The city of Stockton, California, has proven that basic income programs are the future of anti-poverty policy. At least that’s the conceit of those cheering the results of a recent study that tracked Stockton residents who received no-strings-attached cash payments in the years before the pandemic arrived. With prominent politicians, including New York mayoral candidate Andrew Yang, advocating implementation of such programs in dozens of cities and states, this study is supposedly a game-changer. In reality, though, it is nothing of the sort.

The Stockton Economic Empowerment Demonstration (SEED) provided 125 residents of low-income neighborhoods with prepaid debit cards worth $500 each month for two years. It found that recipients used the money to pay for food, utilities, and other merchandise, and the additional flexibility was beneficial for mental health. Better yet, though only 28 percent of recipients worked full time at the beginning of the demonstration, 40 percent did so by the end. Such findings, the study’s authors conclude, show “a causal connection between guaranteed income and financial stability, and mental and physical health improvement.”

Stockton mayor Michael Tubbs celebrated the results, urging the media to “tell your friends, tell your cousins, that guaranteed income did not make people stop working.” According to Annie Lowrey at The Atlantic, SEED has “proved false” the adage that “the best route out of poverty is a hand up, not a handout.” NPR reported that this “high-profile universal basic income experiment . . . measurably improved participants’ job prospects, financial stability and overall well-being.”

Tubbs and his media allies should temper their enthusiasm. The new study should have no bearing on the conversation about basic income—primarily because it is not a basic income experiment.

First, the program can hardly be described as an “experiment.” Its recipient pool included a tiny sample of Stockton residents living in low-income areas. (The “universal” prong of “universal basic income” has already fallen out of favor.) Though residents were indeed randomly selected from within those low-income ZIP codes, 125 people narrowed by geographical scope is far from a representative sample of those who would actually receive basic income if it were instituted as policy. Using a study of 125 people from pre-selected areas as the basis for policies that would implicate millions is absurd.

Calling the program a “basic income” pilot is also hugely misleading. The demonstration provided recipients just $6,000 per year, a substantial supplement to existing income but not nearly enough to qualify as an income floor in a city where the median household annual income is more than $46,000. It bears more resemblance to abandoned cash welfare programs such as Aid to Families with Dependent Children than it does to Andrew Yang’s $12,000-per-year “Freedom Dividend.”

Mr. Tubbs, who now leads an organization called “Mayors for a Guaranteed Income,” insists that the major fear surrounding basic income programs—that they will cause people to work less or stop working entirely—is misplaced. But SEED tells us nothing about potential work effects in the real world. One crucial limitation of the study is that recipients knew that the program was time-limited. We therefore do not know whether basic income would “make people stop working” if it were implemented as policy for the indefinite future. That recipients did not disconnect from the labor market when they knew their benefits were small and temporary is unsurprising and says nothing about basic income as an anti-poverty policy.

Another major concern about basic income as a policy is that it would be enormously expensive, even if it were targeted only at the poor. As a privately-funded endeavor, sponsored largely by billionaire Facebook co-founder Chris Hughes, SEED tells us nothing about how taxes might change to fund such a program. One possibility is a state implementing a basic income program, only to tax its very recipients at a higher rate in order to pay for it, giving money with one hand and taking with another.

Taxes are not the only second-order effect left unexplored by a study so disconnected from basic income’s real-world implementation. For instance, one of the fundamental problems plaguing basic-income programs is that they would cause significant inflation by pumping up demand for certain goods and services. If all consumers suddenly had a significant—and, importantly for business owners, predictable—additional monthly income, basic economic theory suggests that prices will rise to meet the increase in aggregate demand.

There is a real possibility that basic income guarantees are thereby self-defeating. Even if a jurisdiction provided guaranteed income only for low-income households, we should expect price increases among the goods and services low-income people tend to buy, stripping the payments of their value. The SEED study does nothing to assuage such fear, because it is too small a shock to aggregate demand, paid only to a few dozen people for a couple of years. In failing to simulate a demand shock it bears little relationship to a basic income program in practice.

Finally, SEED’s advocates note that the common worry that basic income recipients will spend their money on undesirable goods is unwarranted: “Less than 1%” of SEED money “was spent on alcohol and/or tobacco,” according to the study. This, however, only tracks the expenditures of the SEED debit card, rather than the household expenditures of SEED recipients. Money is fungible. A basic income recipient could spend the cash they received as part of the demonstration on food and spend the money they would normally use for necessities on alcohol.

Though its proponents tout it as an optimistic look at the antipoverty program of the future, the SEED study shows little more than what happens if you give a handful of people extra cash welfare for two years. Basic income advocates, including those running for public office across the country, will tell voters that concerns about basic income’s unintended consequences have been debunked. In truth, they remain as warranted as ever.