RESOLUTION FOUNDATION – WRITTEN EVIDENCE (EUC0107)

The economics of Universal Credit

This submission reflects the expert opinions of a number of welfare experts working at the Resolution Foundation; in particular David Finch and Laura Gardiner. We also appreciate thoughts and contributions from Mike Brewer from the University of Essex, who will be Deputy Chief Executive at the Resolution Foundation from March 2020. At times this submission will draw upon past research published by the Resolution Foundation; full details of these reports are given at the end.

General evidence

Universal Credit (UC) is an ambitious reform project, rightly attempting to remove poor financial incentives and simplify the existing benefit system. Previously, some claimants could receive multiple, interacting benefits and tax credits administered by different authorities; effective marginal tax rates could reach over 90 per cent; and take-up rates of some benefits were low, particularly for those in work. However, attempting to fix these problems has, unsurprisingly, exposed some trade-offs. For example: it is now clear that having multiple sources of benefit income (as was the case for many under the legacy system) provided some level of insurance against any one payment going wrong, and the debate over the 5-week wait and the monthly nature of UC highlights the fact that there are trade-offs involved in choosing the period of assessment when designing any income-related benefit or tax credit.
However, perhaps the most important feature of any income-replacement benefit is that it is readily accessible to those without any other sources of income. UC, of course, was designed to be the safety-net benefit, but the delivery of UC has left many waiting too long for initial payments, or losing access when their needs are acute.

Looking forward, UC is beginning the process of actively migrating claimants from legacy benefits to UC; known as ‘Move to UC’. The above lessons of the past show that the most important thing is to ensure the correct and accurate delivery of this phase of UC, ahead of pressures to meet deadlines and savings targets, which ultimately will affect a wider economic cost if it is not well delivered.
The remainder of this submission looks at the Committee’s questions.

How well has Universal Credit met its original objectives?

As set out in the Department for Work and Pension’s (DWP) business case for UC, the positive economic objectives can be broadly defined as: [i]

Increased take-up of benefits.
Activation of those partially claiming benefits but not in work through additional conditionality.
Increased earnings through improved financial incentives.
Increased earnings through simplification and smoothing of benefit payments.
Operational savings from administration of legacy benefits under UC.

It is too early to conduct a full evaluation of how UC performs against these criteria, primarily because the number of recipients of UC is still a long way below the expected steady state, and those who are receiving it now are not representative of the eventual caseload. Until UC is fully rolled out, and any behavioural changes have had time to take full effect, we will not be able to evaluate the complete effect of UC. However, we can consider emerging evidence and whether this makes achieving the objectives more or less likely.

For example, research by the Resolution Foundation in ‘The long and winding road’[ii] shows that claimants are having difficulties with: managing the payment timeliness of UC, the interaction with their earnings pay cycle, managing their own housing benefit payment, having to wait five weeks for their initial payment, and managing their UC account online. Clearly, these issues were not all a part of the initial design, and their existence questions whether far from smoothing household incomes and providing certainty, that UC has for many done exactly the opposite. On the other hand, UC is still new for many claimants and some of these issues are confined to the start of a claim – a process that some will not have to repeat as, unlike legacy benefits, their UC claim stays with them as they move into and out of work.

The only clear quantitative evaluation of the impact on labour market outcomes [iii] shows that a small subset of unemployed UC claimants were marginally more likely to move into work compared to those claiming Jobseekers Allowance (JSA). Even for JSA claimants, the roll-out of UC means that this evaluation was not representative of all JSA claimants moving to UC. Overall, this evaluation covers just one small part of the expected impacts of UC, and its effect will likely be small in magnitude compared to other groups and longer-term behavioural changes not yet evaluated. The DWP set out their plans for future evaluation of the impact of UC on the labour market in 2018 [iv], where quantitative research using various data sources will assess how other claimant groups compare in UC.
Whether unemployed claimants move into work is only one part of UC’s possible impacts on the labour market, and we need also to consider how it affects the choices made by those already in work over how many hours to work, whether to seek promotion, change jobs or even remain in work.

There are a number of reasons we might expect earnings to be lower for claimants under UC than the legacy tax credit system. The removal of the hours rules under tax credits – where claimants had to work over 16 or 24 hours a week to be eligible – means that some workers are now able to choose to work slightly fewer hours. And, given the taper rate of 63 per cent under UC, such cuts in hours are in effect subsidised by UC: working one hour less a week at the National Living wage is a monthly loss of earnings of £36, but after the UC taper this is a loss of income of only £13. For those earning enough to pay income tax and national insurance contributions, the effective income loss is even less – just £9 per month. The Resolution Foundation report ‘Making the most of UC’,[v] discusses the effect of work allowances, and the removal of the very highest taper rates under the legacy system in more detail. Key, though, is the uncertainty: there is currently limited evidence to suggest what the overall effect of UC on the labour market will be.

We note that there are some aspects of UC’s performance where it would be possible to learn more if the DWP were to publish more information. For example, no data has yet been published which shows whether more eligible benefit units are claiming UC. Such a comparison is technically possible by government by combining administrative data on all legacy and UC claims at single points in time.

Were the original objectives and assumptions the right ones? How should they change?

The original objectives were broadly sensible. If UC is successful in achieving them, then the resulting higher employment and earnings will be better not just for claimants, but also for the taxpayer.

