A new paper by Richard Freeman, Eunice Han, David Madland, and Brendan Duke, Bargaining for the American Dream: What Unions do for Mobility is getting a lot of attention. A key finding is that a parent’s union membership boosts the economic prospects of the children as they grow up and form their own households. The effects are large and especially pronounced for working-class union families.

That result deserves headlines, but I will focus on another significant finding, because it relates to civic engagement. Freeman and colleagues find that labor union membership boosts the economic mobility of all children in the community. They control for a range of relevant factors that might explain away this positive effect (for instance, the makeup of local industry and the progressivity of the tax code.) They find that labor unions have positive effects for non-members.

That finding contributes to a larger literature about the positive economic outcomes of various kinds of civic associations:

• Freeman and colleagues build on the influential research by Chetty, Hendren, Kline, and Saez (2014). Chetty and colleagues found that the odds of moving up the socio-economic ranks are very strongly linked to the community where you grow up, and the main features of the community that matter are: having less residential segregation, less income inequality, better primary schools (measured by income-adjusted test scores and dropout rates), more family stability, and higher social capital. Their main measure of social capital is an index of “voter turnout rates, the fraction of people who return their census forms, and various measures of participation in community organizations.”
• In our own work, we also found strong economic benefits from what we called “social cohesion” at the community level. We defined that as the degree to which residents socialize, communicate, and collaborate with one another. Separately, we looked at the number of nonprofit organization in a community. Both social cohesion and nonprofit density were strong predictors of economic success for communities, even after we adjusted for many other factors like those considered by Chetty et al and Freeman et al. Just as an example of our findings, “An employed individual in 2008 was twice as likely to become unemployed if he or she lived in a community with few nonprofit organizations (the bottom five percent in nonprofit density) rather than one with in the top five percent for nonprofit density, even if the two communities were otherwise similar.”

A union could be considered an example of social capital. (It is an organization of members.) However, Freeman and colleagues controlled for social capital and still found a strong effect for unions. Their method distinguished union membership from civic participation, and the result was a distinct advantage for unions. That raises two questions: 1) Why would civic participation in general have anything to do with economic outcomes at the community level? And 2) Is the case of unions special? For both questions, I would like to focus on the benefits to non-members, because belonging to a union, a church, or an NAACP chapter can have direct and easily explained value for the individuals who join.

Social cohesion, social capital, and nonprofit density (which are overlapping but not identical constructs) could have economic benefits because: active and organized citizens obtain better governance and better laws; they gain skills and values from participation that they also use to help others in their communities; the associations they form reduce community-level problems, such as crime; these associations spread information and raise knowledge; or these associations build norms of trust and collaboration that enable people to contribute to the economy. There is literature to support each of these mechanisms, but no way to be sure whether they contribute to the patterns we see here.

Unions could fit into any of these stories. For instance, unions seek legislation and they may teach members how to collaborate and trust one another. Unions could also be seen as a special case because they have collective bargaining power. Also, people typically join a union because the workplace is unionized, not because they go out looking for a union to join. A knitting club seems very different: highly voluntary, trust-driven, but lacking in explicit economic power. So there could be an economic explanation applicable to unions alone, e.g., by bargaining for higher wages in their own industry, they send positive ripples through the local job market.

Still, I wouldn’t differentiate too starkly between unions and other associations. It is always a bit misleading to see membership as pure individual choice. People join knitting clubs and soccer leagues because someone else has organized these groups (which is hard and skillful work) and has recruited members. So organizing and outreach are always fundamental.

In short, I would posit that all associations–including but not limited to unions–use a set of similar means to improve economic welfare and mobility in their communities. Some of their means run through the state–they obtain better governance–and some of them result from voluntary action apart from the state. Unions do have some special features that allow them to grow to large scale when conditions are favorable and that give them bargaining power. Although their special features are important, unions are also part of a larger story about organized civic life in the US.

See also The Legitimacy of Labor Unions (2001), and my posts on Why the Garden Club Couldn’t Save Youngstown; civic engagement and jobs; and unemployment and civic engagement: the video.