Essays on Labor Market Discrimination on the Basis of Sexual Orientation
- Author(s): Levine, Michael
- Advisor(s): Marks, MIndy
- Sorensen, Todd
- et al.
The following chapters present an examination of the causes and effects of labor market discrimination on the basis of sexual orientation.
The first chapter examines customer-based discrimination, which occurs when prejudice leads individuals to prefer commercial interactions with certain groups of workers to others. Under customer-based discrimination, otherwise identical employees across different classes of workers are not perfect substitutes, as race, gender or sexual orientation becomes a component of workplace productivity.
This is the first paper to identify customer discrimination against gay and lesbian workers. I exploit geographic variation in attitudes towards homosexuality to see if occupational sorting choices vary amongst gay workers in less prejudiced parts of the country versus more prejudiced parts of the country. To this end, I construct a novel state-level index of attitudes towards homosexuality using individual survey responses to questions about taste. Multiple specifications provide evidence that gay men sort away from customer service jobs in areas of the country with strong levels of distaste for homosexuality. I do not, however, find evidence that this occupational sorting exists amongst lesbian women.
The second chapter explores whether firms that partake in taste-based discrimination on the basis of sexual orientation will face a profit penalty, consistent with the Becker model of discrimination. Prior research on discrimination against gay workers has established other aspects of the Becker model, however, the profit penalty is less established in the literature. This paper improves upon existing estimates of the relationship between profits and discriminatory policies by introducing fixed effects to isolate the effect of changes in corporate policy. Additionally, this paper introduces discrete levels of discrimination as explanatory variables to assist in framing the identified effect in the context of the Becker model.
Under multiple specifications, evidence of a relationship between CEI score and profit rate is detected robust to the inclusion of firm fixed effects. These estimates suggest profit rates are improved when firms move from high levels of discrimination to low levels. Consistent with the Becker model, there appears to be no profit increase when low-discriminating firms further advance non-discriminatory policies.