Building the Foundation: A Blueprint for Creating Affordable Housing for DC’s Lowest-Income Residents

For the District’s extremely low-income residents, affordable housing can serve as a foundation for overall stability and well-being. Yet the District’s recent efforts to create and preserve affordable homes, while important, have not substantially expanded the availability of housing affordable to city’s lowest-income residents. By examining the subsidy cost of addressing severe rent burden among DC’s extremely low-income residents, this report contextualizes the District’s present investments and outlines a ten-year blueprint to create 30,000 units of deeply affordable housing.

The full subsidy cost to the District would be $2.6 billion in construction subsidies over ten years, with $732 million per year required for ongoing operating assistance once all affordable units have been created and occupied. This cost is on par with what the District spends on other key city services, such as public education and police and emergency services.[1] The magnitude of the required investment suggests that the District would need to pursue forward-thinking strategies to maximize its subsidy dollars, such as the aggressive utilization of public land, and consider progressively raising revenue to dedicate to deeply affordable housing.

DC’s Extremely Low-Income Residents Face Severe Housing Hardship

Affordable housing can be a foundation for overall stability and well-being for the District’s lowest-income residents. Research shows that having a safe and affordable place to live is essential to health outcomes and achievement in school.[2] Yet few of the District’s extremely low-income renter households (those with incomes below 30 percent of the area median, or $33,000 for a family of four) have access to a home they can afford. In 2016, the most recent year for which data is available, 63 percent of DC’s extremely low-income renters spent at least half of their income on housing.[3] These 27,000 households[4] are considered “severely housing cost burdened” by the US Department of Housing and Community Development and classified as a “worst case housing need.”[5]

The lack of deeply affordable housing impacts nearly every aspect of life for District households working to make ends meet on extremely low incomes. This suggests that the District’s slow progress expanding access to deeply affordable housing is holding the city back in tackling other pressing challenges, such as unequal student outcomes, racial inequity, and homelessness.

One in five children in the District live in a household that is both is extremely low-income and lacks an affordable home,[6] which research shows impedes early childhood development and achievement in school later on.[7] DC’s lack of affordable housing disproportionately falls on residents and communities of color. Of extremely low-income, severely rent burdened households, 88 percent are headed by a person of color.[8] This matches and likely exacerbates the deep racial inequity in DC on nearly every other measure of well-being, from health to financial stability. Finally, extremely low-income families paying unaffordable rents face a high risk of becoming homeless or experiencing other types of housing instability,[9] potentially contributing to the increased pressure on DC’s homeless services system in recent years.

DC’s Slow Progress Expanding Access to Deeply Affordable Housing

There is a clear connection between expanding access to deeply affordable housing and other key community priorities, such as reducing homelessness and addressing the barriers students face in school. However, the District’s recent efforts to create and preserve affordable homes, while important, have not substantially expanded extremely low-income residents’ access to affordable housing.

Despite large new investments in housing programs, the number of deeply affordable units created represents only a small share of the total amount of affordable housing needed for the city’s lowest-income residents. The District has completed or planned fewer than 3,000 rental units for extremely low-income residents since 2015,[10] the year when the city began its substantial and regular investments in affordable housing through the Housing Production Trust Fund, and the year after which data on DC’s affordable housing production is most reliable. Nearly half of those units were targeted to families and individuals already in the city’s homeless services system, through Targeted Affordable Housing and Permanent Supportive Housing.[11]

Notably, the number of severely rent burdened, extremely low-income households certainly underestimates the overall need for affordable housing among DC’s lowest-income residents. The 27,000 figure does not include families and individuals who are currently homeless, or those who are “doubled-up,” sharing housing with another household because they cannot afford rent on their own.[12]

One reason the District’s recent housing investments have not made much of a dent in the housing need among the city’s lowest-income residents is that those resources have not been well targeted to the most severe housing needs. This is clear from the District’s affordable housing pipeline: while three-quarters of renters facing severe housing cost burden in DC are extremely low-income,[13] less than a third of new rental units produced by DC’s housing programs—including vouchers, construction subsidies, real estate development projects, and inclusionary zoning—have been affordable to this group (Figure 1).[14] Most units have targeted households with higher income levels, largely 50 percent or 60 percent of the area median income. Looking only at vouchers and projects that received assistance from DC-controlled subsidy sources, new affordable rental production is only marginally more aligned with the landscape of severe rent burden.

