Skip to content

Spurts in Union Growth: Defining Moments and Social Processes

Literature › Works

Richard B. Freeman (1998). Spurts in Union Growth: Defining Moments and Social Processes. The Defining Moment: The Great Depression and the American Economy in the Twentieth Century (NBER; eds. Michael D. Bordo, Claudia Goldin and Eugene N. White). University of Chicago Press.
Type book chapter
Scope Comparative and methods
Method mixed
Period studied 1880–1995
Regions United States, United Kingdom, France, Norway, Sweden, Denmark, Australia, Canada, other OECD countries
Themes unions strikes, comparative theory, methods data sources
Citation key freeman1998

Argument

Freeman argues that union growth comes in discontinuous spurts rather than smooth logistic growth. The 1930s US spurt was one instance of a general cross-country pattern, not a unique product of the Wagner Act, the CIO or specific personalities. He sets out a simple nonlinear model in which density depends on the contest between union organizing resources and employer opposition, which generates tipping and jumps. He tests a 'top-down' (New Deal/Lewis) account against a 'bottom-up' (worker grievance) account using data on recognition strikes, CIO financial aid and AFL growth, and favours the bottom-up account. He concludes that the New Deal legacy is an outmoded legal framework, not a stable bargaining system.

Evidence and method

US union membership and density series spliced from BLS, Troy, Troy and Sheflin and Friedman (1880-1995); cross-country density data (Visser, Bain and Price, OECD); BLS recognition-strike data and NLRB election statistics; Galenson's data on CIO assistance to affiliates; union-level membership data from Troy.

Findings

  • Union density grew in clear discontinuous spurts, much faster than in other periods when it rose. (p. 4)

    The change in density is larger in the spurt periods (an average annual gain of 2.2 percentage points) than in the nonspurt periods when density increased (an average annual gain of 0.5 percentage points)

  • Excluding spurts, US union history shows gradual erosion of density. (p. 5)

    By contrast, absent the periods of spurt, U.S. union history is characterized by gradual erosion of union density

  • Spurts occurred in similar periods across advanced countries, suggesting a general process. (p. 8)

    Even more striking, union growth spurts have occurred during roughly the same periods in most countries.

  • The cross-country similarity implies a common worker response to the depression rather than country-specific politics. (p. 9)

    The similarity in the pattern suggests that unionization during the depression did not reflect country-specific political events but rather a more fundamental response of employees to the depression situation.

  • Spurts arise endogenously when union organizing resources overtake employer opposition at a critical density. (p. 14)

  • More workers were organized by recognition strikes than by NLRB elections in 1934-39, so the Act changed the process more than the number organized. (p. 18)

    Overall, more workers were organized through recognition strikes during the 1934-39 spurt than were organized through NLRB elections: 1.8 million through recognition strikes versus 1 .0 million via NLRB elections.

  • The CIO was not dominated by unions that depended on its financial support. (p. 22)

    There are more than twice as many members in the unions that received relatively little CIO financial support than in those that received considerable support.

  • AFL unions gained slightly more members than CIO unions in 1935-42, so the spurt was not just a CIO phenomenon. (p. 22)

    Between 1935 and 1942 the AFL gained 2.7 million members compared with a gain of 2.5 million for the CIO (see table 8.9).