DC Flexible Rent Subsidy Program: Findings From the Program’s First Year
Most low-income families can qualify for housing choice vouchers (HCVs); however, although the HCV program imposes no limits on the duration of assistance, eligible households often remain on a waiting list for years before receiving assistance. Rapid rehousing (RRH) programs, on the other hand, offer short-term rental subsidies to households experiencing homelessness, but studies examining whether RRH improves participants’ long-term housing stability show mixed results. Shallow subsidies offer a middle ground between these traditional programs in both the length and amount of housing assistance. In 2017, the District of Columbia’s Department of Human Services (DHS), the DC Interagency Council on Homelessness (ICH), and District of Columbia Department of Housing and Community Development (DHCD) implemented a 4-year pilot for a shallow subsidy program called the DC Flexible Rent Subsidy Program (DC Flex). In October 2020, HUD published a study ― DC Flexible Rent Subsidy Program: Findings from the Program’s First Year ― to track the experiences of participants and staff during the program’s initial year of implementation. The Urban Institute and The Lab @ DC, an applied scientific research team in the Office of the City Administrator, collaborated to examine the impact of this new housing subsidy on homelessness and housing stability in Washington, DC.
With a $5 million appropriation from the District of Columbia Council, Mayor Muriel Bowser signed legislation in 2016 to fund DC Flex to assist a maximum of 125 households. Households began applying for DC Flex from December 2017 through July 2018. To encourage residents to apply, DHS sent letters by mail, text, and email to households that had previously applied for housing assistance. Community-based organizations also referred their clients to the program. The program limits participation to extremely low-income households — those earning up to 30 percent of the area median income — that have children and at least one employed adult. Households must also have recently applied for or received emergency or temporary housing assistance and be living in a legal rental unit in the city with a lease in their name.
DC Flex participants receive access to a checking account and an escrow account that holds the full $7,200 subsidy balance. Participants can use the subsidy only to pay rent, but the flexibility of the program allows families to decide how much to allocate. Each month, program administrator Capital Area Asset Builders, a District financial education nonprofit, transfers funds from the escrow account to the checking account to restore the balance to the total amount of the participant’s rent. This cycle repeats until the escrow account is depleted. Unlike HCVs, the subsidy does not change based on income or household size. Much like a cash reserve, DC Flex enables low-income families to increase their savings to have a financial cushion if their employment or income is disrupted. As long as they continue to meet the eligibility criteria, participants can remain in the program for the duration of the pilot.
The Lab @ DC conducted a quantitative analysis to study three outcomes of interest: the rate at which participants experienced homelessness, the rate at which participants used homelessness services along with emergency shelter and transitional housing, and the impact of DC Flex on participants’ Temporary Assistance for Needy Families (TANF) program benefits. In collaboration with District agencies, The Lab @ DC helped launch the program as a randomized controlled trial (RCT). In total, 3,692 District households applied; however, a number of applications were screened out because they were incomplete, duplicates, or did not meet the eligibility criteria, leaving 719 applicants. DHS and The Lab @ DC conducted five lotteries among eligible applicants between January 2018 and July 2018. A total of 439 applicants were not chosen for the fifth lottery; these became the control group, or “business as usual group,” and could access traditional housing assistance programs instead. The research team used a blocked randomization method to ensure that households in each lottery had characteristics similar to those of the control group. A total of 51 applicants were excluded from the analysis because they did not remain in the program for the full 12 months. Although the lotteries assigned 229 households to the DC Flex treatment group, an additional eligibility verification step deemed some households ineligible, and others decided not to enroll, thus leaving 102 families to officially enroll and begin receiving subsidies in September 2018. By October 2018, an additional 18 households enrolled, bringing the program to a total of 120 filled slots, but only the 102 households that had already begun receiving subsidies are included in the quantitative evaluation. The Urban Institute led the qualitative evaluation of the program, which included two participant focus groups, interviews with participants and staff, observation of meetings, and a survey of program participants. The sample sizes for the quantitative and qualitative evaluations are slightly different, so the responses to the survey do not necessarily convey the sentiments of all participants.