Figure 1.

These outcomes reflect the fact that DC’s housing subsidy programs are tasked with multiple objectives, from production to preservation, rental to homeownership, deeply affordable supportive housing to middle-income housing. In an environment where subsidy resources are constrained and the amount required to meet even one type of need is large, these various goals compete for resources. To date, the District has had no clear means of determining which goal to prioritize, either in which housing subsidy tools it funds, or how it awards those funds to projects.  As a result, funding availability and type often determine which housing needs are addressed, rather than the other way around—prioritizing housing needs, and then matching funding to the need. For instance, the District’s largest housing tool, the Housing Production Trust Fund, has continually failed to comply with statutory rules dictating which income levels its resources must serve, due in part to a lack of funding for operating assistance to link to Trust Fund projects.

Another reason why the District’s housing subsidy resources should be prioritized for the lowest-income residents is the fact that higher-income groups (50 to 60 percent of the area median income) are relatively well-served by the District’s non-subsidy housing tools, such as real estate development and inclusionary zoning, given the level of severe housing cost burden at those income levels. This means that the District is using its subsidy resources such as the Trust Fund to serve groups that could be served with non-subsidy housing tools, whereas deeply affordable housing nearly always requires subsidy.

Housing Subsidy Tools to Create Deeply Affordable Rental Units

There are two ways to create rental housing affordable to extremely low-income residents:  providing tenant vouchers to help residents pay rent in a private market apartment, and building new subsidized affordable buildings.[15]  Providing vouchers and constructing new subsidized buildings each have strengths and limitations; some are highlighted in Table 1.

Tenant Vouchers

The typical source of tenant-based vouchers in the District is the Local Rent Supplement Program (LRSP), a locally funded program based on the federal Housing Choice Voucher Program. Chronic federal underfunding of voucher programs led the District to create the LRSP in 2007. LRSP has two main components: tenant-based vouchers, and project-based rental assistance (discussed in next section). Tenant-based vouchers work by paying the difference between what an extremely-low income household can afford (typically 30 percent of income), and the fair market rent for a private-market unit. A unit’s fair market rent is designed to match its market price, and depends on its size and the neighborhood where it is located.[16] Tenant-based LRSP, including Targeted Affordable Housing, now assists approximately 2,700 households, with an annual funding level of $45.1 million.[17] The District has added new funding for tenant-based LRSP irregularly since the program began (Figure 2).

Figure 2.
New Subsidized Housing Construction

New affordable housing construction in DC is financed with a mix of sources: locally-funded “gap financing” through the Housing Production Trust Fund, along with housing tax credits, tax-exempt bonds, and private debt and equity. The Trust Fund is DC’s largest subsidized affordable housing tool and plays an outsized role in affordable housing development in the city. The Trust Fund provides grants and low-cost loans to help pay for acquisition, construction, or renovation of affordable buildings. That one-time, up-front financing assistance is critical to closing the gap between what it costs to build an affordable building in the District, and the funds available from federal and private sources and housing tax credits.[18] Since 2015, the Housing Production Trust Fund has been funded at $100 million annually. In prior years, the Trust Fund’s resources were smaller and more volatile, relying primarily on the portion of DC’s deed recordation and transfer tax revenue that is dedicated to the Trust Fund each year.[19]

Affordable housing has ongoing operating costs as well, such as utilities and maintenance, that one-time financing assistance from the Trust Fund does not pay for, and which rents affordable to extremely low-income renters typically cannot cover. Operating costs can range from $6,100 to $7,100 per unit per year;[20] but extremely low-income households facing severe rent burden can pay only about $5,000 per year in rent on average.[21] Another cost that must be covered are contributions to the building’s replacement reserve—typically a minimum of $300 per unit per year—which is used to pay for eventual replacement of major building components such as roofs and HVAC systems.[22] Finally, a project must also generate income to pay down principal and interest on its construction loans. Lenders require projects to have a healthy margin between what the project expects to take in in income, and what it will pay in debt service.[23]

The Local Rent Supplement Program’s second component, project-based LRSP, fills the gap by providing ongoing operating assistance to affordable buildings—typically those also assisted by the Trust Fund. Nearly every recently produced Trust Fund unit affordable to an extremely low-income household was paired with project-based LRSP.[24] Without some amount of operating assistance, it is unlikely that affordable projects including extremely low-income units would have “penciled out.”

