Lesbian Day of Visibility

It’s Lesbian Visibility Week so we thought we’d share some of our projects preliminary findings about welfare equality for lesbians. Often LGBT+ data analysis just lumps LGBT+, or LGB, people together, ignoring the vast differences between them. This has been a problem for lesbians throughout the centuries of LGBT+ activism – the privilege of gay men means it is their voices and experiences that get heard and dominate analysis. In this project, we want to push intersectional analysis as far as the data will allow. An easy first-step for this is to focus on the experience of lesbians, who experience the compounding discrimination of being homosexual and women, and compare it to heterosexual women and gay men.

The data is from the Understanding Society an annual longitudinal household survey in the UK which has collected data on sexual orientation since 2011. To maximise the sample size we combine waves 2011 through 2020 keeping only the most recent record per person.

Sexual OrientationMale Female 
Heterosexual22,60493%27,41693%
Gay or Lesbian4012%2861%
Bisexual2671%4722%
Other2351%3451%
Prefer not to Say7673%1,0143%
Total24,274100%29,533100%
Sample size used in analysis

As shown above, crosstabulation between sex and sexual orientation shows that about 2% of the male sample identifies as gay and a further 1% identify as bisexual. By contrast around 1% of the female sample identifies at lesbian and a further 2% identify as bisexual.

 Heterosexual WomenLesbian WomenGay Men
Percent home owners (inc mortgage)70%67%62%
Percent LA or HA renting 18%16%18%
Percent own home worth over 200k38%34%37%
Percent receiving investment income £500+ PA7%5%10%
Percent receiving benefits  £50+ PW44%28%25%
Key outcomes for heterosexual women, lesbian women and gay men

We then compared a number of welfare outcomes between lesbian women and heterosexual women, and lesbian women and gay men. Due to recent welfare reform, and longer-term retrenchment in the UK, personal assets including housing have become more important to an individual’s welfare situation, so we are investigated both welfare access and asset ownership.

In our sample lesbian women are less likely to own their home and are also less likely to live in socially rented housing (from a local council or housing association) than heterosexual women, but these finding are not significant so are not generalisable to the wider population. 

There is, however, a significant (p < 0.05) difference between the housing values of lesbian and heterosexual women. Similarly we find no significant difference between the housing tenure of lesbian women and gay men but we do find a significant difference between their housing values. This is an important finding, which has not been explored widely in the literature on asset-based welfare, or housing economics.

When looking at investment assets we find that in our sample lesbian women are less likely to receive £500 or more per year from investments but this difference is not significant. There is however a significant difference between the investment income of gay men and lesbian women, with gay men more likely to receive £500 or more per year from their investments. 

In terms of benefit receipt, we find no significant difference between gay men and lesbian women when considering receiving benefits of £50 or more per week, but we do find a significant difference between heterosexual and lesbian women. Heterosexual women are more likely to receive state benefits. This may be because heterosexual women are more likely to claim Child Benefit.

In conclusion we find some evidence that lesbian women are disadvantaged in personal assets, compared to both heterosexual women and gay men. We also find that lesbian women are less likely to receive state benefits than their heterosexual counterparts. Further analysis will attempt to disentangle these differences from related control variables which also vary across these groups including income, partner status, family and household size and age.

Theme by the University of Stirling