Prichard, R (on the application of) v The Secretary of State for Work and Pensions

This is my decision after a ‘rolled-up hearing’ of the Claimant’s applications for permission to apply for judicial review and, if permission is granted, for judicial review. The claim concerns a couple who claim welfare benefits, when one member of the household is above state pension age (‘SPA’), and the other is below SPA. I will refer to such a couple as a ‘Mixed Age Couple’ (‘a MAC’). To make it easier to follow the judgment, I will refer to the older partner in a MAC who is above state pension age as ‘the OP’ and to the younger partner in a MAC who is below SPA as ‘the YP’. The parties agree that in a MAC the OP is more likely than not to be a man, and the YP, a woman.

The Claimant (‘C’) challenges the Welfare Reform Act (Commencement No 31 and

The challenge is based on an alleged breach of section 149 of the Equality Act 2010 (‘the 2010 Act’), and on the Human Rights Act 1998 (‘the HRA’).

The Claimant argues, first, that the Defendant (‘D’) ‘failed adequately to consider the equality impact of making the 2019 Order’ (that is, the impact of commencing a provision of primary legislation which had already been enacted by Parliament). Second, C seeks a declaration that the 2019 Order and paragraph 64 of Schedule 2 to the 2012 Act are incompatible with his Convention rights, because, contrary to article 14 read with article 8 and article 1 of Protocol 1(‘A1P1’), they discriminate against people in his circumstances on the grounds of their sex, their age, disability, and because they are in a couple, and that discrimination cannot be justified.

The Defendant (‘D’) resists the section 149 claim with four broad arguments.

D resists the claim for a declaration of incompatibility with three broad arguments.

The 2012 Act created Universal Credit (‘UC’). UC is payable to households, not to individual claimants. C will reach SPA on 6 July 2020. His wife will not reach SPA until 20 July 2026. As a result of the changes made by section 4(1A) of the SPCA (which were brought into force by the 2019 Order), C and his wife will have to continue to claim working age benefits (that is, UC) until 2026, despite the fact that he will have reached SPA. He claims that his household will be £65,000 worse off than it would have been had the change not been made. Had section 4(1A) not been brought into force, C and his wife would have been eligible for PC when C reached SPA. It is important to note, however, that, as D emphasised, and as C accepts, C will not suffer an actual loss of benefits. The loss is a notional loss, measured by reference to the household’s notional entitlement after 26 July 2020, had section 4(1A) not been brought into force.

C’s case is that he and his wife both have serious and chronic illnesses. They receive disability benefits. It is said that neither C nor his wife is now, or is likely in the future to be, subject, under the UC regime, to any requirement to look for work. The couple they belong to is therefore described by the shorthand phrase, ‘no-conditionality MAC’. Further, C will not, in any event, be subject to a requirement to look for work after 6 July 2020, because he will then have reached SPA.

D points out that C and his wife both currently receive UC. No conditions apply to them because C is paid a Carer’s Allowance for his wife, and she is paid a Carer’s Allowance for him. Although it is C’s case that he cannot work, the First-tier Tribunal has found that he did not have, either, limited capacity for work (‘LCW’), or limited capacity for work and work-related activity (‘LCWRA’). C did not appeal against that decision.

D does not accept that C would, if he were entitled to SPC, be entitled to its severe disability component (the ‘extra amount for severe disability’). D argues that his wife’s continuing receipt of a Carer’s Allowance would prevent that.

C’s wife is entitled to a Carer’s Allowance because she cares for him for 35 hours a week. Those hours can include overnight care, or care during the day. D accepts that, because C’s wife is paid a Carer’s Allowance for looking after C, she cannot ‘reasonably be required to work’ under the statutory scheme. For that reason, she is not subject to any work-related requirements. D argues, however, that it does not follow from the payment of the Carer’s Allowance that C’s wife cannot work, nor that she should not be encouraged to work.

The parties agree that, even though C’s wife cannot be required to work, help would be available to her in a JobCentre if she wished to look for work. C points out that such help is in practice available to claimants whether they claim UC or PC. Such help is provided by D under his common law powers, not under the legislative regime governing UC.

The Government’s changes to the welfare system started with a consultation document, 21^st Century Welfare (Cm 7913, July 2010). It generated over 1600 responses. It was followed by a White Paper, ‘Universal Credit: welfare that works’ (Cm 7957, November 2010). This explained why fundamental change was thought to be needed. The Government wanted to reform the benefit system to make it fairer, more affordable, and better at tackling poverty, worklessness and welfare dependency. The new system would promote work and personal responsibility (paragraph 1, executive summary). UC would radically simplify the system and its administration (paragraphs 2, 7, 1, 15 and 16, ibid). The complexity of the system, and the disincentives to work, meant that people were not prepared to take the risk of starting work (ibid, paragraph 4). Worklessness was bad for families and for society as a whole (ibid, paragraph 5). UC would reintroduce a culture of work (ibid, paragraph 6). It would improve incentives for work (ibid, paragraph 9). It would remove distortions in the current system (ibid, paragraph 10). The incentives to work would be backed up by ‘a strong system of conditionality’ (ibid, paragraph 13). The incentive fraudulently not to declare small amounts of work would be reduced because claimants would be able to do small amounts of irregular work without financial loss (ibid, paragraph 16). As many as 350,000 children and 500,000 working age adults could be moved out of poverty (ibid, paragraph 18).

Paragraph 21 of Chapter 2 described the Government’s commitment to supporting disabled people to ‘participate fully in society including remaining in or returning to work whenever feasible’. Paragraph 24.b of Chapter 2 described the Government’s desire to support carers and to improve their opportunities to ‘maintain links with the world of work’. Chapter 5 dealt with the ways in which UC would reduce fraud and administrative error. Chapter 7 described the impact which UC would have, summarised at the beginning of Chapter 7.

Mr Latto, a senior civil servant in the Department of Work and Pensions (‘DWP’), has made a witness statement in which he summarises the changes and their background. I summarise the main points.

The 2012 Act made a major change to the system of welfare benefits by replacing six separate benefits, and tax credits, with UC. Mr Latto describes the policy which underlies the 2012 Act. One important aim is to encourage as many people as possible to work. Working enables people to save for their retirement, maintains a supply of labour and generates revenue in tax and national insurance contributions. It is important for society, and for those concerned, that everyone is encouraged to work at least until SPA.

The Government abolished the default retirement age (that is, the age at which all workers could be made to retire) in 2011. Nevertheless, SPA is a turning point in the benefits system. After SPA, claimants are no longer required to work as a condition of receiving benefits. The assumption is that people who are below SPA will provide for themselves and their households through paid work if they are able to.

Some benefits are means-tested. Means-tested benefits help those who are looking for work, who cannot work, or who are working but who are not paid much. Means-tested benefits are paid to households. UC is an example. Other examples, paid to those who are past SPA, are PC and pensioner HB. Other benefits are contributory, and are based on individual entitlements. The new state pension, which is payable to those who reach SPA after April 2016, is such a benefit. Contributory benefits (Jobseeker’s Allowance and Employment Support Allowance) are also paid to some people below SPA.

In principle two types of means-tested benefit are available to a MAC; PC, the purpose of which is to give financial support to a person who has left the labour market permanently, and UC, which is designed to support those of working age, and to maintain an attachment to the labour market.

There is a policy choice about how to support a MAC. The essential choice is whether or not to treat the YP as if he or she, like the OP, had left the labour market permanently. Under the old scheme, the Government chose to support MACs through benefits for pensioners.

That choice, however, conflicted with the aim of increasing participation in work. There are three main reasons why. First, PC is paid at a higher rate than UC, which is set at a rate which is designed to give claimants an incentive to work, and UC’s system of work allowances and tapers ensures that claimants are better off in work. Second, no conditions related to work are attached to the receipt of PC and of pensioner HB Third, PC provides a positive disincentive to work because earnings above about £10 per week for a couple reduce it pari passu. The enactment of the 2012 Act was an opportunity to ‘align the system of means-tested support with the strategic objective of maximising labour participation, and to ensure that the younger partners in MACs were subject to the same labour-market conditions as other people of the same age’.

