From: allin
A Universal Basic Income (UBI) experiment, funded in part by Sam Altman, was conducted by Open Research, a nonprofit group founded in 2015 out of Y Combinator [00:22:57]. Sam Altman contributed 60 million total to this experiment [00:22:53].
Experiment Design and Participants
The experiment took place between November 2020 and October 2023 [00:23:11]. It involved 3,000 low-income adults in Texas and Illinois [00:23:14]. These participants made an average of just under $30,000 per year [00:23:16].
- Treatment Group: 1,000 participants received 12,000 annually, representing nearly a 50% increase in pay for doing no more work [00:23:27].
- Control Group: 2,000 participants received $50 per month over the same period [00:23:36].
Researchers collected various data points, including blood draws for health impact, tracked time usage (work, play, etc.) via a custom app, and checked credit reports and bank balances [00:23:41].
Key Findings
The research found “almost no lasting impact on everything they tested from overall health to work to education” [00:23:56].
Key quotes from the paper include:
- “The cash transfer resulted in large but short-lived improvements in stress and food security” [00:24:09].
- “We find no effect of the transfer across several measures of physical health” [00:24:14].
- “We also find that the transfer did not improve mental health after the first year, and by year two we can again reject very small improvements” [00:24:17].
- “We also find precise null effects on self-reported access to healthcare, physical activity” [00:24:28].
Additional specific findings:
- Participants were 5% more likely to start a business by the third year [00:28:17].
- People worked slightly less, with a 2% decrease in labor participation, which was considered negligible [00:28:24].
- People in their 20s showed a 2% increase in enrolling in postsecondary education, also a very tiny impact [00:28:31].
- Major benefits to stress and mental health were observed in year one, but by year two, these reverted to the baseline [00:28:37].
- The study showed that people who received the 45,000, while the control group (receiving 51,000 [00:30:47]. This suggests the control group “genuinely bettered themselves over the course of the study” [00:31:06].
Discussion and Implications
The results reinforced the intuition that UBI might not achieve its desired long-term effects. Happiness levels and health levels tend to “mean revert” to their natural state, even after significant physical or financial changes [00:24:54]. In the absence of purpose and community, such transfers may not meaningfully shift life outcomes [00:25:07].
Arguments Against UBI
- Normalization of Spending: The initial happiness from increased income normalizes, leading to different or increased spending, as human desire for improvement is innate and shifts thresholds [00:25:57].
- Economic Impact: Providing 350 billion/month or 7.3 trillion [00:27:04]. The study did not address the mechanism for funding UBI at scale [00:27:16].
- Inflationary Effects: Government-driven money stimulation can lead to inflationary effects, as seen during the COVID-19 pandemic with various financial market bubbles and aggregate inflation (e.g., food prices still up 30-40% since COVID) [00:27:25].
- Demotivation from Work: Just handing people money without requiring work does not motivate them to work harder; it can lead to less work [00:31:15]. This could remove the “bottom rungs of the ladder” where people gain important skills and work ethic [00:31:29]. Historically, cash transfers like those in the Great Society or COVID stimulus checks did not solve poverty and could trap people in dependence [00:29:17], leading to increased spending on leisure, alcohol, and meme stocks rather than productive activities [00:29:45].
- “Virtue Signaling”: Some view UBI proposals as virtue signaling, potentially creating dependence and serving the interests of tech moguls and government officials rather than society [00:30:03].
- Against Human Nature: The belief is that humans are fundamentally driven to find purpose and progress, and a UBI model where machines do everything for them may not lead to happiness. The example of “nepo babies” who are often miserable despite not having to work was cited [00:38:18].
Comparison to Existing Welfare Programs
The U.S. welfare budget was 12,000 per person annually, similar to the UBI study’s amount [00:32:51]. While welfare can be essential, the discussion emphasized that it should ideally be tied to motivation and work, as experienced by individuals who grew up on welfare but were still motivated to work hard [00:33:51].
Alternatives and Future Considerations
Instead of UBI, alternative ideas or improvements to social support include:
- Investment/Business Funding: A “2.0” study could explore putting UBI money into a “perfect portfolio” or a “business formation fund” with educational components, allowing people to take out a sustainable amount annually [00:36:19]. However, existing mechanisms like Roth IRAs and 401ks already offer tax breaks for saving and investing [00:36:55].
- Retirement Accounts at Birth: An interesting idea is to give every child a retirement account at birth, maturing into an index fund that unlocks at age 65-70 [00:39:18]. This provides a safety net without removing motivation to work during prime years [00:39:44].
- Vocational Skills and Education: Focusing on teaching vocational skills, how to get a job, build a business, and succeed in the workplace is seen as more beneficial than simply providing cash [00:41:15].
- Conditional Support: Providing funds conditional on pursuing socially beneficial activities, like paying someone $1,000/month and for school to become a nurse, could address societal problems while fostering motivation [00:41:44].
The discussion also touched on the potential impact of AI on jobs, with the view that AI-based tools and automation, like previous technological advancements (e.g., tractors), would lead to new industries and jobs rather than eliminate the need for human work [00:37:38].