UA v Her Majesty's Revenue and Customs (TC): [2019] UKUT 113 (AAC) [2019] UKUT 113 (AAC)

UPPER TRIBUNAL
ADMINISTRATIVE APPEALS CHAMBER
[2019] UKUT 113 (AAC)
Between
UAAppellantHer Majesty's Revenue and CustomsRespondent
Date 2 April 2019Category: Marriage civil partnerships and living together, Tax credits and family credit, Tribunal procedure and practiceSubcategory: Marriage civil partnerships and living together joint claims, Tax credits and family credit couples and joint claims, Tribunal procedure and practice evidence
[1]This case is a classic example of why it is important for tribunals to be alive to cultural issues when determining appeals, especially when assessing a party’s evidence about intimate matters relating to personal relationships (in the present case, whether a married couple had separated). To rework an old saying, before you judge a woman, walk a mile in her shoes. A summary of Upper Tribunal’s decision[2]I allow the Appellant’s appeal to the Upper Tribunal. The First-tier Tribunal’s decision involves an error on a point of law, as the HMRC accepts. That decision is accordingly set aside. The case needs to be reheard by a new Tribunal.[3]The fact that this appeal to the Upper Tribunal has succeeded on a point of law is no guarantee that the re-hearing of the appeal before the new First-tier Tribunal will succeed on the facts. The new Tribunal may reach the same, or a different, decision to that of the previous Tribunal. It all depends on the findings of fact that the new Tribunal makes. The issue before the First-tier Tribunal[4]On 31 August 2017 HMRC made a decision under section 18 of the Tax Credits Act 2002 that the Appellant – who was then an HMRC employee herself and had been claiming tax credits as a single person since 2009 – was not entitled to tax credits for the 2016/17 tax year. In essence, the decision to remove tax credits was based on the finding that the Appellant was still married to, and more importantly not separated from, her husband. It is not in dispute that the Appellant and her husband were married in an Islamic ceremony in Pakistan in 2005. They are first cousins (the Appellant’s mother is the sister of the husband’s father, and so the husband is both the nephew and the son-in-law of the Appellant’s mother).[5]On 5 June 2018 the First-tier Tribunal (‘the FTT’) upheld HMRC’s decision. The FTT found, as expressed in the summary on its decision notice, that the Appellant was a member of a couple for the 2016/17 tax year and so not entitled to make a claim for tax credits as a single person. The grant of permission to appeal to the Upper Tribunal[6]On 12 November 2018 I gave the Appellant permission to appeal to the Upper Tribunal, and in doing so made some preliminary observations. I cite them at length as they conveniently set the context: “The First-tier Tribunal’s decision: an outline[4]The FTT accepted that the Appellant was married and had not been separated by court order. It followed that the live question was whether she was separated in circumstances in which the separation was likely to be permanent. The FTT found that the Appellant was not separated and, even if she was separated, the separation was not likely to be permanent. In relation to the issue of actual separation, it seems the FTT took into account a number of indicators such as(i) the husband not having any permanent address;(ii) he has keys to the Appellant’s house and kept belongings there;(iii) he has his post delivered to the Appellant’s address (which included bank accounts, loans and car registration documents);(iv) his name was on the electoral roll at the address;(v) the Appellant had not applied for Council Tax single person’s discount; and,(vi) they had a joint bank account (paragraphs [6]-[19] SOR [statement of reasons]). UA v HMRC (TC) [2019] UKUT 113 (AAC) 3 CTC/2349/2018 The grounds of appeal: the Appellant’s case[5]The Appellant’s case is that she is separated from her husband. She argues the FTT failed to give full and proper consideration to her case and, in particular, to the cultural dimension. She explains that she is married to her first cousin and so the families are interconnected. She says that the cultural pressure to have a successful marriage and the shame associated with separation “can lead to actions that other cultures may view as strange or not compatible with a separation but the mask of marriage can hide a very different reality, especially for a woman from my culture.” Provisional analysis: some observations[6]A difficulty that may face the Appellant is that the FTT’s various findings summarised in paragraph 4 above are ultimately all questions of fact not law. It is well established that issues of fact are for the first instance FTT to determine. Thus, the Upper Tribunal must respect the FTT’s fact-finding role so that, if the FTT has approached the task of fact-finding in a rational manner, has given an adequate explanation for its decision, has properly understood and applied the law, and has acted fairly, then the Upper Tribunal cannot interfere.[7]That said, the FTT appears to have given relatively little regard to whether the Appellant and her husband were actually separated within the sense envisaged by the legislation (and contrariwise their status as a married couple). As I observed in an earlier case, “the test of separation must focus on the relationship between husband and wife, and a married couple may be separated whilst living in the same household, whether or not they have a court order to that effect” (DG v HMRC (TC) [2013] UKUT 631 at paragraph 36).[8]Furthermore, it is inherent in these types of situations (where people once married then became separated) that – whether or not they remain in the same household – the couple may continue to have some involvement with each other (see Upper Tribunal Judge Jacobs in SA v HMRC (TC) [2017] UKUT 90 (AAC) at paragraph 10 (a case where separation had been established, and the issue was likely permanence)). This is especially so if there are children involved, as here. In the present case if may be arguable there were contra-indicators on the issue of separation, such as(a) the Appellant’s explanation in relation to the family trips being booked and paid for separately;(b) apparently no financial contributions from the husband;(c) the Appellant also said she no longer wore her wedding ring (p.52) and(d) she had been in receipt of tax credits as single person previously. Has the FTT adequately explained how it balanced such factors against the matters accepted as findings of fact?