UK Social Security and Child Support Commissioners' Decisions

[2008] UKSSCSC CCS_203_2008 (14 July 2008)

 


CCS/203/2008

DECISION OF THE CHILD SUPPORT COMMISSIONER

The decision of the Child Support Commissioner

1. My decision as Child Support Commissioner is as follows. The decision of the Birmingham appeal tribunal on 4 September 2007 under file reference 024/07/03772 is wrong in law. I therefore allow the appellant's appeal against that decision and set the tribunal's decision aside.
2. The assessment of the appropriate level of child support maintenance for the two boys in this case has been the subject of repeated tribunal hearings in the past. In my view I should therefore avoid sending the case back for yet another hearing if at all possible. Fortunately, whatever their other differences, the parents have kept meticulous records of overnight contact and the tribunal was able to make agreed findings of fact. On that basis it is sensible for me to substitute my own decision for that of the tribunal (Child Support Act 1991, section 24(3)(a)).
3. The decision that the tribunal should have made is as follows:

The father's appeal dated 7 February 2007 is dismissed.

The Secretary of State's decision of 19 December 2006 that the father was not entitled to a reduction for shared care with effect from 19 July 2006 and with effect from 13 September 2006 was correct.

This is because in the 12 months ending with the relevant week the father did not have "day to day care" amounting to 104 nights or more in accordance with the statutory definition of that term.

The parties to this appeal and the CSA's two schemes

4. In the technical language of the child support legislation, the appellant is the parent with care under the maintenance assessment, the first respondent is the Secretary of State and the second respondent is the non-resident parent. For convenience, however, I shall refer to them in this decision as the mother, the Secretary of State and the father respectively. For all practical purposes the Secretary of State's functions in relation to child support are carried out by the Child Support Agency (the CSA). This case concerns the shared care of the couple's two sons, who live most of the time with their mother but a significant amount of time with their father.
5. The CSA is at present required by legislation to operate two child support schemes in parallel, an "old scheme" and a "new scheme". In broad terms, cases that predate 3 March 2003 are old scheme cases and cases since that date are governed by the new scheme. This is an old scheme case, because the initial maintenance assessment was made with an effective date of 11 December 2002, before the Child Support, Pensions and Social Security Act 2000 came into force in 2003.
6. There are significant differences between the two schemes. The distinction is crucial in this case as the tribunal worked on the erroneous assumption that this was a new scheme case and so applied the wrong law to the issue of shared care.

The background to this appeal to the Commissioner: the March 2006 tribunal

7. For present purposes the background to the present appeal starts with a separate tribunal hearing in March 2006, which is not now under appeal. On 27 April 2005 a decision maker had made a maintenance assessment for £167.42, with an effective date of 13 October 2004. On 12 October 2005 a decision maker made a revised maintenance assessment for £112.24 a week from the same effective date of 13 October 2004. The reduction was because of an allowance for day-to-day care on the part of the father. The mother appealed.
8. On 21 March 2006 a Birmingham tribunal allowed the mother's appeal. That tribunal decided that for the 12 months between 15 October 2004 and 14 October 2005 the boys had stayed with their father for 100 nights, below the 104 nights per year threshold for shared care in old scheme cases. The tribunal therefore directed that the maintenance assessment made on 12 October 2005 should be recalculated on that basis. No appeal seems to have been made against the March 2006 tribunal decision.
9. On 18 or 19 December 2006 a decision maker then made further maintenance assessments for various amounts with effective dates of 19 October 2005 (for £131.18), 28 June 2006 (for £130.55), 19 July 2006 (for £121.82) and 13 September 2006 (again for £121.82). These assessments included no allowance for any shared care.

