England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM
MR JUSTICE SWIFT
[2019] EWHC 1450 (Admin)

		Royal Courts of Justice Strand, London, WC2A 2LL
		1 December 2020

Lady Justice Simler:

Introduction

1. This appeal concerns a 13 year old school girl, referred to as ZK, who brings these proceedings by her mother and litigation friend, HK. ZK is totally blind as a result of a brain tumour, and partially deaf, in consequence of the same tumour. In consequence of her severe visual impairment and resulting special educational needs, she requires a high level of support. She has been and is being educated in mainstream schools in the London Borough of Redbridge (referred to as "Redbridge"), the respondent to this appeal.
2. The essential challenge pursued on ZK's behalf by HK is to the failure by Redbridge to adopt a "centralised model" of specialist educational support for children with visual impairment in mainstream schools whereby the local authority recruits and employs a central pool of specialist teaching assistants who are then seconded out to the school which the child attends. Instead they adopt a "decentralised model" under which it is the school that recruits and employs a specialist teaching assistant, if and when it has a child attending with that particular need. The decentralised model is said to be incapable at a systems level, of meeting the special educational needs of disabled children like ZK who have a high level of special needs support, including support from specialist teaching assistants, who are in turn, supported by qualified teachers of the visually impaired ("QTVIs"). Redbridge contracts with a specialist provider, the Joseph Clarke Educational Service ("JCES"), to provide the services of QTVI teachers, and to train the specialist teaching assistants. Under the Redbridge model, however, it neither employs those QTVIs, nor employs or trains the teaching assistants who work with visually impaired pupils in mainstream schools. Instead they are employed by the schools in which they work. This is the "decentralised model" under challenge.
3. In her first witness statement HK identified the problems with a decentralised model as including that because teaching assistants are employed by the school and responsible only to the head teacher of a school, there is no central input from the local authority which has the responsibility to secure the provision of special educational support for the disabled pupil. That means that teaching assistants often have no specialist training, for example braille knowledge and/or qualifications and almost never have skills in STEM subjects; and there is no opportunity for continuous professional development for general teaching assistants under a decentralised model. It is her case that this model limits the pool of appropriately skilled staff available to support children like ZK, leaving them vulnerable to having no adequately skilled support as required on transition to a different school (for example, transition from primary to secondary school) or if the designated teaching assistant is absent and in consequence, is inherently unlawful and at odds with Redbridge's compliance with the statutory obligations it owes to disabled pupils with special educational needs. HK also said that the decentralised model results in a lack of choice of schools in Redbridge for pupils like ZK, and a risk of them being forced out of mainstream schools into specialist education at long distance from home with the difficulties (social and travelling time) that entails.
4. Swift J rejected the judicial review claim. His essential reasons so far as relevant to the grounds of appeal advanced on behalf of ZK can be summarised as follows:

i) Irrationality: the judge accepted that the arrangements Redbridge has in place may not be perfect, fool-proof, or fail-safe. He acknowledged evidence both from HK, and from the mother of another child who suffers severe visual impairment that was critical of how educational matters relating to their children had been addressed on occasion (and that might afford the basis for specific challenge). However, this challenge was a systemic challenge at a generic level, to the general arrangements in place. He concluded that at the generic level, the arrangements made by Redbridge for the provision of services from JCES to support the teaching of visually impaired pupils in mainstream schools are not such as to be irrational, and are not such as to warrant an inference that they have been entered into without due regard for relevant considerations.

ii) Illegality: the judge did not consider the arrangements in place under which specialist teaching assistants are employed by schools and trained and supported by JCES gave rise to any inherent likelihood that Redbridge would fail to comply with its legal obligations. He was satisfied Redbridge's arrangements, in particular for the management of transfers between schools at the end of Year 6 for the beginning of Year 7, are sufficient when considered at a generic level, and do not entail the inherent likelihood that Redbridge would fail to comply with its legal obligations.

iii) Unlawful discrimination: the judge dismissed ZK's unlawful disability discrimination claims. Whilst he accepted that these claims raised a different question from the earlier grounds, the outcome was the same; the disadvantages claimed were the same as those relied on in relation to the irrationality and illegality grounds. The claims failed on their facts and the judge was satisfied that the arrangements were not such as inherently to give rise to disadvantage. The issue of justification did not arise.

iv) Public sector equality duty under s.149 Equality Act 2010: the judge rejected ZK's argument that Redbridge failed to have due regard to the special educational needs of VI pupils, and the need to eliminate discrimination between such pupils and other pupils in mainstream schools, by maintaining the decentralised model. The fact that the very purpose of the arrangements Redbridge makes with JCES from year to year is to ensure that what is required to meet the assessed needs of visually impaired pupils is available, means the very purpose of the exercise was to seek the elimination of discrimination between those pupils and other pupils. That purpose was said to run through to Redbridge's decision on what services should be secured from JCES. Accordingly this claim also failed.

5. Mr Nicholas Bowen QC who appears with Mr David Lemer on behalf of ZK (but did not appear below), challenges those conclusions on three broad grounds as follows:

i) The judge wrongly found that Redbridge's decentralised model was not irrational and/or unlawful because he failed to apply the correct test for determining whether a public body's policy is unlawful, erred in his interpretation of section 42 of the Children and Families Act 2014, and made a number of findings of fact which were not open to him on the evidence.

ii) The judge erred in finding that the decentralised model did not cause disadvantage to children with severe visual impairment such as ZK.

iii) The judge erred in inferring compliance with the duty under section 149 of the Equality Act 2010.

6. The appeal is resisted. Ms Deok Joo Rhee QC who appears with Mr Tom Tabori for Redbridge contends that the judge made no error of law; and in relation to all three grounds, contends that the appeal turns largely on challenges to the judge's factual findings which were entirely open to him on the evidence in the case and cannot be impugned.
7. We are grateful to all counsel for their written and oral submissions in this sensitive case.

The legal framework governing special educational needs

8. A child or young person has special educational needs if he or she has a learning difficulty or disability which calls for special educational provision to be made for him or her.
9. The current statutory provisions governing special educational needs and disability provision are in Part 3 of the Children and Families Act 2014 ("the 2014 Act") which replaced the previous scheme in Part 4 of the Education Act 1996. Local authorities are under a duty to exercise their functions with a view to ensuring that all children and young people with learning difficulties or disabilities in their areas are identified. The parent of a child or young person may request an assessment of the educational, health care, and social care needs of a child or a young person (and there is a right of appeal against a refusal to assess): sections 36 and 51 of the 2014 Act.
10. In the light of an assessment, if it is necessary for special educational provision to be made, the local authority must secure preparation of and once prepared must maintain, an education and health and care plan ("the EHC plan") (which replaced statements of special educational need under the 1996 Act). The EHC plan will specify, among other things, the special educational provision required for the child or young person: section 37 of the 2014 Act. There are rights of appeal against the content of an EHC plan: section 51 of the 2014 Act.
11. Regulations governing the form of an EHC plan, the Special Educational Needs and Disability Regulations 2014, provide by regulation 12(1) for separate identification in the EHC plan of inter alia the following: "(b) the child or young person's special educational needs (section B); …. (e) the outcome sought by him or her (section E); (f) the special educational provision required by the child or young person (section F); …(i) the name of the school … to be attended by the child or young person and the type of that institution or, where the name of a school or other institution is not specified in the EHC plan, the type of school or other institution to be attended by the child or young person (section I);…."
12. Section 42 of the 2014 Act imposes a duty on the relevant local authority to secure special educational and health care provision in accordance with the EHC plan. It provides, so far as is relevant:

"42(2) The local authority must secure the specified special educational provision for the child or young person. ……

(6) 'Specified', in relation to an EHC plan, means specified in the plan."

13. As Lewis J observed in relation to the section 42(2) duty in R (Simone and others) v Chancellor of the Exchequer and another [2019] EWHC 2609 (Admin) at [8]:

"It is well-established law, and accepted by all parties, that a local authority must ensure that the special educational provision specified in the EHCP is provided to the child or young person. Furthermore, that obligation is not dependent on available resources: the local authority is obliged to secure that the specified special educational provision is made available (see R (N) v North Tyneside BC [2010] EWCA Civ 135 and the reasoning, in relation to a different provision, of the House of Lords in R v East Sussex County Council ex p. Tandy [1998] AC 714). "

There is no dispute between the parties that this statement reflects the well-established law relating to section 42(2) of the 2014 Act (and its predecessor sections in the earlier legislation) which is part of a scheme for achieving the "best possible educational and other outcomes" for children and young people with special educational needs and disabilities (see section 19(d) of the 2014 Act). It imposes a mandatory duty to secure the special educational provision specified. The word "secure" is an ordinary English word and needs no gloss. What is plain is that the duty has no "reasonable endeavours" escape clause available to excuse failure to secure the provision specified. It was accepted by Ms Rhee QC on behalf of Redbridge that once an EHC plan is in final form and specifies a particular provision, the duty to secure the special educational provision specified in it is absolute and must be met by the local authority.