Operating assistance payment amounts through the Local Rent Supplement Program are based on the same fair market rent levels as tenant-based LRSP vouchers. Reducing project-based LRSP payment standards from market rents to a lower level could reduce the long-term ongoing cost to the District of a deeply affordable unit. However, that would increase the up-front gap financing cost to the District through the Housing Production Trust Fund. Providing that additional assistance to make up for lower operating income might in turn require the Trust Fund to pay for a greater portion of a unit’s total cost (currently capped at 49 percent).

While the District is now making an annual investment of $100 million per year in the Housing Production Trust Fund, this falls far short of the true need for housing subsidies, and these dollars are not currently well targeted to extremely low-income residents. The District has failed even to meet the statutory requirement[25] that 40 percent of Trust Fund resources obligated each year support housing for extremely low-income households. The District recently strengthened its funding award criteria to prioritize Trust Fund dollars for housing for low- and extremely low-income residents,[26] but the city has not reversed its poor track record. In the five years between 2015 and 2019, the District will have met the Trust Fund’s extremely low-income targeting requirement only once, according to city projections (Figure 3).[27] Neither the predominance of preservation projects nor legacy projects in the pipeline fully explains this shortcoming.[28]

Figure 3.

Because deeply affordable units typically require operating subsidies in addition to up-front financing assistance, as outlined above, the city’s underfunding of project-based LRSP has impeded its ability to improve the targeting of Trust Fund resources. The District has funded new project-based LRSP subsidies only irregularly in recent years (Figure 2), and has not provided the amount needed to match with its annual $100 million investments in the Trust Fund.

Yet at the Trust Fund’s current $100 million funding level, directing even 40 percent of resources to deeply affordable housing as required would create only around 220 units of such housing each year—only a small fraction of the full need.

Table 1.

Estimating the Required Investment in Deeply Affordable Housing

What will the District need to spend to truly address the affordable housing needs of its lowest-income residents? This analysis outlines a blueprint in which the District creates 30,000 subsidized affordable rental units for extremely low-income DC households over ten years. This would address the needs of the 27,000 current severely rent burdened households, plus a conservative estimate of growth in the number of extremely low-households over the decade.[29]

For the purpose of estimating the cost of the blueprint, we assume that the city would provide one-half of the needed affordable housing for extremely low-income residents with tenant vouchers, and create the other one-half by constructing new affordable buildings. Cost estimates using several other ratios are available in the Appendix.

Determining the ideal mix of construction and vouchers would require additional research outside the scope of this paper, such as an analysis of the housing market’s ability to absorb vouchers, identifying services and programs that would increase voucher holders’ success in the market, a capacity analysis of the city’s developable land, and of the capacity of affordable housing developers and of private construction financing partners. Ultimately, the appropriate mix would depend largely on whether other policy decisions that would enhance each tool are adopted.

We assume the annual cost of a tenant-based voucher is $19,300 per year. This is the average rent subsidy cost of a Targeted Affordable Housing (TAH) slot for families, and the cost of a TAH slot for individuals.[30] Because this estimate does not attempt to account for the difference in costs by unit size, we use aggregate or average per-unit cost estimates throughout.

The estimated cost to construct a new unit of housing affordable at extremely low-income levels includes two components: gap financing assistance (low-cost loans) for one-time capital costs such as acquisition and construction, and operating assistance to cover ongoing costs such as maintenance and utilities.

