Fair Housing for Same-Sex Couples


Joyce Clapp and Melissa Roberts, Project Coordinators


In January 2015, Greensboro city council members voted unanimously to add protections for sexuality and gender identity to the city’s Fair Housing Ordinance. Greensboro is the first city in North Carolina to protect gay and transgender citizens from discrimination in housing. The ordinance adds to gender, gender identity, gender expression, sexual orientation to the other federal protected categories already in place. This proposal outlines a project to audit rental housing in Greensboro for discrimination of same-sex couples. There is very little research (national or local) in the area of housing discrimination by sexual orientation. Friedman et. al. (2013) found evidence that “housing providers are less likely to respond to same-sex couples than to heterosexual couples” in online solicitations for rental housing, yet there is little data available beyond this study. It would be very beneficial to have baseline data for the City of Greensboro as these new ordinances are implemented. Our past studies have shown that these violations occur in up to 30% of rental transactions. This project seeks to document the incidence and prevalence of LGBTQ housing discrimination and to use this data to educate landlords and property managers. As discriminatory practices are exposed and ameliorated, housing choices for LGBTQ and minority home-seekers will be improved.

Fair housing testing in Greensboro has never before been conducted for sexual orientation or sexual identity discrimination. This study examines housing discrimination against the gay and lesbian couples. The study took place in two phases to gauge the prevalence of sexual orientation discrimination by housing providers in Greensboro, and then to pilot a protocol for face-to-face paired testing. In phase one, a correspondence test with nearly 266 properties was conducted to establish the prevalence of sexual orientation discrimination by landlords and property managers noting discrepancies in access, cost, information, and encouragement when possible. The second phase included 5 face-to-face paired testing and serves as a pilot for eventual audit testing of rental housing on the basis of sexual orientation. Each tester received a stipend of $50 for each completed test. To ensure consistency and accuracy, data from report forms were then compared line by line against tester narratives. Funding received from the Undergraduate Research, Scholarship and Creativity Office at The University of North Carolina at Greensboro was instrumental in the completion of the correspondence phase. Funding received from the Human Relations Department at the City of Greensboro was instrumental in the completion of this phase.

Phase I

In sum, rental inquiries were made by email or online contact forms to all 266 housing providers in Greensboro for whom a means of online contact was available. A database of all known providers of rental housing in the city of Greensboro was created in preparation for this project. This database began with an existing list of already known housing providers in the city, which included individual landlords, apartment complexes, and property management companies. All housing providers received inquiries from the same email accounts representing the same fictitious couples in the same order.



Overall, 38.0% of properties did not respond to any correspondences. Yet, the 165 properties to respond to at least one of the solicitations (email or online form), generated a total of 661 email responses. In all, 92 properties (55.8% of responding properties) showed equal access for all couples. A third of responding properties (52 total) did not respond to all queries. Of these, 9 did not respond to male-female inquiries: 4 of those favored male-male couples by responding only to them, and 5 favored female-female couples only. Thirty-three properties (20%) did not respond to male-male inquires, and thirty-four properties (20.6%) did not respond to female-female inquires. About 15.8% (26) of responses were inconclusive.


A variety of measurements of cost were recorded including fees related to lease initiation (application fees, administrative fees, other fees) and monthly fees (rent, any included utilities, etc.), as well as any specials offered to the fictitious couples. In determining equality of costs quoted, all measurements of costs were considered. Thus, housing providers who indicated different rental ranges for different couples, or advised of a special to waive administrative fees for some but not all couples, were indicated as providing unequal costs to couples. In all, 7.9% (13) apartments quotes the same or equal information to all inquiries. Most properties, 82.4% (136), were indeterminate with costs not provided to everyone. Finally, 9.7% (16) properties clearly provided unequal costs to inquiries.


Information provided was gauged based on housing provider response regarding availability, rental qualifications, and unit information (number of bedrooms/bathrooms, etc.) A variety of measures were considered in determining the qualifications noted, including mentions of credit checks, background checks, fees, income, and references. Providers who gave different information to any of the couples on any of these measures were coded as providing differential outcomes.


Multiple variables were considered in measuring encouragement provided to each couple, these included follow-up, and offers of inspection and contact. Regarding follow-up, the number of emails/follow-ups sent from housing providers to each couple were reviewed; housing providers who made further attempts at contact with some couples but not all, as gauged by the number of emails sent, were found to have provided differential encouragement. Offers by housing providers to inspect the (tour) the property, or to make contact by calling the provider or coming to the office were also compared across couples for each housing provider. In all, 55.8% of tests showed equal encouragement. Two-in-five tests (40.0%) were inconclusive in terms of encouragement. A small fraction (1.8%) of cases showed less encouragement for male-female couples, while 1.2% showed less encouragement for all same-sex couples. There was a single case (>1.0%) that showed less encouragement for male-male couples and likewise one case (>1.0%) showed less encouragement for female-female couples.

Phase II: Face-To-Face Testing

Phase II of the study consisted of a pilot test or face-to-face testing for five housing providers in Greensboro. As in correspondence testing, four couples each comprised of two testers (two female/male; one male/male; one female/female) visited each selected property. Testers received a small stipend of $50 for each completed test, as well as for required tester training.

Properties tested for the Phase II pilot study were randomly selected from those with differential rates of access in Phase I. Each test cycle began with an Advanced Call made by the Testing Coordinator to confirm availability and obtain basic information about any available units. Once availability was confirmed, the four tester couples were provided with the name and basic information for each property, a requested rent range (which remained the same for all couples) as well as the suitable time-window for testing.


Typical measures of access for fair housing studies center around the access or lack of access to housing providers by housing testers: whether partners were able to call agents, meet with agents, inspect the advertised housing unit and/or other units. For Phase II of this study these same measurements were used. None of the couples for our housing tests faced difficulties in obtaining access to housing providers; for all five properties, all measures of access were granted to all couples making inquiries.  No differential access outcomes were observed for any of the five tested properties regardless of condition tested.


Data collected regarding quoted costs included: application fees, credit check fees, security deposit, monthly rent, and any other fees noted (such as cleaning, administrative, or other fees). Generally, all couples were provided the same costs during tester audits. However, there were several discrepancies. Regarding application fees, two properties provided information regarding promotions and fee reductions to some, but not all, couples. Housing provider basis for determining security deposit amounts, and the upper range of possible security deposit, varied to some degree in one test. All couples received the same information regarding credit check fees, and review of questionnaires, narratives, and additional materials related to other fees also appears to indicate no differences in these quoted costs between couples.


Comparison of information provided to testers included that related to availability as well as qualifications. Regarding availability, testers recorded whether the agent indicated an apartment was available on the date, and the size and price requested. All couples across all tests were advised that an apartment was available. Data was also collected from verbal information provided to testers regarding requirements for rental, including: amount of income, credit qualifications, employment background, rental background, criminal history or background check.


Measures of encouragement included variables related to the initial contact with agent, inspection of unit(s), and agent follow-up. These were measured via data on couple wait times, whether they were asked to complete an application during the audit, whether they were offered the opportunity to inspect applicable units, if agent stated they may be ineligible to rent, and whether they were asked about a follow-up from the agent. All couples across all tests were advised that an apartment was available, minor variation occurred with wait time, encouragement to complete an application, ineligibility, and request to follow-up.

Full Report Available: 2-28 Final Draft LGBTQ Report