Judgments - Chief Adjudication Officer v. Stafford and Another

(back to preceding text)

    37. Paragraph (3B) of regulation 5 was inserted by regulation 19(b) of the Income-related Benefits Schemes (Miscellaneous Amendments) Regulations 1995 (SI 1995/516) as from 10 April 1995 in response to a decision by Social Security Commissioner Goodman in R(IS) 15/94, in which he held that the hours of the claimant's wife, a school receptionist who worked an average of about 24 hours in term time but did no work during school holidays, should be averaged under regulation 5(2)(b)(i) over the whole of her cycle of work of one year. The result was that the average of her weekly working hours was reduced to less that 16, and the claimant was found entitled to income support throughout the whole year. Mr Pannick for the respondent explained that the policy purpose of the insertion of paragraph (3B) was to enable many claimants who had children to claim family credit. He said that it was thought to be right that eligibility to that form of benefit should be calculated by excluding periods of school holidays and similar vacations from the period which was to be taken into account for the purpose of working out the average.

    38. The relevant provisions of the Jobseeker's Allowance Regulations 1996 are to be found in regulation 51 which is headed "Remunerative work" and in regulation 52 which is headed "Persons treated as engaged in remunerative work." They provide as follows:

    "51(1) For the purposes of the Act 'remunerative work' means -

    (a) in the case of the claimant, work in which he is engaged or, where his hours of work fluctuate, is engaged on average, for not less than 16 hours per week; and

    (b) in the case of any partner of the claimant, work in which he is engaged or, where his hours of work fluctuate, is engaged on average, for not less than 24 hours per week;

    and for those purposes, work is work for which payment is made or which is done in expectation of payment.

    (2) For the purposes of paragraph (1), the number of hours in which the claimant or his partner is engaged in work shall be determined-

    (a) where no recognisable cycle has been established in respect of a person's work, by reference to the number of hours or, where those hours are likely to fluctuate, the average of the hours, which he is expected to work in a week;

    (b) where the number of hours for which he is engaged fluctuate, by reference to the average of hours worked over -

    (i) if there is a recognisable cycle of work, and sub-paragraph (c) does not apply, the period of one complete cycle (including, where the cycle involves periods in which the person does not work, those periods but disregarding any other absences);

    (c) where the person works at a school or other educational establishment or at some other place of employment and the cycle of work consists of one year but with school holidays or similar vacations during which he does no work, by disregarding those periods and any other periods during which he is not required to work . . .

    52(1) Except in the case of a person on maternity leave or absent from work through illness, a person shall be treated as engaged in remunerative work during any period for which he is absent from work referred to in regulation 51(1) (remunerative work) where the absence is either without good cause or by reason of a recognised, customary or other holiday . . . "

    39. Regulation 53 of the 1996 Regulations is headed "Persons treated as not engaged in remunerative work." It is to be noted that the provision relating to the treatment of a person working at a school or other educational establishment or at some other place of work whose cycle of work consists of one year but with school holidays or similar vacations during which he does not work was included in regulation 51 (remunerative work) and not in regulation 53 (persons treated as not engaged in remunerative work). Nor was it included in regulation 52 (persons treated as engaged in remunerative work).

    40. Mr Drabble for the appellant laid great stress on the fact that each Act contained distinct and separate empowering provisions for the making of regulations as to the meaning of the expression "remunerative work" and the circumstances in which a person is or is not to be treated as "engaged in" remunerative work. He said that the appellant was only to be treated as engaged in remunerative work during weeks when he was not in fact working and for which he was not paid if the exercise of some deeming provision led to that result. He maintained that the respondent's argument that the appellant was to be treated as working throughout the whole cycle of work of which the school holidays formed part was based on a misunderstanding of the regulations, for two reasons. In the first place, the regulations which provided for the treatment of school holidays for the purpose of the average were an exercise of the power to define remunerative work, not an exercise of the power to deem persons to be engaged in remunerative work when they were not working. They did not affect the position of a person who was not engaged in remunerative work at all during the holidays. In the second place, if the regulations were to be read as an exercise of the power to deem as well as the power to define, the extent of the deeming should be confined to those weeks which were brought into account for the purpose of averaging and should leave out of account those weeks which were directed to be left out of account for the purpose of the averaging exercise.

    41. The argument that the effect of the regulations ought to be determined by reference to the enabling provisions which are to found in the statutes has more force in the case of the Jobseeker's Allowance Regulations 1996 than it does in the case of the Income Support (General) Regulations 1987. This is because the layout of the 1996 Regulations indicates that the draftsman of those Regulations was making a conscious attempt to deal separately with the definition of remunerative work on the one hand (in regulation 51) and with persons treated as engaged in, and not engaged in, remunerative work (in regulations 52 and 53) on the other hand. The draftsmen of the 1987 Regulations and of the various amendments which were made to it were not so scrupulous. There is no regulation which is headed simply "Remunerative work". Regulation 5, in which the critical provision about the treatment of holidays and other periods during a cycle of one year during which a person in not required to work is to be found, is headed "Persons treated as engaged in remunerative work". But it contains within the same regulation a variety of provisions made in the exercise of each of the relevant enabling powers. Paragraph (1) contains a definition of remunerative work. Paragraph (3) provides that a person who is absent from work without good cause or by reason of a recognised, customary or other holiday is to be treated as engaged in remunerative work during any such period. Paragraph (3A) provides that a person shall not be treated as engaged in remunerative work on any day when on maternity leave or absent from work when ill. There is a separate list of persons who are not to be treated as engaged in remunerative work in regulation 6.

