Women and the labor movement: occupational sex composition and union membership, 1983-2005

This dissertation brings together two distinct strands of labor market research. Studies on union decline seek to explain the dramatic drop in union membership and influence in the post-World War II era. Studies on occupational sex segregation seek to document how and understand why most women tend to work in occupations with other women. These two broad processes have received much of scholarly attention individually; but there has been little effort to understand how they might be linked. Historical accounts suggest that many unions practiced social closure, ignoring or actively avoiding the causes of working women. And as women have entered the labor market in large numbers in recent decades, they have generally taken positions in the so-called hard-to-organize sectors--industries and occupations that have little historical exposure to unionism, and are regarded as fundamentally incompatible with traditional unionism. My research links the two bodies of work by examining the extent to which varying levels of unionization across different sectors of the labor market is directly attributable to the varying sex composition of those sectors.

I rely on data from the 1983-2005 Current Population Survey Merged Outgoing Rotation Group Files, and use statistical purging techniques which allow me to examine what union coverage rates would have been had the so-called hard-to-organize sectors not been disproportionately comprised of female-dominated occupations. Because processes of unionization are quite different in the public and private sectors, I estimate separate models for these two groups. The private sector findings provide modest support for my arguments, and indicate that union coverage rates in some of the hard-to-organize sectors would have been slightly higher had these sectors been comprised of an even mix of male- and female-dominated occupations. And in the sectors that are more compatible with traditional modes of unionism, union coverage rates would have been lower had these sectors not been heavily male-dominated. The public sector findings tell a different story, and suggest that some public sector unions organize along traditional lines vis-à-vis occupational sex composition, while others have found ways to organize on new, female-dominated terrain.