The Frank Gallagher Thesis Shaped Welfare For Decades. The Data Just Demolished It.

For decades, governments assumed poor people couldn’t be trusted with unconditional cash. A wave of new evidence says they were wrong.

There is a documented phenomenon in the research literature: drug-related hospitalizations spike measurably in the first days of the month, when government disability and welfare checks land. Among SSI recipients with preexisting addictions, within-hospital mortality rises by roughly 22 percent. The data is real.

So is the leap politicians made from it — and that leap is what’s collapsing.

For decades, the fear that poor people will waste unconditional cash on drugs and alcohol has shaped welfare policy across the Western world. A wave of new evidence suggests the fear was never grounded in the reality most people actually live.

A 2007 study in the Journal of Public Economics tracked California hospital admissions and found a 23 percent spike in drug-related hospitalizations in the first five days of the month, driven almost entirely by SSI and disability check arrivals. Among recipients with preexisting addictions, within-hospital mortality rose by 22 percent. The checks weren’t creating new addicts. They were giving existing ones the means to go further. 

Policy Built on a Punchline 

In honor of the despicable, low-life, perturbing, hilarious, and unavoidable character from the TV show Shameless, I’m calling this phenomenon “The Frank Gallagher Thesis,” because in the earlier seasons of the hit show, we watch Frank defraud the system by collecting welfare checks from a late aunt before going to the local watering hole to run his tab. Just like some people did in this most recent study around a Vancouver community-based cohort of people who use drugs.

But what’s now collapsing under a pile of new evidence is the leap from the above-discussed conduct among some check receivers to an entire philosophy of governance.

The Frank Gallagher Thesis is the reason a single mother in Ohio may be subjected to drug screening requirements tied to assistance programs, and a twenty-year-old sleeping in his car in Portland can access a shelter bed but not the cash that would let him secure a deposit on a room.

In the U.S., at least 15 states have passed laws requiring drug testing for welfare recipients, nearly all of which have produced positive rates indistinguishable from the general population and probably cost more to administer than they saved. 

In Australia, the government proposed random drug testing of youth welfare recipients in the name of “love.” In New Zealand, benefit claimants can be required to take drug tests as a condition of job referrals. In the United Kingdom, where the original Shameless ran for 11 seasons, and the tabloid press has perfected the “scrounger narrative,” drug testing provisions were written into the 2009 Welfare Reform Act but never implemented — the political theater of the proposal being the point. In Ontario, a Conservative government drew up plans for mandatory drug testing and abstinence requirements… and so forth. You get the point. 

But what if we were to provide financial assistance to homeless people without a counterpart? Will they go and spend it all on drugs, like the doctrine largely assumes?

In fact, it’s being done — and the evidence against that assumption is mounting quickly.

What Actually Happens When You Give People Cash 

In February 2023, without much national attention, the state of Oregon began wiring $1,000 a month to 120 young people between the ages of 18 and 24 who were living on the streets.

The program, called Direct Cash Transfer Plus, was a collaboration between the Oregon Department of Human Services and Point Source Youth. This national nonprofit had already piloted a similar model in New York City. It operated in three counties — Multnomah, Clackamas, and Deschutes — through partnerships with local organizations, including the Native American Youth and Family Center in Portland.

The payments ran for two years.

Results were published in January 2026: of the 63 participants (out of 120) who completed both baseline and follow-up surveys, 94% reported being in stable housing. Nearly three-quarters were employed or enrolled in school. The rate of high school diploma completion had risen measurably.

Participants described using the money for rent deposits, car insurance, groceries, laptop computers for community college classes, the bus fare, or gas money to flee domestic violence.

One participant, in an anonymous survey response published by the state, put it plainly:

“The DCT program is how I’m still alive. If I did not have such support in the beginning of the program I would still be stuck in a homeless pit.”

Another wrote that she had left sex work, enrolled in school, and started a small beading and art business with the financial breathing room the payments provided.

And there’s a broader pattern from other experiments like this one. 

Oregon’s program was the second state-level implementation of a model first tested in New York City, where 29 young adults received $1,100 a month for two years through the Trust Youth Initiative. At the end of the program, 92% reported that the payments helped them secure stable housing.  In Washington State, a different model — based on one-time emergency payments of $1,000 to $2,000 — kept 93% of recipients from returning to homelessness within a year.

In Denver, where more than 800 homeless adults received monthly transfers, the share of participants reporting that they live in their own housing rose from low single digits at intake to roughly 45% within a year, depending on the group. 

By January 2025, at least 160 guaranteed income pilot programs had been launched across the United States.

The studies varied in design, population, and scale, but the central finding kept replicating: when you give people in crisis unrestricted cash, the overwhelming majority spend it on the unremarkable necessities of survival, such as food, shelter, transportation, utilities,  and their circumstances measurably improve.

What they don’t do, with striking consistency, is what the Frank Gallagher Thesis predicts. 

The Evidence Is No Longer Anecdotal 

No conversation about unconditional cash ever gets very far before someone asks:
“Won’t they just spend it on drugs?”

The most-cited evidence against that assumption comes from the 2020 New Leaf Project, a randomized controlled trial conducted in Vancouver. Researchers gave 50 recently homeless individuals a one-time cash transfer of 7,500 Canadian dollars and tracked outcomes over twelve months. The cash group moved into stable housing faster than the control group and spent the majority of their money on rent, food, and clothing. There was no increase in spending on alcohol, cigarettes, or drugs. In fact, early reporting from the project suggested a decline. 

At a global scale, the evidence is even harder to dismiss. A World Bank review of cash transfer programs across low- and middle-income countries found no systematic increase in spending on alcohol or tobacco, and in many cases, consumption actually declined.

In the United States, the pattern has held. 

The Baby’s First Years study, a rigorous multi-site trial that gave 1,000 low-income mothers either $333 or $20 per month, found no significant increase in alcohol, cigarette, or opioid use among the higher-payment group.

And then there is the Chelsea program in Massachusetts, which provided some of the most striking data of all. Participants who received unconditional transfers of up to $400 a month saw an 87% reduction in substance-use-related emergency room visits, a 62% reduction in behavioral health visits, and a 27% reduction in emergency department visits overall.

So, the bottom line is that across every major study, in every city that has tried some version of this experiment, the Frank Gallagher scenario — the one in which recipients take the money and self-destruct — has not materialized.

And at this point, that absence is no longer anecdotal. It is empirical.