14. Section 43 of the 2014 Act imposes a duty to admit on the school or institution named in section I of an EHC plan.
15. Where parents or young people disagree with the local authority about the contents of sections B and/or F of an EHC plan, or about the placement named in section I, they have a right of appeal to the First-Tier Tribunal (the Special Educational Needs and Disability Chamber). The Tribunal can rewrite the contents of the EHC plan in accordance with its judgment as to education needs (section B), provision required to meet needs (section F) and appropriate placement (section I).
16. Section 44 provides for reviews and re-assessments to be conducted in relation to EHC plans maintained by a local authority, every 12 months.

The factual background

17. ZK has an EHC plan first issued by Redbridge in July 2017. Her most recent EHC plan for the purposes of these proceedings is dated 13 September 2018 and was amended following an appeal to and judgment of the First-tier Tribunal. Her EHC plan specifies Oaks Park High School as the mainstream secondary school at which ZK should be placed (section I). As to the remaining sections, section A ("All about me") is based on information provided by ZK's parents and sets out her background and aspirations, including the following:

"Teaching assistants working with ZK need to be trained by RNIB (in Partners in Learning), and able to use specialist equipment including Braille Note Touch, Embosser. They need to be fluent in UEB Braille 2".

18. Section B (which describes ZK's special educational needs) describes ZK as being within the average to higher/above average range for her intellectual ability; her English and Maths skills are age appropriate. She is also described as:

"learning Grade 2 Unified English Braille (UEB), making good progress and reading well with this new code as it is introduced to her and can read braille independently. …She is developing her touch-typing skills and records her work using a Perkins Brailler [a form of typewriter]".

19. Section E ("My Outcomes") describes her long-term aspirations, including that ZK wants to be an independent learner and to live as normal and independent a life as possible. Specifically, it identifies that by the end of Key stage 3 (the end of Year 9 when ZK would be rising 14) she would have literacy and numeracy skills at a level where she could "use fully contracted grade 2 UEB braille in written work…"
20. Section F ("My Special Education Provision") specifies in detail the special education provision that must be secured for ZK (pursuant to the section 42(2) duty) by reference to her day to day educational needs and the resources necessary to meet those needs. In terms of the specific resources required to be secured to meet the outcomes, it includes the following:

"ZK requires access to a full-time teaching assistant (32.5 hours per week dedicated teaching assistant support).

From September 2018, in addition to the full-time LSA (learning support assistant) ZK shall have an additional adult dedicated to her (for 25 hours per week) to allow sufficient time to prepare braille resources in advance of each lesson. …

The teaching assistant intervention shall be from teaching assistants trained in contacted UEB braille [this should read contracted UEB braille] to allow her full access to the curriculum and the full range of public exams.

From September 2018 ZK will have a total of 58.5 teaching assistant hours support per week. …

An hour per week should be given to any staff learning Braille and teaching specialist skills in Braille. This will be particularly relevant within secondary school for those staff who do not have the appropriate level of Braille.

In the first term, in that setting the QTVI may also be required to provide marking and overwriting of ZK's Braille work in the event that the teachers/teaching assistants do not have the necessary expertise in Braille. This will require an hour per day which can be aggregated over the week…

Within secondary school there will be a requirement for specialist Braille skills for specific subjects to be taught or work prepared including in the subjects of maths, chemistry, languages and music. …"

21. In addition, section F of the EHC plan specified QTVIs to provide advice to ZK's teachers and teaching assistants for 13 hours each week and with effect from September 2018 (i.e. on her transfer to secondary education) ZK's school would need access to a QTVI for at least 15 hours per week.
22. There was a dispute before us (though apparently not before Swift J) about the precise nature of the requirement contemplated by the reference in section F to the teaching assistant support for ZK being from "teaching assistants trained in contracted UEB braille". Mr Bowen QC contended that what was required were teaching assistants who were not only trained to at least grade 2 UEB braille but had also been awarded the formal qualification confirming the attainment of that level of competence, a process that takes at least a year. He relied on what he submitted is Swift J's finding to this effect at [4] where he held: "The teaching assistants must be trained in contracted UEB. Contracted Braille is the Grade 2 UEB code, which is materially different from the Uncontracted, Grade 1 code." For Redbridge Ms Rhee submitted that Swift J did not make a finding that the qualification was required; there was no agreement between the parties that this should be specified in the EHC plan, and instead it was agreed and specified that the teaching assistants would be trained in contracted UEB braille. By implication, she submits, it was understood that the teaching assistants may not yet have obtained their grade 2 qualification.
23. I do not accept Mr Bowen's submission that Swift J made a finding that ZK's teaching assistants should have the grade 2 qualification. He found that they should be trained in contracted UEB braille. There was evidence to support that finding. First, it appears from the witness statement of Conny Kaweesa, a Conciliation and Tribunal Officer who had conduct of the statutory appeal to the Tribunal on behalf of Redbridge, that this was a contested issue in the appeal lodged on 3 September 2017 on behalf of ZK. In particular she makes clear that it was disputed by Redbridge that ZK's teaching assistants would need to have a braille qualification in grade 2 UEB braille. Secondly, the EHC plan, read as a whole, indicates that there was no requirement for the teaching assistants to have achieved the grade 2 qualification at the point when they started working with ZK in Year 7: the references to "grade 2" in the EHC plan does not appear in section F, but rather in other sections that do not deal with the provision of resources required. For example, it appears in section A, dealing with ZK's aspirations; section B, dealing with ZK's own needs (but without reference to her teaching assistants); and section E, dealing with outcomes. This is also consistent with the fact that ZK was herself developing the use of grade 2 UEB braille but had not achieved the qualification, and with the recognition in section F that her teachers and teaching assistants might not have the necessary expertise in braille and might therefore require support from QTVIs for marking and overwriting of her braille work. Accordingly, I proceed on the basis that ZK's EHC plan required teaching assistants trained in UEB contracted braille but did not require the grade 2 qualification to have been achieved by September 2018.
24. At the date of the judgment below, ZK was in Year 8, her second year of secondary education having transferred from primary school at the beginning of the 2018-2019 academic year to Oaks Park secondary school.
25. Swift J described the substance of Redbridge's "decentralised model" as relied on by ZK, as the set of arrangements Redbridge had put in place to obtain the resources necessary to meet the needs of visually impaired pupils who attend mainstream schools within the borough:

"14. A starting point is funding. Mainstream schools receive funding support to meet the cost of special needs education from the Department for Education. This funding meets the first £6,000 of support costs. When, as is the case for a pupil such as the Claimant, the cost of the support required exceeds that amount, the additional costs are met by the local authority.

15. The present position in Redbridge is that there are 168 pupils in mainstream schools in the borough who receive some form of support from JCES. This number includes those with special educational needs arising from visual impairment; it also includes a number of pupils who require support to access education but who are not the subject of an EHCP. There are 33 pupils whose primary need for special provision arises from visual impairment. However, there are presently only 3 pupils in mainstream secondary schools in Redbridge who use Braille (one learning Braille, the other two competent Grade 2 Braillists), and 3 pupils in primary schools (who are both learning Braille). Redbridge relies on these figures to describe the need for specialist Braille support in mainstream schools as a low-incidence need. The Claimant does not dispute that description.