The changes made by section 4(1A) of the SPCA affect MACs who make a new or a repeat claim on or after 15 May 2019. They will not affect C until he reaches SPA. He will not lose any income in the statement which he now receives. The changes apply only to means-tested benefits, and not to the state pension. Before the changes, MACs were entitled to PC and pensioner HB, or could opt to claim UC instead. PC and pensioner HB are still both available to single pensioners and to couples both of whom have reached SPA. As a result of the changes, a pensioner who has a partner who is below SPA cannot claim PC or pensioner HB until his partner has also reached SPA.

Mr Latto describes the development of the MAC policy before Royal Assent in paragraphs 31, 32, 33, 35, 36, 38, 39 and 41. It is clear from this material that the MAC policy and its implications were thoroughly examined, including the question whether its effect should be mitigated by delaying commencement of the policy until the state pension ages had been equalised (which was due to happen in 2018) (see paragraphs 38 and 39). A ministerial submission in December 2010 specifically examined the options for MACs. At that stage, officials recommended that in cases where the YP would be subject to work-related conditions, the couple would be required to claim UC, but that a higher allowance be introduced in UC to ensure that couples were not substantially disadvantaged, while being subject to the work-focused support offered by UC. Mr Drabble relied on this document. A ministerial submission on 14 June 2011 specifically considered the position of a MAC in which one partner was a full-time carer (paragraph 39). A compromise to address the position of a MAC in which the YP was a full-time carer was considered in a ministerial submission on 6 July 2011, after a request for advice from the Minister (paragraph 40). Ministers asked for further advice. Before this could be provided, the Secretary of State decided that a pensioner premium should not be included in UC because of its cost. If it were provided, there was also a risk that it would undermine work incentives for other groups (paragraph 41).

An Impact Assessment of UC was published in October 2011. DWP also assessed the equality impact of UC in November 2011 (‘the 2011 EIA’). The 2011 EIA updated an assessment in March 2011. The overarching aim of UC was ‘to remove the financial and administrative barriers to work inherent in the current welfare system’ and to ensure that work pays (paragraph 5). Section 2 described the consultation process. Respondents had been asked to express their views about the potential equality impacts of the reforms. DWP might publish further impact assessments (‘IAs’) as the detailed policy emerged. The evidence would continue to be considered and stakeholders would be consulted through DWP forums.

Disability was considered in paragraphs 27 to 47 of the 2011 EIA. Paragraph 19 defined ‘participation tax rates’ (‘PTRs’), a measure of changes in the proportion of earnings which people keep when they move into work. Paragraph 30 referred to a table which showed the positive impact of UC on people moving into ten hours’ work a week at the minimum wage. The improvement in the proportions of people with a PTR of less than 60% was very significant both for disabled people and for those who are not disabled, and the proportion of both groups with a PTR of more than 70% much smaller. The improvements for those who were not disabled were somewhat better. Those improvements were a consequence of the features of UC which were described in paragraph 31. The reason for the difference in PTRs between disabled people and those who were not disabled was that disabled people tended to claim other benefits outside UC. The interaction between contributory benefits and carers’ allowance also accounted for the small proportion of disabled households with PTRs of more than 90%. The higher earnings disregards for disabled people mean that their work incentives are significantly improved by the changes.

Paragraph 32 considered the impact of UC on MDRs (marginal deduction rates: see paragraph 19) on people who work. UC virtually eliminated the highest MDRs of 80% for both disabled and non-disabled households. Paragraph 33 concluded that under UC, disabled people who are able to work will see significant improvements in their incentives to work. They would no longer face the punitively high PTRs and MDRs which characterised the current system and would have broadly the same PTRs as people who are not disabled.

Paragraph 38 explained that under UC, a person could not qualify for both a limited capability for work element and a care element, because both elements reflected an inability to work. But households would be entitled to a limited capability for work element for one member and a carer element for the other member. Paragraph 39 recognised that once transitional protection had ended, the entitlement of the most severely disabled would increase, but that of other disabled households would decrease. This reflected the facts that disabled households were less likely to be in work and that UC benefits low-paid workers most. Disabled people were slightly less likely to be moved out of poverty than those who were not, but that analysis did not take into account the improved incentives to work provided by UC (paragraph 41). Some disabled people would be entitled to less under UC, but the Government believed that was justified because money was being targeted at the most severely disabled. Disabled adults who could work would be entitled to higher UC under the new scheme (paragraph 45).

The tapers and disregards improved incentives to work which in turn would promote the equality of disabled people by supporting disabled people who were able to, to work a few hours, especially those with a fluctuating capacity to work. The new system would be simpler than the old (paragraphs 42 and 43).

UC improved incentives to work for disabled people to roughly the same extent as it did for non-disabled people. Because disabled people on UC were more likely to be out of work, it had, on average, less impact on entitlements than for other groups. The combination of increased take-up and changing entitlements (including an increase for 0.8m households with a disabled member) contributed to a reduction in poverty among disabled households (paragraph 46). UC would ease some of the barriers to work faced by disabled people. Many claimants would have smaller PTRs and MDRs and could benefit from the removal of non-financial barriers to work. The changes reduced the ‘perceived risk of employment for disabled people’, especially those who worked a few hours or whose ability to work fluctuated. More support would be needed in some areas, and DWP would consider how best to provide that (paragraph 47).

Gender was considered in paragraphs 48-74. The analysis was generally divided into single men, single women and couples. There was a positive impact on PTRs for people moving into work for ten hours a week on the minimum wage, for both men and women. The improvement was greater for women because of their lower starting point. The improvements were due to two features of UC, one of which was support with childcare costs. That help would be available for couples and for lone parents. The removal of the current requirement to work 16 hours a week would give an important incentive to lone parents and single earners to take their first steps to get a job (paragraphs 49-51). Childcare support would be available to about 80,000 more households (paragraph 52). Both men and women would have significantly greater incentives to work (paragraph 54). UC would make incomes more stable and support more consistent, which would reduce the risks of moving into work (paragraph 55).

The entitlements of single men and women were much less likely to change than those of couples, because more single people receiving UC are out of work than couples. Almost half of couples would have a bigger entitlement, but couples were also more likely to get less. Those whose entitlements changed were likely to face larger changes than single people (paragraph 57). Table 6 set out those effects in some detail.

Workless households were less likely to see changes in their entitlements (paragraph 59). Overall, the average change in entitlement was higher for couples than for single men and women. About 69% of couples are families with children. A higher proportion of households, and in particular, couples, with children would get more under UC.

Couples with one partner under, and one partner over, the qualifying age for PC might get less, as they would have qualified previously for more generous allowances available under PC (paragraph 62).

UC would have a substantial positive impact on poverty, reducing those in poverty by about 900,000 people. The effect on single women and single men was roughly equal. 50,000 of the 350,000 children taken out of poverty were in households with a woman lone parent. This analysis did not take into account the effect of improving incentives to work (paragraph 64). Most lone parents were women, and their rate of employment was lower than the average. The reason for that in many cases was that it was difficult to fit work round childcare. 80% of lone parents worked, or wanted to. UC was an opportunity to narrow the employment gap and to promote equality (paragraph 65). The new system was expected particularly to help lone parents. Childcare support would help this. This was an opportunity to promote gender equality by helping parents into work (paragraph 67).

There were risks of negative impacts. There was a risk of decreased work incentives for second earners in couples (mostly women). Having no parent in work had an impact on young people’s lives and on their attitudes to work. Helping at least one parent into work could help to break the cycle of worklessness in a family (paragraph 68). The overall assessment was that UC improved incentives to work for women and for men roughly equally. There would be strong positive impacts for women (reducing poverty and improving work incentives (paragraph 71)). Small differences in impact as between men and women were related to other differences associated with gender, for example, because single men on UC were more likely to be out of work than single women. The changes associated with UC might improve family life by enabling families to strike the balance they preferred between work and family life (paragraph 73).