Many factors influence the cost of constructing a new unit of affordable housing, including unit size, building type, land costs, and site characteristics. Because there are typically fewer than ten new construction projects in DC’s subsidized affordable housing pipeline that close in any given fiscal year, the composition of those projects has a large impact on the average cost per unit for that year. Because of that variability, all per-unit estimates of new construction projects in this analysis are based on three fiscal years of projects, those closed or projected to close in fiscal years 2017 to 2019.[31]

The average total cost to build a unit in a subsidized affordable building in the District is $377,600, but the District does not pay a project’s entire development cost alone: DC funding sources such as the Housing Production Trust Fund are designed to be “gap financing” tools, meaning that projects must line up other funding sources to pay for at least 50 percent of the total cost. On average, DC gap financing covered 40 percent of the total development cost of recent projects, yielding a DC subsidy cost of $152,100 per unit. We assume that the District’s contribution toward the total development cost would remain constant at 40 percent, which is akin to assuming that other financing sources would also scale up as production expands.[32]

As described earlier, project-based operating assistance payment amounts through the Local Rent Supplement Program are based on fair market rents, as are tenant-based vouchers, so we assume that the annual cost of a project-based subsidy is the same as a tenant voucher, $19,300. We also assume that every unit will require a project-based operating subsidy. We note that there are options the District could explore to reduce the per-unit cost of operating assistance, but that would require a corresponding increase in up-front financing assistance.

For the purpose of this estimate, we assume that per-unit subsidy costs—both construction and operating assistance and tenant vouchers—will increase by 2.4 percent annually. This is average year-over-year inflation (CPI-U) forecasted for the next decade by the Congressional Budget Office.[33] We note that per-unit subsidy costs are influenced by a variety of factors that may change differently or increase faster than consumer inflation.[34]

Subsidy Cost of Addressing the Housing Needs of DC’s Extremely Low-Income Residents

Estimating the level of local spending required to create 30,000 deeply affordable units over a decade is intended to contextualize the District’s present investments and to map out a ten-year blueprint to address severe rent burden among DC’s lowest-income residents. As outlined earlier, the District’s present level of investment in subsidized housing for extremely low-income residents falls far short of providing the amount of affordable housing needed for that income group. This is evidenced by the tens of thousands of extremely low-income households currently facing severe rent burden.

Meeting the full housing needs of DC’s extremely low-income residents over a decade would require a public investment of $2.6 billion in construction subsidies over ten years, and $732 million per year in ongoing operating assistance once all affordable units have been created (Table 2). Because operating assistance is needed only once a unit has been created, and is needed for every year after that, the amount of operating assistance for newly built units and vouchers would increase over time, while the annual construction subsidy required would remain the same except for inflation-related cost increases (Figure 4).[35]

Again, this estimate is based on a plan in which the city provides one-half of the needed affordable housing with tenant vouchers, and creates the other one-half by constructing new affordable buildings. Cost estimates using several other ratios are available in the Appendix.

Figure 4.


Table 2.

Challenges and Potential Solutions

Executing the blueprint to address severe rent burden among DC’s extremely low-income residents would require an unprecedented commitment of public funds far above what DC currently spends on affordable housing—and indeed, what it spends on other core city services. Furthermore, producing such a substantial number of new affordable housing units, whether vouchers that would need to be absorbed by the private housing market, or built units that would need to be constructed on available land, comes with large implementation and practical challenges.

Voucher Absorption

Extremely low-income households with vouchers face a variety of barriers in the rental market and are often at a competitive disadvantage with other prospective tenants in the District’s rental market. While it is illegal in the District for landlords to refuse to accept vouchers, this type of discrimination is not uncommon.[36] Moreover, there are multiple other grounds on which voucher holders are often turned away from apartments, such as low or no credit scores, arrest or conviction records, or past evictions and rent arrears.

Improving voucher holders’ ability to quickly secure a lease in the unit of their choice would be critical to the success of a substantial voucher expansion. This would likely involve a mix of approaches, including regulatory changes restricting landlords’ ability to turn away those with criminal records or past evictions, stronger enforcement of existing laws against source-of-income discrimination, and creating new services and programs to help voucher holders succeed in the market, such as housing search assistance, credit repair programs, and mobility counselling.

