6. Additional Assessment of Options
In this section, we assess all options against a set of criteria to ensure that the full range of policy aspects is considered when developing the income supplement. This also provides a guide for discussion and decision on the income supplement policy.
The criteria we employ to provide additional assessment are closely aligned to the principles for Social Security that were set out in Box 2. This approach has already been used to inform a number of decisions in relation to setting up Social Security Scotland., Here, we are using it in a specific policy context, which required us to revise and reprioritise the criteria applied to the options.
In developing the detail of the criteria, we considered the first two policy objectives for the income supplement, as well as the first test prescribed in the TCPDP.
Simplicity and Transparency. In the context of this criterion, we assess how straightforward and transparent each option is from a family's perspective.
Take-up and Consistency. Under this criterion, we consider how likely each option is to ensure that the income supplement is taken-up and its impacts are felt consistently across Scotland.
Employment and Earnings. This criterion considers the potential behavioural aspects of the income supplement by discussing how recipients may change their labour market behaviour, on the basis of each option.
These additional criteria are now discussed in more detail.
6.1 Simplicity and Transparency
In the context of this analysis, we assess how simple and transparent each option is likely to be for families.
Overall, most of the options considered in this paper are relatively straightforward.
Options 1 and 2 in particular should be relatively easy to understand as they attach eligibility to an existing benefit. Option 3 could be seen as more complex as entitlements differ depending on household characteristics. Option 4 would be the most complex of all, as it introduces a separate means test which may be relatively difficult to understand when set against an option based on qualifying benefits. Whilst we have not modelled this, there may be some differences between the new benefit and the existing UK benefit system in terms of eligibility criteria, for example around the treatment of savings. Although Option 5 is also attached to an existing benefit, households may not associate CTR with child-related support. Also, some of the factors around consistency discussed in Section 6.2 may make it less simple and transparent than other options that are based on qualifying benefits.
An automatic payment assumed in our modelling of Options 1a, 2a, 3a and 5a would be the most straightforward from the families' perspective. It would not require recipients to engage further with the benefit system – i.e. to make an application to receive the income supplement. However, automatic payments could be perceived as being less transparent than application based ones, if not all eligible families are aware of the income supplement or clear on the eligibility criteria. This could be potentially addressed by clear communication about additional entitlements available.
On the other hand, an application process assumed for Options 1b, 2b, 3b, 4b and 5b could increase complexity for families. The more targeted the option, the more detailed the information required in the application process and therefore, these options may well be less straightforward for families. However, the increased complexity could be addressed by an application process that enables different channels of communication and information about entitlement that adapt to the needs and requirements of individual claimants.
6.2 Consistency and Take-up
This assessment considers the consistency of coverage of children in poverty, as well as the level of take-up that each option could achieve.
Consistency Option 1 is based on a near-universal benefit that most households with children receive. Although Child Benefit is now subject to a means test, we would not expect different household types or households in different regions or local authorities to have different experiences of the benefit.
Options 2 and 3 would require full UC rollout to achieve consistency for all households. The UK Government has stated that UC will be fully rolled out by December 2023. However, further delays would limit the initial reach of these options. Unless there is a solution that allows families on legacy benefits to be reached before UC is rolled out, there could be inconsistencies across different family types and regions in terms of their eligibility for the income supplement.
By definition, Option 3 provides different levels of support to different family types, based on characteristics informed by the TCPDP 'priority families'. The purpose of identifying priority families was to ensure the needs of particularly disadvantaged groups were taken into account in policy development. Whilst this is an intentional feature of policy design it could have two significant drawbacks. First one of principle because it could be seen to differentiate between children in poverty; and second one of pragmatism because it would not be easy or appropriate to identify all priority family characteristics.
Option 4 would introduce an entirely new benefit and certain households may be less likely to apply, although whether there would be any inconsistencies across the eligible population that do not already appear in the qualifying benefits for Options 1, 2 and 3 is difficult to tell.
