meta-analysis · 2024
UBI: A Global
Meta-Analysis
A synthesis of findings from 10+ universal basic income experiments conducted across six continents between 1968 and today — covering health, employment, poverty, education, and how recipients actually spend the money.
Published: 15/03/2024 · Sources cited throughout · Full references at bottom
§ 1 — the experiments
A half-century of testing
From a Manitoba town in 1974 to an ongoing 12-year trial in rural Kenya, researchers and governments have been running controlled experiments to answer one deceptively simple question: what happens when you give people money, no strings attached?
Finland
Nation-wide Randomized Trial
sample: 2,000 unemployed adults
No employment drop. Life satisfaction 7.3 vs 6.8 (control). Mental stress: 16.6% vs 25%.
Kela / Ministry of Social Affairs ↗Stockton, USA
SEED Pilot
sample: 125 residents
Full-time employment +12 pp (treatment) vs +5 pp (control). Income volatility halved.
Stockton SEED ↗Canada (Mincome)
Dauphin, Manitoba
sample: Entire town of ~10,000
Hospitalizations −8.5%. High school completion rose. Modest work reductions from students and new mothers.
Forget (2011) / BBC ↗Namibia
Basic Income Grant (BIG)
sample: ~930 residents, Otjivero village
Poverty 76% → 37%. Child malnutrition 42% → 10%. Crime fell 42%. School dropouts 40% → 5%.
Namibia BIG Coalition ↗Kenya
GiveDirectly — world's largest UBI study
sample: ~20,000 people, 245 villages
19% more new businesses. Better food security during COVID-19. Long-term recipients save and invest significantly more.
GiveDirectly (2023) ↗India (Madhya Pradesh)
SEWA / UNICEF Pilot
sample: ~6,000 across 20 villages
73% reduced debt. Girls' secondary school attendance +30%. Children at healthy weight 39.2% → 58.7%.
SEWA / UNICEF Evaluation ↗United States
Negative Income Tax Trials
sample: ~8,500 families, multiple sites
Modest work hour reductions (−5% to −9% primary earners). High school completion rose 5–11% in some sites.
Multiple academic evaluations ↗Iran
National Cash Transfer Program
sample: 70 million people (near-universal)
No negative labor supply effect. Service workers worked ~36 more minutes/week. Youth channeled funds into education.
Salehi-Isfahani & Mostafavi-Dehzooei / WEF ↗Alaska, USA
Permanent Fund Dividend
sample: All Alaska residents (~750,000)
Poverty reduced 20–40% (relative). Employment rate unchanged. Most equal income distribution in the US.
Marinescu (2018) / Earth4All ↗Ontario, Canada
Basic Income Pilot
sample: 4,000 low-income residents
80% felt health improved. 74% ate healthier. Canceled by new government — 80% saw anxiety return immediately.
Maytree Foundation / BICN ↗§ 2 — health & well-being
Money, it turns out,
is medicine.
Across nearly every experiment, recipients reported measurable improvements in mental health, physical health, and nutrition. Financial stress is one of the most pervasive drivers of poor health outcomes — and a guaranteed income removes it.
Only 16.6% of Finnish UBI recipients reported high mental stress, compared to 25% in the control.
Canada's Mincome experiment saw an 8.5% drop in hospital admissions — mainly mental health and accidents.
In just one year, the share of underweight children in Otjivero village fell from 42% to 10%.
Reported health improvement (% relative reduction or gain)
Each bar represents the magnitude of the primary health metric improvement per experiment. Hover for detail.
“The greatest benefit was the fact that I was treated as a human being, not a client.”— Finnish basic income recipient, describing reduced stigma and bureaucratic pressure[18]
§ 3 — employment effects
The fear that never
materialized.
The most common objection to UBI is that people will stop working. The empirical record says otherwise. In most experiments, labor participation was neutral to positive — and where it declined, the reasons were largely positive (education, caregiving).
Employment change under UBI (percentage points)
Positive values = more employment. Negative values = less. Green = positive outcome. Red-toned = reduction. Hover for context.
the evidence says
“No systematic evidence of lazy behavior.”
A comprehensive review by Banerjee et al. (2017) of seven cash transfer trials in developing countries found that poor people did not reduce their work hours when given unconditional cash — in some cases they worked more.[10] In Uganda, similar grants led to 17% more hours worked and 38% higher earnings.