16. Redbridge contracts with JCES to obtain the services necessary both for VI pupils in mainstream schools within the borough, and for other pupils in mainstream schools who have identified special educational needs in consequence of less severe forms of visual impairment. JCES provides what is described as a specialist outreach advisory service for children and young persons in mainstream schools. The services are provided under an annual contract. I have been provided with a copy of the contract dated 1^st September 2018 which covered the period 1^st April 2018 to 31^st March 2018 [This should be 31 March 2019]. The contract is made between Redbridge and the Whitefield Academy Trust – JCES being part of that Academy Trust. The contract includes a detailed service specification, and provision (at clause 3) for variation of the specification by agreement, and obligations of cooperation between JCES and Redbridge. One example of the variation provision is that the September 2018 contract was itself amended with effect from January 2019. Under the contract, JCES provides the services of a number of QTVI teachers; as amended, the contract provides for provision of the equivalent of 2.6 full time QTVI teachers. JCES also provides the services of a habilitation specialist (equivalent to 0.25 of one full-time employee) and provides administrative support staff (again equivalent to 0.25 of one full-time employee).

17. One responsibility of the QTVI teachers is to train staff in the mainstream schools that VI pupils attend. This includes but is not limited to, training the specialist teaching assistants. The training concerns all matters relating to the teaching of VI pupils, including teaching Braille and how to use specialised equipment: see the contract specification at paragraph 5.2.2. Redbridge's evidence is that this part of the support includes providing annual courses for teaching staff, and INSET training at specific schools based on the particular needs of the VI pupils at those schools. Redbridge's evidence also points out that the training for teaching assistants is tailored to the specific needs of the child that teaching assistant has been recruited to support – for example, taking account of whether the child uses Braille, and taking account of what is needed by way of lesson preparation for that child. The QTVI teachers provided by JCES provide an outreach service, i.e. they visit relevant mainstream schools to provide on-site training and assistance to the relevant teaching staff. The habilitation specialist also provides advice and assistance to relevant schools designed to develop the pupil's independent skills. This advice includes audit of a school premise."

26. There was evidence put forward on behalf of ZK below contrasting the arrangements Redbridge has put in place with arrangements in some local authorities that directly employ ready trained specialist teaching assistants, and also QTVI teachers, so that the teaching assistants are ready to be deployed into any mainstream school due to receive a visually impaired pupil. Evidence from Mr Vic Gibson, (a QTVI who, until his retirement in 2016 worked for Peterborough City Council) described Peterborough's Sensory Support Service which employed a number of QTVIs and specialist teaching assistants, most working in schools which visually impaired pupils were encouraged to attend. Mr Gibson also explained that when such pupils chose not to attend those schools the Sensory Support Service provided an outreach service to the schools which they did attend. This or some other type of "centralised model" is relied on by ZK as enabling a visually impaired pupil's transition from one mainstream school to another, with a suitably trained teaching assistant able to be deployed immediately to the new school.
27. There was evidence that a number of local authorities do have centralised arrangements – in the sense that they employ specialist teaching assistants themselves. However, the evidence showed that many local authorities do not, and instead have arrangements along the lines of those in place in Redbridge. RNIB data (published in 2015) indicated that 45% of local authorities did not themselves employ the teaching assistants provided to visually impaired pupils. However, the 55% of local authorities that did employ teaching assistants directly only accounted for the employment of 495 teaching assistants. The vast majority of teaching assistants, some 2,600, worked in the areas of the remaining 45% of local authorities, and were directly employed by the mainstream schools in which they worked.

The challenge to findings of fact made by Swift J

28. Before turning to the grounds of appeal pursued on ZK's behalf, it is necessary to consider the other main findings made by Swift J in relation to the disadvantages and failings relied on by ZK in the Redbridge arrangements: the time taken to recruit and train a teaching assistant (the time-lag issue); the risk that specialist teaching assistants could be redeployed to other duties by individual schools; the availability of cover for absence of specialist teaching assistants; and the contention that choice of schools for visually impaired pupils is reduced. These findings form the factual foundation for the ultimate conclusions reached by the judge. Many are challenged by Mr Bowen on the basis that they were not open to the judge in this case.
29. The time-lag issue: This is dealt with by Swift J at [19] to [24] as follows:

"19. ….The most obvious context (and the one which was the focus of the submissions made) will be when a pupil transfers from primary to secondary school, in the ordinary way at the end of Year 6. When this happens, and assuming that the specialist teaching assistant who has been with the VI pupil at primary school does not transfer her employment to the receiving secondary school, the secondary school will have to have sufficient time to recruit and train a suitably specialist teaching assistant. The time required – in particular the time required for training – is likely to depend on the circumstances and needs of the specific child.

20. I note that in June 2018 the Care Quality Commission and OFSTED undertook a joint inspection of Redbridge's provision for pupils with disabilities and special educational needs for the purposes of assessing the extent to which Redbridge had responded to the reforms in provision of disability and special educational needs put in place by the Children and Families Act 2014. One of the conclusions reached was that Redbridge had in place "appropriate arrangements" to support the transition from primary to secondary education at the end of year 6, and that the "majority" of pupils were well supported at this stage.

21. Based on the evidence in this case, I am satisfied that the arrangements that Redbridge has in place are such that at the time of the annual review mid-way through Year 5 (i.e., the academic year prior to the final academic year in primary education), parents of children with high level special needs are asked to decide on a preferred secondary school. Parents are given two months to consider and reach a decision. Redbridge's evidence includes summaries of how this process has worked in two other cases. There is also evidence of how the process worked in the Claimant's case. So far as concerns the Claimant's position, I have seen a document dating from October 2017 which set out options for the way in which specialised teaching assistant provision would be made, depending on which of two schools the Claimant transferred. That document was then the subject of discussion at a meeting in November 2017.

22. Based on this evidence (both the generic evidence in the form of the inspection report, and the documents that evidence specific application of the general arrangements), I am satisfied that the "decentralised model" does not stand in the way of sensible arrangements being made to deal with the transition from primary to secondary education. Forward planning is required, but suitable arrangements are in place to permit time for the arrangements that need to be made at the receiving secondary school.

23. It seems to me that forward planning would also be needed even if Redbridge had a "centralised" model in place along the lines the Claimant advocates. The needs of any VI pupil may vary from child to child: for example whether the child uses Braille and if so to what level; this means that the skills required from each teaching assistant will also vary. I make this point only because a significant part of the argument before me relied on the time that would be needed to train a teaching assistant in UEB Braille either from scratch to Grade 1 or from Grade 1 to Grade 2. I have referred to the present position in Redbridge already. In Redbridge there are now five Braillists, but prior to these pupils there had been no pupils using Braille in mainstream schools in Redbridge for some twenty years. There was further evidence from JCES to the effect that although JCES provides services to four boroughs including Redbridge, there is only one pupil outside Redbridge who uses Braille. If there were a centralised system I suspect that that too would require recruitment by reference to anticipated or notified future needs, since it would be unrealistic to expect any local authority to retain as a matter of contingency, a number of teaching assistants covering a range of skills to address a low incidence but high level need such as VI.

24. A different factual scenario might be where a VI pupil moved into the Redbridge area and joined a mainstream school part-way through a school year. If this happened without prior notice, it is likely the receiving school might struggle to find a suitably qualified teaching assistant in short order. However, the present arrangements are not in my view unlawful because of the possibility that they may not be able to address such a set of circumstance without delay. For the reasons already indicated, I have significant doubts that any realistic centralised model would be able to fare much better in this set of circumstances. In any event, were such a situation to arise in Redbridge, it would always be possible for Redbridge to bridge any gap in provision by seeking further services from JCES."