The father's appeal against the decision notified on 19 December 2006

10. On 7 February 2007 the father appealed against the decision letter dated 19 December 2006. The decision maker took the view that the issue was "res judicata" (a matter already decided), as the March 2006 tribunal had determined the issue of shared care. The decision maker therefore argued that the tribunal had no jurisdiction to hear the father's appeal.
11. A District Chairman pointed out that the appeal concerned a decision dated 19 December 2006, which post-dated the March 2006 tribunal (with the implication that the issue was not "res judicata") and asked the Secretary of State for a supplementary submission. This submission explained that two quite separate decision-making processes had taken place within the Agency, coincidentally both occurring on 18 December 2006.
12. First, it was said that a decision maker at the CSA Dudley office made supersession decisions with effective dates of 28 June 2006, 19 July 2006 and 13 September 2006. These decisions were based on other reported changes of circumstances, and did not relate to shared care. The Secretary of State's representative now says that there were only two separate supersession decisions, effective from 19 July 2006 and 13 September 2006.
13. Second, and on the same day, a decision maker at the CSA's Central Appeals Unit implemented the March 2006 decision, and so removed the shared care allowance, and then carried that reduction forward to the supersessions made at Dudley. The decision maker's supplementary decision repeated the contention that the father's appeal against the assessment with an effective date of 13 September 2006 was "res judicata".
14. Meanwhile, the mother had sent in evidence to the effect that the father's proportion of shared care amounted to 102 nights for the year from 13 September 2005 to 12 September 2006 (i.e. the 12 months immediately before the date identified by the father in his letter of appeal).

The September 2007 tribunal allows the father's appeal

15. The father's appeal was heard at Birmingham on 4 September 2007 before a lawyer chairman. Both parents and an Agency presenting officer attended the hearing. The father also submitted his own schedule of shared care dates. This ran for a period of almost two years, from 15 October 2004 through to 18 September 2006. So the first year of this schedule referred to a period that had been dealt with by the March 2006 tribunal. However, for the period from 12 October 2005 the father's schedule kept a running tally of the annual number of nights of shared care in the year immediately before each date in the diary. This showed cumulative totals, depending on the precise end date for the 12-month period chosen, which ranged between 96 nights and 121 nights.
16. The tribunal chairman allowed the father's appeal. In the Decision Notice issued on the day, she ruled that the decision made on 19 December 2006 should be revised to the extent that with effect from 19 July 2006 the assessment should reflect shared care of 2 nights a week (i.e. 104 nights a year). The reason for this was an intended change in the frequency of contact under a court order made in September 2006. The case was sent back to the Agency for recalculation accordingly.

The mother's appeal to the Child Support Commissioner

17. The mother applied for leave to appeal. She argued that the period in dispute was the period from 20 July 2005 to 19 July 2006, that it had been agreed that the father did not reach the 104-night threshold for that 12 month period and so therefore he did not have shared care. She objected to the use of the September 2006 court order as it did not relate to the relevant period and a copy had not been before the tribunal in any event.
18. The tribunal chairman provided a full Statement of Reasons for her decision. The mother repeated her grounds of appeal, adding that the September 2006 court order had in any event only had the effect of changing Thursday overnight contact to Sunday overnight contact and so did not affect the overall frequency of contact. A District Chairman refused the mother leave to appeal. A Commissioner then granted leave to appeal in the light of the Secretary of State's helpful observations on the mother's application for permission to appeal. In response the father has made no submissions.

The Commissioner's reasoning

19. I agree with the submission of the Secretary of State's representative that the tribunal decision is erroneous in point of law, but not for all the reasons that he gives. The principal error of law was the tribunal's reference to regulation 7(4) of the Child Support (Maintenance Calculations and Special Cases) Regulations 2000 as the provision governing the assessment of shared care in this case. This was a mistake as the 2000 Regulations apply only to new scheme cases, and this is an old scheme case.
20. The present case is therefore governed by the Child Support (Maintenance Assessments and Special Cases) Regulations 1992 (or the "MASC Regulations"). The shared care rules in the 1992 and 2000 Regulations are similar but not identical. For that reason alone I must set aside the tribunal's decision as erroneous in point of law. I now turn to the particular legal issues raised by this appeal.

Which decision on a maintenance assessment was under appeal?