Land Costs & Availability of Developable Parcels

The high cost and limited availability of developable land in the District hinders the production of affordable housing and contributes to high per-unit subsidy costs. In a 2014 survey of affordable housing developers, 91 percent of respondents said that high acquisition costs limit production significantly.[37] No other factor, including insufficient public subsidies, was cited more often. Affordable housing developers are often at a disadvantage with market-rate developers, who can usually both pay more for the land, and close on the land more quickly because they do not need to apply for a public subsidy.[38]

Another consequence of high land costs is that developers must often seek economies of scale by building more densely on the parcel.[39] However, multi-family development is substantially curtailed in many parts of the city by zoning restrictions.[40] Moreover, denser development increasingly faces neighborhood opposition, particularly when the development includes affordable or supportive housing.[41]

The primacy of land as both a constraint and a cost driver suggests that setting aside and opening up land for affordable housing production should be at the center of DC’s affordable housing strategy. District-owned land presents a unique opportunity to aggressively utilize public assets for affordable housing production. The District mandates that 20 to 30 percent of housing units in DC-owned parcels sold for residential development be set aside as affordable housing in perpetuity.[42]

Leveraging public lands this way has multiple benefits. First, the District can often use the value of the land to underwrite the cost of below-market rate housing, substantially decreasing or even eliminating the need for other subsidies.[43] Second, the fact that the city already has control of the land means that the District can make strategic locational choices that would not otherwise be possible. For instance, the District could use city-owned parcels to maximize affordable housing in neighborhoods where purchasing land on the open market would be prohibitively expensive and where little affordable housing currently exists.

Many of DC’s largest publicly-owned parcels have already been awarded to developers, often with a relatively conservative amount of affordable housing in the development plan. The District should aggressively leverage its remaining supply of public land, together with additional density and subsidy dollars, to create deeply affordable housing far beyond the minimum percentage currently required.

The District could also explore building the stock of public land by acquiring land in neighborhoods that are relatively low-cost but close to gentrifying areas, either on the open market or through eminent domain. DC could hold the land itself as a “land bank,” or turn it over to a community land trust that would keep the housing permanently affordable under community governance.

Limited Subsidy Dollars

Dedicating $2.6 billion in construction subsidies over ten years and maintaining annual operating subsidies of $726 million would certainly require a dramatic rethinking of the city’s budget, and likely its revenue system as well. The operating assistance alone at full buildout would be equivalent to the fifth largest DC agency in terms of local fund budget.[44] The investments outlined under the blueprint represent six times the amount the District currently budgets for gap financing for extremely low-income housing, and nine times what DC budgets for ongoing operating assistance and vouchers (Table 3).

This points to the need for the District to pursue forward-thinking strategies to reduce costs, such as the aggressive use of public land. A wide variety of regulatory and industry solutions, such as “green-tape” permitting and innovations in construction techniques, are beyond the scope of this DC housing subsidy-focused paper, but have been explored and studied elsewhere.[45] Yet ultimately, the District would likely need to consider raising additional revenues to achieve the full investment.

Table 3.

Administration & Asset Management

Overseeing such a large number of new subsidized units would require a substantial expansion of District housing agencies. For instance, agencies would need additional staff to select and underwrite affordable projects, to enforce affordability rules and track loan repayments, and to manage voucher and operating assistance payments. The DC Housing Authority would be responsible for nearly twice as many tenant-based vouchers as it currently administers.[46] Recent audits of the city’s Housing Production Trust Fund suggest that additional asset management capacity at the Department of Housing and Community Development is needed to oversee the current portfolio Trust Fund-assisted projects,[47] meaning that a larger portfolio would require even more extensive efforts to strengthen agency capacity. Finally, the new affordable projects built under the blueprint would, like all buildings, age over the years and eventually need to be rehabilitated or refinanced. The District would need to be prepared to commit additional subsidies in the future to keep these buildings in good physical condition, financially stable, and continually affordable.


Table A1.










Table A2.











Table A3.














Table A4.















