The U.S. Census Bureau in July began asking Americans about their sexual orientation and gender identity — a watershed moment that marks the first time the federal government has tried to capture data on LGBTQ+ Americans in its large real-time national surveys. 

The results so far are preliminary, but they do indicate that the disparities queer Americans experienced prior to the pandemic have continued to endure 18 months in. For some, those disparities have grown deeper.  

According to the data, which captures results from July 21 to September 13, LGBTQ+ people often reported being more likely than non-LGBTQ+ people to have lost employment, not have enough to eat, be at elevated risk of eviction or foreclosure, and face difficulty paying for basic household expenses, according to the census’ Household Pulse Survey, a report that measures how Americans are faring on key economic markers during the pandemic. 

While think tanks like the Williams Institute at the UCLA School of Law and advocate-led research groups have previously studied LGBTQ+ poverty, no large government population surveys, like those conducted by the census or the Treasury Department, have attempted to capture the real-time economic experiences of LGBTQ+ people. 

Previously, those analyses were limited to studies of “same-sex couples,” a question the census began analyzing with limited success in 1990, but that leaves out significant portion of LGBTQ+ people. Lack of accurate data on the population as a whole — and particularly on transgender people, a group that has been chronically under surveyed — hampered any federal response to persisting inequities, advocates say. 

“Having this on [the Pulse survey], both as a way to understand what’s going on during the pandemic, but also hopefully as a starting point to more federal data collection, is really an important moment,” said Bianca D.M. Wilson, the senior scholar of public policy at the Williams Institute. 

The data has only begun to be collected, and it’s still too early to tell whether the differences between groups are representative of the LGBTQ+ population overall or just those who were surveyed by the census at a given moment in time. 

While researchers cautioned against drawing major conclusions, the trends that emerge in the data are consistent with what other surveys have found prior to the pandemic as a result of employment discrimination, underpay, discriminatory lending practices and other policies that have limited economic mobility for queer people.

According to The 19th’s analysis of the first four releases of data from the census survey, as much as 23 percent of LGBTQ+ people and 32 percent of trans people reported having lost employment in the month before the census conducted its questionnaire. About 15 to 16 percent of non-LGBTQ+ people reported the same. 

About 12 percent of LGBTQ+ people said they sometimes or often did not have enough to eat. For non-LGBTQ+ people, the figure was between 6 and 7 percent, and for trans Americans, it was as high as 24 percent. About 31 percent of queer people also said they had difficulty paying for basic household expenses; for non-LGBTQ+ people it was 23 percent.

Housing insecurity was prevalent across all groups, with more than 40 percent of people — both LGBTQ+ and non-LGBTQ+ — saying they were very or somewhat likely to face eviction by the end of September or October. 

It’s unclear how accurate the data for transgender Americans is because the sample sizes are much smaller. But it does follow what is already known: Roughly 29 percent of respondents to the 2015 U.S. Transgender Survey, done by the National Center for Transgender Equality and seen as the only comprehensive study of its kind, said they lived in poverty. About 30 percent said they had experienced homelessness in their lifetimes.

“These are sort of the systemic disparities that we observed pre-pandemic, that the pandemic has not only deepened for both groups, but also sort of widened,” said David Schwegman,  assistant professor of public policy and administration at American University, who has conducted research on “same-sex couples” and housing discrimination.  

Wilson at the Williams Institute said that absent this kind of large-scale data collection about LGBTQ+ people, policymakers couldn’t truly answer big questions about whether attempts to address economic stress exacerbated by the pandemic — like the now-expired federal eviction moratorium — were working for everyone.

But data collection is only one step toward equity.

Dean Spade, an associate professor at Seattle University School of Law who has also advised the upcoming National LGBTQ+ Women*s Community Survey by think tank Justice Work, said that real change requires more than just counting trans and LGBTQ+ people at the federal level.

Counting marginalized people to better understand the issues they face doesn’t necessarily mean their suffering will be addressed through policy, he noted — and trans people are accustomed to social services leaving them out or not being designed with them in mind. It’s why trans people, for example, are helping each other pay for medical procedures that aren’t covered by insurance, housing those experiencing homelessness and creating mutual aid networks, Spade said.

“We’re helping each other survive right now,” he said.

And there are still significant challenges with the data as it is. Samples sizes are small, an issue that has barred marginalized communities, including Asian women, Native Americans and Pacific Islander women, from representation in real-time data on some national surveys. 

Those small sample sizes make it difficult to draw big conclusions from the data until months down the line. The Census Bureau said in a statement that it currently doesn’t have additional analysis to offer on the data, though it did publish a report on the first set of LGBTQ+ data this summer, finding that LGBTQ+ people are more likely than non-LGBTQ+ people to face economic hardship.

“The primary focus has been on collecting and releasing data in a timely manner but there are plans in the future to release data products that will provide additional context,” the bureau said in a statement.

The other challenge has been crafting questions in a way that takes into account knowledge gaps people may have about what terminology best describes them. 

The census survey, for example, asks respondents to choose which best represents how they think of themselves: “gay or lesbian”; “bisexual”; “something else”; “I don’t know”; or “straight, that is not gay or lesbian.” In past attempts to phrase these questions, heterosexual people have been found to incorrectly mark themselves, economists said, so additional phrases have been added to improve clarity. 

The survey also asks if people describe themselves as male, female or transgender, and some transgender people may not want to identify themselves given the rise in anti-trans bills across the country, Schwegman said. 

Spade pointed to smaller studies led by advocates as important pools of information that can’t be found anywhere else, since they ask questions about daily threats like over-policing and poverty. 

“I think that those kinds of studies can be, to a lot of us, more valuable than something larger that didn’t ask the questions or that missed whole groups of people in our community,” he said.

The real-time data from surveys like the current census one, which will be collecting responses from July 21 to October 11, could help impact policies in real time. The problem for pandemic-related policies being negotiated in Congress this fall is that this data may be coming too late, Wilson said. 

“It’s 18 months into the pandemic, and had that been the starting place, we would not be looking at a sample size that would create problems for all the analyses that we want to do to understand a trans-specific experience,” Wilson said.