One fifth of workers would choose to stay in their job even if offered up to 120% of their salary as a guaranteed income, according to a new paper by Julian Jacobs of the University of Oxford and Google DeepMind. The survey behind it covered 8,234 working-age adults in the UK, US, Denmark and India.
Only 26% of respondents value their job mainly for pay and benefits, the paper says. Yet around 75% say they would leave their job if offered 100% of their current wage, about three times as many, by our arithmetic.
Short explainer · 1 in 5 Would Turn Down 120% of Pay
A vertical news Short summarising this story. Figures come from Julian Jacobs’s paper on arXiv; our arithmetic is labelled.
Business Pill · AI Agents Explained
An evergreen Business Engineer explainer on AI agents and the work they automate. It teaches the general idea only and does not discuss this paper.
The key insight: In Julian Jacobs’s survey, about three times as many workers would quit at full wage replacement as say they work mainly for pay, by our arithmetic, while one in five would turn down even 120%.
The paper sorts workers into four “Work Values Profile” groups. It says one third carry an Intrinsic motivation, reporting “a sense of purpose, learning, or a feeling of societal contribution” from their work. Nearly one quarter (23%) have a Community orientation, 17% are Status-Driven, and 26% are Transactional.
It also reports that 81% of respondents said they were satisfied with their jobs. The survey was fielded in June and July of 2026 by the polling organisation More in Common, and the paper says the data collection was supported by Google DeepMind.
Julian Jacobs’s survey of 8,234 working-age adults in the UK, US, Denmark and India: what workers value most in their job, and the share who would keep working even at 120% of their salary as guaranteed income. Figures from the paper; chart by FourWeekMBA.
The Basic Income Test
Respondents were asked, step by step from 60% to 120% of their current salary, what guaranteed income would lead them to quit. The paper says a slight majority would leave by 80% wage replacement, around 65% by 90% and around 75% at 100%.
The 20% who refused every offer up to 120% was “roughly consistent across the incomedistribution”, the paper says: 21% for low-income, 18% for middle-income and 22% for high-income workers. By country, the share was 28.8% in India, 17.9% in the US and Denmark, and 16.0% in the UK. By our arithmetic, India’s share is 1.8 times the UK’s; the paper notes the India sample was online-only.
Intrinsically motivated workers demanded a median guaranteed income of 80% of salary to quit, against 70% for Transactional workers, according to the paper. Status-Driven workers needed the least, with a median of 60%.
The Business Lever: Pricing the AI Labour Transition
The paper argues that “Guaranteeing a basic income may indeed be necessary for managing any future AI labor shock, yet it could ultimately fail to compensate workers for any loss of the non-monetary benefits they derive from work”. It says these non-monetary motivations are evenly spread across income levels and between white- and blue-collar workers.
It also finds that people who think AI is more likely to automate their job demand a lower wage replacement, by two to four percentage points per standard deviation of perceived threat. The paper says the model “cannot infer the directionality of this relationship”.
As we read the paper, the gap between why people say they work and what would make them stop is the core finding: most value work for more than pay, yet most would still leave at full pay.
As we read it, the one-in-five holdouts matter because the paper finds them at every income level, so any loss of what they get from work is not confined to one group.
As we read it, the evidence is stated preference: answers to a hypothetical guaranteed income, gathered by More in Common and supported by Google DeepMind.
Julian Jacobs, arXiv abstract, 29 Oct 2026
“Crucially, these non-monetary work motivations are evenly distributed across the incomedistribution and between white- and blue-collar workers.”
Three Implications
FOR POLICYMAKERS The paper argues a basic income may be necessary for an AI labour shock but could fail to compensate workers for the non-monetary benefits of work.
FOR EMPLOYERS The paper finds 81% of respondents said they were satisfied with their jobs, and a majority are strongly motivated by non-monetary parts of their work.
FOR AI LABS The paper says people who see a higher AI threat to their job demand a lower wage replacement, though it cannot say which way the link runs.
The framework says “Full-range cognitive workers earn a structural premium that automation cannot touch.”
As we read it, the paper adds a non-wage side to the labour question: what people lose from work is not only pay. The framework is the owner’s argument and does not discuss this paper or its figures.
What Is Not Established
The figures come from stated answers to a hypothetical offer, not from people actually given a basic income. The survey’s auction stops at 120%, so the paper cannot say at what level the 20% of holdouts would leave. The paper reports that its views do not necessarily reflect those of the University of Oxford or Google DeepMind. It is a preprint posted on arXiv.
This Week In The AI Supercycle
Anthropic’s Reality Check, The Capital Squeeze, & The New World Order
This week’s issue: for two years the AI build-out was priced as a growthstory. This week it started being priced as a credit story.
A survey of 8,234 adults in four countries finds that about a quarter work mainly for pay, around 75% would still quit at full wage replacement, and one in five would keep working even at 120%. The paper’s author concludes that a basic income alone may not replace what work provides.
A note on sourcing. All facts and figures come from Julian Jacobs’s paper on arXiv (29 October 2026) and its abstract, read on 31 October; the survey results are the author’s. We did not contact the author, the University of Oxford or Google DeepMind.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
