Does Union Membership Produce Healthier Workers?

By Pamela Babcock Jun 19, 2012
Reuse Permissions

More unionized workers in the U.S. rate themselves healthy than their nonunion counterparts, although the difference is admittedly small—just 3 percentage points, a recent study found.

The Duke University study, Union Membership and Self-Rated Health in the United States, appears in the March 2012 issue of Social Forces, a sociological research journal published by Oxford University Press. In it, 85 percent of union workers reported being in good health, compared to 82 percent of nonunion workers.

“We’ve long known that unions are good for the wealth of workers, but now we know they’re good for the health of workers, too,” said study co-author David Brady, Ph.D., a Duke associate professor of sociology.

The study is based on results from more than 11,000 full-time union and nonunion workers who rated their general health in the General Social Survey, a National Opinion Research Center survey from 1973 to 2006.

Brady admitted that while the 3-percentage point spread “doesn’t seem like a lot,” it represents 3.7 million American workers and rivals other factors that can help predict health.

“The effects of being a union member are as big as the effects of being five years younger or being married compared to being divorced,” he said.

Jason Beckfield, Ph.D., a Harvard University sociology professor who studies inequality and social policy, said the study “certainly raises a lot of questions about what the connections” between union membership and self-rated health “might be.”

But Paul Fronstin, Ph.D., director of the health research and education program at the Employee Benefit Research Institute in Washington, D.C., said the difference hardly seems noteworthy.

“Even if (they rate their health) as better, you’re only talking about 3 percentage points,” Fronstin said. “The fact is they are both in the 80s, which makes them both relatively good health.”

Weighing Self-Rating Health

Timothy G. Wiedman, PHR, an assistant professor of management and human resources at Doane College in Crete, Neb., said the study does a good job of comparing workers with largely similar characteristics and controlling for factors such as age, sex, education level, marital status, race, income and whether they work in similar occupations.

But he and some others questioned the reliance on self-rated health. Because participants were asked, “Would you say your own health, in general, is excellent, good, fair or poor?” they are looking “at workers’ perceptions, not their actual health,” Wiedman noted. He added that some have raised “serious doubts about the wisdom of relying exclusively on self-reported health information.”

While self-rated health is not a perfect indicator, it’s probably better than a lot of people give it credit for, Brady said.

“You never have perfect social science evidence. You can’t run a laboratory experiment where you randomly assign people to be a union member or not,” Brady said. “But the evidence we’re able to bring, I think, is pretty convincing.”

More Findings

The study, which Brady co-authored with Duke doctoral candidate Megan M. Reynolds, found that while unionization benefits all workers, it has a particularly large effect on men, those without a college degree and those without a high income. Here are some other findings:

  • Men who are members of a union are 1.27 times more likely to be healthy than nonunionized men.
  • Among those with a bachelor’s degree, unionized workers are 1.19 times more likely to be healthy than nonunionized workers.
  • Among the bottom 75 percent of the income distribution, unionized workers are 1.17 times more likely to be healthy than nonunionized workers.


Brady conceded that reaction to the report has been favorable among labor groups because “pro-labor voices are happy to hear about this study and are talking about it.”

On June 11, 2012, the AFL-CIO NOW blog featured the study, while Dean Baker, co-director of the Center on Economic Policy Priorities, wrote on his group’s blog that the results weren’t surprising because “we already knew that unions increase wages, especially for less-educated workers. They also strongly increase the probability that a worker will have benefits like health care insurance, a pension or paid sick days and family leave.”

Beckfield said one surprise is there’s not a lot of evidence that the association between union membership and self-rated health has changed in magnitude over time, despite the fact that union membership has dropped dramatically and the composition of people in unions has changed. In other words, the people who were in unions in the 1970s likely are not the same people who are union members now.

“I would have expected a change, but it’s not there,” Beckfield said.

Potential Shortcoming?

Wiedman noted that a potential shortcoming is that the study didn’t control for access to health benefits.

According to the Service Employees International Union, in 2009, 92 percent of union employees in the U.S. had access to health care benefits compared to 68 percent of nonunion workers. Given a 24-point advantage in access benefits for union members, Wiedman asked, “Isn't it likely that access to health care benefits is a more important factor than mere union membership?"

Brady said he and Reynolds couldn’t control for health benefits because they didn’t have enough respondents with information on such benefits and couldn’t test this rigorously.

But, he added: “If union workers are more likely to have health benefits and those health benefits result in better health, then we need to appreciate that one of the advantages of unionization is that one is more likely to have health benefits. Unionization is the reason for the better health benefits, and we should give unions credit for negotiating for and securing those benefits.”

Pamela Babcock is a freelance writer based in the New York City area.

Reuse Permissions


HR Education in a City Near You

Find a Seminar

Job Finder

Find an HR Job Near You


Find the Right Vendor for Your HR Needs

SHRM’s HR Vendor Directory contains over 3,200 companies

Search & Connect