[9]Also, it is not immediately obvious that the FTT applied the principles established in R(TC) 2/06, which seems to suggest an objective approach in relation to the question of being separated in circumstances like to be permanent: “before any conclusion can be drawn other than that the separation is likely to be permanent, the tribunal must consider why the separation has occurred, and what indications there are that the couple may be reconciled. Even then, after balancing those indications against any contra-indications, the tribunal must conclude that there is at least a 50 per cent chance of a reconciliation before it can conclude that a single claimant and his or her spouse are to be treated as a married couple” (para. 18).[10]The absence of any specific reference to R(TC) 2/06 is not necessarily fatal, as on appeal the issue is whether the FTT applied the correct principles. It UA v HMRC (TC) [2019] UKUT 113 (AAC) 4 CTC/2349/2018 appears the FTT took into account the Appellant’s assessment of the likelihood of reconciliation from her comment that she sees the couple getting back together in the future (pp.59 and 297) and also the fact that she and her husband had a third child, four years after their alleged separation, as indication of possible reconciliation. These matters, of course, might support the FTT’s findings even if the SOR is in some way lacking.[11]The FTT’s decision to draw an adverse inference from the husband’s non-attendance at the hearing (SOR at paragraph [4]) may be problematic. It is true that following the FTT’s directions notice (p.293) the Appellant did not communicate any difficulties in arranging her husband’s attendance prior to the hearing. If she had, then the tribunal might have drawn her attention to the provisions of rule 16. On the other hand, that did not prevent the tribunal from summonsing the husband of its own motion, especially as it considered his attendance would have assisted. It is possible that the availability of the rule 16 remedy could render the inference inappropriate. However, it is not clear if that had any influence on the outcome of the case based on what is said at paragraph [4] of the SOR. Arguably paragraph [4] does not make it clear whether the inference was put into the scales against the Appellant. The FTT seems to have concluded that the message that the Appellant sent to her husband got through to him but he decided not to turn up. The inference the FTT drew from this is stated to be that ‘[the Appellant]’s husband does not support her appeal’ and it seems to imply that this was a point against the Appellant. But it could also be argued to be a point in her favour because it suggested the husband could not be bothered to put himself to any trouble to help her, and hence perhaps indicating the depth of their estrangement. Barring compulsion under a rule 16 direction, what possible incentive could there be for the husband to attend the hearing? It may be the FTT could be faulted for failing to explain their thinking behind this point.[12]The cultural issues identified by the Appellant in her grounds of appeal may well touch on some of the issues above. In that context, see especially Upper Tribunal Judge Markus QC’s decision in JH v HMRC (TC) [2015] UKUT 397 (AAC) at paragraphs 6-11 (FTT must not impose some objective standard of reasonableness when seeking to establish plausibility and credibility). See also my decision in VMcC v SSWP (IS) [2018] UKUT 63 (AAC), where I “expressed some concern that the First-tier Tribunal had not properly considered the Appellant’s explanations for certain events in the context of the community to which she and her family members belonged” (at paragraph 58 onwards).” The proceedings before the Upper Tribunal[7]Mrs M Spalding has provided a detailed and helpful written submission on behalf of HMRC, which supports the Appellant’s appeal to the Upper Tribunal. This is, I should add, entirely without prejudice to the position that HMRC may adopt on the issues raised by the substantive appeal to the FTT, which will be revisited at the re-hearing I direct below. Mrs Spalding accepts that the FTT identified the correct legal test and approached its determination with some care. However, Mrs Spalding’s submission is that the FTT erred in law by placing undue weight on certain aspects of the relationship between the Appellant and her husband but failing to address the Appellant’s evidence as to how the relationship had changed since 2009. It is true, of course, that the weight to be attached to any given evidence is axiomatically a matter for the judgment of the fact-finding first instance tribunal. However, Mrs Spalding’s further submission is that the FTT did not indicate which part of the Appellant’s evidence it considered unreliable and so failed to provide adequate reasons for the UA v HMRC (TC) [2019] UKUT 113 (AAC) 5 CTC/2349/2018 decision. Mrs Spalding therefore suggests that I allow the appeal, set aside the FTT’s decision and direct a fresh re-hearing before a different FTT.[8]The Appellant understandably supports that proposal.[9]I agree with Mrs Spalding and the Appellant that the FTT’s decision involves the agreed error of law. Despite the Tribunal’s otherwise careful approach, I conclude that its decision cannot now stand. I therefore allow the appeal and set aside the Tribunal’s decision of 6 June 2018. I could just leave it at that. However, it seems to me that this case raises several wider issues on which tribunals may need some guidance. The Upper Tribunal’s further analysis Introduction[10]There are three further points that merit exploration. These are(i) the FTT’s direction for the husband to attend the hearing;(ii) the proper test for assessing whether a married couple “are separated in circumstances in which the separation is likely to be permanent”; and(iii) the cultural issues surrounding the assessment of the Appellant’s evidence.

