Job Search and the Gender Wage Gap

A recent body of literature argues that at least some part of the gender wage gap may be due to differences in the bargaining and negotiation behavior of men and women. If women are less likely to bargain over wages, they will take home a lower fraction of their match surplus. This would contribute to the gender wage differentials observed in the data. Alternatively, wage differentials could be due to differences in sorting into different types of jobs or firms. A well-established literature finds that a considerable fraction of the gender wage gap is due to the occupational choices of men and women (see, for example, Groshen, 1991; Blau and Kahn, 2000, among others). Song, Price, Guvenen, Bloom, von Watcher (2015) find that pay differences between firms are an important source of earnings inequality. Both lines of research suggest that systematic differences in sorting by gender can lead to a gender wage gap. Recently, Card, Cartoso, Heiningand, and Klein (2013) quantified these two separate channels in the Portuguese data: the first one is the sorting channel—the possibility that women are less likely to work for higher-wage firms. The other is the bargaining channel—the possibility that women are less aggressive in bargaining than men. The sorting channel can work through occupational choice as well. Similarly, bargaining is only one aspect of the search process. More broadly, differences in search behavior between men and women, including differences in bargaining, search effort, and search efficiency, may be a key contributor to the gender wage gap. Both the sorting channel and the job search process are closely linked to a worker’s ability to change jobs. In this paper, we focus on gender differences in the job search and job-finding process. Using a unique new household survey, we study the relationship between job search behavior and outcomes by gender. We document new facts on their differences and quantify their effects on the gender wage gap. We find that a large gap in the wages of men and women exist not only in their current wages, but also in their wages at the time of their hiring and the wages of both recently-offered and recently accepted jobs. Gender differences observable characteristics, including demographics, the occupation of the job in question, and firm characteristics, can account for about half of the wage gap of the current job and about two-thirds of the pay gap of job offers. Accounting the differences in prior work history accounts for much of the remaining gap. An important caveat to these findings is that women pay a somewhat larger penalty than men for receiving or accepting an offer from non-employment, even after controlling for observables and work history. Our evidence on job search behavior suggests why differences in in work history may play an important role. We find that women tend to search more while employed and tend to search more intensely, but that they tend to fare slightly worse than men in generating job offers. A key reason for this difference is that men tend to have stronger informal recruiting channels. They generate more unsolicited offers and are more likely to have an employer contact through a referral. The stronger informal networks among men may reflect stronger ties to the labor market. Both may work together to contribute to the gender wage gap. We find that men are more likely to engage in bargaining than women, but only slightly more so. Overall, our results thus far suggest that gender differences in job search behavior and outcomes affect the gender wage gap through their effects on the work histories of men and women.