[1] In FY 2018, the District’s approved General Fund operating budget for public education (DC Public Schools and DC Public Charter Schools) was $1.6 billion, and its budget for police and emergency services (Metropolitan Police Department, Fire and Emergency Medical Services Department, and Police Officers’ and Firefighter’s Retirement) was $869 million.
[2] For a review of the literature, see:
Zippel, C. (2016). Broken Foundation: Affordable Housing Crisis Threatens DC’s Lowest-Income Residents. DC Fiscal Policy Institute.
[3] DCFPI analysis of 2016 American Community Survey 5-year Public Use Microdata Sample. Throughout this paper, certain households headed by college students are deliberately excluded from this analysis, as are households with zero or negative income.
College student-headed households are excluded because of the likelihood that such households may appear severely rent burdened because their rent is paid by a family member outside the household, and because such households are unlikely to have a sustained, long-term need for subsidized housing, so are not comparable to other extremely low-income households. These households are defined as those headed by college or graduate student aged 35 or younger, without children, who within the past year worked fewer than 35 hours in a typical week, or worked fewer than 13 weeks. Such households represent approximately 6 percent of extremely low-income renter households.
Households reporting zero or negative income are excluded due to the difficulty of determining whether such households appear extremely low-income due to temporary business or accounting losses, or truly have no income. For instance, a household whose income is entirely composed of capital gains would appear to have zero income in the ACS, because the ACS does not ask about capital gains income. Also, households that report zero or negative income but report paying rent may be mistakenly reporting rent paid on their behalf by someone else.
While this paper focuses on the severely rent burdened households (dedicating half or more of income to rent), this is not to say that extremely low-income residents who are moderately rent burdened (dedicating 30 percent to 50 percent of income to rent) do not also experience hardship or trouble affording basic necessities due to lack of affordable housing.
[4] DCFPI analysis of analysis of 2016 American Community Survey 5-year PUMS. This figure is not significantly different at a 90 percent confidence level than the figure in our 2016 report, Broken Foundation: Affordable Housing Crisis Threatens DC’s Lowest-Income Residents, which used 2014 American Community Survey 5-year PUMS.
[5] Watson, N.E., Steffen, B. L., Martin, M., & Vandenbroucke, D. A. (2017). Worst Case Housing Needs: 2017 Report to Congress. Washington, DC: US Department of Housing and Urban Development.
[6] 23 percent. DCFPI analysis of 2016 American Community Survey 5-year PUMS.
[7] For a review of the literature, see:
Zippel, C. (2016). Broken Foundation: Affordable Housing Crisis Threatens DC’s Lowest-Income Residents. DC Fiscal Policy Institute.
[8] Persons other than non-Hispanic whites. DCFPI analysis of 2016 American Community Survey 5-year PUMS.
[9] For a review of the literature, see:
Zippel, C. (2016). Broken Foundation: Affordable Housing Crisis Threatens DC’s Lowest-Income Residents. DC Fiscal Policy Institute.
[10] The main dataset used for this analysis is the Affordable Housing dataset available through Open Data DC, which shows projects completed, under construction, or in the pipeline 2015 through March 2018. This dataset is featured in the Deputy Mayor for Planning and Economic Development’s Economic Intelligence Dashboard. Because the Affordable Housing dataset does not include fields for project type or tenure (new vs. preserved, rental vs. homeownership), projects were matched to two additional datasets: the Department of Housing and Community Development’s Development Finance Division Pipeline Dashboard, and the Inclusionary Zoning Dashboard. Remaining missing data on type and tenure was filled using secondary sources, including agency press releases, news reports, and developer websites. The number of replacement public housing units for the extremely low-income level was found in materials available from the New Communities Initiative and DC Housing Authority, and subtracted from a project’s total new extremely low-income rental units, to avoid counting these “preserved” units as “new”. Finally, the number of Local Rent Supplement Program vouchers funded in the FY 2015-2018 Budget & Financial Plans was added to the total count of new extremely low-income units.
Projects subsidized by DC-controlled funding sources (as in Figure 2) were defined as those with the Department of Housing and Community Development listed as the related agency, either solely or jointly with other agencies. Projects associated with independent housing agencies (DC Housing Finance Agency or DC Housing Authority) not in partnership with a DC executive agency were not included.
