UBI-eh?

Dr Joe Chrisp is a Research Associate in the Institute for Policy Research (IPR) at the University of Bath. Professor Nick Pearce is Director of the Institute for Policy Research (IPR) and Professor of Public Policy at the University of Bath. Prof Matteo Richiardi is a Professor in Economics & Director of CeMPA, University of Essex. They are the authors of a new paper ‘UBI-eh?’. This report analyses the cost and distributional consequences of proposed incremental reforms to the UK social security system.

Proposals for Universal Basic Income (UBI) have risen in political salience in the last decade, in both OECD countries and the Global South. According to the definition of the Basic Income Earth Network, a UBI is ‘a periodic cash payment unconditionally delivered to all on an individual basis, without means-test or work requirement’. UBI is advocated as a solution, variously, to: poverty and destitution, dependence on precarious employment and labour exploitation, the impact of automation on labour markets, and unsustainable environmental resource extraction. It is motivated by normative concerns ranging from ‘real freedom’ to republican liberty, equality and sufficiency.

Policymakers, community activists and advocates of UBI have increasingly used pilot projects to generate empirical evidence on the effects of basic income schemes, and to pragmatically experiment with their administration and implementation. Many of these schemes are ‘UBI-like’: that is, they meet one or a number, but not all of the characteristics of a UBI as defined by the Basic Income Earth Network. Typically, they are unconditional with respect to job search requirements and sanctions but not in terms of targeting, whether by income or status. This pragmatism and flexibility with respect to the precise definition of UBI is typical of the ways in which the policy is understood and advocated in electoral politics.

Scholars have undertaken several microsimulations of the costs and benefits and distributional consequences of UBI proposals in different national contexts. These microsimulations can be used to quantify the costs, distributional effects on different households, and impacts on poverty and inequality of UBI schemes. In a new report published this week, we pursue an alternative strategy. Instead of modelling a UBI scheme that could be introduced as a single or ‘big-bang’ reform, we consider a series of incremental, pragmatic reforms that might be considered steps and stages on the road to a UBI as well as worthwhile in their own right.

This approach reflects research into the political economy and politics of UBI, which finds that advocacy for UBI is highly multi-dimensional, dependent on context, and largely pragmatic when faced with institutional and electoral constraints. Political parties that express support of UBI tend to portray it as a long-term goal, particularly when those parties have a reasonable chance of governing, and thus combine UBI advocacy with a series of more modest reforms pitched as steps towards a UBI or, at the very least, consistent with the vision and principles of the policy. These typically include the removal or relaxation of various forms of conditionality attached to benefits, whether behavioural conditions such as job search requirements or means testing. However, steps towards a UBI also often include raising the level of minimum income benefits, the simplification or integration of benefits or the increased capacity to combine benefits with earnings among a wide variety of individual reforms.

Such a perspective also aligns with other proposals in the UK policy context from the Joseph Rowntree Foundation and Trussell Trust for an ‘Essentials Guarantee’, the New Economic Foundation’s National Living Income proposal and the shift by the Scottish government from exploring the feasibility of a basic income pilot to the establishment of a Minimum Income Guarantee research group. Although in the abstract such proposals stand in stark contrast to the universal nature of a UBI, they are very similar to both the kinds of manifesto commitments found in UBI-supporting party documents and the policies tested in so-called basic income pilots.

Thus, the motivation of these microsimulations is to explore the cost and distributional consequences of such policy proposals, including examining individual steps that are relatively modest in nature and a combined or cumulative approach that would constitute considerable movement in the direction of less conditionality within the UK social security system. While these reforms would maintain Universal Credit and household means testing as a core part of the social security system, they would greatly increase the level of support to low-income households and ease the burdens associated with many of the rough edges of the existing work-first benefit system.

In the report, we use UKMOD - an open-source tax-benefit microsimulation model for the UK - and the 2019 Family Resources Survey data uprated to account for distributional income changes since then in combination with policy information from 2022-23. We start by auto-registering all resident households in the Universal Credit system even if they are not eligible to receive a transfer. We model this by increasing take-up to 100% and take this as a presumptive first step for all future reforms. As part of this policy reform, we therefore also assume the removal of sanctions such that households cannot be denied access to the benefits they are entitled in reference to the means test. Based on estimates of existing take-up, this would cost £12.8bn and all subsequent calculations of cost and tax changes were done in reference to a baseline of 100% take-up. It is worth stating that absent of achieving the goal of 100% take-up, the other costs are likely to be an overestimate but equally so will the reductions in poverty.

