Syndicate content

Being open-minded about universal basic income

Ugo Gentilini's picture

In a world riddled with complexity, the simplicity of universal basic income grants (BIGs) is alluring: just give everyone cash. Excerpts of such radical concepts have been put in practice across the globe, with the launch of a pilot in Kenya, results from India, a coalition in Namibia, an experiment in Finland, a pilot in the Unites States, a referendum in Switzerland, and the redistribution of dividends from natural resources in Alaska and elsewhere.

In high-income countries (HICs), the main rationale for BIGs is related to automation, artificial intelligence, stagnant real wages, and the rise of the ‘precariat’. While the alarm bells that ‘the machines are taking over’ have rung for centuries, advances in robotics may keep slicing labor-capital ratios in large-scale industries and a variety of services. Machines, the argument goes, can take the jobs, but should not take the incomes: the job uncertainty that engulfs large swaths of society should be matched by a welfare policy that protects the masses, not only ‘the poor’. Hence, BIGs emerge as a straightforward option for the digital era – one seemingly backed by Silicon Valley and trade unions alike.

In developing countries, conversations are of a different nature. Most social protection systems in middle-income countries (MICs) are still relatively new, and have yet to undergo the centuries of societal struggles and bargains that cemented welfare regimes in advanced economies. But investments in social protection in MICs seem to occur at comparatively earlier levels of development. For example,  mid-19th century poverty levels in the UK and the United States resemble those in India today, sans an equivalent of NREGA or similar rights-based programs – and certainly no biometric Aadhar platforms to help deliver assistance. A BIG in a MIC might still be a long shot, but not that far off from a historical perspective. Iran, for instance, even implemented a variant of BIGs, which rapidly eroded due to inflation and eventually scaled-back for fiscal constraints.

In low-income countries (LICs), social protection systems are truly at their infancy, and have only recently replaced decades of international humanitarian approaches. Here only one-tenth of those living in the poorest quintile is covered by social assistance. Limited finance is no small reason for such performance: pervasive informality and low government revenues (often below 15 percent of GDP) hinder domestic resource mobilization for, among other things, social spending. For example, the budget of a European hospital is almost nine times greater than a LIC’s average spending for the whole social protection sector (i.e., $200 million). A sudden introduction of BIGs in a LIC may be closer to moon-shooting than leap-frogging.

Moreover, there are disputes over the definition and measurement of poverty in contexts where virtually ‘everybody is poor’; targeted approaches have helped reduce the inefficiencies of old-fashioned subsidy programs (which pursued objectives similar to those of a  BIG), but there is a lively debate on the trade-offs and methods of targeting (e.g., see here and here); also, the political economy of redistribution has been largely underexplored.

A unifying constraint in countries across the income spectrum is red tape. Social assistance is underpinned by processes of targeting, application, eligibility verification, registration, recertification, and monitoring – with a vast set of programs sometimes having their own individual processes. These are often introduced for good reasons – such as prioritizing the most vulnerable – but at times complexity and opportunity costs may stifle assistance instead of enhancing it. BIGs may not waive some of those functions, especially in the early phases of introduction (e.g., identification, registration, and recertification); but no doubt the bureaucratic burden would be lessened overall, coming as relief for the bandwidth of poor people.

Importantly, BIGs are a melting pot with many variants leading to dramatically different schemes depending on how they are conceived and designed. For example, arguments have been made that less bureaucracy and more predictability may play in favor of increased work efforts, not less. But ultimately, much of the discussion around labor incentives – as well as sustainability – may boil down to the size and objectives of assistance: should BIGs be large enough to meet basic standards of living as the Suisse? Or should they replace most of the current public programs, thus becoming a kind of voucher, to shop for assistance of choice? Or should they provide a more modest amount of cash, which substitutes for some similar transfer program but that is still provided alongside other public interventions?

