The BaseResearchNational Bureau of Economic Research
The Employment Effects of a Guaranteed Income: Experimental Evidence from Two U.S. States
Six authors report a randomized trial in Texas and Illinois: 1,000 low-income adults aged 21 to 40 received $1,000 a month with no conditions for three years, and 2,000 controls received $50. Recipients had about $1,900 a year less income besides the payments and were 4.2 points less likely to be employed (pooled estimate), with no gain in job quality.
Why it matters
The same thing, for different readers.
What it means for you
Low-income adults given $1,000 a month with no conditions worked somewhat less and spent more, but did not get better jobs. The study is not about AI.
For lawmakers
Bears on income-support ideas for AI job loss: unconditional cash cut work moderately and did not improve job quality, but it tested low-income adults, not displaced workers.
For researchers
The 4.2-point employment fall pools unemployment-insurance records with survey answers (surveys alone: 2.3 points). Partner-hours results were not pre-specified and lose significance after the paper's multiple-testing correction.
The question it raises
The authors read the fall in work as recipients valuing time away from it, raising whether less paid work is a cost to offset or a choice to respect.
In their own words
Every quotation was found, word for word, in the text we saved from the source. That shows the words are there; it does not make them true, or prove the speaker meant them as quoted.
From the source’s own text.
“Overall, our results suggest a moderate labor supply effect that does not appear offset by other productive activities.”
“The transfer caused total individual income excluding the transfers to fall by about $1,900/year relative to the control group and a 4.2 percentage point decrease in labor market participation.”
“Despite asking detailed questions about amenities, we find no impact on quality of employment, and our confidence intervals can rule out even small improvements.”
“While decreased labor market participation is generally characterized negatively, policymakers should take into account the fact that recipients have demonstrated–by their own choices–that time away from work is something they prize highly.”
How the finding was reached
In our words, from the source. Weigh the finding by how it was made.
- What was done
- Randomized trial of a program implemented by two nonprofits in ten counties of north central Texas and nine of northern Illinois. 1,000 adults aged 21-40, with 2019 household income up to 300% of the federal poverty level, were randomly assigned to $1,000 a month with no conditions from November 2020 to October 2023; 2,000 controls received $50 a month. Outcomes come from state unemployment-insurance records (for participants who consented to linkage and could be matched), enumerated surveys (response rates 97% at midline, 96% at endline), online surveys and a phone-app time diary. Pre-registered.
- What it found
- Relative to controls, total individual income excluding the transfers fell by about $1,900 a year (5.6% of the control mean). Employment fell 4.2 percentage points in the pooled records-and-survey estimate (2.3 points in surveys alone, 6.6 in records alone), and hours worked fell by 1-2 a week (1.3 in the survey measure). The extra time went mainly to leisure. The authors find no change in job quality and rule out wage gains above 58 cents an hour, and no significant effect on degree attainment. Spending rose by about $287 a month; well-being gains in year one faded by years two and three.
- What it does not show
- Participants were low-income adults aged 21-40 in parts of two states, not workers displaced by technology, and the paper does not study AI. Payments lasted three years; the authors say follow-ups are needed on what persists after payments end, on children, and on effects if a program ran at scale. Unemployment-insurance records miss some contract and gig work, and the 4.2-point figure depends on pooling them with surveys. The partner and household hours results were not pre-specified and are not significant after the paper's multiple-testing correction. A working paper, not peer-reviewed.
- Status
- A working paper: not yet peer reviewed
The record
- Questions
- What happens to work, income and the economy?
- People
- Eva Vivalt (Wrote it, University of Toronto) · Elizabeth Rhodes (Wrote it, OpenResearch) · Alexander W. Bartik (Wrote it, University of Illinois Urbana-Champaign) · David E. Broockman (Wrote it, University of California, Berkeley) · Patrick Krause (Wrote it, OpenResearch) · Sarah Miller (Wrote it, University of Michigan Ross School of Business and NBER)
- What we read
- We read the whole source.
- Found by
- A research pass: work, income and the economy, gathered October 13, 2026, via
research-work(tier 1: the source’s own channel or an official record) - Cite this
- National Bureau of Economic Research, “The Employment Effects of a Guaranteed Income: Experimental Evidence from Two U.S. States”, August 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/nber-2026-08-guaranteed-income-employment-effects