Age was assessed in paragraphs 86-95. Particular characteristics tend to be associated with age; for example, two thirds of people between 25 and 49 are in households with children. A table segmented by the age of the head of household showed that there was an improvement in PTRs for all age groups. This substantial improvement was a function of the design of UC (paragraph 88). UC virtually eliminated the highest MDRs of more than 80% for all age groups. Households of all ages would see significant improvements in their incentives to work (paragraph 91).

Table 12 showed the impact of UC on levels of entitlement, segmented by the age of the head of household. Paragraph 93 summarised the main points. Under 25s were most likely to see an increase, and least likely to see a decrease, in their entitlements. Those between 25 and 49 were more likely to see an increase, and less likely to see a decrease, than those over 50, largely because of the likelihood that there would be children in the household. Those over 50 were most likely to see a reduction in their entitlement, and least likely to see an increase, and the reductions were likely to be greatest. Households with heads of household over 50 would now include MACs. The entitlement of MACs might reduce as they would not qualify for the more generous allowances available with PC, leading to the high average reduction in entitlement for this age group.

There would be a substantial positive impact on poverty in households in which the head was under 50. The impact was smaller where the head of household was over 50 (paragraph 94). Opportunities to promote equality were considered in paragraph 95. Those under 25 would benefit particularly from the reduced taper and integration of support. Online access for older people was considered in paragraph 96. MACs would see a decrease in entitlement as they would no longer get PC. The change would only apply to new claims. This would directly affect older couples, but the Government believed that it was justified in order to avoid people under pension age being supported by benefits but not being subject to appropriate work-related conditions (paragraph 97). Paragraphs 109-110 explained what further steps would be taken to evaluate the impacts of UC.

On 26 October 2011, the Minister for Employment wrote to the members of the Welfare Reform Bill Committee describing what would happen when the YP was a carer for an OP with severe disabilities. The letter confirmed that MACs would not have access to PC, but to UC and to additional amounts in UC, where they were eligible for those. On 29 November 2011, the Minister wrote to Baroness Hollis, responding to a request made in the debate on 24 October 2011. The letter included worked examples comparing PC and UC for MACs. Two examples concerned a YP with disabilities.

Mr Latto gives the history of the 2012 Act’s passage through Parliament. On 28 April 2011, an Opposition amendment which would have removed the MAC change was defeated in the House of Commons, a similar opposition amendment was not pressed at the Report stage in the House of Commons on 13 June 2011, and at the Report stage in the House of Lords on 12 December 2011, an Opposition amendment to the MAC change was not pressed. The 2012 Act received Royal Assent on 8 March 2012.

Mr Latto also describes debates in the House of Commons and the House of Lords in which the position of a YP caring for an OP was specifically discussed (paragraphs 37, 42 and 46). In the debate on 12 December 2011, Baroness Drake and Lord McKenzie of Luton made the point that incomes for MACs could be up to £100 a week lower under UC. In his response, Lord Freud said that it was the Government’s policy that MACs should claim UC, not PC. He referred specifically to MACs in which one partner was disabled, pointing out that UC contained additional amounts for those with limited capability for work, and with substantial caring responsibilities for a severely disabled person. It was the Government’s view that people of working age who are able to work should prepare for or look for work in return for receiving support from the state.

The Bill received Royal Assent on 8 March 2012. In May the Minister asked for advice on the provision for MACs if the YP had LCWRA or was a full-time carer, and for advice on whether such MACs should be able to claim PC, rather than UC. Officials provided indicative costs. The Minister considered a submission about when to commence the MAC provisions in July and August 2012 and accepted advice that commencement should be delayed.

A further IA dated 7 December 2012 (‘the 2012 IA’) set out the Government’s ‘current assessment of the broad impacts of’ UC as set out in the eight sets of regulations listed in paragraph 1. Annex 1 set out changes to the policy since the 2011 IA. It was estimated (paragraph 18) that the changes would lead to savings of about £2.2bn per year. UC was a fundamental reform which would lead to reductions and increases in entitlements (paragraphs 26 and 29). Its simplicity would increase the take-up of benefits (paragraph 26). Some of the larger notional losses would be felt by MACs. The reforms meant that they would be eligible for UC, not PC, in order to ensure that the YP ‘remains focused on a return to work’.

Transitional protection would ensure that there would be no cash losses for households actively moved from legacy benefits to UC (paragraph 36; and paragraphs 55-57). Paragraph 37 described the changes in income which would affect many households, and the expectation that ‘work is required for those who can’. The pattern of changes was driven as much by simpler rules and greater take-up as by membership of a particular demographic group.

Paragraph 40 described features of the structure of UC which influenced some of the changes in entitlement. These included allowing people to keep more of the money they earn in order to make work pay, and simplifying benefit rates for those with a limited capacity for work, as well as simplifying the structure of disability payments. UC had simple rules for calculating entitlements, but the move away from the complexities of the current system entailed some changes to entitlement which were the result of interactions between different changes (paragraph 41).

Table 4 showed that it was not straightforward to compare current eligibility and changes in income (paragraph 43). The entitlements of most workless households would not change, but some, including MACs, would be affected. The workless households which would get higher entitlements would do so as a result of support being focused on the most severely disabled (paragraph 45).

Paragraphs 51-54 considered the impact on household income for protected groups. Households with a member in a protected group tended to be affected according to whether they worked, and their income.

The average effect on the income of a household with a disabled person in it was smaller than for other households, reflecting the likelihood that those households were likely to be out of work. Other changes to support for people with disabilities were designed to be cost-neutral across the group. The average change for disabled households was an increase of £8 per week, compared with an increase of £16 per week for other households. The changes in entitlement tended to be greater as because support for disabled people was being re-allocated, with the focus of the support on the most disabled (paragraph 51).

The impacts on different age groups resulted from changes in policy. The policy of including MACs in UC meant that the OP was more likely to see a reduction in entitlement (an average reduction of £27 per month in household income), whereas when the head of the household was under 25, the average change was an increase of £20 per month (paragraph 53).

The average impact on single men and women was the same (a small increase). The impact on households with men was on average the same as the impact on households with women (paragraph 54).

UC removed many of the complexities and inconsistencies in the current system, replacing them with increased support for low-income families and consistent support as income rises. This simplification would mean losses for some households. The new system created greater incentives to work particular numbers of hours (paragraph 58). UC rewarded all hours of work. It was reasonable to expect people to respond to incentives to work (paragraph 59). Chart 2 showed the long-term impact of the changes in income for every decile. It showed that UC benefitted low-income families. Those with the lowest incomes gained proportionately most (paragraph 60). The most significant reductions were in the sixth and seventh deciles (those most likely to be receiving working tax credits and nothing else) (paragraph 61). UC improved social welfare by focusing improvements in income on those who valued small changes most (the poorest) (paragraph 63). UC would reduce child poverty and improve work incentives by allowing people to keep more of their earnings as they moved into work, and by introducing a smoother and more transparent reduction in benefits as earnings go up (paragraph 64).

Paragraphs 66-68 explained the three ways in which UC improved work incentives. The current system mainly rewarded those working 16 or 30 hours a week. Conditionality ended for most people once they earned above a minimum which could

A key aim of UC was to encourage those who were out of work to take their first steps into employment. Work allowances and the taper were designed radically to improve the incentives to work for few hours per week (paragraph 71). The changes were designed to enable people to see and understand those improvements (paragraph 72). Table 5 showed the positive changes in PTRs for first earners in workless households taking up ten or more hours’ work per week. Table 7 showed the employment incentives by household type if a first earner started to do 10 hours’ work a week. The pattern of PTRs among protected groups was virtually the same at 10 hours’ work a week (under 60% for virtually everyone) (paragraph 62).

UC also reduced very high MDRs for some groups in the current system (paragraphs 84-85). A system of conditions would also encourage and support as many people into work as possible. The greater simplicity and transparency of the new system would make the reduction in MDRs more effective in providing incentives for work (paragraph 89).

UC would help people move into work by reducing uncertainty. People would no longer have to have their benefits re-assessed by as many as three government agencies (paragraph 99). Claimants would have one relationship with one organisation, one assessment, and one payment. Claimants would have immediate on-line access to work out the effect of working any number of hours. Many changes would be handled automatically. Support for council tax would stay outside the benefit system, administered by local authorities (paragraph 101).