Option 5 is least likely to ensure a consistent impact across Scotland. CTR is targeted at households with high council tax liability relative to their income. However, it may not capture all households with children in poverty as their income may not be low enough to entitle them to CTR, particularly if they are in one of the lowest council tax bands. In addition, households not liable for Council Tax would not be eligible for the income supplement. Examples include parents (most likely lone parents) who live with or have moved back to their parents' house. In addition, students and mothers under 18 are exempt from council tax and therefore would not be in receipt of CTR.
Council Tax liability, and therefore CTR entitlement, is local authority/property band specific. It is therefore possible for a household with the same level of income to be eligible for CTR in one local authority/property but be eligible for a lower amount or not eligible at all in another local authority. Therefore, moving across local authority boundaries or to a different property could affect CTR entitlement.
Overall, for all policy options, the automated payment route can be considered to be more consistent than the application route, as it minimises the chance of missing families with children in poverty. Certain households may be less likely to apply for benefits and this could still be reflected even in automated payment options where eligibility is based on a qualifying benefit.
Take-up If the income supplement payment is automated, the issue of take-up becomes more about the take-up of the qualifying benefit for Options 1a, 2a, 3a and 5a, rather than the take-up of the income supplement itself.
Option 1a is likely to reach most households if the payment is automated, as it is linked to Child Benefit, which is already paid to most families with an estimated take-up rate of 93% in the UK in 2016/17. Although widespread awareness of Child Benefit and the lack of stigma associated with it would suggest that take-up could be higher than for other options, it is not certain this high take-up would be retained. It should be noted that take-up of Child Benefit has been on a downward trend in recent years.
For Options 2a and 3a, whilst the take-up rate of UC is not yet reported, take-up of Child Tax Credit (which UC child entitlement is replacing) in 2016/17 was estimated at 83%. Analysis by the Office for Budget Responsibility and the Joseph Rowntree Foundation suggests that UC take-up is likely to be higher than for the legacy benefits it replaces. As a result, take-up for these options may eventually be higher than current Child Tax Credit take-up.,
For Option 3a, it is useful to consider that Child Tax Credit take-up by larger families was 90% (as opposed to 80% for families with one child) and take-up by lone parents was 96% (as opposed to 72% for couples with children). We have not modelled these aspects because of the risk of introducing too much precision into the analysis but these patterns may have implications for policy impact and spend.
Although Option 4a is not linked to any existing benefit, it is largely targeting similar groups to UC and Child Tax Credit, which are households on low incomes but not necessarily all in poverty. The take-up could be higher than for Options 2a and 3a because it would not be linked to the UC conditionality regime, although eligibility rules may be perceived as being less straightforward and there would be an issue about raising awareness so that families apply.
Although for Option 5a estimates of CTR take-up are not readily available, DWP estimates suggest that in 2009/10, take-up of Council Tax Benefit (which CTR replaced in 2013) for working-age claimants was between 72% to 81%.
If the income supplement is delivered through an application process, as per Options 1b, 2b, 3b, 4b and 5b, there is the additional issue of how many eligible households apply. Take-up is difficult to predict and model as there is a wide range of factors that can affect it. These include attractiveness of the benefit, awareness of the benefit, awareness of entitlement to the benefit or any perceived stigma attached to it. In addition, complexity and transaction costs associated with applying for benefits is considered an important factor in explaining take-up.
6.3 Employment and Earnings
When considering changes to the benefit system, it is important to factor in not only the immediate impacts on poverty and income distribution, but also the impact that it can have on recipient behaviour. This is particularly the case in relation to how people engage with the labour market. However, whilst there is evidence that individuals do respond to changing incentives in the tax and benefit system, quantifying this is a challenging task.
For simplicity purposes, the options we considered in Section 5 assume that the income supplement is a flat payment, and that eligibility ends when a household is no longer entitled to a qualifying benefit (Options 1, 2, 3 and 5) or when income reaches a certain threshold (Option 4). This is often referred to as 'cliff edge' means-testing. Further consideration of behavioural responses may be required when refining the final income supplement policy.
