Testing for Gender Effects Using Natural Experiments
Saturday, Jan. 4, 2020 10:15 AM - 12:15 PM (PDT)
- Chair: Cheryl Doss, University of Oxford
Gender Difference in Labor Demand Behavior
Abstract"In this study, we investigate whether employers’ hiring decisions vary by the gender of the firm’s owner. Our preliminary results show a positive and statistically significant female-owner effect on employment, where female-owned firms are likely to hire about 9% more employees than male-owned firms. We use the instrumental variable (IV) estimation of the labor demand model to control for endogenous capital input due to unobservable factors correlated with firm value. The data used is the 2007 Survey of Business Owners (SBO). The SBO is a 5-year period survey of operating firms and companies in the United States, conducted by The Census Bureau. Surveyed firms are randomly selected from the list of firms’ filed tax returns to the Internal Revenue Service (IRS). The Census Bureau obtains the sample firms’ number of employment, payroll data, and receipts from IRS tax returns. Other information related to the firm owners’ demographics and their business operation is obtained via mail. Of the total 2,165,680 firms represented in the 2007 SBO sample, there are 663,385 single-owner firms. In our dataset, about 33% of the firms are female-owned. The positive female-ownership effect on employment is consistent across different groups such as age, education and industry, though a remarkable heterogeneity on the effect in terms of size and statistical significance is observed. Our analysis shows that younger and more educated female business owners tend to hire more employees than the male owners. The female owners with Bachelor and post-graduate degrees hire 10.6% and 10.3% more employees respectively. The female owners aged between 35-44 and 45-54 hire 11.0% and 15.0% more employees respectively.
The labor demand model with the Average Treatment Effect (ATE) for the female-owned firm is estimated using a log of the number of employees as the dependent variable, and initial capital stock and labor
Gender Identity and Wives' Labor Market Outcomes in West and East Germany between 1984 and 2016
Abstract"We exploit the natural experiment of German division and reunification to investigate how the institutional regimes of the formerly socialist East Germany and the capitalist West Germany shaped gender identity prescriptions in terms of family breadwinning. Germany was divided in two states in 1949 (West: Federal Republic of Germany/FRG and East: German Democratic Republic/GDR), separated by the wall between 1961 and 1990, and reunited in 1990. We argue that during 40 years of separation, public policy and promoted ideologies regarding family models and female employment led to particular cultures concerning paid and unpaid work – for people who shared the same history and culture before the division. Moreover, because the West German institutions were imposed on East Germany following reunification, we can examine how distinct gender identity prescriptions between East and West Germany persist under a common, and slowly emancipating, institutional system.
Based on representative longitudinal data on German married couples from the German Socio-Economic Panel (SOEP), we explore how men’s breadwinner and women’s homemaker role affected wives’ labor market decisions in West and East Germany for three periods: 1984–1990, 1997–2006 and 2007–2016. By scrutinizing if, and how, couples avoided situations where the wife out-earns her husband, we present evidence suggesting that the market economy of West Germany prior to reunification fostered a strong male breadwinner model, whereas the East German socialism did not. Density discontinuity tests and fixed-effects regressions suggest that only couples in West Germany diminished the wife’s labor market outcomes in order to avoid a female breadwinner arrangement. For 1984–1990 and 1997–2006 (but not 2007–2016), the distribution of the wife’s share of household income dropped sharply at 50 percent (where she would start to earn more than him),and wives who out-earned their husband in year t–1 were less likely to do so in year t.
- J1 - Demographic Economics
- J6 - Mobility, Unemployment, Vacancies, and Immigrant Workers