What the policy is
American public housing has a complicated history. The Housing Act of 1937 created the first federal program, authorizing local housing authorities to build and manage subsidized units for low-income families. Post-WWII expansion produced the large high-rise developments that defined the program in the public imagination — Pruitt-Igoe in St. Louis, Cabrini-Green and the Robert Taylor Homes in Chicago — which became notorious for concentrated poverty, physical deterioration, and crime. By the 1990s, many were being demolished under the HOPE VI program and replaced with mixed-income developments.
In 1974, Congress created the Section 8 program — now the Housing Choice Voucher (HCV) program — as a different approach. Rather than building government-owned units, vouchers give low-income households a rental subsidy to use in the private market. The tenant pays 30% of their income toward rent; the voucher covers the difference, up to a fair market rent ceiling set annually by HUD.
Today, federal housing assistance operates through three main channels: approximately 1 million public housing units owned by local housing authorities; roughly 2.3 million Housing Choice Vouchers; and about 1.2 million project-based rental assistance units held under long-term contracts with private owners. Together these programs serve around 5 million low-income households — out of an estimated 20 million eligible. The United States does not guarantee housing assistance; it rations it, primarily through multi-year waiting lists. The median wait for a voucher in a high-cost city exceeds three years; in the tightest markets, waiting lists have been closed for over a decade.
The policy debate has shifted over time. The failures of large-scale concentrated public housing projects led most economists and housing experts toward preferring tenant-based assistance. More recent research — particularly the Moving to Opportunity experiment — has deepened that preference by showing that where low-income families live shapes long-run outcomes in ways that go far beyond housing costs alone.
How it works
Public housing and vouchers work through different economic channels, each with distinct strengths and failure modes.
Public housing is supply-side intervention: government builds, owns, and manages units, renting them at below-market rates to qualifying households — typically those earning below 30–50% of Area Median Income. Rents are set at 30% of household income, so very-low-income families pay very little. The government absorbs capital and operating costs. The core economic problem with large-scale concentrated public housing is that it creates negative peer effects and network externalities — poor job contacts, underperforming schools, and social norms associated with high-poverty environments — that can undermine the economic mobility of residents over the long run.
Vouchers are demand-side: the government gives qualifying households purchasing power to deploy in the private rental market. This preserves residential choice and allows portability — households can move to lower-poverty neighborhoods with better schools and labor market access. Economically, vouchers are generally more cost-effective than new construction, especially in high-cost markets, because they compete for existing housing rather than funding new construction at government cost. The key limitation is supply dependence: vouchers only work if landlords accept them and if enough units exist at or below the fair market rent ceiling. In tight markets with low vacancy rates, many landlords decline vouchers, and fair market rent caps often exclude high-opportunity neighborhoods — confining voucher holders to the same high-poverty areas as public housing residents.
The most important recent evidence comes from the Moving to Opportunity (MTO) experiment — a 1990s randomized trial in which public housing residents were randomly assigned vouchers to move to lower-poverty neighborhoods. Chetty, Hendren, and Katz (2016) found that children who moved before age 13 earned approximately 31% more as adults, were more likely to attend college, and had better health outcomes than comparable children who did not move. Children who moved as teenagers saw negligible effects; adults who moved saw no earnings gains. This age gradient has become one of the most cited findings in housing economics: it is not housing stability alone that matters, but the quality of the neighborhood environment during childhood.
Who wins and who loses
| Group | Effect | Detail |
|---|---|---|
| Very-low-income households on waiting lists | Cost | The majority of eligible households receive no assistance. Multi-year waits are the norm; in high-cost cities, waiting lists stretch over a decade. Access is determined largely by when a family applied, not by need. |
| Current public housing residents | Mixed | Deeply below-market rents provide genuine financial stability. But concentrated poverty environments, deferred maintenance, and limited access to high-opportunity neighborhoods impose quality-of-life costs that partially offset the subsidy. |
| Voucher holders | Benefit | Portable subsidy gives residential choice unavailable in project-based assistance. But tight-market landlord refusals and fair market rent ceilings that exclude high-opportunity areas limit portability in practice. |
| Private landlords | Mixed | Vouchers provide stable, government-backed rental income. But inspection requirements, administrative delays, and the absence of source-of-income protections in most states give landlords reason to prefer unsubsidized tenants. |
| Local housing authorities | Cost | Chronically underfunded capital budgets require deferred maintenance across the public housing stock. HUD's estimated capital needs backlog exceeds $70 billion, and the gap grows each year. |
| Neighboring residents and communities | Mixed | Well-managed, smaller-scale affordable housing has neutral to positive neighborhood effects. Large-scale concentrated poverty developments have documented negative externalities on surrounding property values and public services. |
What the evidence shows
The MTO experiment shows that early childhood neighborhood environment is a powerful determinant of long-run economic outcomes — and that housing policy's most important lever may be enabling children to grow up in higher-opportunity areas.