However, the business case did not consider aspects of the delivery design of UC, or its implementation, that might affect claimants financially, and have subsequent economic costs or benefits. We have already mentioned that one key objective for UC should be that it functions effectively as a safety-net benefit, and it is not clear to us that it is doing so as effectively as the legacy system. More generally, an objective for a benefit aimed at those on no or a low income should be to reduce income uncertainty. Some evidence suggests that this is, in fact, increasing uncertainty for some claimants. We think that this should be considered in any overall assessment of the benefits of UC, as any increase in income uncertainty is simply shifting economic and social risks onto claimants or onto other parts of the public sector.

More generally, UC was designed when out-of-work poverty and unemployment were high. Today, it is in-work poverty and the lack of pay progression and productivity growth that dominate as issues for benefit claimants and society as a whole. In this context, the specific UC objective to reduce the proportion of workless households by ensuring the primary earner in a couple had a strong financial incentive to work now appears misplaced, and risks being a distraction. Instead, as we have recommended since the original implementation of UC, introducing second earner work allowances would better target work incentives to help reduce in-work poverty.

What have been the positive and negative economic effects of Universal Credit?

These have been discussed above, although we reiterate that it is hard to accurately quantify these effects on aggregate at present.

What effect has fiscal retrenchment had on the ability of Universal Credit to successfully deliver its objectives?

Equal reductions in benefit awards to both UC and legacy benefits should not affect any relative comparison between the two systems: the relative economic merits of UC should be achievable in either case. However, the reduction in generosity has reduced the extent to which UC strengthens work incentives – for example with fewer work allowances – and hence potentially undermines the scale of the anticipated economic benefits.
On the other hand, however, we should not ignore the public perception of UC, and how that itself might be affecting whether UC delivers its benefits by indirectly affecting claimants’ behaviour. Limiting the payment of child elements of UC and Child Tax Credit to two children, has drawn significant criticism because of the likely affect on child poverty. Alongside issues such as benefit sanctions, this has changed the way the public view UC by association.

Which claimants have benefited most from the Universal Credit reforms and which have lost out?

The Resolution Foundation provided an updated analysis of the financial winners and losers of UC in ‘The long and winding road’.[vi] This showed that working parents and those with high housing costs will be, on average, better-off; whereas claimants under 25, those without children, those with a disability, and home-owners will be, on average, worse-off. Taking a wider view though; the combination of UC with other welfare reforms – such as the two-child limit and the freezing of local authority housing allowance rates – mean that some of the UC winners will actually be losers once all benefit changes are considered. This assessment, though, does not include consideration of any second-order behavioural effects.

How has the world of work changed since the introduction of Universal Credit?

As mentioned above, UC was designed when out-of-work poverty and unemployment were high. Today, it is in-work poverty and the lack of pay progression and productivity growth that dominate as issues for benefit claimants and society as a whole.
There has been a small shift in low-paid individuals being self-employed, which may be connected to the rise of flexible ways of working. Although this represents a small share of total UC claimants, it means it is even more important to carefully consider policy for this group. As self-employed claimants are more likely to have volatile earnings, the monthly income assessment for UC claimants can end up financially penalising the self-employed because the Minimum Income Floor applies in the given month, even if the claimant’s annual earnings are above the Floor. In ‘Universal Remedy: ensuring Universal Credit is fit for purpose’,[vii] we set our recommendation that UC self-employed persons have longer assessment periods to mitigate this risk.

Does Universal Credit’s design adequately reflect the reality of low-paid work?

A significant number of low-paid workers are not paid on a regularly monthly cycle and, as explored our UC report [viii], this creates a very volatile stream of income that low-earners often struggle to manage. In the Resolution Foundation report ‘Irregular Payments: Assessing the breadth and depth of month to month earnings volatility’,[ix] we showed how the lowest earning employees had particularly high earnings volatility. UC will both smooth and exacerbate these volatile earnings for different claimants. For example, Tax Credits had an annual assessment system that meant if a claimant saw their pay fall by five per cent, they would not receive more in benefit until the reassessment. The monthly assessment periods under UC adjust the benefit received within a month. Conversely, a claimant that has volatile pay, and is paid irregularly (not on the same day each month), will find that the payment timings might mean income in one month being twice that of the next month. There is no obvious answer to how an integrated in-and-out-of-work benefit system can resolve this. However: moving assessment windows and warning claimants paid every four weeks of how that will impact their UC awards would help resolve some of these issues.

If Universal Credit does not adequately reflect the lived experiences of low-paid workers, how should it be reformed?

Given the volatility of earnings for low-paid workers, it is important the Government bears more of the financial risk that claimants face when making an initial claim to UC, and when transitioning from legacy benefits. Moving to a faster payments system and running-on tax credit payments would help achieve this. Allowing claimants to move the assessment period (and payment date) to suit their circumstances would also reduce unnecessary income volatility.
Two other areas stand-out for UC claimants: self-employment, and childcare. As mentioned above, reducing the frequency of assessment periods for the self-employed would better align with the volatile earnings that often come with being self-employed. On childcare, Government can build on the availability of the Flexible Support Fund, and pay for childcare costs up front in UC. This would make moving into and taking up more work much more accessible for those caring for children.
These points relate to how the system works as much as the rules on entitlement, which leads to an important point: for Universal Credit to work effectively it must be responsive to public feedback on how the process of making a claim and receiving payments are working. In any case of change – as with the continued managed migration – the DWP should proceed carefully to avoid unintended consequences.