The headlines tell the story: “Half of all renters can’t afford the rent.” “Renters, get ready to take it on the chin.” “The rent crisis is about to get a lot worse.”

As the nation’s homeownership rate declines—the result of stagnating wages, rising home prices, and the growing number of families still facing foreclosure—poor and middle-class families are competing for scarce rental housing. The nation’s homeownership rate fell to 63.4 percent in the second quarter of 2015, the lowest level since 1967. Almost all the new rental housing built in the past decade has been high-end luxury units. In 2014, the median rent of newly constructed units was $1,372, about half the median renter’s monthly income. The number of renter households paying over half their income for rent is expected to increase from 11.8 million in 2015 to 13.1 million in 2025.

Shaun Donovan, Obama’s first secretary of the Department of Housing and Urban Development (HUD), warned: “We are in the midst of the worst rental affordability crisis that this country has known.”

Skyrocketing rents and the shortage of affordable rental housing are not only a hardship for families but also a problem for business. Across the country, employers complain that high housing costs make it very difficult for them to attract and retain employees at all levels. Moreover, when families are paying so much of their incomes for housing, they have little income left to spend in local businesses.

There are two ways to solve this problem: One, raise wages, and two, provide more subsidized housing to help fill the gap between incomes and rents. In recent years, the federal government has done neither.

Since 2009, the federal minimum wage has been stuck at $7.25 an hour, despite widespread public support to raise it. To boost wages, the federal government has adopted another approach: The earned income tax credit (EITC), a wage supplement for the working poor. The EITC is quite popular with elected officials across the political spectrum. It has become, without much fanfare, the nation’s most effective anti-poverty program. In 2015, 27 million working American families received about $65 billion in EITC benefits. The average annual benefit was more than $2,400.

The EITC is an entitlement, available to all low-income households in which at least one person holds a job. That stands in contrast to federal housing subsidies: Federal funds to help low-income families pay the rent are disbursed via a waiting list or a lottery, not an entitlement. Only 4.8 million households—roughly one quarter of those eligible for assistance—are lucky enough to receive any housing help from HUD. The three major federal housing programs include public housing developments owned by local governments (1.1 million units), privately owned subsidized rental projects (1.2 million units), and housing vouchers which families use to pay for rent in the private rental market (2.2 million). (Another 300,000 low-income households are served by smaller HUD programs.)

The HUD budget for these programs in 2015 was roughly $37 billion. Together, federally subsidized housing for the poor represents only 3.5 percent of the 134 million housing units in the country.

The three-quarters of America’s low-income families who don’t receive government rental subsidies are forced to fend for themselves in the private housing market. Because of the severe shortage of affordable housing, many of them pay over half their incomes just to keep a roof over their heads, live in overcrowded or substandard housing, and face a growing wave of evictions and homelessness, as sociologist Matthew Desmond documents in his powerful new book, Evicted: Poverty and Profit in the American City.

Politically, it has been extremely difficult for housing advocates to convince Congress to maintain, much less increase, the budget for HUD’s subsidized housing programs, despite the growing number of Americans living in poverty and rising housing costs.

But it turns out that HUD is not the nation’s largest housing assistance agency. The biggest slice of federal housing subsidies comes from the Internal Revenue Service—and the beneficiaries are not the poor. In 2014, the federal government provided $130 billion in tax breaks for homeowners—almost three times the size of the HUD budget. That includes $68 billion in tax deductions for mortgage interest, $32 million in tax breaks for local property taxes, and $24 billion in tax breaks on the capital gains when these homeowners sell their homes.

These tax breaks primarily benefit the wealthiest Americans with the most expensive homes. The richest 20 percent of households receive 73 percent of the benefits of just the mortgage interest deduction, worth about $50 billion a year. The wealthiest one percent—those with incomes over $327,000 (for one-person households) and over $654,000 (for four-person households)—get 15 percent of the benefits of the mortgage interest deduction. Low-income homeowners get crumbs: Those with incomes between $10,000 and $20,000 get an average of $294 a year in mortgage interest tax breaks; those earning $20,001 to $30,000 get an average of $447 a year.

Is there a way to combine the best features of the EITC and the housing voucher program to address the housing crisis for America’s working families? The answer is yes: by providing an EITC supplement to help cover the costs of housing. As a nation we’ve long been using the tax code to subsidize housing, especially for the most affluent. Revising the EITC provides another way to use the tax code to subsidize housing—except to target it to those who need it the most, based on both their incomes and the cost of housing where they live.

Housing Conditions of the Working Poor

Housing is the largest part of most Americans’ household budgets. The federal rule-of-thumb is that households should not have to pay more than 30 percent of their household income for housing, but many Americans, and most low-income families, pay much more than that.

The number of renter households paying more than 30 percent of their income for housing increased from 14.8 million to 21.3 million between 2001 and 2014. In 2014, almost half of all renters paid more than 30 percent of their income in rent. The number of these households paying more than half of their income for housing jumped from 7.5 million to 11.4 million during that period—an all-time high. More than one in four renters paid over half their incomes for rent.