Whilst there is extensive literature on behavioural responses to changes in income tax, the evidence is more limited on responses to changes in social security specifically, although similar considerations apply. Box 12 discusses some of the concepts and provides a brief overview of what we know about how different groups respond to changes in social security support.
Box 12: Behavioural responses – insights
For some households, a large part of the additional earnings from taking up employment or working more hours can be lost through higher taxes or reduced benefits. Although not discussed specifically here, costs associated with employment (e.g. childcare or commuting costs) can have similar effects. As a result, some households may gain little or, at the extreme, nothing at all by entering work or working more.
There are two broad ways in which the tax and benefit system can influence households' behaviour when it comes to work incentives. Conceptually, a household can be seen as having to make two decisions 1) whether to work at all and 2) whether to increase earnings through more hours or higher pay. Some tax and benefit policies will have an impact on the first one whilst others on the second one, and many will have an impact on both.
For some means-tested benefits, where eligibility is determined by income, a very small increase in earnings may result in a loss of the full amount of the benefit. This is known as 'cliff-edge' withdrawal and can in turn distort work incentives for those with earnings potential near the threshold where eligibility ends. One way to avoid a 'cliff edge' is to withdraw benefits gradually as earnings rise by applying a taper rate, where for an additional £1 earned the benefit entitlement is reduced by less than £1. However, the actual behavioural impact would depend not only on the relative changes in incentives, but also the number of people whose incentives have changed and their level of responsiveness to incentives. In theory, gradual phasing-out of support should reduce the severity of the impact on work incentives but at the same time it would also bring more people into eligibility, thus weakening incentives over a wider range of earnings, and could affect more people than a cliff-edge withdrawal.
Different groups may respond differently to work incentives. Some evidence shows that responsiveness (both in terms of work participation and hours) varies with the level of education and income suggesting that low earning parents and parents with lower levels of education tend to be more responsive. This evidence is consistent across lone parents and both partners in couples with children. However, some evidence finds that, among men with lower levels of education, only participation response is sensitive to changes in work incentives and that male hours are mostly irresponsive.
Lone parents (most of whom are women) face significant barriers to entering work (e.g. due to high childcare costs). In empirical literature, there is a broad consensus that lone mothers' decisions around whether to enter work are the most sensitive to incentives across all groups. This observation holds, albeit to a lesser degree, for married/cohabiting women and married men with lower levels of income/education, and the range of estimates available in literature is fairly wide for these two groups. A range of empirical studies find that, in terms of hours of work, men are much less responsive to incentives. Lone mothers' hours responses are similar to those of married women – both are moderately sized and may depend on the age of their children.,
A few studies have looked at how responsiveness varies with the age of youngest child and there is some evidence that this is higher for women with younger children suggesting that they are more responsive to changes in benefit payments. This may be because women who are already in work gain more flexibility around hours when their children enter nursery/school but the responsiveness falls as children get older., In addition, responsiveness may also depend on the level of earnings as high-earning parents can afford to pay for childcare. The evidence is, however, mixed and some studies suggest that mothers with all children of school age are more responsive to incentives than mothers with children aged under 5.
Modelling the impact on work incentives is beyond the scope of this paper. However, what can be offered is some commentary around how the options may differ in terms of the impact on the labour market, as well as the scope for refining these options to minimise any negative impacts.
Under Option 1, eligibility ends when the household no longer receives Child Benefit. We have assumed that those who currently receive a reduced amount due to the High Income Child Benefit Charge (i.e. households with individuals earning between £50,000 and £60,000) would still receive the full income supplement under this option. If such an approach was adopted, the households affected by the withdrawal of the income supplement are those with individual earnings in excess of £60,000. For a household with two children, assuming a £10 weekly income supplement payment per child, the annual reduction would be in excess of £1,000. For families on lower incomes, who may be more responsive (see Box 12), this option is unlikely to distort incentives to increase hours or earn more, although incentives to move into work could be affected since out-of-work support would increase.