21. The tribunal found that the 19 December 2006 decision applied to four different effective dates and that the father's appeal was made in relation to the last of these dates (13 September 2006). The Secretary of State's representative takes a different view. He argues that the father was in effect appealing against all four assessments made on 19 December 2006 (i.e. those dated 19 October 2005, 28 June 2006, 19 July 2006 and 13 September 2006).
22. I disagree with this aspect of the analysis of the Secretary of State's representative. In answer to the question on the appeal form "Which decision are you appealing about?", the father stated categorically "The refusal to allow shared care w.e.f. 13.9.06". When the form asked him to explain why he thought the decision was wrong, he stated "I have had shared care for 12 months prior to 13.9.06 and since that date" (emphasis added).

What then was the "effective date" of the maintenance assessment under appeal?

23. The process of ascertaining the "effective date" is governed by some complex rules. In particular, if a party applies for a supersession, then a superseding decision on the basis of a change in circumstances takes effect from the week in which the application was made (Child Support (Maintenance Assessment Procedure) Regulations 1992 (SI 1992/1813), reg. 23(4)).
24. But what was the effective date for the decision under appeal? It is clear from the Record of Proceedings that the tribunal was alive to the importance of establishing the appropriate effective date (and also the "relevant week"). The tribunal specifically adjourned the hearing for 10 minutes for the presenting officer to make enquiries by telephone to clarify the point. The chairman's note then records that the September decision was itself a revision (not a supersession) of the July decision, and that a telephoned notification had been received from the father on 19 July 2006 of a change in his housing costs and council tax.
25. The tribunal accordingly found that the decision with an effective date of 13 September 2006 was itself a revision of an assessment effective from 19 July 2006 (the amounts were certainly the same). This also tallied with the parties' common recollection. The tribunal accordingly found that 19 July 2006 was the effective date of the decision under appeal. In my view the tribunal reached the correct conclusion on this issue.
26. Tribunals have been criticised in the past for not making appropriate enquiries about such matters (see e.g. CCS/3671/2002 at paragraph 12). Rather than adjourning to another date, and cause yet more delay, this tribunal took the sensible course of a short adjournment to allow enquiries to be made. It may not have had documentary evidence to support its conclusion, but it had the best evidence available. It adopted a pragmatic approach that, in my judgment, was sufficient.

Was the decision appealed against "res judicata"?

27. The original decision maker and the Secretary of State's representative both suggest that the appeal was "res judicata", although the presenting officer abandoned this challenge at the hearing before the tribunal. "Res judicata" is an unhelpful Latin tag for a legal doctrine meaning that the matter has already been decided. In this context it is also known as issue estoppel. The basic idea is simple – an issue that has already been determined cannot be re-opened and re-litigated.
28. "Res judicata" is a concept that has its roots in the adversarial court system and does not translate easily into an inquisitorial tribunal system. Indeed, Mr Commissioner Jacobs ruled that neither "res judicata" nor any analogous principle applied in the original child support scheme, before the decision making reforms in the Social Security Act 1998 (see CCS/1535/1997). The same Commissioner also made it clear that his reasoning applied to the pre-1998 Act system and not to the post-1998 arrangements for decision making (CCS/2403/1998 and CCS/262/1999).
29. However, the new system is governed by section 46A of the Child Support Act 1991. This makes it tolerably clear that while appeal tribunal decisions are final (subject to any appeal, etc: see s.46A(1)), findings of fact are not conclusive for later decisions (see s.46A(2) and CIS/1330/2002). No regulations have been made to any contrary effect (contrast regulation 10 of the Social Security (Decisions and Appeals) Regulations 1999 (SI 1999/991) in the context of decisions about incapacity for work).
30. In my view the use of the term "res judicata" in the child support context is therefore both unhelpful and inaccurate. In any event, on the facts of this case it is plain that the father was not seeking to re-open e.g. the assessment with an effective date of 19 October 2005, but only that effective from 13 September 2006. That (and the decision of 19 July 2006) had not been the subject of the March 2006 tribunal, which was concerned with a different period, as the tribunal chairman in the present case noted. So there was no basis for arguing that the father's appeal was "res judicata", either in law or on the facts.