The conditions attached to receipt of benefit would change. Claimants would be required to make a claimant commitment. Different groups would be given different conditions; an expectation to look for work, and take a job, and support to do those things (paragraph 101). Conditions would be imposed on an extra one million claimants. UC would be calculated per household, but conditions would be imposed on claimants individually, bringing partners of those who were out of work back into contact with the labour market (paragraph 102). Some claimants, for example those with regular and substantial caring responsibilities or people with limited capacity for work would not be subject to conditions (paragraph 103).

UC was a fundamental change to the benefit system. It was very difficult to analyse how claimants would respond to the different changes it would make (paragraph 104). With that caveat, the four potential effects on the labour market were described in paragraphs 107-118. Three are relevant in this case: the changes in financial incentives to work (paragraphs 107-108); the increased simplicity and transparency of UC, and the smoothing of a move into work (paragraphs 109-114); and the changes in conditionality (paragraphs 115 and 116). It was difficult to assess the extent to which those changes would add to the responses to other financial and non-financial incentives in UC (paragraph 116).

Annex 3 described the plans for monitoring the effect of UC.

Annex 4 considered the risks and opportunities to promote equality.

Disabled people would benefit particularly from the removal of non-financial barriers to work. UC removed the distinction between in and out-of work support, and recognised the presence of limited capability for work. That considerably reduced a significant barrier to work (especially for those wanting to work part-time or whose ability to work fluctuates). Work-related conditions would be extended to more claimants. Just over a quarter of the partners who would be affected were disabled. The work-related conditions which would be applied to people with limited capability for work would potentially increase equality for them in the labour market. Conditions would not be applied to those with limited capability for work-related requirements. The system would be simpler for claimants. Most transactions could be done on-line. Uncertainty would be reduced, and transparency increased because only one claim would be made, in one integrated system. Because UC would remove complexities and inconsistencies, some disabled people could be entitled to less under UC. The

UC would particularly benefit those who wanted to work a small number of hours. They would be helped with childcare. Lone parents, the majority of whom were women, would have better incentives to take their first steps into work. Helping more parents to get the support to take a job would promote gender equality. Appropriate work-related conditions would apply to more claimants. 70% of the partners affected by this were women. UC was an opportunity to promote equality in work and to narrow the employment gap. Because the focus of UC was to reduce workless households, there was a risk of reducing incentives to work for second earners in couples (mostly women). Having at least one parent in work was what mattered.

MACs would be entitled to less under UC because they would not qualify for PC which is more generous. The Government believed that this was justified to ensure that people under pension age were supported by the benefit system while having the appropriate level of work-related requirements. This would only apply to new claims.

In March 2013, the Minister was given information about what percentage of MACs also received a disability benefit or incapacity benefit (40%) and the percentage which received an additional amount for carers (30%). The Minister again asked for advice on whether MACs in which the YP was not subject to conditions (for example, because she was a carer) should be exempt from UC, be able to claim a premium, or be able to claim PC. Implementation of the MAC provision was not imminent, so the issue was not considered further. In April 2013, Ministers decided to defer the replacement of pensioner HB with a housing credit until at least 2017/18. The MAC policy was reviewed in June 2014. The note prepared for the Minister referred to the earlier requests for advice. The Minister agreed with officials’ recommendation not to change the policy. Also in June 2014, officials provided the Minister with worked examples comparing the effect of PC and UC. Several were cases in which the YP was a full-time carer. The Minister had no comments.

On 20 July 2016, the Minister announced a revised timetable for UC. The ‘migration’ to UC of those claiming ‘legacy’ benefits was postponed until 2022, as was the inclusion of pensioner HB in PC. On 15 December 2016, a Ministerial submission summarised the development of the MAC policy to date (see paragraph 59 of Mr Latto’s witness statement). The Minister agreed to continue with the policy as so summarised. In November 2017 the Minister considered issues about pensioner HB. On 23 November 2017 the Minister announced changes to UC which entailed a further postponement of the implementation date for the MAC policy until 1 February 2019. The Secretary of State confirmed on 1 December 2017 that he was content to make an order commencing the MAC policy from 1 February 2019.

On 23 April 2018, a ministerial submission described the background to the MAC policy. It set out the likely savings (£68m rising to £394m in 2023-24, ‘largely because [UC] rates are significantly lower than [PC] rates’). The average notional loss in benefit was estimated at around £72 per week in 2019/20. In some cases it could be £100. The savings had already been scored as part of the business case for UC. Annex A set out worked examples showing the potential effects of the policy. In one, the OP was disabled, and the YP receiving a carer’s allowance; precisely the circumstances of C and his wife. The example showed that such a MAC would be worse off under UC. An Equality Analysis dated 12 April 2018 (‘the 2018 EA’) was attached to the submission. The submission asked the Secretary of State to make her decision having considered the 2018 EA and the Family Test Assessment. It emphasised that it was the Secretary of State’s personal obligation to consider the impact of the policy on people with protected characteristics.

The 2018 EA accurately summarised section 149. It referred to the 2011 EIA. Officials had also taken into account United Nations Conventions on the Rights of Persons with Disabilities, for the Elimination of Discrimination against Women, and of all forms of Racial Discrimination. The 2018 EA described the MAC policy. Its fundamental rationale was to ensure that the YP had appropriate work-related conditions and relevant support to find work. It would ensure that the YP would continue to have access to support from personal work coaches until they reached pension age. As many people as possible should benefit from such support. The policy was also fair as between people of working age, who would be treated in the same way whether or not their partner was of pension age. The likely savings were described. There would be few direct, as opposed to notional, losers. The average notional loss was £72 per week rising to £107 per week in 2023/24. The 2018 EA explained that analysis for some characteristics was either not available or limited because information on them was not collected as part of the claim process. Survey data was unreliable for MACs because of the small sample sizes involved. Disability was an example.

DWP did not hold data on how many PC claimants were disabled for the purposes of the 2010 Act, so receipt of disability benefits was used as a proxy. Just under half of all households claiming PC received such a benefit. 38% of MACs received a disability benefit. Disabled recipients of PC were therefore less likely to be in MACs compared with other couples who received PC. This suggested that disabled people would not be disproportionately affected by the policy change. It was possible that a wider group of disabled recipients of PC might be disproportionately affected, if the definition in the 2010 Act was used instead of the proxy measure. There was not enough information, but, in any event, no reason to think that the position would be different from the position shown by reference to the proxy measure.

To assess the effect on sex, the current make-up of claimants was considered. Almost two thirds of current beneficiaries were women. About 13% of male claimants were in a MAC, and about 7% of women claimants. That showed that the change had the potential to affect a slightly higher proportion of men than women. Overall, the policy was not considered to have a disproportionate impact on people based on their sex.

The policy was, by definition, aimed at MACs, so older benefit claimants would be disproportionately affected compared with younger claimants. The fundamental rationale of the policy was to ensure that the YP was not adversely affected by a lack of support to gain employment with its attendant benefits, as in the current system. Although many older claimants might potentially be disproportionately affected by a notional loss of benefit income, ‘overall, a disproportionate negative effect may be justified by the potentially offsetting positive effect of the policy in increasing the work chances of the [YP] (where there is eligibility to UC) through it, access to support in finding work’. Data from 2017 showed that 68% of the OPs were under 70, and only 3% 80 or over, suggesting that, among pensioners, the policy would mainly affect younger pensioners.

Overall, men were potentially slightly more likely to be affected than women, although men and women would be equally affected by the policy’s actual application. People who were disabled were slightly less likely to be affected than those who were not, although the data were not accurate. Older claimants would be disproportionately affected by the policy, but this might be justified by the positive impact of the fundamental rationale for the change. The position would be monitored.

