to raise those advertised wages in response to hiring difficulties. 

This is particularly challenging for smaller companies. “Big firms, they’re fairly well-informed,” Friedrich says. “It’s the smaller mom-and-pop shops, maybe smaller retail stores or some similar lower-skill workplaces, that don’t get the message immediately.” 

Friedrich says that their model’s predictions, validated by patterns in the survey data, tell a clearer story about why firms often don’t raise wages to prevent or quickly resolve hiring difficulties: facing uncertainty, firms choose low initial wages to avoid overpaying and then are slow to update these offers. This “information friction” prolongs hiring gaps, including among attractive companies, and can lead firms to ultimately set wages even higher than they might have if they’d acted sooner. 

“Maybe I don’t know how tough the competition is out there—and because I don’t want to overpay, my strategy is to feel the market’s temperature first by offering a low wage,” Friedrich explains. “If I get turned down, I’ll readjust wages up gradually. But I’m going to struggle [to fill vacancies] in the meantime, and I’m going to end up paying more anyway in the end.” 

Wage intelligence needed 

Friedrich sees the research as proof of concept for the role of information frictions in labor constraints. He hopes that other researchers will study them further. 

“The next wave of research can look at particular segments of the labor market and ask how strong the friction is,” he says. 

In the meantime, businesses hoping to improve their hiring can start by investing in “wage intelligence” tools and services to get up-to-date information on competitive pay rates. These services can be expensive but are often worth the cost, especially for rapidly expanding companies. “Getting access to this information is really important when you have a lot of profit opportunity now,” Friedrich says. 

Meanwhile, policymakers could provide more-timely and -comprehensive public reporting on wages as well as support pay-transparency legislation—both of which would make wage intelligence more accessible to businesses that can’t afford to pay for it. 

“Public-wage statistics are so coarse and fairly delayed,” Friedrich says. “They’re making life pretty hard for smaller firms. The private sector shows that it’s possible to collect this information more in real time, so I think more could be done.”