The New York Times story comparing the earnings of private-sector employees with state government employees confirms what the Iowa Policy Project found in a new report the same week: When you control for education, most state government employees are paid less than their private-sector peers.
Despite the similar findings, our study and the Times study used different approaches.
First, the Times study was more limited in scope, looking only at state government workers, while our study looked at state and local government employees. This matters because local government workers, as our analysis showed, are compensated even more poorly than state government workers.
Our study also accounted for factors that are known to affect compensation beyond education, such as work experience, firm size, sex, race, and annual hours worked. In addition, our study attempted to account for the effect of the public sector’s more generous benefits packages. The Times story looked only at wages.
Second, the Times used slightly different methodology and a different data source than we used. The difference in categorizing education was the most important difference in methodology between our report and the Times’ story. Workers with associate’s, bachelor’s, master’s, professional and doctoral degrees were all lumped together in the Times report as one group, and workers with some college but no degree, a high school diploma, or workers with less than a high school education as the second category.
Categorizing workers that way distorts the picture. It inflates wages in the public sector, as a much larger proportion of state government workers have master’s, professional and doctoral degrees, and a smaller proportion of state workers have an associate’s degree, than in the private sector. Figure 1 in our analysis demonstrates the key differences in educational distribution among the sectors.
Our study also looked into each educational category, comparing, for instance, public-sector workers with a bachelor’s degree with private-sector workers with a bachelor’s degree. Thus the table that accompanied the Times story is not comparable to the results in Table 1 of our study.
Using the data set with which I conducted our study — the Census Bureau’s Current Population Survey — I compared median pay between the two sectors across the broad educational categories used in the Times story.
Despite the different methodologies and the different datasets, the numbers tell the same story: State government workers with a college degree or more — over two-thirds of Iowa’s public sector workforce — are paid less than college-educated private sector workers.
The Times study found that college-educated state employees earn 6.9 percent less than college-educated private-sector workers in Iowa. The analysis I conducted with our dataset was virtually indistinguishable, finding a 6.1 percent differential.
Among less-educated workers, our findings confirmed the Times, though the 19 percent state-government advantage reported by the Times outstripped the advantage in our data set by 8 percentage points.
The Times has made a valuable contribution to the discussion on public-sector and private-sector compensation. In addition to broadening the discussion by virtue of its vast readership, the Times also independently confirmed the analysis by IPP and the several different state-level analyses conducted by the Economic Policy Institute. When the educational differences between the private- and public-sectors are accounted for, it is clear that most public employees could be making more money in the private sector.
Posted by Andrew Cannon, Research Associate