The RCT sought to demonstrate the impact of DC Flex on participants’ housing stability and use of homeless services compared with the control group. Out of the 102 families receiving DC Flex subsidies, 1.8 percent of them experienced homelessness while enrolled. Researchers also measured use of Continuum of Care (CoC) services such as emergency shelter, transitional housing, RRH, and assessments for short-term housing. About 22 percent of the 102 participants used at least one CoC service, but without DC Flex, researchers estimated that the share of participants needing CoC would have increased by approximately 28 percentage points. In addition, the Urban Institute determined that 30 percent of participants in the control group received RRH during the first year compared with only 9 percent of DC Flex enrollees. The researchers also found that DC Flex had no statistically significant effect on the rate at which participants accessed and used TANF program benefits.
With a 61 percent response rate to its survey administered midway through the first year, from February to March 2019, the Urban Institute gauged participants’ satisfaction, use of the flexible accounts, and plans for future use. A total of 94 percent of respondents were very satisfied or satisfied with the program, with some citing the opportunity to receive assistance for a maximum of 4 years (27%) or the flexibility to decide how much of the subsidy to use each month (27%). Many participants would prefer to use the funds for expenses other than rent. Most survey respondents (76%) indicated that they had sufficient funds to pay their rent for the month, and 80 percent said that they had a plan for paying their rent after using all of their subsidy. Forty-three percent of respondents indicated that they used the money to cover all or most of their rent, and nearly 30 percent indicated that the amount they used toward rent depended on their income and expenses in each month. A few participants (9%) used the funds only as a last resort.
The Lab @ DC analyzed the bank account history of DC Flex households for the full 12-month period. By the ninth month, slightly more than 40 percent of participants had depleted all their DC Flex dollars, and approximately 59 percent had used all their funds by the end of the year. This finding could indicate that households entering the program owed back rent and needed DC Flex dollars to pay arrears. Other participants may have focused on immediate stabilization rather than long-term planning. The researchers suggest that these trends could shift during the second program year, after participants have paid off debts.
In terms of their plans for the funds, 69 percent of respondents to the Urban Institute survey said that they were very likely or somewhat likely to spend the full $7,200 subsidy before the end of the program year. On average, however, the account data indicate that participants had $543 in unused funds after the first year. DC Flex funds can be rolled over to the following year if the family remains eligible for the program, so some families may have decided to reserve funds to provide a financial cushion for the future. Interviewees and focus group participants indicated that the subsidy amount was not enough for extremely low-income households to afford rent in Washington, DC. Many participants and some program staff from DHS, ICH, and DHCD recommend increasing the subsidy; targeting the program to people with slightly higher incomes; or allowing participants to use the subsidy in neighboring suburbs in Maryland and Virginia, where monthly rents are lower.
The study team determined that the District successfully launched DC Flex through implementation, enrollment, and account management. Although a single year is not long enough to determine the program’s impact on the city’s homelessness trends, researchers at the Urban Institute recommend DC Flex as an alternative to other housing services. Focus groups revealed that the program helped families maintain housing stability, especially after they became ineligible for other subsidy programs or exited RRH. However, more indepth and longitudinal studies will be needed to gauge whether the program is more successful than traditional subsidies at stabilizing families or even ending the use of CoC services. Researchers originally had planned to include employment and income data based on District administrative records that were scheduled to be gathered in March 2020. The team delayed the collection of this information when the COVID-19 pandemic overwhelmed the District with unemployment insurance applications. Future evaluations of the remaining years of the pilot will examine whether DC Flex participants’ economic well-being and housing stability have improved compared with nonparticipants by highlighting indicators such as evictions, rental assistance applications, and employment. Continued evaluations to understand these trends will be necessary to determine the effect of DC Flex on housing stability and its potential as a model for other jurisdictions.