Exploring the Distributional and Work Incentive Effects of Plausible Illustrative Basic Income Schemes

Among the many concerns about universal basic income (UBI or basic income), two of the most commonly expressed are that it would have undesirable distributional consequences (by failing adequately to compensate the recipients of withdrawn benefits) and that it would erode work incentives (via higher tax levels). This paper addresses these concerns, building on our previous working paper The Fiscal and Distributional Implications of Alternative Universal Basic Income Schemes in the UK in two main ways:
• By exploring the distributional consequences in greater depth
• By exploring outcomes of UBI schemes in relation to static financial work incentives
We examine three schemes, argued to be the most plausible of those modelled in the previous working paper, and pitched at three levels of generosity. These are:
• UBI set at the level of the tax saving implied by personal income tax allowance (PITA)
• UBI set at the level of existing benefits
• UBI set at the level of existing benefits with premiums for individuals determined as disabled or severely disabled
This report makes a number of original contributions to the literature. In terms of the distributional consequences of the schemes modelled here, we determine the proportions of households in different income and demographic groups that would expect to gain or lose out financially. The proportions of households gaining and losing from reform is important, as positive distributional effects on aggregate can mask significant losses for some households, including the most vulnerable. Another major contribution is that we analyse the schemes’ distributional consequences with respect to the characteristics of disability status and gender at the individual and household levels, going beyond the ‘standard’ distributional categories presented in existing studies.
Our main findings suggest that when we pay for a UBI by withdrawing a large number of benefits and increasing payroll taxes, large numbers of households will inevitably experience significant losses of income. More significantly, despite the generally progressive character of the schemes modelled here, these losses are not concentrated among richer groups; on the contrary, they are proportionally larger for the bottom three income quintiles. While the specific patterns of winners and losers varies with the details of each scheme, when we eliminate the mainstay of means-tested support in line with the UBI payment, women lose out on aggregate compared to men – and disabled people lose out compared to non-disabled people unless additional premiums are paid on top of the uniform UBI. These are important concerns that anyone hoping to design ethically desirable and politically feasible UBI schemes need to address.
Turning to our contribution in relation to financial work incentives, we construct indicators of ‘participation tax rates’ (PTRs) and ‘marginal effective tax rates’ (METRs) which describe financial incentives to work at all, and financial incentives to progress in work or increase work effort marginally, respectively. We also construct indicators of the proportions of households facing improved, deteriorating or unchanged financial work incentives as a result of the reforms.
We find that on average, PTRs and METRs increase as a result of all three illustrative basic income schemes, a consequence of tax increases and the elimination of the personal allowance. However, this does not mean that the schemes modelled here would necessarily have negative consequences with respect to labour market participation. We find that the lower income quintiles, workless households, and households in receipt of at least one means-tested benefit tend to contain larger proportions of households facing improved PTRs. It is highly plausible that the effects of stronger work incentives on particularly sensitive groups may outweigh the more generalised effect of weaker work incentives over the wider population.