Option 2 has been modelled assuming that the income supplement is fully withdrawn once the household no longer receives UC. Unlike for Option 1, this threshold will be different for different types of households. This is because the number of children, housing costs and other UC entitlements will vary which means the point at which the Maximum Award is reduced to zero through tapering will be different for different households (see Annex I). For a family with two children, assuming a £10 weekly income supplement payment per child, the annual loss in benefit income would be in excess of £1,000, as per Option 1.
Under Option 3, the payment is higher for certain types of families – lone parents, families with young children, larger families, families where a child or adult is disabled and young mothers. This means that these groups will be entitled to a higher payment if they have one or more of these priority characteristics and as a result they will stand to lose a higher amount if their eligibility to UC ends. Some evidence (as presented in Box 10) suggests that parents (mothers in particular) of young children may not be as responsive to incentives if their earning potential is low which may soften the impact on employment, although lone parents' responses were found to be more sensitive.
The considerations are similar for Option 4, where eligibility for the income supplement ends at net household earnings of £25,000. There are a number of variations that could be considered when developing a new means-tested benefit. For example, the income threshold could be based on income from earnings only or income from a range of sources, including other benefit income. Similarly, the income threshold could account for the size and composition of the household. Therefore, the impact on work incentives will depend on the way the means test is designed and the level of payment. It is noted that the weekly payment required would be higher if the income test was based on a lower earnings threshold. This could in turn result in a higher distortion of work incentives since a larger amount of income supplement would be withdrawn.
Under Option 5, eligibility for the income supplement is assumed to end once a family has stopped receiving CTR. Given how CTR operates, this would happen at different income levels for households in different council tax bands, and would also depend on the local authority in which they live (as the amount of council tax paid for each band differs across authorities). Assuming a £45 weekly income supplement payment per child, the loss in income once the family stops receiving CTR will be very large and will increase with the number of children. A family with two children would lose around £4,700 per year.
Scope for refinement of options There is a case for tapering the income supplement, as it would allow the payment to be withdrawn gradually and avoid situations when a household does not gain from increasing hours or earnings. However, the advantages of tapering should be viewed in context of the size of payment and the ease of designing and implementing a taper.
Of all the options considered, Option 5 presents the clearest case for tapering the benefit because of the high weekly payment (at £45) required to achieve the 3 pp reduction in child poverty. Unless there is a mechanism for tapering, this option could result in negative labour market responses.
The case is less clear for other options where payments are lower (£10) and in particular for Option 1, where the households affected are higher income households who may be less responsive in terms of hours worked (although the increase in entitlement overall could still affect the decision to enter employment for those out of work).
For Options 2 and 3, ensuring that everyone in receipt of UC is paid the full income supplement while tapering the payment at the same time could impact the simplicity of these options. In theory, the most efficient approach would be for the income supplement to effectively simulate an increase in the UC Maximum Award, which is tapered when a household is no longer eligible for UC, thereby creating a uniform taper with the UC. However, this approach would involve identifying and means-testing payments for households outside of the UC system, which is likely to be complex and could create confusion for claimants. Tapering payments for households in receipt of UC creates the issue of overlapping tapers and high benefit withdrawal rates which can impact on incentives. To address this income from UC could be taken into account for taper purposes but this would merely reduce rather than eliminate the problem.
The same issues apply to Option 5 with the additional consideration of the CTR taper. Designing a suitable taper for this option would be the most challenging as the interaction with UC and CTR taper would need to be carefully considered.
For Option 4, any taper would have to be chosen carefully so that, for households already subject to the UC taper, the two do not interact. Alternatively, income from UC could be taken into account before a taper is applied to ensure that the total benefit withdrawal rates do not exceed 100%.
Simplicity is key for good benefit design, and there are advantages in having simple and easily understandable rules to give households certainty about their entitlement, which may be more difficult to achieve with some of the more complex means-testing approaches. Therefore, the relative unclear benefits of tapering must be weighed against the complexity of administering the taper.