Earnings Differentials of Males and Females in - PowerPoint Presentation

Slide1

Earnings Differentials of Males and Females inSame-sex and Different-sex Couples in Canada, 2006-2010 

Richard E. Mueller

Department of EconomicsSlide2

Introduction Gays and lesbians have been able to marry legally since July 2005 – although such marriages recognized before this date in most jurisdictions Discrimination of the basis of sexual orientation is prohibited by Chapter 15 of the Charter of Rights and Freedoms

Have these rights also extended to the workplace in terms of equal earnings?Slide3

Previous LiteratureMost US studies find that gay men earn less, and lesbian women more, although more recent studies show no difference – degree of “gayness” is importantOnly four previous papers in Canada

Mueller (2007) – uses the GSS

Carpenter (2008), Lafrance, Warman and Woolley (2009) and Harris (2012) all use various waves of the CCHS

All basically find the gay penalty and the lesbian

premium

for

those in couplesSlide4

Identification ProblemWhat is gay, lesbian and bisexual?Behaviour-based approach

proportion of same-sex partners?

any same-sex partner?

same-sex partner over past 5 years?

Self-identification

cohabitating with a same-sex partner

direct questioning

Indicator for “gayness” in labour market best, but marriage is a strong signalSlide5

Data Only three data sets up to the task Census – largest samples, but not as confidential and only covers coupled gays and lesbians – hasn’t been used yet

CCHS – broader coverage of GLB, but not as confidential

GSS – only partnered gays and lesbians, but a telephone survey which (arguably) elicits more accurate responsesSlide6

DataUse pooled GSS data from Cycles 20-24 (2006-2010)Keep only thoseaged 20 to 60 (attached to LF)

who worked in previous year,

without missing data points

earning more than $5 dollars and less than

$500

dollars per hour

who are cohabitating (married or CL)Slide7

Data problem“Annual personal income” could be from any source, plus it

’

s a categorical variable

calculate midpoints & transform into 2006 dollars

drop those whose main source of income is not paid employment or SE

do interval regressions (robustness check)

Final sample: 14,021 male (90 gays) & and 11,498 females (118 lesbians) Slide8

Profile of GaysRelative to males in different-sex partnerships, gay males have identical incomes (statistically)

have more education (esp. Bachelor’s degree or above, 51% vs. 32%)

have less experience, work fewer hours per week, but work more weeks (still fewer yearly hours)

less likely to be married, be an immigrant, or have children present

less likely to work in construction and manufacturing, concentrated in prof. servicesSlide9

Profile of LesbiansRelative to females in different-sex partnerships, partnered lesbians incomes about 40% higher

have more education (esp. Bachelor’s degree or above, 57% vs. 33%)

work more hours per week

less likely to be married or have children present

More likely to be in construction and prof. services, less likely to be in health and SASlide10

ResultsBuild up model in a stepwise fashion by adding variables for education, experience, weeks and hours worked, occupation, industry, children, etc.Compared to those of in heterosexual partnerships:

Gay males have an income penalty of 4-5 percent (but not significant)

Lesbians earn about 16 percent more (and this is significant)

Results reasonably robust to changes in model and estimation techniqueSlide11

ResultsBut (even though results are mostly insignificant):comparing married gays and lesbians to either CL gays and lesbians or married heterosexuals reduces their relative incomes limiting the sample to only the self-employed increases their relative incomes

Slide12

ConclusionsGays in partnerships have a relative income penalty of about 5% (but not significant)Lesbians have incomes 16% higherComparisons with earlier Cdn. literature

increases in relative incomes (but comparisons a bit shaky)

– real change or bias due to confidentiality?

Weak evidence of discrimination

higher SE income

lower

incomes for

married gays and lesbiansSlide13

Thank you!