The Secretary of State then raised various detailed concerns about commencing the MAC provisions, as Mr Latto explains in paragraphs 75-86 of his witness statement. She agreed in principle that the provisions should be commenced, noting that the legislation had already been passed, but the precise timing of commencement was considered several times. Ministers wanted to understand more about how the policy would work in practice and the extent to which MACs had already been affected by other changes. She specifically asked about YPs caring for OPs. Officials prepared a further submission about that on 9 May 2018. This analysed the composition of the affected groups and gave advice about the timing of commencement order. The

Section 150(1) of the 2012 Act lists provisions which ‘come into force’ on the day the 2012 Act is passed. Section 150(2) of the 2012 Act lists the provisions which ‘come into force’ at the end of the period of two months beginning with the day on which it is passed. Section 150(3) of the 2012 Act provides ‘The remaining provisions of this Act come into force on such day as the Secretary of State may by order made by statutory instrument appoint’. Section 150(4)(a) and (b) enable an order under subsection (3) to appoint different days for different purposes and, for some provisions, different days for different areas. An order under section 150(3) may ‘make such transitory or transitional provision, or savings, as the Secretary of State may consider necessary or expedient’.

In R v Secretary of State for the Home Department ex p Fire Brigades Union [1995] 2 AC 513 (‘the FBU case’) the Appellate Committee of the House of Lords construed

The issues for the House were whether section 171(1) imposed a duty, or conferred a power, on the Secretary of State (and if a power, what the kind of power it was) and whether, by saying that he would not bring the provisions enacting the scheme into force, and that he would, instead, introduce a different non-statutory scheme, the Secretary of State had acted unlawfully. That case therefore concerned a refusal to exercise a commencement power, whereas this case concerns the exercise of such a power.

The majority of the Committee (Lords Browne-Wilkinson, Lloyd and Nicholls) held that the Secretary of State had acted unlawfully. All three noted that the formula in section 171(1) is common in legislation (p550B-C, 574D-E and 574D). It was conferred ‘for the purpose for bringing the sections into force’ (per Lord Browne-Wilkinson at p

Lord Browne-Wilkinson held that section 171(1) did not impose a legally enforceable duty on the Secretary of State. It conferred a power to bring the provisions into force

It is for Parliament, not the executive, to repeal legislation (p 552D-F). The fact that a scheme approved by Parliament was on the statute book and would come into force if and when the Secretary of State so determined was directly relevant to the question whether the Secretary of State could in the lawful exercise of prerogative powers decide to bring in a further non-statutory scheme and refuse to exercise his discretion to bring the statutory provisions into force (p 554A). The Secretary of State could not validly resolve to give up his statutory duty to consider from time to time whether to bring the statutory provisions into force.

Lord Lloyd said that it might surprise the man on the Clapham omnibus that ‘legislative provisions in an Act of Parliament, which have passed both Houses of Parliament and received the Royal Assent, can be set aside…by a member of the executive’ (p 568H). He agreed with counsel for the respondents that section 171(3) cast light on the purpose for which the power in section 171(1) had been conferred. He construed section 171 ‘so as to give effect to, rather than to frustrate, the legislative policy enshrined in sections 108-117, even though these sections are not in force’. Those provisions were a statement of Parliamentary intention. He read section 170 as providing that sections 117-117 ‘shall come into force when the Home Secretary chooses, not that they may come into force if he chooses’. Section 170 conferred a power ‘to say when, but not whether’. Parliament’s intention in enacting section 170 coincided with the reaction of the bus passenger. The Secretary of State had power to delay the coming into force of the provisions, but no power to reject them or set them aside as if they had never been passed (pp 570-1). He could find nothing in section 171 which justified the decision of the Secretary of State to refuse to implement the statutory scheme. ‘By renouncing the statutory scheme the Home Secretary has exceeded his powers and thereby acted unlawfully’ (p 571E-F). He made clear distinction between delaying commencement and an abdication of that power (p 572C). He left open the question whether the Secretary of State had a duty to bring the sections into force as soon as he judged it appropriate, or merely to keep the exercise of the power under review (p 573D-D).

The question for Lord Nicholls was whether the power conferred by section 171(1) limited the way in which, or the purposes for which, the prerogative could lawfully be exercised in the same field. If the relevant provisions had been repealed, there would be no difficulty. If the power had been exercised, the problem would not arise. He said ‘Parliament enacts legislation in the expectation that it will come into operation. That is so even if Parliament does not itself fix the date on which that shall happen’ (p 547DE). A power like section 171(1) gave the executive flexibility to make the practical preparations necessary to bring legislation into force. ‘The decision is best left to the minister whose department will be giving effect to the legislation when it is in operation. He is given a power to select the most suitable date, in the exercise of his discretion’ (p 574F). The discretion was a wide one. The minister could take into account a range of considerations in deciding whether the moment had come to appoint a day, including financial considerations.

Section 170(1) did not impose a duty on the Secretary of State which was enforceable by mandatory order, but a duty to ‘consider whether or not to exercise the power to appoint a day’ (p575E-F). If the Secretary of State considers the matter and decides not to exercise the power, the power continues to exist. The Secretary of State cannot abrogate it. The power, and the duty to consider whether to exercise it, continue. The power is exercisable, and the duty is to be performed, by the holder of the office for the time being. ‘The obligation will only cease when the power is exercised, or Parliament repeals the legislation’ (p 575H). The executive cannot exercise the prerogative in a manner, or for a purpose, inconsistent with the Secretary of State continuing to perform the duty (p 576A).

On the facts, the Secretary of State had not merely decided, for financial considerations, not to bring the provisions into force for the time being. He had gone further, in two linked ways. First, he had treated the provisions as a dead letter. He had failed to appreciate that, as long as the power was unrepealed, he had a duty to keep the exercise of that power under review. Second, he had replaced the statutory scheme with something else. Pending the exercise of the commencement power, the Secretary of

Section 149(1) of the 2010 Act obliges a public authority, ‘in the exercise of its functions’, to have ‘due regard’ to the equality needs listed in section 149(1)(a), (b) and (c). Those are the needs to:

Section 149(3) explains that having due regard to the need described in paragraph

Section 149(4) provides that ‘The steps involved in meeting the needs of disabled persons that are different from the needs of persons who are not disabled include, in particular, steps to take account of disabled persons’ disabilities.’ Compliance with the duties imposed by section 149 ‘may involve treating some persons more favourably than others; but that is not to be taken as permitting conduct that would otherwise be prohibited by or under’ the 2010 Act (section 149(5)). The relevant ‘protected characteristics’ are age, disability, gender reassignment, pregnancy and maternity, race, religion or belief, sex, and sexual orientation (section 149(7)). They do not include marital status.

By section 149(2), section 149 applies to bodies which are not public authorities, but which exercise public functions. Section 149(9) enacts Schedule 18, which provides for exceptions. By paragraph 4(1), section 149(1) does not apply to a person listed in subparagraph (2) and to the functions listed in sub-paragraph (3). The persons listed in subparagraph (2) include the House of Commons and the House of Lords (paragraph 4(2)(a) and (b)). The functions listed in paragraph 4(3) include ‘a function in connection with proceedings in the House of Commons or the House of Lords’ (paragraph 4(3)(a)).

There have been many decisions in which the duty imposed by section 149 has been considered on the facts of individual cases. C relies on paragraph 26(4) of the judgment of McCombe LJ in Bracking v Secretary of State for Work and Pensions [2013] EWCA

In Bracking, as in many decisions on section 149, there was no argument about the relevant principles (see the second sentence of paragraph 25). The judgment which McCombe LJ cited for the proposition on which C relies (Kaur v Ealing London Borough Council [2008] EWHC 2062 (Admin)), is, in any event, of dubious authority. It was given after the defendant agreed to concede the claim, on the second day of the hearing (see paragraph 10 of the judgment, and see paragraphs 11 and 12 for the Judge’s reasons for giving judgment, nevertheless).

Further, the proposition cited in paragraph 26(4) of Bracking was, in Kaur, expressly based on article 2(b)(ii) of the Race Relations Act 1976 (Statutory Duties) Order 2001,

Article 2(1) imposed a duty on a public authority to publish a race equality scheme, which showed how it intended to fulfil the duties imposed by section 71(1). Article 2(2)(a) imposed a duty on an authority, as part of its published scheme, to state, in particular, those of its functions and policies, or proposed policies, which that person had assessed as relevant to its performance of the duty imposed by section 71(1) of the RRA; and its arrangements for (i) assessing and consulting on the likely impact of its proposed policies on the promotion of race equality; and (ii) monitoring its policies for any adverse impact on the promotion of race equality.

