For much of the last century it seemed that the slice of the total economic pie going to workers was — like the speed of light — constant. No matter what the economy's makeup, labor could collectively depend on taking home roughly two-thirds of the country's total output as compensation for its efforts. Workers' unchanging share, the economist John Maynard Keynes declared in 1939, was "one of the most surprising, yet best-established, facts in the whole range of economic statistics."
But in recent decades, that steady share — which includes everything from the chief executive's bonuses and stock options to the parking-lot attendant's minimum wage and tips — started to flutter. In the 2000s, it slipped significantly. Although the numbers have inched up in the last couple of years, labor's portion has not risen above 59 percent since before the recession.
The decline has coincided with a slowdown in overall growth as well as a stark leap in inequality. "Labor is getting a shrinking slice of a pie that's not growing very much," David Autor, an economist at M.I.T., said. It is a development that is upending political establishments and economic policies in the United States and abroad.
Some economists argue that technological advancements are to blame as employers have replaced workers with machines. Others point to trade powered by cheap foreign labor, a view championed by President Trump that particularly resonated among voters.
Alternate culprits include tax policies that treat investment income more favorably than wages; flagging skills and education that have rendered workers less productive or unsuited to an information- and service-based economy; or a weakening of labor unions that has chipped away at workers' bargaining power and protections.
Over the last 15 years, for example, labor productivity has grown faster than wages, a sign that workers are not being adequately compensated for their contributions. And some industries have fared worse than others. Slices of the pie going to mining and manufacturing narrowed the most, while service workers (including professional and business services) had the biggest gains.
From manufacturing to retailing, giant companies have managed to gobble up a larger and larger share of the market.
While such concentration has resulted in enormous profits for investors and owners of behemoths like Facebook, Google and Amazon, this type of "winner take most" competition may not be so good for workers as a whole. Over the last 30 years, their share of the total income kitty has been eroding. And the industries where concentration is the greatest is where labor's share has dropped the most, according to research that analyzed confidential financial data from hundreds of companies.
"They're very sophisticated and efficient and they don't use as much labor," said Mr. Autor, who worked on the research with a team of economists that included David Dorn, Lawrence F. Katz, Christina Patterson and John Van Reenen.
With efficiencies come advantages.
These superstars can offer more variety, cheaper prices and convenience, but the bigger chunk of profits that they capture is split among fewer workers.
Software platforms and online services, for example, may be expensive to install, but not so costly to expand. Technology can make it easier to exploit smaller competitive advantages. And there is no need for duplicate payroll, shipping or human resources departments.
The idea of superstars vacuuming up a majority of goodies is perhaps more obvious on the individual level. Because of technology like cable and satellite television and the internet, music luminaries like Beyoncé and Taylor Swift or sports phenoms like LeBron James or Cristiano Ronaldo can reach a much larger audience and gain a greater proportion of the revenue generated.
Writing about the advent of superstars in the modern era, the economist Sherwin Rosen noted in 1981 that there was "a strong tendency for both market size and reward to be skewed toward the most talented people in the activity."
What was once true of pop stars can now be seen in more mundane industries. "Over the past several decades, only the highest earners have seen steady wage gains," a report from the president's Council of Economic Advisers concluded late last year. "For most workers, wage growth has been sluggish and has failed to keep pace with gains in productivity."
Research has shown that in pretty much every industry, most of the income differential among workers is not a result of a chasm between the highest and lowest paychecks within one company, but rather differences among companies.
Larger companies tend to pay better than smaller ones, Mr. Autor and the other researchers say, and in the technology sector especially, salaries can be impressive. Workers there are sharing in the larger pie — there are just fewer of them to do so.
The researchers examined six industries that account for 80 percent of private employment in the United States. In each one, they discovered "a remarkably consistent upward trend in concentration." In manufacturing, for instance, the top four companies controlled 43 percent of sales in 2012, up from 38 percent in 1982. In finance, the figure grew to 35 percent, from 24 percent, and in retail trade it went to 30 percent, from 15 percent.
The faster concentration grew, the bigger the drop in labor's share.
"What's different about new superstar firms is they don't have the cadre of middle-class jobs for nonelite workers," said Mr. Katz, an economics professor at Harvard.
"That's very worrisome," he said, adding that "the trend is going on in country after country."
Correction: March 10, 2017
An article on Thursday about the effect large companies may have had on a decline in workers' share of income misspelled the surname of an author of the study on such "superfirms." He is John Van Reenen, not Van Reenan.