In the US, many mines and factories that once employed mainly men without higher education have closed in recent decades. They were replaced by jobs in the care sector - for example, in medicine and education, where women are more often employed. Researchers from the University of Notre Dame, the University of Zurich and Princeton decided to find out how this shift in roles affects voting. The results were published in the American Journal of Political Science.
The scientists hypothesized that men's loss of breadwinner status and the growth of women's employment create two pressures: anxiety about social status among men, and financial difficulties among women, since the new jobs pay worse. As a result, voters may support politicians who promise to restore traditional family and economic arrangements.
First, the authors analyzed data from surveys of Americans about their intentions to vote in congressional elections from 2006 to 2020. These responses were matched with information on employment at the county level from the US Census Bureau. Particular attention was paid to areas where jobs for men without higher education were declining while the share of working women without higher education was growing. It turned out that in such communities people were more often ready to vote for Republicans. A shift to the right occurred both among men with higher education and among women without it. And among women with higher education, no significant changes were found.
Then the researchers studied data from a long-term study of Americans born in 1957-1964. Participants were asked whether they agreed that it is better for a man to have a career and for a woman to take care of the family. It turned out that when a man lost the role of main breadwinner, his support for the traditional family model rose sharply. Among women who worked year-round, views became more progressive, but only if the family's total income remained stable. If family earnings fell by half or more, this effect disappeared.
The authors emphasize that these are statistical associations that by themselves do not prove causation. In addition, the conclusions concern short-term political reactions in specific generations. Over time, as the new roles become familiar, attitudes may change. Much also depends on social policy: in countries with developed support for working parents, for example in Scandinavia, the transition might look different.