Gender Preferences in Job Vacancies and Workplace Gender Diversity
In spring 2005, Austria launched a campaign to inform employers and newspapers that gender preferences in job advertisements were illegal. At the time over 40% of openings on the nation’s largest job-board specified a preferred gender. Over the next year the fraction fell to under 5%. We merge data on filled vacancies to linked employer-employee data to study how the elimination of gender preferences affected hiring and job outcomes. Prior to the campaign, most stated preferences were concordant with the firm’s existing gender composition, but a minority targeted the opposite gender - a subset we call non-stereotypical vacancies. Vacancies with a gender preference were very likely (>90%) to be filled by someone of that gender. We use pre-campaign vacancies to predict the probabilities of specifying preferences for females, males, or neither gender. We then conduct event studies of the effect of the campaign on the predicted preference groups. We find that the elimination of gender preferences led to a rise in the fraction of women hired for jobs that were likely to be targeted to men (and vice versa), increasing the diversity of hiring workplaces. Partially offsetting this effect, we find a reduction in the success of non-stereotypical vacancies in hiring the targeted gender, and indications of a decline in the efficiency of matching. For the much larger set of stereotypical vacancies, however, vacancy filling times, wages, and job durations were largely unaffected by the campaign, suggesting that the elimination of stated preferences had at most small consequences on overall job match efficiency.
We thank Ghazala Azmat, Omar Bamieh, Thomas Le Barbanchon, Pierre Boyer, Clément de Chaisemartin Armin Falk, Lucas Girard, Ingrid Haegele, Kerstin Holzheu, Patrick Kline, Michael Kosfeld, Francis Kramarz, Peter Kuhn, Marco Manacorda, Paolo Mangano, Steeve Marchand, Guy Michaels, Maria Guadalupe, Alexandra Roulet, Ity Shurtz, Benoit Schmutz, Camille Terrier, Andrea Weber, David Yanagizawa-Drott, Lennart Ziegler, Florian Zimmermann, and Josef Zweimüller for helpful comments. We thank Josef Zweimüller for providing us access to the data, and Damian Osterwalder for support in interpreting features of the data. Elke Lujansky-Lammer and Rainer Lichtfeld provided information on the institutional background. We also thank seminar participants at UC Santa Barbara, UC Berkeley, University of Lausanne, CREST, CERGE-EI, Insead, and Ben Gurion, as well as audience at the Brucchi Luchino Conference, the Ski and Labor Workshop, the Annual Meeting of the Italian Association of Labor Economics, IDSC of IZA Workshop, the Spring Meeting of Young Economist, and the Annual Congress of the Swiss Society of Economics and Statistics. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.