[11] The number of Permanent Supportive Housing Units in each project was obtained by cross-referencing the Affordable Housing dataset with the Department of Housing and Community Development’s Development Finance Division Pipeline Dashboard (see note 10), along with agency press releases and funding award announcements.
[12] The data source for this figure, the American Community Survey (ACS), includes two types of observations: housing units, and people living “group quarters” such as homeless shelters, nursing facilities, correctional facilities, and college dormitories. This analysis excludes observations in group quarters due to the unreliability of those estimates for DC due to small sample size, so homeless residents living in shelters or other short-term, group housing situations are excluded. People experiencing homelessness who are unsheltered are not surveyed by the ACS, which surveys only fixed addresses, so this analysis excludes those residents as well.
This figure also does not capture “doubled up” households. The ACS defines a household as a housing unit, and includes all people “living or staying” at an address as part of the household—including those who are living there temporarily, if they do not have another place to stay. Thus two families living together because each could not afford housing on their own, or a single person staying temporarily with friends or relatives, would appear as one household.
While the ACS includes both household-level and person-level data that theoretically could enable splitting doubled-up households into two housing units, doing so would require making assumptions about what constitutes an unsustainable versus a sustainable shared housing situation. That definition has dimensions beyond what is reflected in data—such as the presence of beneficial mutual economic and social support, inter-generational and extended family ties, or inter-personal tension and conflict over shared resources. Moreover, because the ACS does not include data on each household member’s contribution to the rent, it would be impossible to determine whether one or both sub-households would be severely rent burdened if the household was split. Finally, while we explored identifying unsustainably doubled-up households as those who are overcrowded (1.5 or more people per room), that figure was 1,000 or just 2 percent of extremely low-income renter households, and we were concerned about introducing error into the total in order to accommodate a relatively small subset of households.
[13] 75 percent. DCFPI analysis of 2016 American Community Survey 5-year PUMS.
[14] See note 10.
[15] This paper focuses solely on rental housing as widely the best affordable housing solution for those with extremely low incomes, because homeownership at this income level is usually only possible under limited circumstances; for instance, through subsidy-assisted limited-equity coops.
[16] For LRSP and HCVP fair market rents by neighborhood and unit size, see the DC Housing Authority website.
[17] Fiscal Year 2018 Budget & Financial Plan; FY 2016-17 Performance Oversight pre-hearing responses submitted by the DC Housing Authority to the DC Council Committee on Housing and Neighborhood Revitalization, Feb. 2018. This figure differs from the tenant-based LRSP line item in the DCHA budget book, which includes Permanent Supportive Housing rental assistance budgeted through DCHA. We exclude those funds (information provided by the Council Budget Office and Department of Human Services) from the tenant-based LRSP total.
The funding total includes tenant-based LRSP and TAH vouchers that are funded but not yet leased (households have six months to lease an apartment from when the receive the voucher). The 2,700 figure is all units leased as of January 2018.
[18] Housing Production Trust Fund loans are paid back over time according to the loan agreement. Typically, loan agreements specify that Trust Fund dollars will be repaid from a portion of the project’s cash flow. Repayments may be deferred for a number of years.
[19] See the annual analyses of the city’s budget for affordable housing published as part of DCFPI’s Budget Toolkit.
[20] Department of Housing and Community Development, Consolidated Request for Proposals, Fall 2017.
Operating costs include, for instance, utilities, maintenance and repairs, administrative and leasing expenses, security and cleaning services, and taxes and insurance. For a full list of the costs included in the figure, see Form 202 – Application for Financing, Fall 2017.
[21] DCFPI analysis of 2016 American Community Survey 5-year PUMS.
[22] Ibid.
[23] Affordable housing projects typically have debt coverage ratios (net operating income divided by debt service payments) of 1.1 to 1.2.
[24] DCFPI analysis of the Department of Housing and Community Development’s Development Finance Division Pipeline Dashboard and DC Housing Authority Local Rent Supplement Program contracts and pipeline as reported to the DC Council.
[25] DC Code § 42–280(b-1)2.
[26] See Department of Housing and Community Development, Consolidated Requests for Proposals, 2015 through Fall 2017.
[27] FY 2016-17 Performance Oversight pre-hearing responses submitted by the Department of Housing and Community Development to the DC Council Committee on Housing and Neighborhood Revitalization, Feb. 2018.