30. Mr Bowen criticised these findings on a number of grounds. First, in his submission they do not address the fact that parents are entitled to appeal to the Tribunal to challenge the school named in a child's EHC plan and the Redbridge model must therefore be able to but currently cannot take account of this entitlement in terms of any forward planning it undertakes to ensure that the necessary resource has been secured in time. Secondly, he challenges the conclusion that "it would always be possible for Redbridge to bridge any gap in provision by seeking further services from JCES" (at [24] set out above) and submits that the possibility of securing additional services from JCES cannot realistically compensate for gaps in the provision of specialist teaching assistant support given that JCES does not provide in-classroom support, but limits its services to advice, training and the provision of QTVIs. Thirdly and more broadly, he is critical of the fact that the judge did not spell out the evidence relied upon in reaching the conclusions he did on the time lag issue. To the extent that he identified specific evidence (for example the Joint OFSTED and CQC Inspection Report findings set out in a letter dated 25 July 2018) Mr Bowen criticised the judge's approach as superficial: the evidence did not support his conclusions. The fact that at a micro level some students are accommodated is no answer to the point that without a bank of centrally employed UEB qualified teaching assistants ready to start for each low incidence high need VI pupil, there is a built-in, systemic time lag problem with Redbridge's model. Moreover, the judge failed to address other relevant evidence, including a critical paper dated 6 June 2018 (exhibited to the first witness statement of Mr Vic Gibson), prepared by VIEW, a professional organisation representing the education workforce that supports children and young people with vision impairment (or VI).
31. Having considered all of the evidence available to the judge, I have concluded that he made permissible findings, albeit concisely expressed, and was entitled to prefer the documentary and witness evidence of Redbridge, including the findings set out in the 25 July 2018 report, over that put forward on behalf of ZK where there was conflict.
32. Significantly, the judge had evidence from Joena Stanley, a senior EHC coordinator, about the way in which the transition process works in Redbridge, both in terms of high-level timings and in practice in relation to ZK and two other students, referred to as S and A, all of whom had transitioned from one school to another at different stages of their education. The thrust of Ms Stanley's evidence is that, where parental preference is expressed at a sufficiently early stage, training for the teaching assistants employed by the relevant school begins so as to accommodate the request and the likelihood of an appeal reduces. In any event, the early contact (with parents) procedure means that the process generally begins 17 months before a child transfers to secondary school so that even with an appeal, there is time to complete most of the necessary training. Once a school is named, if additional VI specialist teaching assistants are needed, the school will recruit the teaching assistant and JCES will begin training them. Where a need for longer training is identified, this can be accommodated through informal discussions between JCES, Redbridge and the indicated school, and Ms Stanley states that this is what happened in the case of student S.
33. To the extent that training for teaching assistants is ongoing when the child begins secondary school, Ms Stanley states that there is QTVI support provided by JCES available for preparation of grade 2 level resources (in other words fully contracted UEB braille) until the teaching assistants have fully qualified, or to provide advice and support on a weekly basis. Her evidence is that Redbridge's commissioned service can readily provide this. That is supported by the terms of the annually negotiated contract between Redbridge and its specialist provider of special educational needs services, the Whitefield Academy Trust (which consists of two specialist schools, one being JCES, "the Provider"). Although in class support is not generally provided by QTVIs, Clause 1.7 of the contracted service specification provides:

"The Provider will provide such additional Services as may from time to time be required by the authority on account of unforeseen events or occurrences or on account of a particular occasion."

34. The evidence before the judge (from Mr Jilul Hoque and others) was that this was not just theoretical: while the annual budget for VI services spent on JCES had been the same for many years (£112,995 annually), in May 2018 there was an agreement to increase the budget for the financial year 2018/2019, coming into force in September 2018, to £189,943 to cater for the growing demand for support from schools in relation to individual children. So far as ZK is concerned, her EHC plan identified the potential need for one hour per day of additional QTVI support to provide marking and overwriting of her braille work in the event that her teachers or teaching assistants did not have the necessary expertise in braille in the first term at Oaks Park, and the increased budget was in part to cater for this additional spend. In light of this evidence, it seems to me that the judge was entitled to conclude that "it would always be possible for Redbridge to bridge any gap in provision by seeking further services from JCES" and I do not accept the criticisms made by Mr Bowen in this regard.
35. It is true that HK has been highly critical (as reflected in her second witness statement) of the process leading to Oaks Park being named as ZK's secondary school, and she details a series of difficulties she had with other preferred potential schools (Wanstead High School, Valentine High School and Isaac Newton School) leading, she says to the time-lag about which complaint is made. But in ZK's case there was early contact from Redbridge with her parents in January 2016 about transitioning to Oaks Park High School as a suggested secondary school, and there were visits in early 2017 to the school by her parents. The Deputy Head of JCES, Anne Webster, wrote to ZK's parents in February, March and May 2017 about other possible schools, stating that JCES would work with the chosen school over the next year and half to put in place the support and provision necessary to support her transition to whichever school was identified. A letter to ZK's parents dated 11 April 2017 inviting school preferences to be notified indicated that it was "very important that at least two preferences are provided". In June 2017 HK named Wanstead High School only, and in November 2017, Redbridge confirmed that it would fund training of two teaching assistants and the other costs associated with this proposed move to Wanstead High School.
36. Although a final resolution of the choice of school might have occurred around this time and would have allowed ample time for training and support for ZK's transition to secondary school, the process was in fact more protracted in ZK's particular case. However, as the judge permissibly found, this was not a consequence of systemic failings in the arrangements, but as a consequence of a combination of circumstances rendering Wanstead High School unsuitable for environmental reasons, and the very late identification of Oaks Park as the specified school. (This was not agreed until April/May 2018, though I do not suggest any fault is to be attached to either ZK's parents or to Redbridge in this regard). Moreover, there was nothing in the model of arrangements, as the judge found, to preclude Wanstead High School as the chosen school.
37. The evidence indicated that Anne Webster visited a number of the schools suggested by HK during this period, and it is clear that both she and others from Redbridge were proactive in consulting schools (in accordance with Redbridge's duty under section 39(2) of the 2014 Act) to inform them of the funding and other support (through JCES) that would be made available to them, and in seeking to facilitate ZK's placement at the school of her choice, for example, by seeking to resolve difficulties raised by the schools, such as the need for an environmental audit at Wanstead High School, and possible reasonable adjustments that might have to be funded. This latter point is disputed by HK but I note, for example, the email from Conny Kaweesa to Wanstead High School, dated 10 November 2017, which states:

"Liz has asked me to progress discussions regarding skilling up two Teaching Assistants in Braille to be available for [ZK] when she starts at Wanstead High School from September 2018.

I can also confirm/reiterate that:

1. The LA shall fund the reasonable adjustments, which, that Wanstead High School has to make in order to admit [ZK] from September 2018.

2. The LA shall fund the training and skilling up of the two Teaching Assistants, which the school has identified in Braille and cover any associated costs which includes their transport costs.

3. The LA is aware of the family commitments of the two Teaching Assistants who have been identified by the school and is therefore prepared to consider bringing the training to them.

4. The LA is prepared to order the equipment in advance, so that the Teaching Assistants and relevant professionals have an opportunity to test the equipment and to practice.

I understand that Wanstead High School does not object to being named in [ZK's] EHCP from September 2018".

38. Ultimately, Oaks Park High School was identified late in the day. Although it is a mainstream and not a special VI school, Ms Stanley said (at the time of the judgment below) that it had four teaching assistants trained to grade 2 and three teaching assistants from the team learning contracted UEB braille. The lead on the team also had the VIEW/RNIB Partners in Learning qualification. Members of the team attend JCES training courses from time to time. Ms Stanley said that a product of this school-led, JCES-facilitated model, has been the development of Oaks Park High School since September 2013 into a highly resourced and expert VI placement. She said the system adopted by Redbridge enabled parental preference (subject to statutory grounds for declining a particular school choice) to be accommodated while at the same time building capacity for a high level of VI specialist provision in one of Redbridge's schools and the capacity through JCES and the early notification system to build other provision elsewhere in the borough. She said, "Redbridge has never had a situation where a placement for a child or young person with VI was not ready in time".
39. The VIEW paper relied on by Mr Bowen is directed at a different problem, and does not begin to address the difference between a decentralised and centralised model of arrangements, or the extent to which the disadvantages raised by ZK in this case are an intrinsic feature of the decentralised model.
40. Although there is a passing reference in the VIEW paper to covering the costs of a "central, fully staffed and resourced VI service", the paper is plainly directed at changes in special educational needs funding that require local authorities to delegate most of the central special educational needs budget to schools. The paper calls for a comprehensive review of the current system of provision for children and young people with visual impairment on the basis that the current system is said no longer to be fit for purpose, but it is clear that the real issue is one of resource and the perceived need for ring fenced special educational needs funding so that specialist provision is made available. The paper is not directed at the employment structure adopted and whether teaching assistants are employed centrally by the local authority or directly by the schools, and there is no suggestion that one employment model is inferior or flawed, whether on its own or by reference to the other.
41. Moreover the particular problems identified in the VIEW paper (that VI educational services are becoming increasingly absorbed into larger services for children and young people with a range of special educational needs and disabilities, that the vast majority of teaching assistants out of a workforce of about 2,000 are recruited and employed directly by schools and most have little or no supervision from a QTVI or specialist training, that VI services are losing QTVIs, that children with mild or moderate levels of VI are at most risk of being denied specialist support, and that not all local authorities consider VI to be a complex need deserving of an EHC plan) do not apply under the Redbridge model with which this court is concerned. In Redbridge, JCES provides specialist services including training for teaching assistants and QTVIs for children with severe visual impairment. There is no suggestion of a failure to produce appropriate EHC plans. In the case of ZK, her EHC plan spells out precisely what must be provided by way of resources, including specifying specialist training for teaching assistants and advice, support and supervision from a QTVI.
42. Risk of redeployment and lack of cover: Swift J dealt with these issues at [29] to [34] as follows:

"29. ….At a theoretical level this is possible. Because specialist teaching assistants are employed by schools, the school could redeploy a teaching assistant to cover a need arising elsewhere in the school, for example the absence of another member of staff. However, what is important is the likelihood that this possibility will occur in practice, and the chance that that risk would be removed or substantially reduced if Redbridge adopted the Claimant's preferred centralised model.