(i) The FTT’s direction for the husband to attend

[11]On 28 November 2017 the FTT issued a Directions Notice (p.293) to the parties. Direction (2) read as follows:
“(2) The Tribunal have considered the evidence in the appeal bundles and consider it would be assisted if the Appellant’s husband Mr [redacted] were to attend the hearing. [The Appellant’s husband] is to attend as a witness and give evidence. The Appellant is therefore put on notice that she should ensure [he] attends and that if he fails to do so that the Tribunal may draw an adverse inference.”
[12]The FTT GAPS2 records system shows that Directions Notice was issued to the Appellant and the Tax Credits Office on 8 December 2017. There is no evidence or any suggestion that a copy was issued directly to the Appellant’s husband.[13]The next document held on the appeal file after the Directions Notice is the record of proceedings of the FTT’s hearing on 10 April 2018. This records the attendance of the Appellant and two representatives from HMRC. It makes it clear that at the outset of the hearing the Judge raised the issue of the non-attendance by the Appellant’s husband. His careful note records “Directions 293, App[ellant]:- my Mum said she would tell him to come. I never have a conversation with him. I see him at my mother’s and at my house.”[14]The FTT’s statement of reasons dealt with the matter as follows:
“4. Directions had been made on 28 11 17 for [the Appellant’s] husband to attend the hearing. It was stated that if he failed to do so the Tribunal might draw an adverse inference. He did not attend. [The Appellant] told the Tribunal that her mother had said she would tell him to attend. There was no reasonable excuse for his non-attendance. The Tribunal did draw an adverse inference. The inference was that [the Appellant’s] husband did not support his wife in her appeal. This was an inference only. The appeal was not decided on the basis of this inference.”
UA v HMRC (TC) [2019] UKUT 113 (AAC) 6 CTC/2349/2018[15]There are several difficulties with the FTT’s approach, both before and at the hearing.[16]First, while the FTT was plainly entitled to make a direction that the Appellant’s husband should attend the hearing to give evidence (see Tribunal Procedure (First-tier Tribunal) (Social Entitlement Chamber) Rules 2008 (SI 2008/2685), rule 15(a), (b) and (e)(i)), there is no evidence the FTT gave any thought as to how in practice this direction would be communicated directly to the husband and how his willingness to attend would be ascertained. In the absence of direct service on the husband, it is also unclear how he was to be informed of his right to apply to have the direction amended, suspended or set aside (see rule 6(4)). Nor, assuming it were established that the husband had declined to attend, is there any evidence that the FTT considered exercising its power to issue a witness summons to him under rule 16 (or inviting the Appellant to make such an application).[17]Second, the FTT appears to have proceeded throughout on the assumption that it was the Appellant’s responsibility to ensure the attendance of her husband. Yet the core of the Appellant’s case was that she was estranged from her husband. The FTT was in effect requiring the Appellant to arrange the attendance of a potentially hostile witness. The Direction presented the Appellant with a juristic Catch-22. If she persuaded her husband to attend the hearing, then it could be said against her that, by definition, their relationship could not have broken down, or not to the extent that she claimed. If she was unable to persuade him to attend, then apparently “the inference was that [he] did not support his wife in her appeal”.[18]Third, the FTT’s deployment of the notion of an adverse inference was itself confused. As noted, the FTT drew the inference that the husband “did not support his wife in her appeal”. This finding is ambiguous. It could mean (and presumably was intended to mean) that the husband did not support the Appellant’s appeal because they were not in fact separated. Alternatively, it could equally mean he did not support the appeal because he did not consider they were separated, although objectively they were so estranged.[19]Fourth, the FTT’s reliance on the concept of an adverse inference failed to have regard to the principles affirmed in decisions such as R(CS) 6/05 and SSWP v HS (JSA) [2016] UKUT 272 (AAC); [2017] AACR 29. That case law shows that a finding of fact may be made against a party where that party fails to produce evidence that he or she could reasonably be expected to produce. Yet there was simply no adequate evidence for the FTT’s finding that “There was no reasonable excuse for his non-attendance”. The FTT had the barest of evidence that the Appellant’s husband had even been told to attend. At best it had a hearsay assurance that he would be told by a third party, the Appellant’s mother. As Mr Commissioner (now Judge) Jacobs held in R(CS) 6/05 (at paragraph 25):
“As the key factor is the inference that the party’s lack of co-operation is indicative of an inability to answer the opposing case, it is always relevant to consider whether there is another explanation for the lack of co-operation.”
