What Is Universal Basic Income? Paying Everyone, With No Conditions

Four features distinguish the proposal from an ordinary benefit, and dropping any one of them produces something different:

• •Universal: paid to every individual regardless of income or wealth, which removes the administration of means testing and the stigma attached to claiming
• •Unconditional: not tied to seeking work, training, health status or behaviour, which removes both the sanctions regime and the bureaucracy that runs it
• •Individual: paid to a person rather than a household, which matters because household-based benefits can trap a dependent partner
• •Regular and in cash: a predictable monthly payment rather than a one-off grant or an in-kind provision, so recipients can plan and choose
• •Variants that relax these include a negative income tax, which tops up low incomes and tapers away, and a guaranteed minimum income, which is means tested; both achieve some of the same ends at lower gross cost and reintroduce some of the same bureaucracy

The unusual feature of the idea is that it attracts support from opposite ends of the spectrum for incompatible reasons. From the left, it is a floor below which nobody falls, a recognition of unpaid care work, a source of bargaining power for workers who can refuse bad jobs, and a way to abolish the punitive conditionality of modern welfare systems. From the right and among libertarians, it is a replacement for a sprawl of overlapping programmes, each with its own bureaucracy, rules and cliff edges, with a single transparent payment that treats people as capable of deciding what they need; Milton Friedman's negative income tax proposal came from exactly this direction, and Charles Murray has argued for replacing the entire welfare state with a single grant. The two coalitions agree on the mechanism and disagree completely on whether it should supplement or abolish existing provision, which is why the idea polls well in the abstract and struggles when specified.

Three serious criticisms recur. The first is cost: a payment large enough to live on, given to everyone, is an enormous gross expenditure, and although the net cost is much lower once it replaces existing benefits and is clawed back through tax from higher earners, the tax rates implied in most costed proposals are substantially above current levels. The second is work incentives, the concern that a guaranteed income reduces employment. The third, made from the left, is that a universal flat payment is a worse use of money than targeted support, because a disabled person, a carer and a person in an expensive city all have different needs, and a single sum either fails them or is set so high it becomes unaffordable; on this view the policy risks being used as cover for dismantling services that cost more and do more. A fourth practical objection is that it does nothing about housing costs, which in many places would absorb the payment through rent increases.

A body of evidence exists, with the important caveat that almost none of it tests a genuinely universal permanent scheme, since trials are temporary, local and usually targeted, which changes behaviour. The American and Canadian negative income tax experiments of the 1970s found small reductions in hours worked, concentrated among second earners and young people staying in education longer. Finland's two-year trial ending in 2018 gave unemployed people an unconditional payment and found no significant effect on employment either way, alongside clear improvements in reported wellbeing, trust and mental health. Kenya hosts the largest and longest study, run by the charity GiveDirectly across many villages with a twelve-year horizon, and its published results show increased enterprise, investment and consumption without the increase in alcohol spending critics predicted. Cash transfer programmes across the developing world consistently show the same pattern, and a widely cited review found no systematic evidence of spending on alcohol or tobacco rising. The honest summary is that unconditional cash works better than sceptics expect and that no trial has answered the question of what a permanent national scheme would do to labour supply and prices.

Two long-running schemes come closest. The Alaska Permanent Fund has paid every resident an annual dividend from oil revenue since 1982, ranging from a few hundred to over two thousand dollars, and studies have found no reduction in aggregate employment, with a small rise in part-time work. Iran replaced fuel subsidies with a large cash transfer to households in 2011, effectively a national basic income, and analysis found no negative effect on labour supply. Macau pays an annual sum from gambling revenue. The common feature of the schemes that survive is a dedicated revenue source that is not general taxation, which removes the fiscal argument, and this is also why proposals to fund a basic income from resource rents, carbon taxes, land value or data dividends recur. Interest has risen with concern about automation, and it rose sharply during the pandemic, when several countries made large unconditional payments in an emergency and demonstrated that the administrative machinery to do it exists.

A basic income is universal, unconditional, individual and regular, which removes means testing, sanctions and the associated bureaucracy, and it attracts support from the left as a floor under everyone and from libertarians as a replacement for the whole welfare apparatus. Objections centre on gross cost, work incentives and the fact that a flat payment ignores differing needs. Trials in Finland, Kenya, the United States and Canada find at most small reductions in hours worked and clear wellbeing gains, and the durable real schemes, in Alaska and Iran, run on dedicated revenue rather than general taxation.