Earlier this month, we argued that Housing Choice Vouchers, also known as Section 8 vouchers, ought to be provided to every household with a qualifying income. The limited funding for vouchers today leaves millions of people—over three-quarters of those who qualify—without help when official public policy has declared that they need it. We also pointed out that we could pay for the entire voucher expansion—and dramatically increase housing assistance to middle-income homeowners—simply by ending one kind of housing subsidy, the mortgage interest tax deduction, to households making over $100,000 a year.

But there are a number of tactical issues with this plan. For one, there’s the part where the federal government—the same one that has not been in a hurry to spend much money on social services for quite a while—decides to create a new entitlement worth tens of billions of dollars. More prosaically, there’s also the problem of finding landlords who actually accept vouchers. In most of the country, it’s perfectly legal not to, leading to situations where people wait years to get off the voucher waiting list, only to discover that they can’t actually find a place to use the voucher once they have it. And even where Section 8 discrimination has been outlawed, enforcement is, to put it lightly, lax.

But don’t worry, there’s another solution: refundable tax credits. In case you forgot, credits represent money that comes straight off your tax bill, while deductions (as in, the mortgage interest one) reduce the amount of your income that’s taxed. And a refundable tax credit means that if you owe less money in taxes than the credit is worth, the government actually writes you a check for the difference. Surprisingly enough, one of the most notable refundable tax credits—the Earned Income Tax Credit, or EITC, which basically supplements the income of low-income workers—enjoys ideologically broad support, from Milton Friedman to Bernie Sanders.

So what would our housing tax credit look like? Possibly, a lot like the EITC and Housing Choice Vouchers smushed together. Tenants would report their income and rent to the IRS as they did their taxes; those whose income fell below 50 percent of their area’s median income would qualify for refundable tax credits that made up the difference between 30 percent of their income and whatever their rent was, up to some predetermined fair market rent.

Of course, if we wanted to to take this opportunity to improve on the way vouchers currently work, we could do that, too. That might look like tweaking the “area” used to determine “Fair Market Rents,” moving perhaps from metropolitan regions to ZIP codes, or other smaller units that would allow vouchers to be more useful in higher-income, higher-opportunity neighborhoods; or acknowledging that the 30 percent ratio for housing costs doesn’t make sense for everyone; or whatever else.

One complication is that tenants would have to somehow verify their rents. But that wouldn’t have to be too hard—a copy of the lease, or some signed IRS-specific form, ought to be enough. It would, in any case, almost certainly be less of a hassle from the landlord’s perspective than participating in Section 8.

And the advantages are pretty big. For one, although money spent through tax preferences is every bit as real as money spent through programs like Section 8, tax credits are much easier to sell politically: They can, not untruthfully, be described as tax cuts, rather than new spending, even though functionally they are almost identical. Second, perhaps even more importantly, a refundable housing tax credit would require much less cooperation on the part of landlords. In fact, the rent verification system could be designed in such a way that landlords would never need to know up front whether their tenants intended to use the housing tax credit, any more than they currently know whether their tenants use the EITC. In that way, any apartment charging Fair Market Rent or less would be usable by housing tax credit recipients, vastly expanding housing options for low-income renters.

Lastly, a refundable housing tax credit should theoretically cost about as much as a straightforward voucher expansion, which the Congressional Budget Office pegged at $41 billion a year over ten years. And really, if you take away one thing from these posts, it’s that figure. While $41 billion isn’t overwhelming compared to the size of the federal budget—or even compared to what we already spend on housing subsidies to people who arguably don’t need them—its effects would dwarf the marquee affordable housing initiatives coming out of places like New York City and San Francisco. While local efforts to relieve the housing cost burdens of low-income residents are laudable, the scale of the problem in very high-demand regions is simply beyond the budgets of cities to deal with.

If we’re serious about closing the gap between market rents and what low-income people can afford, we have to be serious about the scale of the resources that need to be brought to bear. (We also need to be serious about making sure that gap isn’t larger than it needs to be by combating exclusionary zoning that’s designed to push market prices higher.) Making housing assistance an entitlement through either vouchers or a tax credit doesn’t have to be the only way we do it, but any other solution will have to be of a similar size.

The current voucher program is small-scale, stigmatized, and more complicated than it needs to be. It doesn’t reach all those in need, many landlords simply refuse to participate, and both tenants and landlords have to deal with a bureaucratic system to receive benefits. In contrast, the housing benefits we give to the middle-class and wealthy are large, normalized, and automatic: anyone who itemizes their income tax deductions receives the mortgage interest tax deduction. Maybe we should take advantage of the less burdensome way we dispense housing assistance to those who don’t really need it, and use the same method to provide housing assistance to all of those who really do.