The next step is to finish the migration to Universal Credit from legacy benefits. This is in effect the final step of auto-registering all resident households in the Universal Credit system and indicates the cost and distributional consequences of doing so even for those currently receiving legacy benefits.

The next reform is to remove wealth and savings from the household means test in Universal Credit. The current situation means that any capital or savings – excluding primary housing – above £6k reduces the amount of Universal Credit you receive up until savings of £16k at which point you are entirely ineligible. We then reduce the taper rate to 40% and thus make Universal Credit more universal, reducing marginal effective tax rates for those already receiving it and extending receipt of the benefit up the income distribution. (Of course, the flipside is this would increase marginal effective tax rates for those not currently receiving who would then now be recipients).

We also model increasing the standard allowance in Universal Credit to £120 a week for all single people and £200 a week for all couples. The status quo was £265.31 a month (c.£61.23 a week) for single people under 25 and £334.91 a month (c.£77.29 a week) for single people aged 25 and over. For couples both under 25 the standard allowance was £416.45 a month (c.£96.10 a week), while couples with at least one member aged 25 and over receive a standard allowance of £525.72 a month (c.£121.32 a week). The levels were increased by 10.1% in line with inflation for 2023-24 but this is not part of our analysis.

We then remove the High Income Child Benefit Charge for households with children where one adult earns over £50k. This would reinstate the universality of the benefit so that all households with children would receive it and there would be no increase in marginal tax rates for those earning between £50k and £60k. We also model abolishing the two-child limit in Universal Credit. The current policy restricts the amount of support given to households with more than two children within a household. Abolishing the limit would make payments adjust for the additional needs larger families have and increase support for such households with a low income. Finally, we model an increase the level of Child Benefit for all children to £30 a week. The status quo was that the level for the eldest child was set at £21.80 a week and £14.45 a week for all subsequent children. As with the standard allowance in Universal Credit, benefit levels were increased by 10.1% in line in inflation for 2023-24 but this is not part of our analysis.

In the full report, we show the effect of these reforms on poverty rates. Prior to any intervention the poverty rate is 17.99%, which falls considerably to 16.22% after we assume full take-up when households are auto-enrolled, and conditionality is removed. The current child poverty rate is much higher at 25.25% and the effect of full take-up is even larger, reducing child poverty to 23.15%. While elderly poverty is much lower in the status quo at 13.88%, full take-up has the most dramatic effect reducing the rate to 10.17%.

The subsequent individual reform that most reduces poverty is the increase in the standard allowance. Implementing this would reduce overall poverty rates to 14.5% (a reduction of 1.72 percentage points) and child poverty rates to 20.33% (-2.82pp). Increasing Child Benefit is the next most effective policy reform for reducing poverty, particularly in the case of child poverty. Overall poverty rates fall to 15.39% and child poverty rates fall to 20.61%. Unsurprisingly, the least effective poverty reduction strategy is the removal of the High Income Child Benefit Charge. This actually increasesoverall poverty rates by 0.01 percentage points to 16.23% and only reduces child poverty by 0.06 percentage points to 23.09%. The removal of the savings means test is also underwhelming from a poverty reduction perspective leading to a 0.06 percentage point reduction in overall poverty rates and only 0.04 percentage point reduction in child poverty. (A reminder here that all subsequent measures of poverty after a policy reform assume full take-up). Nonetheless, both reforms may be considered important from the perspective of universality.

We acknowledge some important limitations in our approach. Firstly, we would need to use the latest Family Resources Survey data and indeed the 2023-24 tax and benefit system for more precise costings relevant to a General Election year. As with all straightforward microsimulation analysis, the results also show the static effects of policy changes, not dynamically accounting for behavioural change. It is unlikely in the full cumulative reform scenario that there would be no behavioural changes but we remain relatively sceptical that such changes can be predicted a priori given their complexity for different groups. Relatedly, we do not fully engage with the notion of minimum income guarantees conceived of bringing all residents to a specific level and we also do not fully account for the interactions of wages and welfare. Our analysis focuses on policy reforms alone.

Nonetheless, we believe that our paper serves to contribute (at least) two important things. First, we hope to provide a reliable evaluation of possible policy reforms that any progressive government would be interested in implementing with the aims of reducing poverty, increasing coverage and reducing bureaucratic traps. Second, we aim to reframe the debate around UBI and Universal Credit away from polar opposition towards consideration of the ways in which we can move from the existing Universal Credit system towards a more universal, unconditional and generous social security system. We believe that advocates of UBI should consider such incremental reforms of the UK welfare state as plausible and desirable, and consistent with their normative ambitions.

All articles posted on this blog give the views of the author(s), and not the position of the IPR, nor of the University of Bath.