Stockton — why it worked
Stockton's $500/month removed material barriers to employment. Recipients used the cushion to job-hunt, pay for childcare, or finish a certificate program — then moved into better jobs.[2]
Why did Stockton see gains while Finland saw no change?
Stockton
Recipients were the working poor — precariously employed. The $500 was additive (not a benefit replacement) and acted as a work-enabler by removing financial barriers to finding better jobs.
Finland
Recipients were unemployed in a country with strong social safety nets. The €560 roughly matched what they already received — it mainly removed conditionality and reduced stigma, but structural unemployment remained.
US NIT experiments
Benefits phased out with income, creating marginal tax rates of 50%+. This is a design flaw, not a UBI characteristic — true UBI doesn't phase out, so wouldn't create the same disincentive.
Canada Mincome
The work reduction was concentrated among two groups: new mothers spending time with infants, and teenagers staying in school longer. Both are positive social outcomes, not laziness.
§ 4 — poverty & education
In the developing world,
the results were transformational.
When people start from a point of unmet fundamental needs, even a modest basic income can unlock cascading change — in nutrition, schooling, debt, and community safety.
Namibia
Otjivero village · 2008–2009A village of ~930 people received N$100/month (~$7). Within one year, extreme poverty fell in half, children returned to school, and the community became measurably safer.[4]
Kenya
GiveDirectly · 2017–presentThe world's largest UBI study: 12 years, 245 villages, ~20,000 recipients. Early results show recipients don't just survive — they build. Long-term assurance changes how people think about the future.[5]
India (Madhya Pradesh)
SEWA / UNICEF pilot · 2011–201320 villages. 18 months. The results rippled through food, health, housing, education, and debt — demonstrating how a small but reliable income lifts nearly every dimension of household welfare simultaneously.[6]
§ 5 — how the money gets spent
On what you'd hope for.
A persistent worry about unconditional cash is that recipients will waste it. The data consistently proves otherwise. Across vastly different contexts — a California city and rural African villages — the pattern holds: food, shelter, health, education.
In Stockton: less than 1% went to alcohol or tobacco.[2] In India: no increase in alcohol consumption was recorded.[6]
Stockton SEED — how recipients spent $500/month
Based on debit card transaction data from the SEED evaluation.[2]
| category | Stockton, USA (urban) | rural villages (India / Kenya) |
|---|---|---|
| Food & groceries | ~37% — healthy food, groceries | ≈60–70% — hunger relief, nutrition |
| Housing & utilities | ~11% utilities + part on rent/home goods | 10–15% — shelters, clean water, fuel |
| Healthcare | <5% — basic needs took priority | ~5% — doctor visits, medicine |
| Education | <5% — some school supplies or tuition | 5–10% — fees, uniforms, books |
| Debt repayment | 52% → 62% making regular debt payments | ~5–10% — 73% reduced debt burden |
| Savings & investment | Some saved for future; took career risks | 10–20% — livestock, tools, micro-enterprises |
| Alcohol / tobacco | <1% — less than pre-UBI baseline | No measurable increase |
§ 6 — context matters
Developed vs. developing:
two different interventions.
One clear pattern across all experiments: the same intervention has vastly different effects depending on baseline conditions. UBI in a wealthy country solves a different problem than UBI in a low-income one.
High-income countries
Finland · Canada · Stockton · Ontario · Alaska
- —Basic needs largely covered by existing welfare or income
- —UBI adds psychological security, reduces stigma and bureaucracy
- —Modest material gains; meaningful mental health improvements
- —Employment effects are neutral; UBI enables career pivots
- —Finland did not lift people from absolute poverty — it made them happier
- —Ontario: 80% felt health improved, 74% ate healthier
Low-income countries
Namibia · Kenya · India · Uganda
- —Cash addresses a shortage of goods, not just security
- —Even a small basic income unlocks cascading change
- —Dramatic improvements in nutrition, school attendance, crime
- —Recipients invest in tools, livestock, micro-enterprises
- —Poverty, malnutrition, and dropout rates fell by double digits
- —The same dollar goes much further — and buys back dignity
Ontario: what happens when you take it away
Ontario's pilot was canceled by a new government in 2018 before it could complete. A follow-up survey of participants painted a stark picture of what happens when stability is yanked back.[9]
§ 7 — limitations
What the experiments
can't yet tell us.
The evidence is promising — but it is also partial. Most experiments were small, short, and isolated from the broader economy. A national UBI at scale would introduce dynamics these pilots simply cannot test.