[20]Fifthly, the FTT may well have concluded that “the appeal was not decided on the basis of this inference”, but it is difficult to see the FTT’s conclusion as other than weighing against the Appellant when the reliability of her evidence generally fell to be assessed. This points to a wider issue considered further below. As the Appellant put it in her notice of appeal, “The tribunal did not understand the cultural pressures I am under and face; this is shown by the tribunal asking that my husband is a witness at the hearing”. UA v HMRC (TC) [2019] UKUT 113 (AAC) 7 CTC/2349/2018 (ii) The proper test for assessing whether a married couple “are separated in circumstances in which the separation is likely to be permanent” Introduction 21. A person who is a member of a couple may not make a claim for tax credits as a single person. Such an individual must claim tax credits jointly with their partner (see Tax Credits Act 2002, section 3(3)).

(d) and Schedule 24 paragraph 144(3)):

“(a) a man and woman who are married to each other and are neither— (i) separated under a court order, nor (ii) separated in circumstances in which the separation is likely to be permanent”. 22. In the present case it was undisputed that the Appellant and her husband were married to each other and were not separated under a court order. So, the issue for the FTT was whether they were “separated in circumstances in which the separation is likely to be permanent”
. In summary, the FTT concluded that they were not separated and, if it was wrong about that, then the separation was not likely to be permanent.[23]It follows the first question to be addressed was whether the parties were separated. The FTT made its findings of fact and gave the reasons for its decision on this question in the passage at paragraphs 6-15 of the statement of reasons. Paragraphs 6-14 inclusive are all related to what might be described as ‘hard’ facts (e.g. the parties’ respective living arrangements, the Appellant’s failure to apply for a council tax single person’s discount, her failure to have the locks changed, the arrangements for the couple’s children, the husband’s postal address and his presence on the electoral roll at her address, etc.). There are three difficulties with the Tribunal’s approach. The first relates to the vagueness of the FTT’s fact-finding and associated reasoning. The second concerns the FTT’s approach to contra-indications. The third involves its treatment of ‘soft’ facts. The vagueness of the FTT’s fact-finding and associated reasoning[24]One example will suffice. The question of the parties’ living arrangements was necessarily crucial to the outcome of this appeal. In her original letter of appeal, the Appellant put her case as follows:
“I have not denied being married to [my husband] but that does not mean we are a couple who share a bed and all financial, material and emotional needs of children. [My husband] left me in 2009 and has not permanently resided with me as a couple. I am the breadwinner and provider for my family (which consists of me and my children). [My husband] is my mother’s nephew therefore all matters connecting between me and [him] are addressed through my parents. In 2009 when [he] left me, a mutual acceptance was agreed through family involvement which was the children are my responsibility and would reside with [me] and [he] would visit children as and when he can.”
[25]In other correspondence on the appeal file she added that “he does not reside at my address but visits the children as and when he wants to because he holds a key to my address and has refused to return it therefore I do not know when he will turn up. I have not been able to take any form of legal proceedings due to family involvement and also because of not being able to afford the costs”. UA v HMRC (TC) [2019] UKUT 113 (AAC) 8 CTC/2349/2018[26]The Appellant’s oral evidence at the hearing was noted in the record of proceedings as follows:
“Lived together with H until October 2009. I lived at [redacted address, here given as no.123] in 2009 and up to present time. Up to 2009 H also lived at no.123. After 2009 he left no.123. I do not know where he has lived since then. He has stayed at my mother’s, my brother’s and at no.123. If at no.123, I leave. I cannot tolerate him. Since Oct 2009 I have been the only adult at no.123. Since Oct 09 I have lived at my mother’s and brother’s.”
[27]The FTT’s finding on this issue as was follows:
“6. [The Appellant’s] evidence was that he had left in 2009 and that after then her husband had not had any permanent address. He had stayed at her house, at her mother’s house and at her brother’s house. However, [the Appellant] herself has not always stayed at her own address. She herself has stayed at her mother’s house for extended periods. They have therefore both stayed at the same address. These findings are based on her evidence.”