Arguments for and against
- Directly serves households the market cannot reach: The lowest-income households — earning below 30% of Area Median Income — cannot afford market-rate housing regardless of how much supply is built or how far prices fall. Zoning reform, filtering, and market-rate construction do not reach families earning $15,000–$20,000 per year. Direct subsidy is the only mechanism that reliably provides housing stability for the poorest households.
- Neighborhood effects generate large long-run returns: The MTO evidence shows that housing assistance enabling children to grow up in lower-poverty neighborhoods generates roughly 31% higher adult earnings, increased college attendance, and better health outcomes. The long-run social return on well-targeted vouchers — accounting for reduced public assistance, health care, and criminal justice costs — may exceed their direct fiscal cost.
- Housing stability reduces other public expenditures: Stable housing reduces emergency room visits, school disruption, foster care placements, and incarceration — costly outcomes that fall on public budgets. Evidence from Housing First programs for chronically homeless individuals shows that providing unconditional permanent housing reduces total system costs, sometimes by more than the housing subsidy itself.
- Vouchers convert assistance into economic mobility: Unlike public housing, which ties a family's subsidy to a specific unit in a specific location, portable vouchers allow households to follow job opportunities, access better schools, and move away from high-poverty environments. Properly implemented, voucher programs transform housing assistance from a shelter program into an economic mobility program.
- Severe rationing leaves most eligible families unserved: The US serves roughly 1 in 4 eligible low-income households. The unmet need is not a design flaw — it is a consequence of chronically insufficient funding. Access is rationed by waiting-list position, which is largely determined by when a family first applied. This arbitrary allocation is difficult to defend on efficiency or equity grounds.
- Vouchers fail in tight housing markets: Voucher efficacy depends entirely on landlord acceptance and adequate private-market supply. In high-cost, low-vacancy markets, many landlords decline voucher holders — and in most states, this refusal is legal. Fair market rent ceilings often exclude high-opportunity neighborhoods, meaning families theoretically free to move anywhere are practically confined to the same kinds of high-poverty areas as public housing residents.
- Concentrated public housing creates lasting negative externalities: Large-scale projects that concentrate very-low-income families in a single location generate poor peer networks, underperforming schools, high crime, and limited job contacts. These concentrated poverty effects partially or fully offset the financial value of the subsidy. The evidence from demolitions and relocations consistently finds that children do better when moved out of high-poverty projects — which is an indictment of the projects' neighborhood environment, not of housing assistance itself.
- Capital underfunding degrades the existing stock: The federal government has chronically underfunded public housing maintenance, accumulating a capital needs backlog that HUD estimates at over $70 billion. Units deteriorate and become uninhabitable faster than new ones are added, shrinking the deeply affordable stock even as waiting lists lengthen. This is a predictable consequence of treating housing assistance as a discretionary budget item subject to annual appropriations pressure.
The debate between public housing and vouchers is partly a false choice — both programs serve households the private market cannot reach at all, and both are underscaled relative to need. The more important question is whether the US is willing to fund housing assistance at anything close to the scale the eligible population requires. Currently, roughly 17 million eligible households receive nothing. The research evidence points to a few clear conclusions: well-targeted vouchers are more cost-effective than new construction in most markets but require landlord participation and supply to function; the neighborhood environment during childhood matters enormously for long-run outcomes, making portability a policy priority; and the very poorest households need deeply subsidized units that filtering and vouchers alone cannot provide. A coherent policy would scale the voucher program toward universal coverage for eligible households, invest in the existing public housing capital backlog to prevent further deterioration, and pair both with the zoning reforms and supply expansion that determine whether vouchers have anywhere to be used.