Scale & macroeconomic effects
Most treated groups numbered a few thousand, isolated from the broader economy. A national UBI could cause price inflation (especially housing), alter wage bargaining, or generate fiscal pressures that small pilots cannot reveal.
Duration
Most experiments ran 1–3 years. Long-term behavioral shifts — both positive compounding and potential negative habits — may only emerge over decades. The Kenya 12-year trial will be the first real test of this.
Targeting vs. true universality
Finland, Ontario, and most US pilots targeted unemployed or low-income individuals. We don't know how higher-income individuals would respond, or how universal coverage might shift public perception and political dynamics.
No counterfactual comparison
Experiments compared UBI to the status quo — not to alternative interventions (targeted job training, childcare, healthcare). Whether UBI outperforms equally-funded targeted programs remains an open question.
Political fragility
Ontario's pilot was canceled mid-run. Finland's was time-limited by design. No country has yet run a full, permanent, universal UBI long enough to observe steady-state effects. Political viability is an outcome experiments cannot measure.
Funding mechanism effects
All pilots were donor- or government-funded without requiring direct tax increases. A real UBI funded via income taxes, carbon taxes, or benefit consolidation may alter behavior in ways not captured by these studies.
§ 8 — future research
The questions that
still need answering.
The experiments so far have answered some questions and debunked some myths. But they have also surfaced new ones. These are the research directions that would most advance our understanding.
Long-term & large-scale effects
Most pilots lasted 1–3 years. Longitudinal research is needed to see if benefits compound over time — or if any negative patterns emerge. The Kenya 12-year study will be the first real test. Macro-level simulations calibrated with micro-data from pilots can model economy-wide effects.
Fiscal sustainability & funding design
How UBI is funded matters. Funding via income taxes may affect work incentives differently than a carbon tax or sovereign wealth fund. Pilot programs that incorporate a taxation component, or hybrid models (partial UBI + existing programs), are needed to evaluate cost-benefit trade-offs.
Differential impacts across demographics
Does UBI close racial wealth gaps? Does it increase women's bargaining power at home? How does it affect youth differently from retirees? Disaggregated analysis of existing pilot data, and targeted qualitative research, would help ensure equitable policy design.
Behavioral & psychosocial dimensions
Do people become more entrepreneurial when fear of destitution is removed? Stockton anecdotes suggest yes. Future experiments should incorporate behavioral science measurements: risk appetite, locus of control, hope for the future — quantifying what qualitative reports hint at.
UBI vs. alternatives
No experiment has directly compared UBI to equally-funded targeted interventions. A trial with one group on UBI, another on conditional transfers, and a control would finally answer whether unconditional cash outperforms the counterfactual — or whether it depends on context.
Automation & labor market disruption
A major motivation for UBI is cushioning displacement by AI and automation. Small-scale studies can't replicate wave-scale job loss, but scenario modeling calibrated to existing behavioral data from UBI pilots could inform how a basic income would function during a structural employment shock.
§ 9 — conclusion
UBI can work.
But it isn't magic.
The empirical record — spanning five decades, six continents, and vastly different economic contexts — paints a nuanced picture. Basic income consistently delivers on its core promise of improving human well-being. It consistently defies its loudest critics. But it is not a cure-all, and context shapes everything.
What the evidence shows
- —Mental and physical health improves in virtually every trial
- —Mass workforce exit never materialized in any credible study
- —Recipients spend on essentials — not temptation goods
- —Education and nutrition gains are consistent in developing regions
- —Financial stability enables better long-term decisions
Where results diverge
- —Developed countries: psychological relief over material change
- —Developing countries: transformational impact on basic needs
- —Employment gains depend on design and baseline conditions
- —Generosity matters: too small = limited impact; poorly tapered = disincentives
- —Short-term pilots underestimate long-term compounding effects
What remains uncertain
- —Macroeconomic effects at national scale (inflation, wages)
- —Long-term behavioral change beyond 3-year pilots
- —Political and fiscal sustainability in real implementations
- —Whether UBI outperforms equally-funded targeted alternatives
- —How to preserve unconditionality while ensuring fiscal viability
The assumed drawbacks of UBI — laziness, misuse of funds, social collapse — did not materialize in trial conditions. The hoped-for benefits did materialize, to varying degrees, in nearly every experiment run. That is the empirical record. It provides a stronger foundation for informed policy debate than any purely theoretical argument for or against.