30. The evidence I have seen satisfies me that under the present arrangements the risk of such an occurrence is low. The witness statement from Lesley Carty, the Special Educational Needs Co-ordinator at the school the Claimant attends explains the way in which the specialist teaching assistants at that school work. There are four Braille-trained teaching assistants at the school. They work specifically to support the VI pupils at the school. There is no suggestion that they have been called away from those duties to undertake other work, either regularly or at all. This impression that the risk of this is low is reinforced by the job description for the teaching assistants. Although the document is stated to be non-contractual, the job description makes it clear that the line of reporting for the specialist teaching assistant is through the school's Inclusion Manager (the manager with specific responsibility for the provision of special educational needs at the school). This reporting line must significantly reduce, if not remove, the risk that a teaching assistant would be redeployed away from specialist work. There is also a further practical consideration. The specialist teaching assistants may be employed by each school, but their positions are funded by Redbridge. In the event that a specialist teaching assistant was redeployed away from specialist duties either permanently or from time to time, I see no reason why it would not be open to Redbridge to require the school to stop that practice.

31. All these matters satisfy me that the risk the Claimant points to is theoretical rather than real. A risk of this nature or extent does not render the arrangements presently in place unlawful.

32. It must also be pointed out that a centralised model that the Claimant contends for would not necessarily remove the risk of redeployment. Even though under such arrangements, Redbridge would be the employer, the specialist teaching assistant would be seconded to the school and would fall under the day-to-day control of the senior staff at the school. If a school were minded to redeploy specialist teaching assistants in the way the Claimant suggests (and I note once again, there is no evidence at all that this has happened at the Claimant's present school) it could try to do that even if secondment arrangements were in place. If it did, the local authority employer would no doubt act to bring that state of affairs to an end. But that scenario would be little different from the pupil's perspective from that which could, also in theory, arise under Redbridge's present model. Put shortly, neither set of arrangements removes the possibility that a school might attempt some form of subversion; no set of arrangements would be fail-safe in such circumstances. But there is no evidence to support any contention that that is the basis on which I should approach this matter. The evidence goes in the other direction and indicates that where specialist teaching assistants are in place to assist specific pupils, those arrangements are not abused by schools.

33. There is always a risk that, just like any other teacher, a specialist teaching assistant might on occasion be off work, sick. Any absence of any teacher is capable of having some adverse impact on the education of the pupils affected. However, I do not consider the possibility of sick leave is such as to render Redbridge's present arrangements unlawful. Cover arrangements can be made for short-term absences. At the school the Claimant attends there are four specialist teaching assistants. Apart from that there is the possibility that Redbridge could seek assistance from JCES to cover a sickness absence, perhaps in the form of additional short-term assistance from a QTVI. The fact that Redbridge may not hold in reserve a fully-trained substitute (which I take to be the Claimant's point for this purpose about a centralised system) does not render the present arrangements inherently unlawful. It is in the nature of any arrangement which requires teaching staff to be available that adjustments may have to be made from time to time to deal with unexpected circumstances.

34. Nothing that I have said so far is to be read as meaning that the arrangements Redbridge has in place are perfect, or fool-proof, or fail-safe. I doubt that they are. I have seen evidence both from the Claimant's mother, and from the mother of another child who suffers severe visual impairment. Each is critical of how on occasions, matters relating to the education of their children have been addressed. But the issue in this claim is not directed to the specifics of individual cases. Rather, the challenge is at a generic level, to the general arrangements in place. What happens on any particular occasion may be the subject of specific challenge which may or may not succeed on its merits. However, at the generic level, the arrangements made by Redbridge for the provision of services from JCES to support the teaching of VI pupils in mainstream schools are not such as to be irrational, and are not such as to warrant an inference that they have been entered into without due regard for relevant considerations. This disposes of Ground 1."

43. In relation to these findings, no challenge is made to Swift J's conclusion that there was a theoretical only, but not any actual, risk of specialist teaching assistants being redeployed to other duties by schools.
44. ZK's central criticism is that the judge was wrong to conclude that the decentralised model is capable of managing specialist teaching assistant absences when it is said to be obvious that a centralised model involving a pool of specialist teaching assistants employed by the local authority level would provide much more effective cover than could ever be available from an individual school which may only have specialist teaching assistants for an individual child. Absence is a predictable risk and Redbridge's system does not cater adequately for this, but a centralised system would.
45. The difficulty with this submission is that the centralised model contended for would require a bank of additional specialist teaching assistants (trained to the standard required by each individual pupil's EHC plan no matter how high level and low-incidence the requirements are) held in reserve and able to step in should that pupil's teaching assistant be absent for any reason. Mr Bowen was unable to say how many reserve teaching assistants and trained to what particular level of specialist skill would be required in this model. Further, Mr Hoque's evidence makes clear that having made enquiries of other local authorities with in-house VI services, he found no examples of local authorities maintaining a bank of reserve specialist teaching assistants. That is unsurprising in my view, in circumstances where there is always a shortage of suitably qualified staff and on the face of it, to have a potentially idle pool of reserve specialist teaching assistants to provide cover as and when, appears both unrealistic and unsustainable.
46. Moreover, once that is accepted, as the judge observed, whether teachers or teaching assistants are centrally or directly employed, there is an inevitable risk of absence from time to time and cover arrangements are necessary. The judge recognised that the arrangements in place at Redbridge are unlikely to be perfect or fail-safe, and he acknowledged the evidence of HK (and another mother) criticising arrangements on specific occasions. However, again as he recognised, the claim was not directed at individual failures, but at the systemic level. At that level, cover was available at Oaks Park in terms of there being four specialist VI teaching assistants employed. Moreover, because ZK has two teaching assistants specified for her, one providing classroom support and the other preparing resources, they are interchangeable and can step in to cover an unexpected absence. Ms Stanley also said that if both teaching assistants are unexpectedly absent, JCES support can be called upon to provide advice and support to teaching assistants allocated by the school to cover the absence within the provision for QTVI support made, and we were told that the school can contact JCES directly and is not required to go through Redbridge.
47. Choice of school: Swift J dealt with this issue at [26] to [28] as follows:

"26……..The evidence of ZK's mother is that when the time came for the Claimant to move schools, at the end of Year 6/beginning of Year 7, her school of choice was unable to accept the Claimant because it did not have an appropriately trained teaching assistant. ZK's mother also contends that when she approached other schools, some gave the same response.

27. Focusing on the position of the school of choice, ZK's mother's account of events is disputed in Redbridge's evidence: see the witness statement of Joena Stanley. Her evidence is to the effect that after the school of choice was identified (which was a school close to her home), an assessment was undertaken to determine whether its premises were suitable. The outcome of that assessment (referred to as an "environmental audit") was that the school site was not suitable for the Claimant. In addition, significant building works were due to commence at the school in September 2018. That too made it impracticable for the Claimant to attend that school. Thus, on Redbridge's account the problem was not the lack of a suitably trained teaching assistant; no such teaching assistant was in post, but a suitable teaching assistant could have been recruited and trained.

28. My conclusion is that the Claimant's ability to move to her school of choice was not affected by any matter inherent in the so-called decentralised model. I accept Redbridge's evidence as to the circumstances that prevailed at the school of choice. I do not accept the Claimant's contention that the decentralised model makes mainstream schools less willing (or less able) to accept a VI pupil. The arrangements that Redbridge has in place for managing school transfers between Year 6 and Year 7 do permit a lead time that ought to be long enough to allow time to recruit and train a specialist teaching assistant. Redbridge's evidence on the steps taken in the Claimant's case indicates that Redbridge has good lines of communication with its mainstream schools on special needs matters. I have no reason to think that schools are unaware of the assistance available to them when it comes to securing training for suitable teaching assistants."

