The importance of workplaces in rising income inequality
A working paper released by the National Bureau of Economic Research earlier this week provides new and interesting data on the trends behind rising income inequality. The paper by Erling Barth, Alex Bryson, James C. Davis, and Richard Freeman shows that a large portion of rising income inequality is due to differences in pay across workplaces and not rising differences within individual workplaces. Exploring this trend and its causes is critical for a deeper understanding the structural changes in our economy over the past several decades.
Stories about rising inequality often focus on changes in workers and households—their education levels, whether they are union members, and the structure of their families. But the new report by Barth at the Institute for Social Research, Bryson at the National Institute of Economic and Social Research, and Davis and Freeman at the National Bureau of Economic Research includes an important actor: the worker’s employer.
The characteristics of workers are important for determining levels of pay, but so too are the characteristics of the individual workplaces where workers show up every day. After all, an employee of a more productive or more profitable establishment is going to make more than a similar worker employed by a lower productivity or low-profit establishment. (Note: there’s an important distinction between a firm and an establishment. A firm, or the overall business, can have many establishments. Think of it this way: a fast-food chain is a firm, but an individual store is an establishment.)
What this new paper shows is that within the United States rising inequality has more to do with a worker at one establishment making much more than a similar worker at different establishment and less to do with managers pulling away from line-workers at the same workplace. The report finds that from 1992 to 2007 approximately 67 percent of the increase in the variation in earnings is because of rising variation in pay across establishments.
The reason for the rising dispersion across establishments isn’t clear. Variation increased both across and within industries. Some of the increase is because variation in pay across industries in, say, the finance industry, pay more now relative to establishments in the manufacturing industry for a similar worker. Yet variation in pay at establishments within the finance industry increased as well. So we can’t pin the increase entirely in a shifting industry composition. Something is also affecting dispersion within industries.
The explanation for increasing variation in pay across establishments most likely isn’t a U.S.-only phenomenon. A paper by economists David Card and Patrick Kline at the University of California-Berkeley and Joerg Heining at the Institute for Employment Research finds that rising inequality in wages in western Germany between 1985 and 2009 was also fueled in large part by rising inequality across establishments.
This new paper by Barth, Bryson, Davis, and Freeman is important not because it is definitive proof of a trend. The data are not as high-quality as the German data used by Card, Heining, and Kline in their paper, but both papers are indicative of a trend that needs to be better understood. If the apparent trend is correct, then researchers and policymakers need to pay much more attention to factors affecting employers when considering rising income inequality.