What is the "relevant week" for the maintenance assessment under appeal?

31. The next question for the tribunal was to identify the right "relevant week". This week is important as it is used as the starting point for calculating a parent's income (in this case, in practice, the father's). The concept of a "relevant week" is defined in some detail in regulation 1(2) of the MASC Regulations 1992. Where there is a supersession decision, and that decision has been prompted by a reported change of circumstances, the "relevant week" is the period of seven days before the date on which the application was made (para. (c)(i) of the definition in reg. 1(2)). Where the Secretary of State has initiated the supersession, the "relevant week" is the period of seven days before the official notification under reg 24 (para. (c)(ii)).
32. The Secretary of State's representative argues that the tribunal simply assumed that the relevant week was the week before the effective date in the absence of any evidence about an application for a supersession and that this amounts to an error of law. I disagree. I have already found that the tribunal was entitled to reach the pragmatic conclusion on the available evidence that the effective date was 19 July 2006. It was also entitled to conclude that the relevant week was the period of seven days before that date. So far, therefore, I find no error in the tribunal's approach.

What is "day-to-day care" for the purposes of the shared care rules?

The relevant law

33. This is where the tribunal started to come unstuck. In old scheme cases a complicated formula provides for a reduction in the child support maintenance liability where separated parents both provide "day to day care" for the same child (MASC Regulations 1992, reg. 20). But for these purposes a parent only provides "day to day care" if they meet the 104 nights threshold. This rule is contained in the definition of "day to day care" in reg. 1(2). The material part reads:

" "day to day care" means—

(a) care of not less than 104 nights in total during the 12 month period ending with the relevant week; or

(b) where, in the opinion of the Secretary of State, a period other than 12 months but ending with the relevant week is more representative of the current arrangements for the care of the child in question, care during that period is not less in total than the number of nights which bears the same ratio to 104 nights as that period bears to 12 months."

The tribunal's reasoning on day to day care

34. So in this case the starting point was that the 104 nights test should be applied to the 12 months ending with the relevant week. The tribunal found that the father had overnight care for 101 nights in that 12-month period. On that basis he did not qualify. However, the tribunal noted the way that the cumulative total of annual overnight stays varied according to the end date chosen. The tribunal considered that it was unjust for the outcome of the appeal to depend "on an almost fortuitous choice of dates within a narrow timeframe".
35. The tribunal therefore decided to take advantage of what it saw as the flexibility in the Regulations of using a different period. The tribunal relied on regulation 7(4) of the 2000 Regulations. This reads as follows:

"(4) The circumstances in which the Secretary of State may have regard to a number of nights over less than a 12 month period are where there has been no pattern for the frequency with which the non-resident parent looks after the qualifying child for the 12 months preceding the relevant week, or the Secretary of State is aware that a change in that frequency is intended, and in that case he shall have regard to such lesser period as may seem to him to be appropriate, and the Table in paragraph 7(4) and the period in paragraph 8(2) of Schedule 1 to the Act shall have effect subject to the adjustment described in paragraph (5)."

36. In particular, the tribunal relied on the September 2006 court order as representing a change in the frequency of overnight contact. The tribunal undertook an arithmetical calculation and found that the court order had the effect of providing for exactly 104 nights of shared care over the course of a year. This, the tribunal ruled, was a more appropriate approach and so the father's appeal succeeded.

Why the tribunal erred in law

37. There are several problems with the tribunal's reasoning on this point. First, it applied the new scheme rule (actually devised for a 52 night a year threshold) rather than the old scheme rule (devised for a 104 night a year threshold). In particular, there is no direct equivalent in the old scheme to the provision in the new scheme that operates where "the Secretary of State is aware that a change in that frequency is intended". Second, even if there were such a change, the tribunal did not use a period of less than 12 months. Instead, it used a different 12-month period (by reference to the new court order) – precisely the problem it purported to identify with the default rule. Third, the tribunal relied on the court order as evidence of a change in the frequency of overnight care. It is by no means clear from the Record of Proceedings that the tribunal had evidence before it to that effect. Certainly the mother's case is that the effect of the court order was merely to substitute Sunday overnight stays for Thursday overnight stays, a contention the father has not challenged. Finally, it is well-established as a matter of law that it is wrong to place undue reliance on the terms of a court order – what actually matters is the factual amount of contact that takes place (see e.g. CCS/2885/2005 on the new scheme).