48. Mr Bowen challenges the finding that ZK's ability to move to the school of her choice was not affected by any matter inherent in the decentralised model. He submits that there was clear evidence before the judge that several schools informed HK that they would not be suitable for ZK because of the lack of suitably trained teaching assistants. Particular reliance is placed on communications from the special educational needs coordinator at Seven Kings School on 19 May 2017 and from Woodbridge High on 9 February 2018 to the effect that there were no current teaching assistants available to learn braille and querying the time it would take in any event.
49. Once again I am not persuaded that these criticisms undermine the judge's findings. It is implicit in the passages set out above that he preferred the evidence of Redbridge (Joena Stanley and others) on this issue to that of HK, recognising there was a stark conflict.
50. Moreover, the evidence demonstrates that neither Seven Kings School nor Woodbridge High were identified as preferred schools by HK to Redbridge. I have referred above to the evidence of efforts made by Redbridge to accommodate parental preference in terms of choice of school for ZK. Both Anne Webster and Joena Stanley were extensively involved in consulting with her parents and liaising with relevant schools. Specialist analysis and an environmental audit was conducted by JCES in relation to both Wanstead High School and Oaks Park. There is no reason to infer that the same consultation, liaison and analysis would not have been conducted in relation to Seven Kings School and Woodbridge High, had these schools been identified. It is inevitable in circumstances where a child has a low incidence but high level need such as visual impairment that a school without experience of meeting such a need might require engagement by the local authority about the services that would be available (whether in terms of funding or the provision of services by JCES) to enable such a school to accommodate such a pupil. That need is catered for by the consultation provision in section 39(2) of the 2014 Act referred to above, and the evidence demonstrates Redbridge's proactive engagement in this regard.
51. It is also clear from the evidence of consultation and proactive engagement by Redbridge that it seeks to accommodate parental preference where possible. Inevitably, the earlier that choice is made the easier it will be to accommodate it. Here, the process was protracted, and Valentine School and Isaac Newton School were not notified until relatively late on because Wanstead High was preferred (but ultimately found not to be a practicable choice for environmental, not other reasons). Again, there is no reason to infer that the same extensive engagement would not have taken place with these schools had the choice been made earlier in the process. Moreover, once a chosen school is identified and specified in the EHC plan, it is not open to a school to refuse to accommodate the student in question (see section 43 of the 2014 Act). The section 42 duty arises at that point and the local authority must ensure that the specified provision is secured. There is no evidence to suggest that this did not or would not have occurred.
52. For all these reasons, I am satisfied that it was open to Swift J to conclude that ZK's ability to move to her school of choice was not affected by any matter inherent in the decentralised model, and that this model does not make mainstream schools less willing (or less able) to accept a VI pupil.
53. Having addressed the legitimacy of the factual findings made by Swift J, I turn to address the grounds of appeal.

Ground 1: challenge to finding there is no irrationality or illegality in the decentralised arrangements adopted by Redbridge by reference to the four factual issues relied on by ZK

54. On behalf of ZK a number of points were made in writing to support the contention that Swift J failed to apply the correct test for determining whether the arrangements made by Redbridge are irrational and unlawful. The first of these was an argument to the effect that the judge erred by applying too stringent a test by suggesting that a policy was only irrational if it inevitably led the local authority to act illegally (at [18] and [35] of the judgment). However, in the course of the hearing Mr Bowen made clear that he did not press the criticism of the use of the word "inevitable" in the paragraphs relied on because he accepted that the judge was simply responding to the case advanced on behalf of ZK, which used the word "inevitably".
55. Instead he accepts that the judge's essential conclusions on this issue are to be found at paragraphs [40] to [43] as follows:

"40. In the present case the Claimant's substantive case based on this principle rests on: (a) the obligation at section 42(2) of the Children and Families Act 2014, that local authorities "secure the specified educational provision", i.e. the provision specified in an EHCP; and (b) the factual contention that the lead-time under a decentralised system that is necessary for a school to secure and train a specialist teaching assistant means that, for an initial period at least, there will be a failure to comply with the section 42 obligation.

41. For the reasons already set out, I do not consider the arrangements in place under which specialist teaching assistants are employed by schools and trained and supported by JCES give rise to any inherent likelihood that Redbridge will fail to comply with its section 42 obligation. I am satisfied Redbridge's arrangements, in particular for the management of transfers between schools at the end of Year 6 for the beginning of Year 7, are sufficient when considered at a generic level, and do not entail inherent likelihood that Redbridge will fail to comply with its section 42 obligations.

42. This case concerns provision for severely disabled children with high-level educational needs who are pupils in mainstream schools. The Claimant's circumstances are a particular example of circumstances of such pupils. Where such high-level needs require the presence of a specialist teaching assistant, it is likely that some time will be needed to recruit and train that assistant. But it seems to me that this will be so regardless of whether the local authority concerned works to a centralised model or a decentralised one. The premise for the Claimant's submissions as to the virtues of a centralised system presuppose that under any such arrangements the local authority would employ a sufficient number and range of specialist training assistants so as to be able to cover any/all particular needs that might fall to be met. If that is not the premise then the lead-time problems the Claimant identifies would also be a feature of a centralised model. That premise seems to me to be unrealistic. It would require local authorities to retain that range of reserve teaching assistants indefinitely, on the off-chance that a pupil whose needs required those skills entered a mainstream school in the local authority's area. To take the example I have already mentioned: before the five VI pupils now in Redbridge's mainstream schools, there had been no such pupils in Redbridge for 20 years.

43. I cannot see that compliance with section 42 of the 2014 Act requires local authorities to maintain arrangements of the sort that are inherent in the Claimant's argument. "Secure", as it is used in section 42 of the 2014 Act means provide and maintain. Sometimes provision can take time to arrange, particularly in a case with a pupil with high-level needs. The expectation, even in such cases, is that local authorities will plan so that needs do continue to be met even when a child transfers from one school to another. Compliance with section 42 is likely to require a local authority to engage in reasonable forward planning. However, there may be circumstances which cannot reasonably be expected to be met by reasonable forward planning. If such a situation arises section 42 requires a local authority to ensure that what is specified in the EHCP is provided promptly, and thereafter maintained."