In paragraph 20 of Kaur, Moses LJ interpreted article 2(b)(ii) as a ‘duty to assess the likely impact of those policies on the promotion of race equality…These provisions express the vital principal that the impact of any policy should be assessed and steps to obviate any adverse impact considered before the adoption and implementation of the proposed policy’.

Section 1(1) of the Human Rights Act 1998 (‘the HRA’) defines ‘the Convention rights’. They are set out in Schedule 1 to the HRA (section 1(3)). They include Article

Section 3(1) of the HRA imposes a duty ‘So far as it is possible to do so’ to read and give effect to primary and subordinate legislation in a way which is compatible with Convention rights. Section 3 does not affect the ‘validity, continuing operation or enforcement of any incompatible primary legislation’ or of any incompatible subordinate legislation if (disregarding any possibility of revocation) primary legislation prevents the removal of the incompatibility’ (section 3(2)(b) and (c)).

Section 4(2) applies in any proceedings in which a court determines whether a provision of primary legislation is compatible with a Convention right (section 4(1)). If a court is satisfied of that incompatibility, it may make a declaration of that incompatibility (section 4(2)). If a court is satisfied that a provision of subordinate legislation made in the exercise of a power conferred by primary legislation is incompatible with a Convention right and that (disregarding any possibility of revocation), the primary legislation prevents the removal of the incompatibility, it may make a declaration of that incompatibility (section 4(4)). A declaration of incompatibility does not affect the validity, continuing operation or enforcement of the provision in respect of which it is given and is not binding on the parties to the proceedings in which it is made (section 4(6)).

Section 6(1) makes it unlawful for a public authority to act in a way which is incompatible with a Convention right. But section 6(1) does not apply if, ‘(a) as the result of one or more provisions of primary legislation, the authority could not have acted differently, or (b) in the case of one or more provisions of, or made under, primary legislation which cannot be read or given effect in a way which is compatible with the Convention rights, the authority was acting so as to give effect to, or to enforce, those provisions’ (section 6(2)). ‘Public authority’ includes ‘any person certain of whose functions are of a public nature, but does not include either House of Parliament or a person exercising functions in connection with proceedings in Parliament’ (section 6(6)).

Article 8 of the ECHR provides:

Article 1 of Protocol 1 (‘A1P1’) to the ECHR provides:

Article 14 of the ECHR provides:

A claim based on article 14 requires the court to decide four questions.

Mr Drabble accepted that the test which a court applies in deciding whether a measure in the field of social security is justified is whether the measure MWRF: see R (Carmichael) v Secretary of State for Work and Pensions 2016] UKSC 58; [2016] PTSR 1422 at paragraph 27 and DA v Secretary of State for Work and Pensions [2019] UKSC 21, paragraph 65. At paragraph 120 of DA Lord Carnwath (with whom two other members of the Court agreed) said that ‘It is necessary to distinguish between the general impact of [the measure at issue]…and particular effects on an identifiable group which can properly be the subject of a distinct claim under article 14’. In Langford v Secretary of State for Defence [2019] EWCA (Civ) 1271 the Court of Appeal also took a narrow view of the test stated in paragraph 65 of DA; see paragraphs 52-56, per McCombe LJ. He rejected a submission that the MWRF criterion had to be applied at each stage of the proportionality test. This approach has been followed recently in Drexler v Leicestershire County Council [2020] EWCA (Civ) 502.

Mr Drabble qualified his acceptance of the MWRF test, however, by submitting that in deciding whether or not the measure is MWRF, the court is answering the question whether there is a reasonable relationship of proportionality between the means and the aim. It is clear to me that this approach is inconsistent with the reasoning of the majority in DA.

In his skeleton argument, C made three broad points.

The contention underlying the challenge based on sex is that men are much more likely to be the OP, and on average to die earlier. This means that ‘the 2019 Order’ creates a ‘significantly aggravated risk that male pensioners in MACs will die without ever being eligible for a benefit designed for their age group, or will only be eligible for it for a short time, and will therefore die having received significantly less in benefits than they would have done otherwise’. Disabled people are also, it is said, likely to die earlier.

C argues that because D misunderstood the risk and extent of the disability impact, D was ‘left unable lawfully to consider how it might be tackled’. She was left ‘unable to give informed regard to exempting MACs where the YP was unable to work, either because she was disabled, or a carer. D failed to appreciate that where one member of a MAC is disabled, it is particularly likely that he or she will be unable to work, and that the other member of the MAC will be caring from him or her. Households with disabled people are said to be particularly likely to have no member who is in a position to enter the labour market. ‘It does not seem to have occurred to [D] that the group who would suffer from her policy without even in theory benefitting from it…would be disproportionately likely to include disabled people. It is said that D made her decision on the false premise that only a minority of MACs contains a person who receives disability benefit. It now appears, from statistics gathered after the decision, that the statistics on which D relied related only to the pensioner claimant and not to the MAC. The new statistics show that in a clear majority of (57%) of MACs, someone receives disability benefit. D’s new analysis is, itself, flawed.

Mr Drabble’s principal oral submission was that the very substantial financial impact on this group of disabled people had not been assessed as an impact on disabled people.

Mr Drabble pointed out that section 8(2)(d) of the 2012 Act gives the Secretary of State considerable scope to adjust the rate of benefits to meet particular needs. It was relevant ‘when it came to a decision to commence’ section 4(1A) of the SPCA. The fact that the Secretary of State could include an amount for a carer in UC did not alter the fact that this change took the MAC out of the SPC regime. He accepted that the Secretary of State could not abandon paragraph 64 (and, therefore, section 4(1A)), but ‘You can ensure that the circumstances are such that it is acceptable to commence paragraph 64. You can make sure that you do not do it until there is a regime in place. If you are going to put everyone into UC you have to ensure that the regime does not impose too big a hit’. In deciding whether or not to commence paragraph 64 in particular circumstances at a given time, the Secretary of State had to acquaint himself with the principal adverse effects and treat them as relevant to the discharge of the section 149 duty. The Secretary of State had to assess the risk of an adverse impact and

The facts demanded the creation of an exception from the general rule about MACs for those in the position of C, just as such an exception was created in R (Burnip) v Birmingham City Council [2012] EWCA (Civ) 629; [2013] PTSR 117and in R (Carmichael) v Secretary of State for Work and Pensions [2016] UKSC 58; [2016] 1 WLR 4550. Mr Drabble relied on the statement by Lord Nicholls in the FBU case that the Secretary of State has a wide degree of discretionary judgment as to whether or not it was appropriate in the circumstances to commence paragraph 64. Section 149 required the Secretary of State to have due regard to the impact on people with protected characteristics of bringing paragraph 64 into force at a particular moment, and to consider the size of that impact. The size of the impact could be changed by adjusting other elements of the legislative scheme as they affected C, or by deciding that, given the size of the gap, that the impact was too big. The size of that gap might narrow at some point in the future. He could decide to impose a notional loss on people without protection, but only if he assessed it ‘properly’; or he could decide not to commence the provision now. None of this was analysed in the FBU case. As long as the Secretary of State did not substitute an inconsistent scheme under the prerogative, he could delay commencing paragraph 64.

