The rise of income inequality in the United States since the late 1970s is a well-documented fact, but the reasons for the rise still aren’t well understood. The possible culprits include skill-biased technological change, globalization, the rise of the robots, and an increasingly popular reason: increased “rents” in the U.S. economy.

Rents, in economics parlance, are extra returns above and beyond what we’d expect in a competitive market. A new paper by Jason Furman, chair of the President’s Council of Economic Advisers, and Peter Orszag, former director of the Office of Management and Budget and a current Vice Chairman at Citigroup Inc., presents some evidence that not only have rents increased, but they provide a fundamentally important explanation for rising inequality.

Furman and Orszag’s analysis centers on the idea that the firm, where workers are employed, plays a major role in the level of income inequality. A number of research papers over the past couple of years find that a large share of rising income inequality is due to larger inter-firm inequality. While some of the rising inequality is because of forces related to characteristics of individual workers—such as education level or union status—a large chunk also is due to the characteristics of the firms at which workers are employed.

One possible explanation for why some firms are pulling away from others is because of increased rents that these firms are capturing. Using data from the McKinsey Corporate Analysis, Furman and Orszag look at the distribution of returns on invested capital across firms, and find a high and rising dispersion of returns among firms. The ratio of returns at the 90th percentile to returns at the 50th percentile was once about 3-to-1; now it’s close to 10-to-1. But the level of return is quite high for firms well below the top—about 30 percent for firms at the 75th percentile. These kinds of returns and their distribution is a sign of increasing rents, and the capture of those rents by leading firms.

But these rents don’t necessarily go all to the owners of the firm. In fact, for rising rents to be a major contributor to rising income inequality, the rents have to be shared with workers at those firms. Furman and Orszag point out that for this to be true then there has to be a significant amount of friction in the labor market. And there is quite a bit of evidence for this idea—the two economists point to the large amount of research in recent years showing a decline in job-to-job transitions. The exact source of this declining dynamism isn’t certain yet, though there is evidence that declining demand for labor is responsible.

While the evidence is far from definitive, increased market power could help explain raising rents as well as decreased labor demand. In conjunction with product, frictions in the labor market generate rents. If a firm has monopsonic power, for example, results in firms hiring fewer workers than is optimal for the entire economy. Increased market power and rents are also potential reasons for the decline in the share of income going to labor.

As Furman and Orszag make clear several times in the paper, the story presented here is far from a slam dunk. The evidence for each individual link in the chain isn’t rock solid yet, and digging into the different research areas is vital. Putting more emphasis on rents in the investigation of income inequality prompts a lot of interesting questions.