Intermediatepolicy
Universal Basic Income Pilots: What the Experiments Actually Found
Learning objectives
- Distinguish universal basic income from guaranteed-income pilots
- Summarize the major experiments: Kenya, Finland, Stockton, and others
- Report the findings on work, health, and well-being accurately
- Explain the limitations: scale, duration, and funding questions
- Assess what the evidence can and cannot settle about full UBI
What UBI is — and what the pilots are not
Universal basic income (UBI) is a simple idea: every resident receives a regular, unconditional cash payment, enough to cover basic needs, with no work requirements and no means testing. It differs from traditional welfare in its universality (everyone gets it) and its unconditionality (no strings attached).
Almost none of the famous 'UBI pilots' actually tested UBI. Most were guaranteed-income experiments: time-limited payments to selected low-income participants, funded by philanthropy, with no tax increases to pay for them. Kenya's GiveDirectly experiment, Finland's 2017–2018 trial, and Stockton, California's SEED program all fall in this category.
This distinction matters enormously for interpreting results. A pilot tells you how recipients behave when given temporary extra cash; it cannot tell you how an economy behaves when everyone receives permanent payments funded by higher taxes. The pilots answer a narrower — but still valuable — question.
Kenya: the largest experiment
The charity GiveDirectly ran the largest UBI-style experiment in rural Kenya, delivering long-term unconditional cash transfers to thousands of recipients across hundreds of villages, with randomized control groups. Researchers tracked outcomes over more than a decade — an unusually long horizon.
The findings were broadly positive: recipients invested in assets like metal roofs and livestock, started businesses, and reported better psychological well-being. There were spillover benefits to non-recipients in treated villages, as the cash circulated locally. Long-term transfers outperformed short-term ones, suggesting permanence matters.
But Kenya's context — extreme poverty, a largely informal economy — limits generalization to rich countries. And even here, the transfers were donor-funded; the experiment says nothing about the taxation required to fund UBI permanently at national scale.
Most pilots last 1–3 years with hundreds of participants. Kenya's ran over a decade across whole villages — the closest any experiment has come to testing UBI's long-run effects. Its positive results on assets and well-being are the strongest evidence the pilot literature offers.
Finland and Stockton: rich-country tests
Finland's 2017–2018 experiment gave 2,000 unemployed Finns a monthly unconditional payment instead of standard unemployment benefits. Employment effects were small — recipients worked only slightly more than the control group — but well-being improved markedly: better health, less stress, greater confidence in the future.
Stockton, California's SEED program (2019–2021) gave 125 low-income residents $500 per month. Employment among recipients rose relative to the control group, and participants reported better mental health and greater ability to handle emergencies. The results surprised skeptics who predicted work would collapse.
Both experiments were small and short — and Finland's was not truly universal, since only the unemployed were eligible. Their consistent finding is worth noting, though: unconditional cash did not produce the work-shirking catastrophe critics predicted, and it measurably improved well-being.
The work question
The oldest objection to UBI is that free money destroys the incentive to work. The pilot evidence, taken together, suggests modest effects: some reduction in work hours, concentrated among groups like young people staying in school longer and mothers spending more time with children — outcomes many would consider features, not bugs.
The 1970s US negative-income-tax experiments found work reductions of roughly 5–10% for men and more for married women — meaningful but far from the collapse critics feared. Modern pilots find even smaller effects, possibly because payments are modest relative to living costs.
But pilots cannot capture general-equilibrium effects: if everyone received UBI permanently, wages, prices, and labor supply would adjust in ways no small experiment can observe. The work question at full scale remains genuinely unanswered — the most important thing the pilots cannot tell us.
What the pilots cannot prove
Three fundamental limits apply to every pilot. First, funding: pilots are donor-funded windfalls, while real UBI requires taxation that would change behavior — the pilots measure the benefit side without the cost side. Second, scale: giving cash to a few hundred people in a large economy has no effect on wages or prices; giving it to everyone would.
Third, duration and expectations: a two-year payment is treated as temporary windfall; a permanent, universal entitlement would reshape life decisions — education, family, migration — in ways temporary pilots cannot reveal. Alaska's Permanent Fund dividend, the longest-running unconditional payment, is suggestive but far too small to be a UBI test.
None of this makes the pilots worthless. They have credibly established that unconditional cash improves well-being, does not destroy work incentives at modest levels, and can be delivered efficiently. What they cannot do is settle whether full UBI is affordable or desirable — that remains a question of values and macroeconomic judgment, not just evidence.
Every major pilot gave temporary, targeted, donor-funded cash. Real UBI is permanent, universal, and tax-funded. Extrapolating from one to the other requires assumptions no experiment can validate — beware anyone who claims the pilots 'prove' UBI would work (or fail) at national scale.
Key points
- UBI = universal, unconditional, permanent cash; most pilots tested none of these fully
- Kenya's long-running experiment found positive effects on assets, business, and well-being
- Finland and Stockton found small employment effects and meaningful well-being gains
- Work reductions in pilots are modest — far from the predicted collapse
- Pilots cannot test funding (taxation), scale effects, or permanent expectations
- The evidence supports cash transfers' benefits; full UBI remains unproven either way
Common misconceptions
✓ Truth: They prove that temporary, targeted, donor-funded cash transfers help recipients — a valuable but much narrower finding. Nothing in the pilot literature tests permanent universal payments funded by taxation, which is what UBI actually is.
✓ Truth: The consistent finding across decades of experiments is modest work reductions, not collapse — and some of the reduction reflects socially valued choices like education and caregiving. Critics' strongest argument is about untested full-scale effects, not the pilot results.
Frequently asked questions
What is the difference between UBI and guaranteed income?
UBI is universal — everyone gets it regardless of income. Guaranteed income is targeted at low-income people. Most 'UBI pilots' were actually guaranteed-income experiments, which matters because universality changes the cost and the tax burden enormously.
How would UBI be funded?
Proposals include income taxes, carbon taxes, land-value taxes, or a sovereign wealth fund like Alaska's. The funding question is UBI's hardest problem: a meaningful payment for everyone costs a large share of GDP, and no pilot has tested the taxation side.
Did anyone quit working in the pilots?
Very few. The consistent finding is small reductions in hours, concentrated among students and caregivers. But pilots gave modest, temporary payments — a permanent livable UBI might produce larger effects that no experiment has measured.