The consideration of the position of MACs was in flux when Royal Assent was given in April 2012. It is now a long time since Royal Assent. D had to comply with section 149 in the context which was current when D was considering whether or not bring paragraph 64 into force. D had to decide whether it was appropriate to commence paragraph 64 in 2019. Section 149 required D to investigate the impact on the listed equality needs of commencing paragraph 64 in 2019. Mr Drabble initially submitted that section 149(1)(a) was engaged because there was a discrimination issue. He withdrew that submission in his reply, and relied instead on section 149(1)(b). Sections 149(3)(a) and (b) and (4) were potentially relevant. Putting the disabled into a group which included all able-bodied people and which exposed them to a lower rate of benefit exposed disabled people to a disadvantage under section 149(3)(a). To do so when there is a big hit did not minimise the disadvantage. Section 149 required the Secretary of State to look at the size of the disadvantage at the moment when he decided to commence paragraph 64. The section 149 duty is to be grafted onto the principles set out in the FBU case. Parliament had decided the policy, but had not decided whether it was appropriate in current circumstances to commence paragraph 64. It was not an answer to economic problems caused by commencing paragraph 64 that Parliament had decided to enact it eight years previously. D did an impact assessment. On its face, it was intended to inform D about whether it was appropriate to commence paragraph 64.

The protected characteristics which were relevant were age, disability and sex. Mr Drabble did not make oral submissions about age and sex, relying on his skeleton argument. The focus of his oral argument was disability, which he described as ‘the sharpest analysis’. He referred to some of the documents. He submitted that MACs were not mentioned in the initial impact assessment. The treatment of disability took matters no further. The Ministerial Submission of 3 December 2010 recognised that where the YP was subject to conditions, she would be required to claim UC, but recommended that an allowance be paid in UC to ensure that MACs should not be disadvantaged. There was no discussion in the documents of a MAC in which the YP was not subject to conditions. A memo referred to in paragraph 40 of Mr Latto’s witness statement did mention this group. At that stage, it was accepted that the MAC could be entitled to PC if the YP was a carer. This group was again considered expressly on 6 July 2011. A letter dated 26 September 2011 from the Minister to the members of the relevant Commons Committee said that this could be dealt with by a caring component in UC. In December 2012, there was an impact assessment after Royal Assent. There was no assessment of the impact on this group. MACs are referred to, but not MACs who are not subject to conditions. The focus of the discussion is returning to work. The 2018 EA did not deal with this issue, either.

Mr Drabble submitted that D had misunderstood the risk and extent of the disability impact. D was obliged to inform himself of the impact in sufficient detail, and had not done so. He accepted that the primary legislation had to stay intact, but D had to consider whether or not it was appropriate in the circumstances to commence in 2019 and had to give informed regard to that question. He withdrew the submission in paragraph 22 of his skeleton argument that D was ‘not able to give informed regard to exempting MACs where the [YP] was unable to work (typically because of caring for their disabled partner or because of their own disability)’. D had failed to realise that when one member of a MAC is disabled, it is particularly likely that they will not be able to work, and particularly likely that the other member is caring for them. The households of disabled people are particularly likely to contain no-one who can enter the labour market. It did not seem to have occurred to D that the group who would suffer from the policy without being able to benefit from it (a MAC in which the YP cares for the OP) was disproportionately likely to include disabled people. The change for this group did not cohere with the policy which underlies it.

In his reply, he contended that the wide range of circumstances which D could take into account in deciding whether or not to commence paragraph 64 included that ‘the hit was too big’. There was nothing in the reasoning in the FBU case which required D to commence paragraph 64 when he did. D had to look at the size of the hit and decide in the light of that whether it was appropriate to commence paragraph 64. It was perfectly possible for legislation to sit in the statute book for a long time without being commenced. In April 2011, and in 2012, it was not proposed that this should be a measure for saving costs (see paragraph 36 of Mr Latto’s statement and E 1129) and that had changed. The size of the hit was not part of the wording of paragraph 64. At the point when Parliament decided to put MACs in UC and not to have a pension premium in UC, the system was still evolving. The content of the measure had changed over time and would change in the future. D had to have due regard to the impact of commencing paragraph 64 in 2019 as opposed to later. There was a greater impact now as a consequence of current rates of benefit. That was not precluded by the structure of the legislation. The effect of section 149 was that D could not bring paragraph 64 into force unless he had had due regard to the size of the hit in terms of the equality needs.

C made four main points in his skeleton argument.

Mr Drabble did not develop any argument about HB and council tax support. I consider that they do not give rise to any distinct issues. The fact that C and his wife will not be entitled to pensioner HB is a direct consequence of the fact that they will be in the UC regime, not the PC regime. Council tax support is the responsibility of local authorities, not of D.

Many of the points about differential treatment and justification which are made in C’s skeleton argument overlap with or duplicate the points C made about adverse impact in the context of the section 149 challenge. For example, disabled male pensioners are significantly more likely than ‘their comparators’ (not described) to die without ever becoming eligible for PC. The group of people unable to respond to ‘the work incentivisation rationale and therefore the group harmed without receiving any benefit consists disproportionately of disabled people. C is treated less favourably than a person who is below pension age because he is not allowed to claim the benefit which is appropriate to his age. He is subjected to the same treatment as a younger person, when he should be treated differently (cf Thlimmenos v Greece [2000] EHRR 41). There is direct age discrimination, not on the grounds of C’s age, but on the grounds of his wife’s age. C is treated less favourably than someone who is his age, but whose partner is older.

The ‘fundamental rationale’ for this change, identified in the 2018 EA, is to ensure that the YP has ‘the appropriate conditionality’ applied via UC ‘and receives the relevant support to find work’. It does not apply to C and to his wife, or to the group to which they belong, that is, ‘no-conditionality MACs’ (those members of MACs in which the YP is not subject to conditions because she is disabled, or because she is caring for someone who is disabled). That means that paragraph 64 is MWRF. A less intrusive way of achieving ‘substantially the same’ objective would be either to allow pensioners in no-conditionality MACs to receive PC or to provide no-conditionality MACs with a pensioner premium in UC.

Mr Drabble introduced his oral submissions by describing what paragraph 64 does. Before it was enacted, the OP could, when he reached SPA, claim SPC, even if the YP was under SPA. A couple are now ‘forced’ to claim UC, with substantial financial consequences. The MAC is taken out of the regime for those of SPA and into the regime for those below SPA. This change affects more men than women. It creates a particular problem for a discernible and readily identifiable group of disabled people, that is, all OPs who are disabled and who are cared for by a YP. The YP is not therefore subject to work-related requirements under the regime for UC (‘conditionality’) with the consequence that the rationale for the treatment of MACs generally does not apply. He accepted that, in general, whether a MAC should be assigned to SPC or to UC was policy choice, but submitted that the rationale for assigning MACs to UC when the YP is subject to work-related requirements does not apply to this defined group of MACs. There was therefore no justification for treating C and his wife differently from a MAC who are both above SPA. C and his wife were being taken out of SPC for reasons which did not apply in his case.

The position of that group of disabled claimants is directly comparable to that of the disabled claimants in Carmichael (paragraphs 42-49) (see paragraph 106, above). A rule is applied to both which is applied equally to able-bodied people, but it has a disproportionate effect on the disabled group, an effect which has no relationship with the legitimate aim which the rule aims to achieve. A less intrusive means of achieving the legitimate aim would be to allow the OP in a ‘no-conditionality MAC’ to claim SPC, or to give the OP a pensioner premium in UC. He also relied on R (Hurley) v Secretary of State for Work and Pensions [2015] EWHC 3382 (Admin); [2016] PTSR 636, while accepting that Collins J had not explained how he reached his conclusion in that case.

C’s situation was predictable and commonplace. While some people might ‘flit in and out of conditionality’, there is a stable group of MACs in which OP is disabled and YP is a carer. It followed, as night follows day, that the rationale of incentivising work does not apply to them. The imposition of a notional loss on this group was a detriment to them which was not rationally connected with the aim of incentivising work. Discussion

There are three issues.

As I have said, the judgment in Kaur was given after there was no longer any dispute for the court to resolve, and, it is therefore likely, not after full argument. Moses LJ was considering a different statutory scheme. It is not clear to which ‘policies’ the phrase ‘those policies’ in the first part of the sentence from paragraph 20 of the judgment, which I quoted in paragraph 90, above, refers. I do not consider that on its proper construction, paragraph 2(b)(ii) of the Order imposed any duty on a public authority to assess the effect of each policy it introduced. What it did, rather, was to impose a high-level duty to publish a scheme which said how the authority intended to assess the likely impact of proposed policies on the promotion of race equality and how it intended to monitor its policies for adverse impacts on the promotion of race equality.