This housing/income squeeze is particularly onerous for the poor. Many low-income families not only have little income left over after they pay the rent, but they also live in overcrowded conditions and/or live in housing that is physically substandard, which can cause risks to health and safety.

Having a job doesn’t guarantee respite from the housing crisis. There is no state in the country where a minimum-wage worker working full time can afford a one-bedroom apartment at the fair market rent. In order to afford a modest, two-bedroom apartment, renters need to earn a wage of $19.35 an hour. This “housing wage” is much higher in some areas.

In 2014, 78.6 percent of “extremely low income” working families (defined as households with incomes below 30 percent of the area median income, or AMI) paid more than half their incomes for housing, while 36.5 percent of “very low income” working families (households with 30 to 50 percent of AMI) did so. Together, 54 percent of America’s working poor—7.5 million households—pay over half their incomes for housing, according to the Center for Housing Policy.

Housing Voucher Basics

In the 1970s, the Nixon Administration introduced a new federal program, initially called “Section 8” and now called the “housing choice voucher” (HCV) program. The idea was to give families vouchers—the housing equivalent of food stamps—so they could find and pay for apartments in the private rental market. Its advocates viewed vouchers as a cheaper way to provide housing for the poor than constructing residential developments with government subsidies attached to the buildings rather than the consumers. By 1980, about 600,000 low-income households used housing vouchers. It now serves 2.2 million households (5.3 million people) at an annual cost of about $19.6 billion.

The vouchers subsidize families’ rent so they don’t pay more than 30 percent of their incomes for housing. Households with vouchers have an average annual income of $13,821. Three-quarters of these households are extremely low income. Eighty percent have incomes below $20,000. However, few vouchers serve the working poor. Roughly three-quarters of voucher recipients do not work, primarily because they are elderly and/or disabled.

Each year HUD examines rental-housing conditions in every metropolitan area (and non-metro areas) to determine fair market rents (FMRs)—typically rent levels at the fortieth and fiftieth percentile of rents for apartments with one, two, or more bedrooms—which serves as the maximum rent for households with vouchers.

The average monthly rent for voucher recipients is $1,139. Families contribute on average $364 a month towards rent. HUD makes up the difference, an average of $775. Those families lucky enough to win the housing voucher lottery not only save money on rent, they also live in much better apartments in safer neighborhoods than comparable families without vouchers. About half of voucher households find housing in the suburbs, although there are still many landlords who refuse vouchers and real estate agents who steer voucher holders into low-income neighborhoods. Studies of HUD’s experimental Moving to Opportunity initiative, which provided poor families with vouchers to find apartments in middle-class areas, found that doing so improved the odds that low-income parents would find better jobs, their children would do better in school, and families were better off in terms of mental and physical health as well as other indicators of well-being.

Congress has been unwilling to increase the number of vouchers to serve all, or even most, families who are eligible for them. Although Republicans designed the program, they and many conservative pundits soon turned against it, arguing with little evidence that it destroyed middle-class neighborhoods by exporting poor families to areas where their presence increased crime.

A Popular Anti-Poverty Tool

The EITC has been popular with most Democrats and many Republicans since it was created in 1975. Congress has increased the EITC several times, raising benefit levels and expanding eligibility criteria.

The EITC reduces tax burdens and supplements wages, especially for families with children. It’s a refundable tax credit—workers who qualify for the EITC can get back some or all of the federal income tax that was taken out of their pay during the year, and even cash back. Workers with incomes below about $39,000 to $53,300—depending on marital status and the number of dependent children—are eligible for the EITC. (Even workers whose earnings are too small to owe income tax can receive it).

Under the EITC formula for 2015, a single parent with two children who earns $25,000 would receive $4,092. A married couple with a combined income of $40,000 and two children would receive $2,095. Also, workers who have no children and have incomes below roughly $14,800 ($20,300 for a married couple) can receive a small EITC. An employee’s EITC benefit increases with each additional dollar of earnings until reaching the maximum limit. This creates an incentive to leave welfare for work and for low-wage workers to increase their work hours. The EITC generates some $1.50 to $2 in economic activity for every $1 that goes to working families, according to the Center for American Progress.

In 1986, Congress indexed the EITC for inflation, which it has refused to do for the minimum wage. In 2013, the EITC lifted about 6.2 million people above the poverty line, including about 3.2 million children. In addition, it reduced the severity of poverty for another 21.6 million people, including 7.8 million children, according to a report by the Center on Budget and Policy Priorities.

The EITC is a popular policy tool in both parties for several reasons. For one, it comes in through the back door (as a tax break) rather than through the front door (as a direct grant) like food stamps, Medicaid, welfare, and housing vouchers. Another reason is that it rewards people who work, so it does not carry the stigma attached to “welfare,” which is viewed by many politicians, editorial writers, and opinion leaders as subsidies for the so-called “undeserving” poor.

