Was it Discrimination or the Market? Female Employment and the Wage Gap by Michelle Rendall :: NEUDC 2007 Papers :: Northeast Universities Development Consortium Conference :: Center for International Development at Harvard University (CID)

This paper quantitatively tests how much of the post-WWII evolution in employment and average wages by gender can be explained by a model where changing labor demand requirements are the driving forces. I argue that a big fraction of the original female employment and wage gaps in mid-century, and the subsequent shrinking of both gaps, are explained by labor reallocation from brawn-intensive to brain-intensive jobs favoring women’s comparative advantages in brain over brawn. I analyze the effects of an exogenous “brain-biased” technical change, increasing the relative productivity of brain-intensive to brawn-intensive production processes, on aggregate employment and wage gap trends. Initial results suggest the mechanism to be able to explain 37 to 67 percent of the rise in married female labor force participation, about 89 percent of the rise in single female labor force participation and about 53 to 62 percent of the closing wage gap, with an initially slower growth rate in average female to male wages due to selection bias. Moreover, the model, similar to the data, generates fairly steady married and single men’s labor force participation over time. ∗Department of Economics, University of Texas at Austin, 1 University Station C3100, Austin, TX 78712. Email: [email protected]. All errors are mine. M. RENDALL 2 1 Theories and Facts on Women’s Employment This study provides evidence from United States data and develops a general equilibrium model based on the following three facts of labor demand, labor supply and wages since World War II: 1. Women’s labor force participation, aged 25 to 64, rose from only 32 percent in the 1950s to 71 percent in 2005. More specifically, married women increased their labor force participation by 46 percentage points and single women by 15 percentage points (see figure 1), while men’s labor force participation stayed fairly steady. 2. The gender wage gap, defined as average female to average male wages, changed quickly during the same period. Initially falling from close to 65 percent to a low of 57 percent in the mid 1970s, before starting to close again reaching around 77 percent by 2005 (see figure 1). While some of the initial fall in the wage gap was likely due to World War II, a stagnation is undisputable. 3. A selection bias of women into occupations with initially lower wages, and a subsequent rise of the relative returns to these occupations, coupled with a rise in women’s relative labor supply to these occupations can be clearly seen in the data (the classification of these occupations will be discussed in detail in section 2 and 5). That is, the wage gap closed for two reasons, (1) a rise in the returns to “female-friendly” occupations, and (2) a faster rise in the female to male efficiency unit labor supply to these occupations (see figure 2). While it is a popular perception that anti-discrimination laws focused on gender equality were the main reasons behind women’s changing labor market participation and earnings, economic studies have found various other reasons played an important role in shaping women’s labor market experience, such as changes in women’s work experience, education, and occupational changes (see for example Black and Juhn, 2000; Blau, 1998; Mulligan and Rubinstein, 2005, and references therein). The main driving force in changing female employment and wages should be of particular interest to policy makers. For example, in order for the European Union to reach its 2010 Lisbon target of raising the proportion of women in the labor force from 51 to 60 percent, European Union policy makers can learn from what has driven the United States or other countries’ female