What the tribunal should have done

38. What the tribunal should have done is to apply the definition of "day to day care" in reg. 1(2) of the MASC Regulations 1992 to determine whether or not the 104 nights test had been satisfied. The tribunal found as a fact that the 104-night rule was not met in the period of 12 months ending with the relevant week – the father had overnight contact amounting to only 101 nights during this period. It was right to do so.
39. The question then is whether it is appropriate under the second limb of the definition to consider a period other than 12 months as being "more representative of the current arrangements". If so, I may indeed rely on a different period, but using the same ratio as 104 nights to a year to see if the threshold is met. There is some uncertainty as to the meaning of "the current arrangements". In R(CS) 4/03 Mr Commissioner Turnbull (at paragraph 13(3)) suggested that this referred to the position as at the effective date of the maintenance assessment in question. I agree that that is the most likely meaning, notwithstanding the other doubts I mention below.
40. In this context I reiterate that the schedule of overnight stays was agreed between the parties and accepted by the tribunal. This demonstrated that in the 12 months preceding 19 July 2006 there were 101 overnight stays. It so happened that in the year to 13 September 2006, the next effective date, the running total was 102 nights. It is also true that the cumulative total ranged from 96 to 121 nights.
41. Is there a good reason to choose a period other than the 12 months ending with the relevant week? Is a period other than that 12 months "more representative of the current arrangements for the care of the child in question"? I think not. There are obviously going to be variations over time, especially with different arrangements for term-time and school holidays. But the year to the relevant week (July to July) is as good a measure as any of the extent of shared care.
42. By way of a benchmark I have also examined all the running totals from 15 October 2005 (the day after the last date chosen by the March 2006 tribunal) to 18 September 2006 (the end date in the agreed schedule), a period of about 11 months. If one looked at the running totals in this period only up to 19 July 2006, it appears that the cumulative totals were 103 nights or less on 165 occasions and reached 104 nights or more on 113 occasions. If one goes to the end of the schedule in September 2006, there were 187 counts of 103 nights or less and 152 counts of 104 nights or more. Whichever approach is taken, the fact is that more often than not the total is below 104 nights.
43. I therefore conclude that the standard test under the first limb of the definition of day to day care applies in this case. The father does not meet the 104 nights threshold on that test. The tribunal was clearly troubled by how the outcome of the appeal might depend on a fortuitous lottery of dates. In my judgment, however, it is both appropriate and legally correct to have regard to what actually happened, and not to the terms of a later court order, which in any event related to a subsequent period.

The conflict between Commissioners' decisions R(CS) 4/03 and R(CS) 1/03

44. Had I decided that it was appropriate to select a "more representative" period other than 12 months, the next question that would have fallen for decision is when that alternative period should end. Does it end with the relevant week (here the week immediately preceding 19 July 2006)? As the submission by the Secretary of State's representative notes, there is a conflict of Commissioners' decisions on this point. Although the point is not decisive on the approach I have taken in this case, it may be helpful to express a view as that conflict of opinion has been highlighted and may need to be resolved in another case.
45. In R(CS) 4/03 (originally decided under reference CCS/128/2001), Mr Commissioner Turnbull considered the history of the statutory definition of "day to day care" and how it had been amended, concluding that the alternative period chosen under limb (b) of the definition need not end with the last day of the relevant week. The Commissioner set out his reasons for preferring this construction in some detail at paragraph 13 of his decision. In particular, the Commissioner noted that the words "but ending with the relevant week" had appeared in the original formulation of limb (b) but had later been removed as part of another amendment. Mr Commissioner Turnbull took the view that the most likely explanation for this was that there had been an intention to change the meaning of the provision. That appeal was decided on 20 December 2001.
46. On the other hand, in R(CS) 1/03, which was in fact decided after R(CS) 4/03 (on 8 July 2002), Mr Commissioner Jacobs arrived at the contrary conclusion. According to Mr Commissioner Jacobs, the position is as follows:

"18. The correct approach is this. Begin by identifying the relevant week. The Secretary of State has identified that as the week of 21 to 27 June 2000. Then, take an overview of the shared care for the period of 12 months ending with that week. If there are no significant changes within that period, day to day care can be determined by reference to the whole period. If there has been a change, the part of the period before the most recent change should be eliminated. Day to day care can then be determined by reference to the remainder of the period. Finally, any changes after the effective date and before the date of decision can be identified and taken into account under paragraph 15 of Schedule 1 to the Child Support Act 1991. The only problem that this analysis creates arises if there has been a change between the relevant week and the effective date. The period will be short, so the chances of this happening are small. But it can happen. An imaginative use of paragraph 15 can prevent any unrealistic assessments based on out of date arrangements for day to day care. I respectfully disagree with CCS/128/2001 – I am not persuaded that the amendment to the definition could have the effect suggested by the Commissioner."

47. The position used to be that where there were two conflicting decisions of equal status, and the earlier decision was fully considered in the later decision, then the later decision should be followed unless the judge was convinced that that later decision was wrong (R(IS) 13/01). Recently, however, in CIS/3101/2007, Mr Commissioner Jacobs has drawn attention to the Court of Appeal's decision in Re Taylor (a bankrupt) [2007] Ch 150, in which a different view has been adopted. The proper approach is to form a view on the merits of the arguments and the relevant authorities. Be that as it may, I prefer the construction advanced by Mr Commissioner Jacobs in R(CS) 1/03 in any event. My rationale is as follows.
48. In R(CS) 4/03 Mr Commissioner Turnbull gave four reasons for his conclusion. The first was that the repeal of the words "but ending with the relevant week" in limb (b) indicated an intention to change the meaning of the definition. This seems to me the strongest reason, but it is also at least arguable that these words were regarded as unnecessary surplusage.
49. The second reason relates to an argument in favour of Mr Commissioner Jacobs's reading, namely that on the view expressed in R(CS) 4/03 a decision maker or tribunal could not take a different 12 month period, but oddly only a shorter or longer period. Mr Commissioner Turnbull suggested that one could circumvent this outcome by choosing a period of slightly more or less than 12 months. This seems less than wholly convincing.
50. The third reason was that this construction would enable a more representative assessment to be made, especially where changes took place after the relevant week but before the effective date. However, this argument does not have regard to Mr Commissioner Jacobs's telling argument that the same end could be achieved by judicious use of the (more open-ended) staging provision in paragraph 15 of Schedule 1 to the Child Support Act 1991.
51. Mr Commissioner Turnbull's final reason was that the case for using the same end date was much weaker under limb (b) than limb (a), where the argument was based on convenience. However, again this approach does not have regard to the availability of paragraph 15 of Schedule 1 to the Child Support Act 1991.
52. So far as the assessment of shared care in old scheme cases is concerned, I therefore prefer the approach set out by Mr Commissioner Jacobs at paragraph 18 of R(CS) 1/03. In new scheme cases, of course, different provisions apply (see CCS/2885/2005).

A concluding comment

53. It is in the nature of bright line rules that a person may fall on "the wrong side of the line". However, the old scheme 104 nights test is a bright line rule with a degree of flexibility. Even with that flexibility, the father did not fall on the "right side of the line". Whether it is appropriate to have a child support system that seems positively to encourage tribunal appeals over whether a particular threshold of shared care has been met, and arguably also satellite litigation in the family courts over contact disputes, is a matter for policy makers and not Child Support Commissioners to resolve. The debates on shared care in the context of the reforms in the Child Maintenance and Other Payments Act 2008 show that there is no consensus on such issues.

Conclusion

54. For the reasons above I allow the mother's appeal. My decision is as set out above at paragraphs 1 to 3.

(signed on the original) N J Wikeley

Deputy Commissioner

13 July 2008