Proposal: An EITC Housing Supplement

For all its benefits, the EITC has its shortcomings that Congress should address. Most serious is that the EITC’s benefit levels are the same across the country, even though the cost of living—especially housing—varies dramatically.

By contrast, HUD’s housing choice voucher program varies the amount of subsidies to households based on local conditions by calculating fair market rents—typically rent levels at the fortieth percentile for apartments with one, two, or more bedrooms—in different parts of the country. For example, according to HUD, the fair-market rent for two-bedroom apartments is $832 in Memphis, $908 in Phoenix, $1,093 in Chicago, and $1,424 in Los Angeles. The household income needed to afford a typical two-bedroom apartment ranges from $33,280 in Memphis to $36,320 in Phoenix, to $43,720 in Chicago to $56,960 in Los Angeles.

To remedy this flaw in the EITC, Congress should revise the EITC’s benefit levels to account for differences in the cost of living, particularly housing costs. This approach would reach many more families and require much less bureaucracy than the housing voucher program. Cushing Dolbeare, founder of the National Low Income Housing Coalition, first proposed this idea in 2001, and several researchers at the University of North Carolina explored it two years later, but it gained no political traction at the time. As the housing crisis has worsened, and the EITC has gained in popularity, it is time to give the idea a second look, and make it simpler.

A formula similar to HUD’s FMRs can be used to determine the size of the EITC housing supplement in each area. Like HUD’s FMRs, the EITC housing supplement would vary from area to area depending on market conditions.

The EITC serves a much wider range of families than the housing voucher program, including more who work and more who earn over the minimum wage. As a result, if the goal is to reduce rent payments to no more than 30 percent of household income, the subsidy required for most EITC recipients would be significantly less than the typical per-household voucher subsidy.

The EITC housing supplement could be set at the difference between 30 percent of the household’s income (including the EITC benefit) and the local fair market rent. In this way, the benefit would be tied to a family’s earnings as well as their housing costs. If a household’s income is greater than the difference between 30 percent of household income and the local FMR, the family would not quality for the housing component. It is likely that a significant number of EITC recipient households would not qualify for the housing supplement, because they already pay less than 30 percent of income for rent. But for those who face a serious income/rent squeeze, the supplement would make a significant difference.

Another plus is the simplicity of implementation. HUD’s housing voucher program is administered through local government housing agencies, which take applications, make sure that families meet the income guidelines, establish a waiting list, and inform participants of the maximum subsidy they qualify for. (Participants can rent apartments above the FMR if they are willing to pay more from their own pockets). By contrast, the EITC housing supplement program would require little bureaucracy. To receive the EITC, families simply fill out an income tax form. Depending on their income, they either pay less in taxes or receive a reimbursement check in the mail.

Moreover, the EITC is an invisible subsidy. Just as it works now, landlords would not know whether would-be renters are receiving any EITC benefits, including the housing supplement. It could thus have the effect of reducing discrimination against low-income households with subsidies.

The table below illustrates how the program would work for retail sales clerks—as single parents with two children and as married couples with two children—in four cities where wages vary slightly but rental housing costs vary dramatically. It identifies whether any subsidy—and, if so, how much subsidy—would be needed to guarantee that no household pays more than 30 percent of its income for rent. The EITC housing supplement would thus vary based on these conditions.

To be sure, there are limitations to the EITC housing supplement. One problem with the EITC is that most families receive the credit as an annual lump sum when they get their income tax refund—a particular burden for low-income families, who typically live month-to-month and especially need the income to pay the monthly rent. The Brookings Institution has proposed a number of other ways to provide workers with periodic payments instead of a once-a-year lump sum payment. Congress could require employers to participate in this periodic payment plan and/or the IRS could work with unions, churches, and community groups to publicize this option and get more eligible workers to use it.

Another likely concern is that the EITC only serves low-income families with at least one job-holder. Jobless families deserve decent housing. For those households, some version of the housing voucher program, as well as funds to expand the supply of affordable housing, will still be needed. Anti-poverty advocates at the Center for Budget and Policy Priorities have suggested that a renters’ tax credit would also help address this problem.

Neither the proposed EITC housing supplement nor the existing voucher program does anything to regulate rents. In a handful of cities, local rent control laws put limits on rent increases, but in most parts of the country landlords are free to set rents based on market conditions. HUD sets the FMRs at the fortieth or fiftieth percentile rent levels. Thus, the cost of the EITC housing supplement will be based on filling the gap between household incomes and a proportion of market rents, whatever they are in different parts of the country.

But for all its flaws, an EITC housing supplement holds much promise. More than six decades after the Housing Act of 1949 established as a national objective the achievement of “a decent home and a suitable living environment for every American family,” we are far from reaching that goal. In fact, federal housing policy is currently skewed toward the affluent, who receive far more in government housing subsidies than the working families and the desperately poor who need help the most. Adding a housing supplement to the popular EITC can get us closer to our national goal by helping the working poor pay the rent.

Thanks to Barbara Sard, Daniel McCue, Richard Rothstein, and Mindy Ault for their comments on earlier drafts of this article.