56. Swift J rejected the submission by Redbridge that illegality could only arise if there was an unacceptable risk of unfairness rather than an unacceptable risk of illegality.  He held that expressing the test by reference to 'unfairness' was appropriate in cases dealing with policies of a procedural nature but that the same test was applicable when judging substantive policies too.  He referred to the judgment of Whipple J in Bayer Plc v NHS Darlington CCG [2018] EWHC 2465 (Admin) where she formulated the question for the court in terms of whether the policy was capable of lawful application, assessing the policy realistically and pragmatically.  Swift J concluded that there was no difference between that test and Richards LJ's notion of inherent illegality in R (Tabbakh) v Staffordshire and West Midlands Probation Trust [2014] EWCA Civ 827, [2014] 1 WLR 4620. 
57. In relation to the conclusions set out at paragraphs 40 to 43 of his judgment, Mr Bowen submits first, that Swift J was wrong to apply the test set out by Whipple J at first instance in Bayer Plc at [198], namely whether the Redbridge policy was "realistically capable of implementation… in a way which does not lead to, permit, or encourage unlawful acts". He recognises that the Court of Appeal ([2020] EWCA Civ 449) upheld the test applied but submits that the factors which persuaded Whipple J to modify the test in Bayer Plc are not present in this case, and the decision is highly context specific. Here, unlike in Bayer Plc there is no separation between Redbridge as the architect of the arrangements and those responsible for implementing them in an unlawful manner; no suggestion that the arrangements leave open the possibility of a person to whom they are directed of choosing between a lawful and unlawful method of operation; and there are not a variety of possible outcomes in relation to the arrangements. Instead he submits that the arrangements in the present case operate in the same way in every case, govern the provision of teaching assistant support for every child with visual impairment in the borough, and will frequently result in unlawful gaps in service provision. Even if the modified test in Bayer Plc is applicable, then Redbridge's arrangements are not realistically capable of lawful implementation for children such as ZK, particularly (but not exclusively) those for whom there is no lengthy advance notice of their arrival at school, or who seek to challenge the local authority's decision to name a particular school on their EHC plan.
58. Moreover, to the extent that Swift J sought to equate the Bayer Plc test with the inherent illegality test, that is wrong and inconsistent with the distinction drawn between the two approaches by Underhill LJ at [207] in Bayer Plc. He submits that in light of a number of cases decided after Swift J's judgment (including FB (Afghanistan) v SSHD [2020] EWCA Civ 1338 and BF (Eritrea) v SSHD [2019] EWCA Civ 872), the appropriate test for determining whether there is systemic illegality in the present case is to ask whether there is a real (as opposed to a fanciful) risk of a breach by Redbridge of its section 42 duty for more than a minimum number of visually impaired children. He accepted during the course of submissions that if "inherent risk" were to be equated with "real as opposed to fanciful risk", he would not take issue with the inherent risk formulation. He also accepted that context is plainly relevant.
59. Secondly, in Mr Bowen's submission, Swift J misunderstood the duty imposed by section 42 of the 2014 Act. He relies on the judge's statement at [43] that "Compliance with section 42 is likely to require a local authority to engage in reasonable forward planning. However, there may be circumstances which cannot reasonably be expected to be met by reasonable forward planning" as indicating that the judge misunderstood the mandatory nature of the duty and imported into it a reasonable endeavours or best endeavours defence. Since the judge's conclusion that the arrangements adopted by Redbridge did not give rise to any inherent illegality is contingent on his understanding of the duty imposed by section 42, if he misunderstood the nature of the duty his overall conclusion is flawed and cannot stand.
60. I do not accept these submissions. My reasons follow.
61. We were referred to a large number of authorities in connection with this ground of appeal and considerable reliance was placed on the most recent decisions of this court in BF (Eritrea) and FB (Afghanistan). I do not think it necessary or useful to analyse the various cases referred to and the different approaches adopted in those cases. The fact that BF (Eritrea) is due to be heard on appeal by the Supreme Court where these tests will no doubt be considered reinforces that conclusion.
62. Context is obviously important in determining the appropriate test to be applied. In this case the test adopted by the judge was that set out at [41] where he considered whether the arrangements in place under which specialist teaching assistants are employed by schools and trained and supported by JCES gave rise to any inherent likelihood that Redbridge would fail to comply with its section 42 obligation. That test applied by Swift J derived from R (Refugee Legal Centre) v Secretary of State for the home Department [2005] 1 WLR 2219 (and R (Tabbakh) v Staffordshire and West Midlands Probation Trust). In both of those cases, as the judge recognised, the focus of the challenge was the fairness of the arrangements made and that explained why in each case the court's reasoning was formulated in terms of the language of fairness (i.e.. the essential question to be asked was whether the system of processing asylum seekers established by the policy under challenge was inherently unfair). There had to be a risk of unfairness inherent in the system itself rather than one arising in the ordinary course of individual decision-making.
63. But as Swift J correctly held, the principle is an applicable standard by which to judge substantive policies as well. What matters in a systemic challenge of this kind is the need to distinguish between an inherent failure in the system challenged and individual examples of failings or unfairness which do not touch on that system's integrity, however difficult it might be in practice to distinguish between those two situations: see to this effect R (Woolcock) v Secretary of State for Communities and Local Government [2018] EWHC 17 (Admin), [2018] 4 WLR 49, Hickinbottom LJ at [68]. In other words, the court must distinguish between examples in the evidence which demonstrate a systemic problem from those which remain cases of individual operational or other failure. That is precisely what Swift J did.
64. Thus I am satisfied that Swift J's analysis of the law was correct and has not been undermined by the additional cases to which we were referred. It is clear that although he referred to Bayer Plc, he did not apply the test set out in that case over the test of inherent illegality. Moreover, even if the test set out in BF (Eritrea) or FB (Afghanistan) (to the extent that they are different) should have been applied, I do not consider it would have made any difference whatever on the facts of this case.
65. In light of the evidence and the factual findings summarised above, it seems to me that Swift J was amply entitled to conclude that the arrangements in place under which specialist teaching assistants are employed by schools and trained and supported, including for the management of transfers between schools at the end of Year 6 for the beginning of Year 7, are sufficient when considered at a systemic level, and do not entail any inherent likelihood that Redbridge will fail to comply with its section 42 obligations.
66. While I can see the attractions of a centralised model which puts the local authority in full charge in terms of preparing to secure whatever provision is specified in a student's EHC plan no matter how complex and low incidence it might be, there is no evidence here of Redbridge's model putting constraints on early planning, or taking it out of the driving seat. There is no evidence that the Redbridge arrangements entail waiting until a draft EHC plan is finalised before taking action in response to it (as Mr Bowen asserted); nor that Redbridge delays commencing an informal dialogue both with JCES and the school in question about fulfilling the requirements specified in the draft plan until such a plan is finalised. To the contrary, the evidence indicates that Redbridge's model is flexible and can be tailored to an individual's needs at the point of delivery. Redbridge is proactively involved with the process from an early stage, and the arrangements in place allow for the necessary preparatory training or other work to take place in time for transition to a different school. As is demonstrated by Joena Stanley's evidence (and as reflected by the approach to ZK's case, and the cases of S and A), Redbridge is proactive in offering its (and JCES') expertise to parents and the notified school, and in working with all parties to secure the necessary provision.
67. Furthermore, the judge acknowledged evidence from HK about individual instances of educational matters not being addressed on occasion as she would have wished (and that this might afford the basis for specific challenge). However, this challenge was a systemic challenge at a generic level, to the general arrangements in place. He concluded that at the generic level, the arrangements made by Redbridge for the provision of services from JCES to support the teaching of visually impaired pupils in mainstream schools are not irrational. I can see no basis for challenging that conclusion.
68. As to the arguments advanced by reference to section 42 of the 2014 Act, although Ms Rhee was anxious to emphasise the absence of any individual, specifically pleaded breaches of section 42, I do not regard that as significant. ZK is entitled to rely on a systemic challenge and her case does not depend on any finding of an underlying specific breach. In any event, this judicial review was commenced before ZK's transition to Oaks Park and that timing may explain the absence of any particular pleaded breach. Moreover her application to adduce fresh evidence made in advance of the hearing of this appeal would have addressed the question of individual breaches of section 42. That application was refused by this court because it raised a wholly new and different case. Nonetheless, it would not be fair to hold that against ZK in the circumstances.
69. I am satisfied that the judge had a full and accurate understanding of the absolute nature of the section 42 duty in this case. He recognised that the section 42 duty only arises once the EHC plan is in place. That means that the choice of school issue relied on by ZK is not relevant to this question, and he correctly focussed on the factual contention that the lead-time under a decentralised system necessary for a school to secure and train a specialist teaching assistant means that, for an initial period at least, there will be a failure to comply with the section 42 obligation.
70. Swift J rejected the premise of ZK's case that section 42(2) of the 2014 Act requires local authorities to employ a sufficient number and range of specialist training assistants to be able to cover any and all particular needs that might fall to be met. It followed that the same early planning would be needed under any model of arrangements adopted by Redbridge in terms of forward planning to recruit and train specialist training assistants. I cannot see any basis for impugning that conclusion. To the contrary, I agree with it.
71. Further, his reference to "reasonable forward planning" does not import a reasonable endeavours defence, nor did Swift J dilute the duty by concluding that "secure" means "provide and maintain". That is a misreading of what he said.
72. Rather, given that all arrangements whether centralised or otherwise are likely to feature lead-time issues, Swift J made clear that while it is for Redbridge to work out how to make special educational needs provision and how to fund that provision, the expectation in every case is that there will be adequate early planning to ensure that each affected child's needs are provided for and continue to be met as they transition to a different school (whenever that occurs) to ensure the necessary provision continues to be secured. As a matter of fact, the judge accepted (as he was entitled to do) there were suitable arrangements in place in Redbridge to permit time for the arrangements that needed to be made at the receiving secondary school. Where preparation for a child is likely to take time, it must start sufficiently early to accommodate that. But where there are circumstances that cannot be met by reasonable early planning, that does not excuse compliance: as Swift J himself observed, "If such a situation arises section 42 requires a local authority to ensure that what is specified in the EHCP is provided promptly, and thereafter maintained."
73. No separate argument was directed at Swift J's conclusion that the arrangements put in place by Redbridge are not irrational. The findings made by the judge entitled him to reach this conclusion for the reasons already set out.
74. For all these reasons I am satisfied that Swift J was entitled to conclude that the arrangements put in place by Redbridge to meet the special educational needs provision in the borough are neither irrational nor unlawful. This ground accordingly fails.

