When policymakers and the public debate sources of growing income inequality, a usual suspect is soaring CEO pay. It’s an understandable culprit. After all, compensation packages for business leaders have soared since the 1970s, and the gap between their pay and average worker wages has grown in tandem.
The rise in earnings inequality from 1981 to 2013 is less about one worker earning more than a co-worker in the same company than about growing differences from one company to the next.
But a recent Minneapolis Fed working paper identifies a far more significant cause: Inequality among firms, not within them. In “Firming Up Inequality,” Fed consultant Fatih Guvenen, with four co-authors, finds that the rise in earnings inequality from 1981 to 2013 is less about one worker earning more than a co-worker in the same company than about growing differences from one company to the next.
Analyzing a massive set of U.S. firm data on worker earnings histories, the economists measure three possible sources of this inter-firm rise in average earnings.
- An increase in the practice of some firms paying higher wages than others do for equally skilled workers (firm pay premium).
- A rise in high-wage workers working with other high-wage workers (segregation).
- Greater correlation of the two: high-wage workers increasingly working at high-paying firms (sorting).
They find no change in pay premiums among firms, so it has no contribution to rising U.S. inequality. This result was “surprising,” noted Guvenen in an email, because a similar study in Germany found that rising firm pay premiums accounted for a third of that country’s rise in inequality. “We find that not to be the case in the U.S.”
But the other two factors, segregation and sorting, did increase and together account for two-thirds of the overall rise in U.S. earnings inequality. This is even more important for small and medium-size firms, accounting for 84 percent of their increase in earnings inequality (and over 90 percent for truly small firms, those employing between 20 and 1,000 workers).
“The large rise in earnings inequality in the United States can be decomposed into three equally important forces. A rise in the segregation of higher-paid workers to the same firms (segregation), that these high-paid workers are typically moving into higher-paying firms (sorting) and a rise in earnings inequality within larger firms.”
Another key finding: While the remaining one-third of increased U.S. earnings inequality is explained by variance within firms, most of it comes from large firms. Top executive compensation in these companies with over 10,000 employees have grabbed the headlines, but there are so few CEOs that their pay packages explain statistically little of the overall inequality increase.
“The large rise in earnings inequality in the United States can be decomposed into three equally important forces,” the economists write in summary. “A rise in the segregation of higher-paid workers to the same firms (segregation), that these high-paid workers are typically moving into higher-paying firms (sorting) and a rise in earnings inequality within larger firms.”
To arrive at these findings, the economists build and analyze a data set that anonymously matches employee earnings records to specific firms, using the W-2 forms that firms send annually to the U.S. Social Security Administration. The data set thus tracks every individual’s wage patterns over time and the pay schedules of individual companies, providing a comprehensive multidecade mosaic of U.S. earnings trends and the contribution that firm pay practices make to rising earnings inequality.
The authors emphasize several advantages in using this particular data set. First, it covers all U.S. workers and firms through 2013. Second, it captures a large share of earnings at the very top, as well as the bottom, an earnings range not often available. Third, it gathers information on entire firms, not simply individual locations.
What’s driving this?
While the economists document the predominant source of rising inequality, their research doesn’t address the forces behind it. They do suggest several possibilities, however.
In some industries, dominant companies have established high barriers to entry, preventing competition from driving down profits. This enables them to pay higher wages to attract top talent. Google is an obvious example: Its search engine is so pervasive that competitors are unlikely to succeed. Not coincidentally, perhaps, their software engineers are among the highest paid.
Another potential explanation: Outsourcing nonessential activities such as custodial work, food service and security has allowed firms to focus on their core business, “perhaps leading to greater cross-firm segregation by worker skill level.”
To the degree that workers learn skills from their co-workers and would benefit by working alongside those with high ability, “rising segregation will dynamically increase inequality.”
The authors also voice concern about the social impact of increased inter-firm inequality due to worker sorting and segregation. Workers with lower skills and lower pay appear to be losing access to jobs at high-paying firms: If a tech company outsources food service, company cafeteria employees won’t gain the pecuniary benefits of working for a high-paying firm, exacerbating overall income inequality.
Moreover, firms are increasingly the source of health and retirement benefits for their workers, and divergence into high- and low-paying companies suggests that some workers will be less protected than others from ill health and old age. A third concern: To the degree that workers learn skills from their co-workers and would benefit by working alongside those with high ability, “rising segregation will dynamically increase inequality.”