I therefore consider that the judgment in Kaur isnot persuasive, and, indeed, that it is clearly wrong. As I have said, it is also plain from the judgment in Bracking (given by McCombe LJ) that there was little, if any dispute about the supposed legal principles which are stated in paragraph 26 of the judgment. It follows that I should approach the proposition stated in paragraph 26(4) of Bracking, which is said to be based on Kaur, with caution. That caution is reinforced by the approach of the Court of Appeal in R (Bailey) v Brent London Borough Council [2011] EWCA Civ 1586, in which the Court of Appeal, after full argument, deprecated minute forensic criticisms of the way in which a local authority had assessed the impact on the equality needs of the closure of some of its libraries.

That caution is further reinforced by the approach of McCombe LJ himself in paragraph 44 of a later decision of the Court of Appeal, Powell v Dacorum Borough Council [2019] EWCA (Civ) 23. He said that the previous decisions about section 149 have to be taken in their contexts. The way in which section 149 will apply on the facts will be different in each case, depending on what function is being exercised, and on the facts. The judgments, including the judgment in Bracking, must not be read as if they were statutes. He referred, with approval, to a similar statement by Briggs LJ (as he then was) in paragraph 41 of Haque v Hackney London Borough Council [2017] EWCA (Civ) 4. See also, more recently, and in the context of the exercise of a different function, McMahon v Watford Borough Council [2020] EWCA (Civ) 597, at paragraphs 68 and 89, per Lewison LJ. Section 149 and commencement orders

On paper, D seemed to be disposed to argue that section 149 does not apply at all to decisions to commence primary legislation. D wisely modified this extreme submission just before, and at, the hearing. There is no express exception from section 149 for a minister making such a decision, in contrast to the exception which applies to both Houses of Parliament in respect of proceedings in Parliament. I accept that section 149 does apply to a decision to commence legislation. But this context informs the regard to the listed equality needs which is ‘due’.

Although the FBU case concerns a refusal to commence legislation, rather than a decision to commence it, the speeches of the majority explain (absent section 149) some of the principles which govern the exercise of a power to commence legislation. I reject the submission (if made) that a minister may decide not to commence legislation if a change in circumstances makes it inappropriate to commence it, if it suggested that a

Some of Mr Drabble’s submissions came close to implying that, contrary to the clear decisions in the long history of this provision that MACs should be eligible for

There are many aspects of the context which inform the degree of regard which was due to the listed equality needs when D took the decision to make the 2019 Order. I list some of them now.

I reject the submission that in making the 2019 Order D failed to have regard to the

I am conscious that I should not fall into the trap of looking at D’s assessments of the impacts of a wide-ranging group of reforms through the very limited focus of a lens provided by the complaint which C makes. I reject the specific criticisms C directed at the 2018 EA, for three reasons. First, the 2018 EA is not sole evidence of the regard which D had to the listed equality needs, for the reasons I have given. Second, in the context of the factors I have listed, section 149 did not require D to analyse the impact of the MAC policy in the way that C submits. Third, in any event, the consideration which D did give to the impacts of the MAC policy enabled D to understand its principal effects on people with protected characteristics, and, in that way, to give the regard to the listed equality needs which was due when D exercised the commencement power.

There is no apparent dispute that this claim falls within the ambit of a Convention right (that is, article 14 read with A1P1). That being so, I can see no point in resolving the limited potential dispute about whether it is also within the ambit of article 8. As D accepts, the real question is justification. That question involves the same factors whether or not article 8 is in play.

D argues that C cannot show that there is a difference in treatment of people in analogous or relevantly similar situations on grounds of sex. The 2018 EA showed that about 13% of men who were claiming PC or PHB, or whose partner did, were in a MAC, whereas about 7% of women who were claiming both benefits, or whose partner was, were in a MAC. Proportionately fewer women are in a MAC because a significant number of women are sole claimants after the death of their partner. Fewer men will be the surviving partner in a MAC because they have a shorter life expectancy than women. Women also on average have lower pensions than men and so are more likely to need means-tested support in retirement. But while the MAC policy affects a higher proportion of male claimants than women, the absolute numbers of claimants of each sex who are affected is the same (because most MACs consist of one person of each sex). This led D to decide that the MAC policy did not disproportionately affect one sex rather than the other. While it was open to D to analyse the effect on the sexes in this way in the 2018 EA, I do not consider that this way of analysing the effect entirely disposes of C’s argument about disparate impact.

D also argues that there is no similar treatment of people whose situations are relevantly and significantly different (and who ought, therefore, to be treated differently) on grounds of disability (ie there is no Thlimennos claim based on disability). D argues that, as in R (Drexler) v Leicestershire County Council [2019] EWHC 1934 (Admin) paragraphs 54-55, a disabled person or a carer in a MAC is not treated in the same way as an able-bodied person on UC, because sums are paid under UC to help disabled people and carers. There was an appeal to the Court of Appeal in Drexler (2020] EWCA (Civ) 502). I gather from the terms of the judgment of the Court of Appeal that leave to appeal on this point was not granted. I accept this argument.

D accepts that it is arguable that there is differential treatment on grounds of age of C and of the age of his wife, and on grounds of relationship status.

In case my conclusion about discrimination on the grounds of disability is wrong, I will consider whether all the differential impacts which C asserts are justified. I have considered the policy documents at some length. It is clear from those that the legislators, and D, were aware of the financial effects of the MAC policy and of the impact of the MAC policy on couples in which one member of the couple was disabled, or a carer. The effects of the MAC policy, generally, and in the case of no-conditionality MACs, are not, therefore, a surprising anomaly which no-one was aware of until this litigation. The general and specific effects were known about throughout. Whether to adopt the MAC policy in the first place, and whether, if adopted, to adapt it in the case of no-conditionality MACs, are paradigm examples of policy decisions. That is why the MWRF test applies when the court considers whether any differential treatment is justified.

The factors which are relevant to the question whether the application of the MAC policy to no-conditionality MACs is MWRF overlap with the factors which are relevant to section 149. They are not the same, because whether differential treatment is justified is an objective question for the court, and does not simply depend on what D thought when D decided to make the 2019 Order (though I recognise that I should defer appropriately to D’s judgment about issues of socio-economic policy, and that D has considered this policy, and its impact, over a long period).

What is said in the 2018 EA to be ‘the fundamental rationale’ for the MAC policy generally is precisely that. The fact that the fundamental rationale for the MAC policy generally does not apply to no-conditionality MACS does not mean that the other policy reasons for the MAC policy, and for UC generally, do not apply to no-conditionality MACs. At least four do apply.

First, the structure of UC gives better financial incentives to the YP in a noconditionality MAC who, like C’s wife, is able to work a few hours a week, to take a ‘mini-job’, and thus to get back into work, and decrease the MAC’s reliance on benefits, than there are in PC. The fact that C’s wife cannot be required to work does not entail that she should, either, not be expected to work if she can, or that, if she can work, she should not receive financial incentives to do so. Both sides agree that, like any other claimant of UC or PC, she would be able to ask for help from a JobCentre to support her to find work, if she chose to do that. Second, a general rule which applies to all MACs is simpler to understand and to administer than a general rule with an exception. Third, C does not suffer an actual, but only a notional loss. Fourth, UC has components for those who are disabled and for those who are carers. Finally, to the extent that this

For those reasons, I do not consider that the MAC policy, in its application to noconditionality MACs, is MWRF. It follows that the differential treatment of which C complains is not a breach of article 14 (whether read with article 8 or with A1P1), and, therefore, that the 2019 Order, which provides for the commencement of paragraph 64 of the 2012 Act (and by that route, of section 4(1A) of the SPCA) is not incompatible with C’s Convention rights. Conclusion

For these reasons, when D made the 2019 Order, D had the regard to the equality needs listed in section 149 of the 2010 Act which was due, and the 2019 Order (and the provisions it commences) are not incompatible with C’s Convention rights. I will give permission to apply for judicial review, but I refuse the substantive application for the reasons given in this judgment.