[28]I readily accept that this was a difficult case. However, the findings in paragraph 6 of the statement of reasons are thin and fail adequately to address the Appellant’s arguments about the nature of the separation. Contra-indications[29]As regard contra-indications, the Appellant’s evidence was consistent that she received no financial support whatsoever from her husband. As she told the FTT, “If he was providing for me, I would not be in debt.” At paragraph 11 of the statement of reasons the FTT found as follows: “11. [The Appellant’s] husband has not paid any maintenance for his wife or children. She has not made any application for maintenance or to the Child Support Agency. She was asked why she had not made any application. Her reply was that this would cause a rift in her own family and she would be disowned by them. There was no reason to doubt this evidence from her. Nevertheless, the fact is that her husband has not paid maintenance for his wife or children and has not paid towards the upkeep of the house”.[30]This would appear to suggest that the couple were not just unhappily married but leading separate lives, but it is unclear what weight, if any, the FTT attached to this potentially significant contra-indication. The First-tier Tribunal’s treatment of ‘soft’ facts[31]Furthermore, in terms of the third difficulty with the FTT’s approach, the only reference to what might be called ‘soft’ facts about the nature of the couple’s relationship was in paragraph 15, which read as follows:
“15. [The Appellant] sees in the future that they will get back together again. She hopes that he will become her ‘soul mate’ again. This is also what her parents and his parents would want. This is in line with her statement at the interview under caution when she was talking about a conversation with her father in law. It was also in line with her evidence to the Tribunal recorded at page 298. It is recognised that in her evidence to the Tribunal, recorded at page 303, she also said that in the long run it looked likely that she would take him back but that for the moment the separation was permanent.”
UA v HMRC (TC) [2019] UKUT 113 (AAC) 9 CTC/2349/2018[32]I should add for completeness that the reference to the Appellant’s oral evidence, as recorded at p.298, read thus:
“At work I portrayed myself as happily married. I did not want to shame my family. They are my only support. They would disown me if I went thru’ a divorce. He was my soulmate – we were happily married. Life was lovely till he started seeing other women. I see future as getting together with him & hope he becomes my soul mate again. This is what my parents + his parents would want.”
[33]Furthermore, the reference to p.303 is to the note from the record of proceedings to the effect that “‘I am happy to take him back’ – in long term it looks like this will happen.”[34]Paragraph 15 does not involve the FTT making any findings of fact as such. It is more by way of the FTT narrating the Appellant’s evidence as to how she viewed the relationship with her husband. Unfortunately, the FTT, by focussing solely on the ‘hard’ facts, to the exclusion of the ‘soft’ facts, failed to consider the question of separation in the round. As I observed in DG v HMRC (TC) [2013] UKUT 631 (AAC) at paragraph 36:
“… The question of whether there is an enduring household is often an important and necessary consideration, but it is by no measure a determinative consideration. There are some married couples who do not share a single common household but, because they are ‘living apart together’, are not separated within the terms of section 3(5A)(a)(ii). Rather, the test of separation must focus on the relationship between husband and wife, and a married couple may be separated whilst living in the same household, whether or not they have a court order to that effect. Ultimately, neither limb of section 3(5A) is conclusively determined by physical separation, although in the typical case physical separation in terms of living apart at different addresses will be very good evidence of the separation required. Instead, section 3(5A) is at root concerned with emotional separation or estrangement.”
[35]There is further invaluable guidance on the proper application of section 3(5A) in SA v HMRC (TC) [2017] UKUT 90 (AAC), a decision of Upper Tribunal

Judge Jacobs, in the passage at paragraphs 10-12:

“10. Section 3(5A)(a) applies to couples who are married but separated. It has to be interpreted and applied in the context that there is a continuing marriage. It is inherent in such a situation that the couple may continue to have some involvement with each other, over family and financial matters if nothing else.”