Ground 2: challenge to finding that there was no unlawful disability discrimination

75. The premise for the unlawful discrimination claims advanced by ZK below is that a decentralised model has a disproportionate adverse impact on pupils in mainstream schools with severe special educational needs such as visually impaired pupils, because those pupils depend on specialist trained teaching assistants, and without their support cannot access education in mainstream schools in the same way either as pupils with less severe special educational needs, or pupils who do not have special educational needs at all. The discrimination claims comprised a claim that the arrangements Redbridge has in place amount to unjustified indirect discrimination against visually impaired pupils such as ZK (contrary to the Human Rights Act 1998 and Articles 14, 8 and/or Article 2 of Protocol I to the European Convention on Human Rights and/or sections 19 and 29(6) of the Equality Act 2010); and a claim that Redbridge's decentralised arrangements are a failure to comply with the obligation to make reasonable adjustments contrary to sections 20 and 21 and Schedule 2 of the Equality Act 2010. In relation to each claim, the disadvantages relied on are the same as those relied on to support the claim that the arrangements are irrational and/or unlawful but include the allegation that the decentralised model makes it less likely that ZK will be able to go to the school of her choice (which is not relevant to the legality claim).
76. In relation to the unjustified indirect discrimination claims (whether under the Convention, the Human Rights Act 1998 or the Equality Act 2010, because there is no difference in substance between the two claims) Swift J held that the relevant comparison was between on the one hand visually impaired pupils in mainstream schools requiring provision of specialist teaching assistance and on the other hand either pupils in mainstream schools with less significant special educational needs, or pupils in mainstream schools with no special educational needs. Neither side challenges that approach and I proceed on that basis.
77. Swift J held that these claims failed on their facts. He rejected the contention that the Redbridge arrangements are an inherent source of disadvantage for the protected group in the sense that they bring with them the disadvantage relied on. He emphasised that although evidence of problems in individual cases where there have been complaints that what has been provided has not met the provision specified in the EHC plan could give rise to an inference of systemic failure, he was nonetheless satisfied that the arrangements in place with JCES for the provision of specialist support to pupils in mainstream schools do make sufficient allowance for the differences between pupils with special educational needs requiring the provision of specialist teaching assistance and other pupils.
78. In relation to the reasonable adjustments claim, ZK had to establish that the arrangements made by Redbridge with JCES for the provision of support for visually impaired pupils in mainstream schools put pupils like ZK at a substantial disadvantage in comparison with other pupils. Swift J held that the claim of failure to comply with the reasonable adjustments duty failed on its facts for the same reason as the other discrimination claims because substantial disadvantage was not made out on the facts.
79. In writing there are two challenges to this part of Swift J's judgment, although neither was developed orally. First, there is a factual challenge to the conclusion that the decentralised model does not in fact disadvantage ZK and other pupils with severe visual impairment in the ways alleged. Reliance is placed on specific examples of gaps in the delivery of learning/educational content, general gaps in the training of teaching assistants and absences as demonstrating that ZK was disadvantaged because she was unable to follow the curriculum without appropriate specialist, appropriately trained support, at times missing school days and becoming distressed and depressed. Furthermore, she had no choice but to go to Oaks Park despite it being far from where she lives. Secondly, in writing reliance was placed on Article 24 of the UN Convention on the Rights of Persons with Disabilities to contend that the judge was wrong to conclude that this provision took matters no further in this case. That was an error because Article 24 is highly material to the court's assessment of the extent to which the decentralised model is discriminatory. So far as concerns the absence of any choice in relation to schools as a consequence of the decentralised model, reliance is placed on HK's evidence that Oaks Park was in fact the only secondary school able to provide some of the specialist teaching assistant support required by ZK demonstrating that the Redbridge model creates a real risk that children with complex special educational needs will be forced to attend special schools because mainstream schools do not have the necessary specialist teaching assistant support in place to accommodate them.
80. I have dealt above with the challenges advanced on behalf of ZK to the underlying factual findings made by the judge. True it is that there is a stark conflict in the evidence put forward on behalf of ZK about the way in which the Redbridge arrangements work in practice and the difficulties perceived on occasions with those arrangements, as compared with the evidence put forward on behalf of Redbridge. The judge made no findings of specific failings or breaches and was not invited to do so. This court cannot make such findings either. Rather as the judge made clear his focus was at a generic level on the extent to which, if at all, the disadvantages relied upon are inherent in the arrangements put in place by Redbridge. He was satisfied that they are not and that sufficient allowance for the differences between pupils with special educational needs requiring specialist trained teaching assistants and other pupils is made by the Redbridge arrangements. There was evidence to support those conclusions as I have explained above. He expressly addressed the contention that ZK's choice of school was narrowed by the arrangements put in place by Redbridge. He concluded that ZK's ability to move to her choice of school was not affected by any matter inherent in the decentralised model, accepting the evidence of Joena Stanley as to the circumstances that led to Wanstead High School concluding that the school would not be able to accommodate ZK. On Redbridge's account the problem was not the lack of suitably trained teaching assistants. Although none was in post a suitable teaching assistant could have been recruited and trained. The problem was rather the environmental audit as explained above. Moreover, in light of the evidence, again summarised above, the judge permissibly concluded that the decentralised model does not make mainstream schools less willing or able to accept a VI pupil.
81. In light of the factual findings made by the judge, he was entitled both to dismiss the discrimination claims for the reasons he gave and to conclude that reference to Article 24 took the case no further. For all these reasons, this ground also fails.

Ground 3: the challenge to the judge's conclusion that the claim based on the section 149 public sector equality duty (or PSED) failed.

82. At paragraph 72 of his judgment Swift J held that this claim failed because the very purpose of the arrangements Redbridge makes with JCES from year to year in order to ensure that special educational needs of the visually impaired pupils in mainstream schools are met is to seek to eliminate discrimination between visually impaired pupils and other pupils in mainstream schools who do not have special educational needs. He held that purpose runs through to Redbridge's decision on what services should be secured from JCES.
83. In writing, but again not pursued orally, ZK submits that the judge was wrong to conclude that the duty had been discharged in this case because he was wrong to rely on Lord Brown's approach in R (McDonald) v RBKC [2011] UKSC 33). McDonald was concerned with a challenge to an individual community care case where the focus was on securing equality of opportunity for the person subject to the plan, and not a systemic policy challenge. Although accepting that compliance with this duty is a matter of substance not form, for ZK it is argued that Redbridge in fact failed to comply with the duty in substance and the judge did not refer to any evidential matters which would demonstrate such compliance. The focus on Redbridge's purpose in making the challenged arrangements is not sufficient to demonstrate compliance with the duty and in the absence of any finding that Redbridge had actual regard to the needs specified by section 149, the judge was wrong to find that Redbridge is operating its decentralised model in compliance with the requirements of that duty.
84. I can deal with this ground very shortly. In McDonald Lord Brown rejected the argument that there was a failure to have regard to the PSED. At [24] he said:

"24.  This argument … is in my opinion hopeless. Where, as here, the person concerned is ex-hypothesi disabled and the public authority is discharging its functions under statutes which expressly direct their attention to the needs of disabled persons, it may be entirely superfluous to make express reference to section 49A and absurd to infer from an omission to do so a failure on the authority's part to have regard to their general duty under the section. That, I am satisfied, is the position here. The question is one of substance, not of form. …"

The logic of the approach identified by Lord Brown in McDonald is that in any case where the public body concerned is discharging its functions under legislation expressly directed at the needs of a protected group it may be unnecessary to refer expressly to the PSED or to infer from an omission to do so, a failure to have regard to that duty. In other words, the nature of the duty to have due regard is informed by the particular function being exercised and not vice versa. In this case, Redbridge makes arrangements to discharge the duty imposed by section 42 of the 2014 Act expressly to secure special educational provision for pupils with special educational needs, including those with a disability. Accordingly I can see no error in the judge's reliance in this regard, on the very purpose of the arrangements that Redbridge makes with JCES to ensure that the provision required to meet the assessed needs of visually impaired pupils in mainstream schools, as specified in EHC plans, is made. That was sufficient to meet the public sector equality duty in this case.

Conclusion

85. For all the reasons set out above, all grounds of appeal fail, and if my Lord and Lady agree, the appeal fails and must be dismissed.

Lady Justice Rose:

86. I agree.

Lord Justice Baker:

87. I also agree.