[28] See DCFPI testimony at the DC Council Committee on Housing and Neighborhood Revitalization performance oversight hearing for the Department of Housing and Community Development, Mar. 2018.
[29] The number of extremely low-income households in need of affordable housing is assumed to grow by 1.1 percent annually. This is a linear projection of the number of households in that income group following the annualized percent increase between 2007-2011 and 2012-2016 (the most recent year for which data is available), times the rate of severe rent burden in 2016. We do not assume the rate of severe rent burden will change over the ten-year period, as that would require making other assumptions. Holding the rate of severe rent burden stable is equivalent to assuming that rents will not grow faster than income for extremely low-income households—which makes this assumption very conservative. Because the extremely low-income level is defined as 30 percent of the area median income, this assumption also assumes that area median income and the distribution of DC households around the area median income will follow past trends as well.
For a step-by-step, see Table A1 in the Appendix.
[30] Interagency Council on Homelessness, Strategic Planning Committee, Jan. 2018.
[31] DCFPI analysis of the Department of Housing and Community Development’s Development Finance Division Pipeline Dashboard as of Feb. 2018. Missing data on total development cost was filled using information from the project’s loan agreement resolution submitted to the DC Council.
[32] DC’s affordable housing pipeline grew soon after larger, annually maintained investments in the Housing Production Trust Fund, but given the unprecedented scale of the subsidy investments modelled here, it is possible that the supply of private financing for affordable housing would not expand as quickly, meaning the District would need to contribute a larger share of total project costs.
[33] Congressional Budget Office, 10-Year Economic Projections, An Update to the Budget and Economic Outlook: 2017 to 2027, June 2017.
Per-unit costs for each year of the ten-year blueprint are shown in Table A2 in the Appendix.
[34] For instance, construction costs are influenced by the cost of materials and labor. Project costs are also influenced by the regulatory environment such as permitting fees and green building requirements. Land costs are impacted by many factors influencing real estate development including market rents, investor behavior, zoning restrictions, and the cost of construction. Ongoing subsidy costs are linked to market rents for both tenant-based vouchers and project-based rental assistance in the Local Rent Supplement Program.
[35] New construction projects can take several years from when gap financing agreements close to when construction is finished; therefore, in year ten of the blueprint, not all 30,000 units not all units would have been completed and occupied. While operating assistance payments cannot begin until a unit is occupied by a qualified tenant, we assume operating assistance would be funded starting in the same year as the gap financing, as housing agencies cannot typically commit funds that have not been appropriated (and it is unclear whether other financing partners would lend to a project based on funds not present in the budget).
[36] The Equal Rights Center. (2013). Will You Take My Voucher?: An Update on Housing Choice Voucher Discrimination in the District of Columbia.
[37] Tatian, P.,  Leopold, J., Oo, E. Joseph, G., MacDonald, G., Nichols, A., Woluchem, M. & Zhang, S. (2015). Affordable Housing Needs Assessment for the District of Columbia: Phase II. Washington, DC: Urban Institute.
[38] Ibid.
[39] Ibid.
[40] DC Office of Planning District Wide Summary Zoning Map, as of Jan. 2018.
[41] For examples, see:
Giambrone, A. (2017, Jul. 27). How an Unofficial Park Became the Crucible in D.C.’s Long-Stalled Public Housing Reboot. Washington City Paper.
Giambrone, A. (2017, Sept. 27). Ward 3 Neighbors Fight Planned D.C. Homeless Shelter With New Zoning Appeal. Washington City Paper.
[42] DC Code § 10–801b(3).
[43] By selling or leasing the land at a below-market or discounted price, the District reduces the total development cost. Those cost savings are used to offset the cost of providing below-market rate housing.
[44] Approved FY 2018 General Fund budget by agency, Office of the Chief Financial Officer via CFOInfo.
[45] See for instance:
University of Minnesota Center for Urban and Regional Affairs, Housing Justice Center, & Becker Consulting. (2015). Best Practices to Reduce the Cost of Affordable Housing.
[46] The DC Housing Authority currently administers approximately 12,00 federally-funded vouchers and 2,700 Local Rent Supplement Program vouchers, according information from the Department of Housing and Urban Development’s Housing Voucher Program Support Division and the DC Housing Authority.
[47] Office of the DC Auditor. (2017). DHCD Should Improve Management of the Housing Production Trust Fund to Better Meet Affordable Housing Goals.
Office of the DC Auditor. (2018). Stronger Management of the Housing Production Trust Fund Could Build More Affordable Housing.

Latest Publications