Can a shallow, flexible rent subsidy prevent homelessness?

Why is this issue important in DC?
Since 2016, the number of families experiencing homelessness in DC has decreased by 49%. ¹ While the District has made substantial investments and significant progress addressing homelessness, a large gap between wages and housing costs means that many families leaving the homelessness system can still struggle to maintain stable housing, even when they are employed. DC has many different tools to address housing affordability, ranging from one-time emergency rental assistance to permanent housing, but it is important that we try to use our resources to provide the right amount and type of assistance for each family. Programs that provide smaller amounts of assistance every month—called “shallow” subsidies—may help, but families can sometimes struggle when their income or expenses fluctuate.

What are we doing?
The District wants to know if a more flexible shallow subsidy that empowers households to make their own budgetary decisions will improve their housing stability and allow the District to serve more families. In 2017, DHS piloted a rent subsidy program called DC Flex, which gives low-income, working families $7,200 to spend per year on rent for four years (or longer if the pilot is extended). Because a family’s ability to pay rent may change as their income and expenses fluctuate month-to-month, ² participants can choose how much of their DC Flex funds to spend on rent in a given month (up to their total rent amount) until the $7,200 is exhausted each year.

To test the impact of DC Flex, we conducted outreach to identify potential applicants and then used a lottery to randomly select 125 eligible household applicants to participate. We will compare outcomes between those who received the funds and those who didn’t each year to measure whether the pilot reduces homelessness or housing instability and whether it increases economic-well being. Applications for the pilot closed in July 2018, and any new openings are offered to people randomly chosen among applicants.

What have we learned?
We found that, in the first year of DC Flex, only 2% of participants became homeless—meaning they needed emergency shelter for at least one night. But, this low rate was similar to the control group of families who did not receive DC Flex. DC Flex did, however, decrease the use of other housing support programs, by 29 percentage points. This decrease is mostly because fewer families in DC Flex accessed rapid rehousing, a widely-used program that provides short- to medium-term rental assistance and supports to families experiencing homelessness. ³

The Urban Institute found that 94% of families participating in DC Flex surveyed were satisfied with the program. Many participants also reported that the program had led to positive changes in their finances. Together, these findings suggest that DC Flex is a viable program model and may be a good fit for many low-income families’ housing needs.

What comes next?
We are working on an analysis of the second year of DC Flex, which overlaps with the first 6 months of the COVID-19 pandemic. The effects of the pandemic will make it harder to measure the impact of DC Flex over multiple years, but we expect to report our findings in 2024. Our results will inform how financial flexibility is incorporated in future housing programs in the District. In 2023, we also began testing the DC Flex subsidy in single adults with no children.

What happened behind the scenes?
The Lab rarely has the staff or funding to gather detailed qualitative information on programs. For DC Flex, we have been fortunate to partner with the Urban Institute to conduct interviews, surveys, and focus groups to help policymakers understand how the program is working and participants’ opinions of DC Flex. This research was made possible through a grant from the U.S. Department of Housing and Urban Development.

¹ DC Department of Human Services (2021).
² Hannagan and Morduch 2015, Farrell and Greig 2015.
³ DC Interagency Council on Homelessness, p.10