[11]The length of the separation may be relevant, but its significance may be difficult to assess. Separation followed by divorce is the usual sequence of events, so the lack of divorce proceedings may indicate that the parties have not yet decided their future marital status. On the other hand, the longer the separation continues, the stronger the indication that the parties may be content with the status quo. The legislation envisages that a couple may remain married but separated. It would be wrong to reason that because they remain married, their separation cannot be permanent. There may, after all, be good reasons –reasons that make sense to the parties – not to divorce. It may be that they cannot afford it. It might, realistically, make no practical difference to their situation. They may be happy as they are and not wish relations to be soured by legal proceedings. They may just find the idea of legal proceedings intimidating. UA v HMRC (TC) [2019] UKUT 113 (AAC) 10 CTC/2349/2018 My point is that the length of the separation must be considered in the context of all the circumstances – as the judge said.[12]It is important not to confuse the existence of the separation and its permanence. The degree to which the parties continue to be involved with each other and in each other’s lives may indicate that they have not truly separated. but if that is not the case, as the judge found in this case, it is wrong to assess the evidence on the basis that continuing involvement is necessarily an indication that the separation is not likely to be permanent. As I have said, continued involvement may be inherent in the situation. There is also this point. Even divorced couples may continue to be involved in each other’s lives. This may be because it is forced upon them by their responsibility for their children or by the financial dependence of one on the other. But it may also be because they continue to share interests and wish to provide mutual support. The model of a divorced couple as warring combatants is not universal. If continuing friendship and support is possible for a divorced couple, it follows that those features must not be taken as necessarily indicating that the separation is likely not be permanent.”[36]Paragraph 16 of the FTT’s statement of reasons declared that “It was based on the facts set out in the preceding paragraphs [note: i.e. paragraphs 6-15] that the decision was taken. There was also further evidence given and facts found on the following points.” This statement was then followed by three paragraphs dealing respectively with the way that the Appellant portrayed herself at work (paragraph 17), the couple’s use of a car (paragraph 18) and a trip to Pakistan (paragraph 19). In each case, however, the FTT expressly disavowed any reliance on these matters in reaching its decision as e.g. the true factual circumstances were too unclear. These three paragraphs were then followed by the final substantive paragraph in the FTT’s statement of reasons:
“20. The role of [the Appellant’s] family was difficult to understand. The accepted evidence was that she was initially reluctant to admit to her parents, and later to her parents in law, that the relationship between her and her husband had broken down. They were all aware of it later. [The Appellant] considers that a public acceptance of their separation would bring shame on her family and they would disown her. Reactions of her parents and family would appear to indicate that they do not accept that [the Appellant] and her husband are separated, or if they did accept this they do not accept the separation is permanent.”
[37]This takes us to the second question that may arise. Assuming for the present the parties were separated, was that separation “likely to be permanent” within section 3(5A)(a)(ii)? The only passages that dealt with this aspect of the case were paragraphs 15 and 20 of the statement of reasons. However, as noted above, paragraph 15 was no more than a summary of the Appellant’s evidence while paragraph 20 seems to accept that the relationship had indeed broken down (the reference to the “accepted evidence”) but it was not likely to be permanent as the wider family either did not acknowledge the separation or did not consider it was likely to be permanent.[38]However, the statutory test under section 3(5A)(a)(ii) is ultimately an objective test, not a subjective test, as shown by R(TC) 2/06 (admittedly a case in which there had been a trial (physical) separation, at least at the outset of the tax credits claim):
“18. While each case must turn on its own facts, it is important to bear in mind that married couples do not separate unless there have been serious problems UA v HMRC (TC) [2019] UKUT 113 (AAC) 11 CTC/2349/2018 in their relationship, or other problems, such that continuing to live together becomes unacceptable for at least one member of that couple. Before any conclusion can be drawn other than that the separation is likely to be permanent, the tribunal must consider why the separation has occurred, and what indications there are that the couple may be reconciled. Even then, after balancing those indications against any contra-indications, the tribunal must conclude that there is at least a 50 per cent chance of a reconciliation before it can conclude that a single claimant and his or her spouse are to be treated as a married couple. It is unlikely that such a reconciliation will occur before the parties have taken steps to deal with the problems that led to the separation in the first place, and have actually begun the process of arranging to live together again. A tribunal should be slow to differ from the claimant’s own genuine assessment of the likelihood of a reconciliation, although, of course, that is a subjective assessment and the tribunal is not bound by it.”