Whether UBI is the right tool depends on societal values, fiscal context, and implementation design. What the research shows is that, given the chance, most people use unconditional income to improve their lives — and the lives of everyone around them.
§ 10 — references
Sources & citations
Kangas, O. et al. (2019). "Effects of the basic income experiment on employment and well-being." Ministry of Social Affairs and Health, Finland. https://ec.europa.eu/social/BlobServlet?docId=20846&langId=en
West, S. et al. (2021). "Preliminary Analysis: SEED's First Year." Stockton Economic Empowerment Demonstration. https://www.stocktondemonstration.org/press-landing/guaranteed-income-increases-employment-improves-financial-and-physical-health
Forget, E. (2011). "The Town with No Poverty: Using Health Administration Data to Revisit Outcomes of a Canadian Guaranteed Annual Income Field Experiment." https://www.bbc.com/worklife/article/20200624-canadas-forgotten-universal-basic-income-experiment
Namibia BIG Coalition (2009). Learning from the Pilot Project. BIG pilot evaluation report. https://centreforpublicimpact.org/public-impact-fundamentals/basic-income-grant-big-in-namibia/
GiveDirectly (2023). Early findings from the world's largest UBI study. Kenya UBI results. https://www.givedirectly.org/2023-ubi-results/
SEWA / UNICEF (2013). India Basic Income Pilot, Madhya Pradesh — project evaluation report. https://degrowth.info/blog/universal-basic-income-in-india-a-promising-experience
Salehi-Isfahani, D. & Mostafavi-Dehzooei, M. (2017). "Cash Transfers and Labor Supply: Evidence from a Large-Scale Program in Iran." Economic Research Forum. https://www.weforum.org/stories/2017/05/iran-introduced-a-basic-income-scheme-and-something-strange-happened/
Marinescu, I. (2018). "No Strings Attached: The Behavioral Effects of U.S. Unconditional Cash Transfer Programs." Roosevelt Institute. https://earth4all.life/views/the-alaska-permanent-fund/
Basic Income Canada Network (2019). Signposts to Success. Survey of Ontario pilot participants. https://maytree.com/publications/critical-lessons-from-ontario-about-how-to-set-up-a-basic-income-experiment/
Banerjee, A. et al. (2017). "Universal Basic Income in the Developing World." Conference paper summarizing global evidence. https://www.theigc.org/blogs/do-cash-transfers-make-poor-work-less
McKinsey Global Institute (2020). "An experiment to inform universal basic income." https://www.mckinsey.com/industries/social-sector/our-insights/an-experiment-to-inform-universal-basic-income
Haushofer, J. & Shapiro, J. (2016). "The Short-Term Impact of Unconditional Cash Transfers to the Poor." Quarterly Journal of Economics. https://www.povertyactionlab.org/evaluation/effects-universal-basic-income-during-covid-19-pandemic-kenya
MIT Sloan (2020). "How a universal basic income stabilized Kenyans in bad times." https://mitsloan.mit.edu/ideas-made-to-matter/how-a-universal-basic-income-stabilized-kenyans-bad-times
The Health Benefits Of Universal Basic Income. PMC / National Center for Biotechnology Information. https://pmc.ncbi.nlm.nih.gov/articles/PMC8249026/
Universal basic income pilots — comprehensive overview. Wikipedia. https://en.wikipedia.org/wiki/Universal_basic_income_pilots
IPR Blog, University of Bath (2022). "Universal basic income experiments: The state of play." https://blogs.bath.ac.uk/iprblog/2022/01/12/universal-basic-income-experiments-the-state-of-play/
Finland's Basic Income Experiment 2017–2018 Results. Finland.fi official infographic. https://toolbox.finland.fi/life-society/infographic-finlands-basic-income-experiment-2017-2018-results/
The Guardian (2020). "Finnish basic income pilot improved wellbeing, study finds." https://www.theguardian.com/society/2020/may/07/finnish-basic-income-pilot-improved-wellbeing-study-finds-coronavirus
Note: This meta-analysis draws on a combination of peer-reviewed academic studies, official government evaluation reports (e.g. Kela/Finland, Ministry of Social Affairs), and credible institutional sources. Non-peer-reviewed sources (government reports, pilot surveys, NGO evaluations) are indicated as such in the citations above. Cross-verification with primary studies was performed where possible.
stay updated
support us
we are a registered Canadian non-profit.
donation information is soon to come.