Gill, R (On the Application Of) v Cabinet Office [No.3]

The claimant, chair of the Sikh Federation UK, challenged the omission of a Sikh tick-box option from the ethnic group question in the Census Order 2020. The core legal issue was whether the Office for National Statistics had applied an unlawful, unpublished or inconsistent policy when deciding which response options to include.

• Ground 1: The court found no Lumba-style failure to apply a published policy — the ONS had applied the published prioritisation tool methodology (updated) rather than an unpublished alternative; references to "public acceptability" in consultation documents related to topic/question design, not to the prioritisation of tick-boxes.
• Ground 2: The 2011 prioritisation tool was properly reviewed and updated for 2021; changes were minor and did not amount to application of a materially different, secret policy.
• Ground 3: There was no inconsistent application of criteria between topics and response options once the difference in purpose between topic selection and tick-box prioritisation is understood.
• Ground 4: Reliance on the Kantar focus-group research was lawful; alleged methodological errors were not shown to be so serious as to vitiate the conclusions.

Accordingly, the application for judicial review was dismissed.

Background and parties: The claimant (Amrik Singh Gill) challenged the Census (England and Wales) Order 2020 for failing to include a Sikh tick-box response option under the ethnic group question. The defendant was the Cabinet Office acting through the Office for National Statistics (ONS) and UK Statistics Authority. The ONS had consulted and used a prioritisation tool (PT) to assess proposed additional tick-boxes; it also commissioned qualitative research from Kantar.

Nature of the claim / relief sought: Judicial review seeking a declaration and quashing order contending the ONS applied an unlawful or unpublished policy and otherwise applied its criteria inconsistently or irrationally.

Procedural posture: This was the third judicial review on this issue: a prior judicial review in 2019 (Gill v Cabinet Office [2019] EWHC 3407 (Admin)) was dismissed as premature; a second application was refused permission; the draft and then final Order were laid/made in early/mid 2020 and this challenge followed the making of the Order.

Issues framed: (i) whether the ONS failed to apply its published "public acceptability" (PA) evaluation criterion to tick-box assessment (Ground 1); (ii) whether the ONS applied an unpublished or materially different prioritisation tool rather than the updated 2011 PT (Ground 2); (iii) whether the ONS applied criteria inconsistently between topics/questions and response options (Ground 3); and (iv) whether reliance on the Kantar report was unlawful because of methodological flaws or legally unsustainable conclusions (Ground 4).

Court's reasoning and conclusions: The court emphasised the distinct purposes of topic selection and tick-box prioritisation and accepted ONS evidence that different, bespoke criteria apply to each. The May 2016 consultation documents and subsequent materials were read in context and, on a reasonable-reader approach, did not show any commitment that the Topic Criteria (including the PA Criterion as defined for topic selection) would be applied to the separate exercise of prioritising tick-box response options. The ONS had clearly stated it would apply an updated prioritisation tool; the updated 2021 PT (published June 2019) represented only minor and explicable changes. The qualitative Kantar research was a legitimate input; asserted errors were either misunderstandings of the qualitative methodology or not so serious as to vitiate the conclusions. The court therefore rejected the claim on all four grounds.

Wider procedural and remedial points: The court considered the discretion on relief and the operation of s.31(2A) Supreme Court Act 1981. Even had an error been found, the court indicated (and applied reasoning) that the exceptional disadvantage and administrative detriment of quashing the Census Order close to the census date would weigh heavily against granting such relief. The claimant's allegations of lack of candour were not upheld.

The claim is dismissed. The court concluded that the ONS applied the published and applicable evaluation methodology (an updated version of the 2011 prioritisation tool) to response-option prioritisation and did not apply a hidden or inconsistent policy; qualitative research reliance was lawful and methodological criticisms did not demonstrate material irrationality. The claimant’s grounds failed, and no relief was granted.

This is first-instance judicial review litigation in the Administrative Court. The judgment records two earlier related judicial review proceedings: JR1 (Gill on behalf of Sikh Federation UK) v Cabinet Office [2019] EWHC 3407 (Admin) (dismissed by Lang J as premature and in breach of parliamentary privilege) and a second application in April/May 2020 where permission was refused on the papers (Foster J, 7 May 2020). The draft Census Order was laid on 2 March 2020 and the Census (England and Wales) Order 2020 was made on 20 May 2020; the present claim was issued 11 June 2020 and heard on an expedited basis in October 2020.