[39]The wider family’s assessment of the likelihood of a reconciliation may be just as subjective as that of a party to the relationship. This may be especially so when cultural pressures in a marriage with close intra-family ties are taken into account. (iii) The cultural issues surrounding the assessment of the Appellant’s evidence[40]It is trite law that the assessment of the facts is a matter for the first instance tribunal. But it is not always that straightforward, as Upper Tribunal Judge Markus QC helpfully explained in JH v HMRC (TC) [2015] UKUT 397 (AAC), where the issue was whether an unmarried couple were still living together or had separated (with regard to the slightly different test under section 3(5A)(b)): “5. The Upper Tribunal will be slow to interfere with the First-tier Tribunal’s findings of fact. It may only do so if those findings were made in error of law. This includes making perverse or irrational findings on material matters, which includes findings which are not supported by the evidence; failing to take into account material matters; or taking into account immaterial matters. See R (Iran) v Secretary of State for the Home Department [2005] EWCA Civ 982 at [9] – [11].[6]The assessment of the credibility or plausibility of a witness’s evidence is primarily a question of fact for the tribunal. In HK v Secretary of State for the Home Department [2006] EWCA Civ 1037 Neuberger LJ said, at [30], that rejection of an account on grounds of implausibility must be done ‘on reasonably drawn inferences and not simply on conjecture or speculation’. In addition, a tribunal may properly draw on its common sense and ability, as practical and informed people, to identify what is or is not plausible.[7]In Gheisari v Secretary of State for the Home Department [2004] EWCA Civ 1854 (with which Neuberger LJ agreed) the Court of Appeal emphasised that an account that is unlikely may nonetheless be true, just as a likely account may turn out to be untrue. Faced with an account which a tribunal considers to be improbable, its task is to appraise the evidence and the individual who gave the evidence, and decide whether it is true – Gheisari at [12], [13] and [16]. It may not be necessary for a tribunal to carry out a strict two stage test (improbability followed by truth), but: ‘What would be wrong would be to say …. that because evidence is inherently unlikely it inevitably follows that it is wrong. An unlikely description may, upon a consideration of the circumstances as a whole, including the UA v HMRC (TC) [2019] UKUT 113 (AAC) 12 CTC/2349/2018 judge’s assessments of the witness and any explanations he gives, be a true one.’ (Pill LJ in Gheisari at [21])[8]The above discussions were made in the context of asylum appeals, where inherent improbability may be particularly unhelpful because ‘[m]uch of the evidence is referable to societies with customs and circumstances which are very different from those of which the members of the fact-finding tribunal have any (even second-hand) experience’ (HK at [29]). It follows that inherent improbability may be more helpful in cases where the evidence is closer to the experiences of the tribunal, but it will nonetheless only be a component of the overall task which is to decide whether a witness’s account did occur not whether it was likely to have occurred. The general approach set out in HK and Gheisari is apt in cases such as the present. I note that it was followed by the Court of Session Outer House in an appeal concerning a decision relating to the educational needs of a learning disabled child: G v Argyll and Bute Council [2008] CSOH 61 at [157].[9]It also follows from this approach that it will generally be inappropriate for a tribunal to appraise evidence by reference to what a reasonable person would have done. The question for the tribunal is what the individual in question is likely to have done, not what some other (hypothetical or actual) person would have done. As both Sedley LJ and Pill LJ said in Gheisari, the fact-finder should appraise the person giving the evidence (paragraphs [13] and [21] respectively).”[41]In that case Judge Markus concluded that the FTT had erred in law for the following reason (at paragraph 10; ‘Mr W’ was the claimant’s alleged partner in that appeal):
“… The tribunal did not accept the appellant’s evidence because it considered that it was improbable. It did so on the basis of what the tribunal would have expected a person in her position to have done, and in one instance the tribunal expressly applied the test of a ‘reasonable’ person in the appellant’s position. The tribunal did not, whether as part of a single fact-finding process or by considering the evidence in stages, consider whether the appellant’s account was true rather than improbable. Had it done so, the tribunal would have had to consider matters such as the appellant’s particular circumstances and her explanation for her actions or those of Mr W, and would have had to assess what it was likely that she or he would have done.”
[42]In another passage which has potential resonance in the context of the present appeal,

Judge Markus observed as follows (at paragraph 11(c)):

“… In the context of alleged relationship breakdown, such an approach is particularly inappropriate as different people will behave in such markedly different ways and it is unrealistic to expect people to make choices according to some objective standard of reasonableness.”
[43]However, in fairness to the FTT that decided the appeal in the present case, it did not fall into the trap of erroneously applying an overarching test of objective reasonableness in assessing the plausibility and reliability of the Appellant’s evidence. Indeed, in several instances it made findings of fact that were expressly stated to be based on an acceptance of the Appellant’s evidence (see e.g. some of its findings about the parties’ living arrangements). The difficulty with the FTT’s approach is a more nuanced point. When assessing evidence about parties’ living UA v HMRC (TC) [2019] UKUT 113 (AAC) 13 CTC/2349/2018 arrangements, it is important that tribunals make appropriate findings while viewing the facts through the prism of any relevant cultural context.[44]As I said when giving permission to appeal in VMcC v SSWP (IS) [2018] UKUT 63 (AAC) (cited at paragraph 26 of that decision):
“I am concerned that the FTT may not have given proper and full consideration to the Appellant’s case. In particular, I think it possible that the Appellant's background as coming from the Traveller community may not have been taken into account. The case has equal treatment implications. For example, cultural differences may significantly affect the way that evidence is viewed. A state of affairs that may be regarded as literally incredible by one community may be regarded as quite ordinary by a member of another community. For example, the Appellant says she was disowned by family members for becoming pregnant. Such disowning was very common at one time in the majority population; it may well still be common in minority communities. The Tribunal appears to have ignored that aspect of her explanation for the events in question.”
