There are a lot of possible places to start when looking at the literature on gender or racial wage discrimination/differentials. Preferably, it'd be good to look at the literature on labor supply as a whole (e.g. Mincer equation, work-leisure tradeoffs etc.), but we could be here all day discussing papers related to modelling wages in general (I am happy to provide some if you would like.) I will do my best to summarize the theoretical and empirical literature I give reference to below. Let us look at differentials.
Becker (1973), A Theory of Marriage: Part I was some of the earlier work on family economics, where men and women would tend to pick each other for marriage based on their relative strengths. To put it roughly, though people tend to like marrying others who are the same height, race, education level, etc., the theoretical model set up by Becker predicted that because families care about work (employment), leisure, and household production, there'd end up being some sort of division of labor so that all these needs were met.
Mincer and Polachek (1974), Family Investments in Human Capital: Earnings of Women and others would build off of this earlier body of work when looking at the effects of education and work training on wages. This paper in particular argued that the expectations of market activity, particularly regarding raising children, would affect how much women would pursue education or work training, and would effect how employers would want to pay women over time. If women are more likely than men to leave the workforce to spend time raising a child, then their wage-age profile will change. Intuitively, even before a child is born, women might get paid less because future expectations of productivity affect wages and individual investment into skills.
Becker (1971), The economics of discrimination is a gigantic book written as a lot of the basis for modelling economic discrimination. There's a lot to sift through and I'm not super well-read in that area, but the gist of the model set up here is you can distinguish between prejudice which is where employers may have a preference for certain type of worker that shows up in the utility function, and discrimination, which is the explicit unequal treatment of equals; that is, unequal outcomes. Although there may be prejudice from employers, depending on the search model used, if there are enough jobs by non-prejudiced employers, we may not see any discrimination in outcomes.
Even after controlling for confounding variables that could affect the wage gap between women and men, or between races, there is some discrimination in outcomes on the basis of race and gender.
Dougherty (2004) Why is the rate of return to schooling higher for women than for men? uses the Oaxaca decompositions in E. Sommer's answer to look at why the rate of return to education seems to be higher for women than for men. In this paper, one theory looked at was that better educated women are more willing/able to compete with men to push back against gender discrimination, hence why it seems the rate of return is higher. The paper has some support for this finding.
Lang and Lehmann (2011) Racial Discrimination in the Labor Market: Theory and Empirics notes that while taste/preference frameworks combined with search models are able to explain a lot of differences between wages and unemployment duration, these models are not so good at showing differences in unemployment rates themselves. Statistical models (employers have some sort of imperfect information that leads to discriminatory outcomes through, for example, self-confirming stereotypes) can explain magnitudes of wage differences well, but not much about employment. This paper tried to combine elements of the two approaches.
This above article is a particularly good one to look at (and should not be too much of a problem to sift through if you have a good math background). I don't think I can do justice trying to summarize the breadth of their results here in a reasonable amount of time, so check it out on your own if you like.
There are a lot of good articles in labor economics that look at gender and racial discrimination but those are a few I've picked out to give some brief summaries on. Some of the ones I've not mentioned, but which are important to the literature are listed below. Just because I chose not to summarize them does not mean the below articles are particularly less important. I just have my own time constraints. My presentation of the existing literature is also not comprehensive, but hopefully a good starting point for others.
Hope some of these resources prove useful to you and future users.
Further Reading
Oaxaca-Blinder Decomposition:
Blinder, Alan S. "Wage discrimination: reduced form and structural estimates." Journal of Human resources (1973): 436-455.
Oaxaca, Ronald. "Male-female wage differentials in urban labor markets." International economic review (1973): 693-709.
Labor Supply/Compensating Differentials:
Oaxaca, Ronald L., and Michael R. Ransom. "On discrimination and the decomposition of wage differentials." Journal of econometrics 61, no. 1 (1994): 5-21.
Rosenzweig, Mark R., and T. Paul Schultz. "The demand for and supply of births: Fertility and its life cycle consequences." The American Economic Review 75, no. 5 (1985): 992-1015.
Henderson, Daniel J., Solomon W. Polachek, and Le Wang. "Heterogeneity in schooling rates of return." Economics of Education Review 30, no. 6 (2011): 1202-1214.
Goldin, Claudia, Katz, Lawrence F, and Kuziemko, Ilyana, “The Homecoming of American College Women: The Reversal of the College Gender Gap”, Journal of Economic Perspectives 20(4), 2006, 133-156
Blau, Francine D., and Lawrence M. Kahn. "The US gender pay gap in the 1990s: Slowing convergence." ILR Review 60, no. 1 (2006): 45-66.
Mobius, Markus M., and Tanya S. Rosenblat. "Why beauty matters." American Economic Review 96, no. 1 (2006): 222-235.
Lang, Kevin, Michael Manove, and William T. Dickens. "Racial discrimination in labor markets with posted wage offers." American Economic Review 95, no. 4 (2005): 1327-1340.
