Universal basic income experiments have a history of producing findings that satisfy neither proponents nor critics completely — and Uruguay's IBU pilot, the largest and most methodologically rigorous conducted in Latin America, is consistent with that pattern. The results are real, replicable, and instructive. They are also qualified in ways that complicate the political narratives that had formed around the programme before it concluded.
The pilot design, developed by a team from the Universidad de la República and the London School of Economics, used a stratified random assignment process that allocated recipients across the lower three income quintiles and included a matched control group of 11,800 non-recipients tracked through the same administrative data systems. The UYU 18,000 monthly payment — approximately USD 450 at the programme's inception, adjusted for inflation annually — was unconditional, universal within the sample, and explicitly not means-tested.
Material deprivation outcomes were unambiguous. Food insecurity among recipient households fell by 67 percent relative to controls over the five-year period. Housing instability — measured as arrears on rent or mortgage payments — fell by 41 percent. These effects were larger and more durable than prior conditional cash transfer programmes had produced, a finding the report attributes to the absence of compliance requirements that had historically caused eligible families to lose benefits during periods of administrative disruption.
The entrepreneurial effect was the finding that attracted most academic attention. Recipients were 2.8 times more likely than controls to have started a formal business by year three — an effect the researchers attribute to the income floor's reduction of the downside risk associated with leaving wage employment. The businesses formed were disproportionately in services and small-scale manufacturing, and their three-year survival rate exceeded the national average for new firm formation by 18 percentage points.
"The question was never whether it worked," said lead researcher Professor Cecilia Rossel. "The question was always what we meant by working. By almost every measure we chose, it did."