[45]Similar cultural differences apply in the context of relationship breakdown and divorce. In general, social attitudes to relationship breakdown are far more relaxed than they used to be, and in many communities the stigma that historically attached to divorce has all but disappeared. But this trend is by no means universal. As the Judicial College’s Equal Treatment Bench Book (revised March 2019) observes, “there are many different views on the acceptability of divorce in different cultures” (at paragraph 110). The position in Islam is explained in a publication by the Nuffield Foundation, written by Sonia Nûrîn Shah-Kazemi (Untying the Knot: Muslim Women, Divorce and the Shariah, 2001, pp.6-7):
“In Islam, marriage is regarded as a contract; consequently, divorce has always been permissible in the Shariah, which, however, advocates conciliation (mediation) prior to, and as an alternative to, litigation… Divorce is not an entirely unilateral prerogative of the husband, but can also be initiated by the wife on application to a Qadi (religious judge). The existence of this legal framework for the dissolution of marriages has had a fundamental impact on the attitudes towards divorce held by Muslims. It is said in a well-known hadith (a saying of the prophet Muhammad), ‘Among all things that God has made legal, divorce is the most hateful.’ This hadith encapsulates the complex attitude towards divorce; on the one hand, it is permissible to end a marriage, that is to say that the hardships of an intolerable life-long union need not be endured. On the other hand, divorce is a final resort, and to be avoided if at all possible.”
[46]The notion that ‘Among all things that God has made legal, divorce is the most hateful’ is reflected in the Appellant’s evidence. In her request for a mandatory reconsideration she wrote as follows:
“I have kept my separation very private and discreet other than my family no one else has been aware of this, I only told my in-laws in 2016, because I stand the chance of losing my family support here in England. My mother cares for my children while I go to work and if I lost her support I would not be able to work and provide for my children and I would have no-one to turn to.”
[47]Likewise, in the Appellant’s notice of appeal to the Upper Tribunal, she writes that “my cultural background is such that I have a choice of appearing to be married and being accepted by my family or divorcing and leading a life that others UA v HMRC (TC) [2019] UKUT 113 (AAC) 14 CTC/2349/2018 understand as that of a single parent, but that would also lead me to being disowned by my family.” This is a dilemma that many people in other communities, including many judges, will never have to face and may have some difficulty in comprehending. The Appellant has further eloquently explained that that the cultural pressure to have a successful marriage and the shame associated with separation “can lead to actions that other cultures may view as strange or not compatible with a separation but the mask of marriage can hide a very different reality, especially for a woman from my culture.” She also points out, in relation to a marriage to a first cousin, that conduct which may appear to be that of a dutiful wife may be equally explicable on the basis of family duty or respect for one’s parents or other elders. The following three examples will suffice:• The Appellant’s evidence was that the parties had effectively separated in October 2009. One of the factors relied upon by the FTT in finding that the parties had not separated was that the couple had a third child born in 2014. There was no mention by the FTT of the Appellant’s oral evidence at the hearing that “2013 mother in law visited for family wedding. We slept in same room – he in bed, I on floor, 3rd child conceived then – in same room for several months – for show for mother in law”. Nor is there any reference to the more detailed explanation about the living arrangements during this visit that the Appellant gave in the interview under caution.• A further factor was that “In January 2010 [the Appellant] prepared a letter for her husband regarding a penalty notice and gave her own address as his address”. There was no mention of the Appellant’s oral evidence that “I prepared the letter on work computer – appealing a penalty notice – my mother told me to prepare this letter”.• Another factor was that the Appellant’s husband had his post delivered to her address. There was no mention of the Appellant’s explanation both in writing and orally that “my family made this agreement and I had to agree to this. I have tons of post for him. Some he picks up. Some he tells my son to collect and bring to my mother’s for him. I have raised this with my Mum and she said he said he didn’t have a permanent address.”[48]These examples illustrate the importance of interpreting evidence about an alleged separation with an awareness of its cultural context. Finally, what happens next? The new First-tier Tribunal[49]This all means there will need to be a fresh hearing of the appeal before a new First-tier Tribunal. Although I am setting aside the previous Tribunal’s decision, I should make it clear that I am making no finding, nor indeed expressing any view, on whether the Appellant was properly entitled to tax credits as a single person as claimed. That is a matter for the good judgement of the new tribunal. That new tribunal must review all the relevant evidence and make its own findings of fact.[50]In undertaking this process the new Tribunal will doubtless be assisted by the helpful submission from Mrs Spalding. Conclusion[51]The Appellant’s appeal is allowed. I conclude that the decision of the First-tier Tribunal involves an error of law. I allow the appeal and set aside the decision of the tribunal (Tribunals, Courts and Enforcement Act 2007, section 12(2)(a)). The First-tier Tribunal decision is now of no effect. The case must be remitted for re-hearing by a UA v HMRC (TC) [2019] UKUT 113 (AAC) 15 CTC/2349/2018 new Tribunal subject to the directions above (section 12(2)(b)(i)). My decision is as set out above. Signed on the original Nicholas Wikeley on 2 April 2019 Judge of the Upper Tribunal

Cited in 2 later judgments