Striking Minneapolis parks workers, unionized with LIUNA Local 363, march to the park headquarters on July 24, 2024. Photo by Max Nesterak/Minnesota Reformer.
A remarkable phenomenon that’s emerged in the past 50 years is that Americans who earn a college degree will live considerably longer than those without.
Economists Anne Case and Angus Deaton have shown that the mortality gap between Americans with and without a college degree has widened from about 2.5 years in 1992 to over 8.5 years post-pandemic. As a result, Americans with a college degree are on a similar upward trajectory of life expectancy compared to those in other rich democracies, whereas Americans without a college degree are distinct in their stalled and falling life expectancy.
In a paper recently published in the academic journal Social Science & Medicine, my colleagues and I show an important but unappreciated contributor to widening mortality inequality: the decline of labor unions.
Labor unions were once seen as a major holder of power in American society, with one-third of the private sector unionized in the mid-1950s. Since then, labor power and membership have declined rapidly, so that today only about 6% of the non-agricultural private sector is unionized. Economists, political scientists and sociologists have long recognized the economic importance of unions — that they reduce overall inequality and boost the earnings of otherwise lower paid and less powerful workers.
When we gave a broad reading of the literature on unions, we suspected that a unionized career had an impact on America’s worsening mortality crisis. In addition to higher wages, unions also provide a wide range of superior benefits to otherwise less powerful workers: health insurance, time off and a pension, occupational security, a path for upward attainment, and workplace safety, to name a few.
Unions also tend to push for more egalitarian and protective policies which might have downstream health consequences. And, unions tend to foster community and social connection — you’ll see union picnics at many of the Twin Cities’ decent-sized parks every summer — and even marital stability, all of which are well-established predictors of mortality. We therefore set out to assess the extent to which union membership can help explain disparities in mortality.
To do so, we used a dataset called the Panel Study of Income Dynamics, or PSID, which is the longest running longitudinal dataset in the United States and widely considered one of the highest quality datasets in the social sciences. The PSID began in 1968 with a nationally representative sample of respondents. It tracked a variety of demographic and economic information from people and their families and followed those who split off and descended from these original families. The survey is still running and completed its most recent wave of data collection in 2023.
We followed people who were in early adulthood in the initial waves of the data collection. We tracked the amount of time that these respondents were members of a union during their working years. We then tracked mortality to assess whether people who spent more of their working years as union members had lower risks of mortality in comparison to otherwise similar non-unionized individuals.
We found that union membership was a key predictor of mortality. Each additional year that a respondent spent as a union member associated with about 1.5% lower odds of mortality after the age of 40. Furthermore, these union mortality effects were concentrated among the groups that Case and Deaton have flagged as at risk and largely driving America’s mortality crisis: people without a college degree, and the white men who once comprised the bulk of union membership.
Is 1.5% lower odds of death significant? We compared the size of this union protection to a broader and potentially more fundamental effect, that of employment. The influence of union membership on mortality was about half the size of employment more generally. This in itself was unsurprising given the well established importance of work as a fundamental social determinant of health.
Yet the fact that our observed effect of union membership was comparable to employment suggests that the decline of union membership and power over the past 75 years has had much broader reaching consequences on American health and well-being than previously appreciated.
Unions have historically been one of the few institutional anchors of support and stability for less powerful American workers. Their demise has unsurprising consequences beyond mere pay, but these broader consequences on life expectancy have not been fully appreciated by the research community, let alone by a political system that has spent decades eroding the power of unions.
What are the broader implications of our finding?
We urge public health officials to consider worker power and worker voice a fundamental social determinant of health, one which has helped create the mortality crisis that we currently face.
Can we realistically call for a re-invigoration of the American labor movement that occurred in the mid 20th century? We are skeptical. Unions face substantial opposition and pushback from employers and a byzantine system of policies and regulations that make organizing extraordinarily difficult.
Workers and labor leaders need to be creative in reinvigorating collective action, and we are heartened by the ingenious and creative actions that many are taking. To them, our research shows the fundamental payoff of their efforts.
Labor organizing puts more money into the wallets of ordinary workers — and gives them more years to enjoy with their loved ones.