The Hilltop neighborhood in Tacoma, Washington is one of the poorest in the state — with many families living in poverty and struggling to maintain housing. Such instability inevitably impacts the public schools’ capacity to function; in 2006, Hilltop’s McCarver Elementary School witnessed a student turnover rate of 179 percent. This, as researchers at the Urban Institute have pointed out, is nothing unique to McCarver. Low-income students frequently move around, and family homelessness is a growing problem. A new report from the National Center on Family Homelessness reveals that one in 30 children experienced homelessness in 2013.

What is new is the policy solution being tested out by the Tacoma Housing Authority and Tacoma Public Schools — largely backed by the Gates Foundation — to see if traditional Section 8 housing vouchers can be leveraged to provide affordable housing and boost local school performance.

The Gates Foundation is investing in a host of such experimental partnerships to research the link between housing instability and educational achievement. With the potential to improve school performance, boost employment levels and reduce dependency on housing vouchers, these programs have decided political appeal. Yet while the Tacoma experiment offers a good model of cooperation among agencies, there’s little evidence that the approach can reduce poverty, as policymakers hope it will.

The McCarver Special Housing Program stipulates that in order to remain eligible for a housing voucher, parents must keep their kids enrolled in McCarver Elementary School throughout the five-year study, regularly demonstrate active involvement in their children’s education and attend a host of job training programs. With the help of local caseworkers, they must also achieve “economic self-sufficiency” (earning enough to afford a two-bedroom apartment, which is about $771 per month in rent) by the end of the fifth year.

For the first year, rent is massively subsidized — just $25 per month — and in each year thereafter it increases by 20 percent. The hope is that if housing can be a stabilizing force, especially in the first year, then parents can focus on their economic future and their children’s schooling.

So far, the McCarver program has yielded some interesting results. Across all participating households, average monthly income rose from $436 in 2011 to $836 in 2013. Student turnover within the experimental cohort is also significantly less compared to the rest of the school.

The program began with 49 families in 2011, joined by seven more in 2012. Of those 58 families, 39 are still participating.

Some departed for positive reasons, like finding a new job in a different part of town — but others for less encouraging reasons. “Some families really struggled and couldn’t complete the requirements, so the families were essentially dropped from the program,” says Justin Milner, an Urban Institute researcher who studied the McCarver project. (The McCarver program doesn’t count the 19 families that dropped out when calculating the cohort’s average monthly income increases.)

Although Tacoma is a relatively poor city, and even in wealthier areas, many highly educated people cannot find stable work, the McCarver program frames the economic issue as one where parents lack the skills and training necessary to obtain employment — not that there aren’t solid jobs to get. When I asked Greg Claycamp, an official with the Tacoma Housing Authority, what would happen if families were unable to pay full rent at the end of the five years because there are no decent employment prospects available, he could not confirm that families would continue to receive housing assistance. “I think we want to be very careful about not penalizing a household that turns out to have really complex challenges,” Claycamp says. “But for now no final decision has been made.”

Researchers Sendhil Mullainathan and Eldar Shafir, in their book Scarcity: Why Having Too Little Means So Much explain how many social policy designers “assume the problem is a lack of understanding or of motivation. So they follow up with attempts to educate or to sharpen incentives.” Mullainathan and Shafir have found that such carrot-and-stick approaches might actually lead to greater harm.

This highlights a central tension with these new partnerships. Exploring the relationship between education and housing is a smart thing to do — giving kids and parents the opportunity to learn in safe, caring and even rigorous environments is something that all democratic societies should be promoting. Leveraging these relationships as a means to reduce poverty and reliance on welfare, however, is quite another objective.

In mid-October, Claycamp joined housing authority officials from around the country in Washington D.C. to attend a conference at the National Council of Large Public Housing Authorities (NCLPHA) where this question of cross-collaboration between school districts and housing authorities was a major theme. Another NCLPHA conference devoted specifically to this question will be taking place in D.C. in February.

“Our focus on this started a couple of years ago, given the difficult environment in Washington and budget constraints,” says Sunia Zaterman, executive director of NCLPHA. “We’re serving at best a quarter of the households that are in need, so we want to reduce inter-generational poverty to free up resources for the next ones on the waiting list.” Gates-funded experiments are cropping up all over, from the Pacific Northwest to Akron, Ohio to New Haven, Connecticut. Not all follow the Tacoma model — some are focused on implementing early childhood learning initiatives and some are geared toward mentorship and tutoring. Right now, Zaterman says, housing authorities are just interested in expanding research studies to learn more about what these diverse partnerships are capable of doing, although she stressed no one-size-fits-all approach would be feasible or wise.

The Gates Foundation operates with the stated belief that quality education is “the best way to break the cycle of poverty.” Yet despite their strong embrace of metrics and data, this does not appear to be an evidence-supported claim. Tax and transfer programs, like that of Social Security, which reduced elderly poverty from 47 percent in 1967 to 15 percent today, have proven to be far and away the most effective poverty reduction policy.

Robust educational support and greater mentorship opportunities are good things, and have been shown to lead to better educational outcomes. So exploring how to improve a child’s education by leveraging school and housing strategies is promising, and warrants further investigation. “Even in our days of doom and gloom, this has been very uplifting,” Zaterman notes.

But as more organizations, think tanks and working groups devote their energy to researching joint housing and education programs, it’s important to pursue these programs for the right reasons. Improving educational achievement is an important aim in and of itself. Democratic nations depend upon the presence of an active and educated citizenry. But in the absence of a strong economy and without an ample cash-transfer system that can assist the poor who lack wealth and savings, families are unlikely to break the cycle of poverty. There are great possibilities, and also practical limitations, to these new initiatives.