These questions should not be addressed in isolation, but organically. Because of their foundational nature, BIGs offer a clean slate to revisit system-wide issues: this means, among others, considering BIGs within an optimal composition of social assistance and insurance (e.g., BIGs are similar to social pensions for seniors, although extended to those above 18 years of age); connecting the discussion to the flexibility and formality of labor markets; and anchoring the narrative to the financing and tax discourse (e.g., Friedman’s version of a BIG was indeed a negative income tax).

Over the next two years, we have a great opportunity to unpack the rationale, contexts, and analytical and practical implications of BIGs as an option on a policymakers’ social assistance menu. As the Safety Nets Global Solutions Group in the World Bank, we are excited to contribute to this agenda, and look forward to collaborating and engaging with all those interested in the matter.

To get in touch: and


Submitted by Nithin on

We need to start being a little more forward looking in terms of evolution of welfare state in our work. I am pleased to see you bring the issue of nature of work and its consequences for the welfare state. Universal income is one simple - albeit probably expensive - solution. The real challenge to address the revolution in labor markets - and more appealing to me at least - is to work on a sophisticated system that uses variety of instruments such as in-work (labor policies, tax credits), direct assistance and insurance. We have a long way to go in terms of seeing the welfare state in this framework. We say we do, but practically the "resolution" of our "view" is low. Some folks are quite adept with "vanilla" type cash transfers, some are experts on pensions and others in "in-work" benefits. But I hazard to claim that very few in our practice yet have the skills to deal with tax expenditures. Therefore my sense is that we need more staff who can integrate these bits to achieve, what could be potentially a more optimal solution than a simple universal income scheme. The reason is that simplicity of a universal income scheme in a fiscally constrained world (most of it at least ) would probably generate assistance that is inadequate, and/or incentive distorting. So "Occam's razor argument" may not follow in this context. But would be interesting to watch this space and hope we all can develop further on being more expert integrators of different types of instruments for achieving social justice...

Submitted by Ugo on

Thanks Nithin, really appreciate your insightful thoughts.

Submitted by MA on

Thank you, this is very timely and many of us look forward to thoughtful analyses of the various proposals. Will there be a chance to consider how these proposals compare to cash transfers or voucher schemes?

Submitted by Laurie on

Implicit in the basic income idea is that this is a transfer of cash so that individuals and households will spend basic incomes in the market on the basis of their own preferences. Sounds good in principle, but the idea has different meanings. For example egalitarians see it as a route to gender equality and eliminating income inequalities; libertarians on the other hand view basic income as a way of reducing the state and 'burdens' associated with labour market regulation. Which ever 'lens' one chooses, there are two fundamental challenges that cut across all countries irrespective of level of per capita income (1) basic income is a strategy for distribution without a strategy for production whereas existing social security systems in MICs and HICs - and safety nets in LICs - are expected to complement labour market participation and not to replace it; (2) markets do not work well, even in high income countries( and not all social risks - such as the need for long term or child protection measures - are amenable to social insurance instruments. Thus there are are least five policy areas, in HICs, where distributing cash may not suffice: health, social care, housing and food. For health and social care it is widely accepted that a public system is needed, and needs to be financed by taxes or social insurance by the state. Likewise housing is poorly served by the market with high and low rents which generates exclusion through shortages of supply. This calls for state intervention beyond a basic allowance for housing costs. In LIC settings, market imperfections are numerous. Thus any consideration of basic income has to be considered alongside the broader provision of public goods. The idea of universal basic income hold much appeal, but is not obvious that it will give a much better outcome than improvement and expansion of existing social protection systems. A more worthwhile endeavour might focus on productive employment, the relationships between taxation and benefits (e.g., EUROMOD), and pre-distribution programmes( education and TVET, early child development, child protection, and long term care for the growing elderly population). Such an approach might help the World Bank move away from its traditional SP silos (pensions, cash transfers) and being to look at holistic systems of social protection. This where the need for fresh thunking and new approaches is greatest.

Add new comment