    42. On further examination the precise separation for which Mr Drabble contended is not maintained, even in the case of the 1996 Regulations. Where the definition of remunerative work in regulation 51 requires the working out of a number of hours of work over by reference to an average of hours worked an element of deeming is inherent in the definition of "work". The effect of the averaging exercise is to require it to be assumed that the person is working for the average numbers of hours for each week in the cycle, even although for some of those weeks he may be working for far fewer than the 16 hours which disqualifies a person from benefit. For example a person whose average is, say, 25 hours per week but who works for only one hour in a given week in the cycle is deemed, as a result of the averaging exercise, to be engaged that week in remunerative work for not less than 16 hours.

    43. I am content to accept, as a general proposition, that guidance as to the meaning of a provision in a statutory instrument may be obtained from the wording of the enabling power under which it was made. But the reliability of that guidance is diminished if one finds, as one does in this case, that a variety of enabling powers were available to the draftsman of the statutory instrument and there were no compelling reasons for maintaining a strict separation between the exercise of each of them. The fact that it is accepted in this case that both sets of Regulations have the same effect makes Mr Drabble's argument even harder to accept. Mr Pannick was entitled to ask that it be tested on the weaker alternative. When it is applied to the 1987 Regulations it is not sustainable.

    44. I see no alternative therefore in this case to the normal approach to the construction of a statutory instrument, which involves examining the relevant provisions in the context of the regulations as a whole and determining their effect according to the ordinary meaning of the words used. For convenience I shall conduct my examination under reference to the Income Support ("IS") Regulations, as this was the regime which was enacted first. I shall give references to the corresponding Jobseeker's Allowance ("JSA") Regulations where this is appropriate.

    45. The appropriate starting point is IS regulation 5(1) [JSA regulation 51(1)]. This provision sets out the requirements which must be met with regard to each of the two words in the phrase "remunerative work". First there is the work element. This is to be measured by reference to the number of hours of work a week in which a person is engaged. Then there is the remuneration element. This is to be determined by reference to the question whether the work in question is work for which payment is made or which is done in expectation of payment. The question of remuneration has not given rise to any difficulty in this case.

    46. The choice of method which is to be used to determine the number of hours of work a week in which a person is engaged depends upon whether or not there is a recognisable cycle of work. The Social Security Tribunal, having examined the appellant's contract of employment, found as a fact that the number of hours for which he was engaged fluctuated. This was because in the year 1995/96 he worked for 20 hours a week for 38 weeks and he did no work at all for 12 weeks. But the tribunal also held that he had a recognisable yearly cycle of work. It follows that IS regulation 5(2)(b)(i) [JSA regulation 51(2)(b)(i)] applies to this case. The calculation which it lays down is based on the average number of hours for the complete cycle.

    47. The regulations do not explain what they mean by the word "cycle" in this context. But I think that the concept of a cycle of work in which the number of hours for which the person is engaged fluctuate necessarily involves the idea that there will be periods in the cycle when the person works for more hours per week than the average, and that there will be other periods in the same cycle when he works for fewer hours per week than the average or, it may be, does not work at all. The fact that the cycle may include periods when the person does not work at all is made clear by the words which appear in parenthesis in paragraph (2)(b)(i) [JSA regulation 51(2)(b)(i)] which require there to be included in the period of one complete cycle, "where the cycle involves periods in which the person does not work," those periods but disregarding any other absences. I do think that it can be doubted therefore that the reason why the periods of school holidays when the person does not work form part of his cycle of work for the purpose of calculating the number of hours for which he is engaged in work is that he is regarded for the purpose of the regulations as engaged in work during those periods even although he is not in fact working. It is on this vital point that I respectfully disagree with my noble and learned friend, Lord Scott of Foscote. To use a colloquial expression, the person is regarded as being "in work" during the periods of the cycle when he is not working, although he does no work during those periods.

    48. This brings me to paragraph (3B) of IS regulation 5 [JSA regulation 51(2)(c)] to which the opening words of IS regulation 5(2) say that paragraph (2) is subject. It provides that for the purposes of paragraph (2)(b)(i) of IS regulation 5 (calculating the average of hours worked if there is a recognisable cycle of work) a person's recognisable cycle of work at a school, other educational establishment or other place of employment is one year - as it was held to be in this case - and where the cycle includes periods of school holidays or similar vacations when he does not work - as it did in this case also - those periods "shall be disregarded in establishing the average hours for which he is engaged in work." The wording of JSA paragraphs 51(2)(b)(i) and 51(2)(c) is different, but I agree with my noble and learned friend, Lord Millett, for the reasons which he has given that the result is the same.

    49. Two points in particular are to be noted about the wording of the last few words of IS paragraph 5(3B) which I have placed between quotation marks. The first is the extent of the disregard which is directed by the paragraph. It extends only to the calculation of the average hours for which the person is engaged in work during the cycle. The period of the cycle of work is not to be treated as having been altered. What is altered is the periods within the cycle which are to be taken into account for the purpose of the averaging exercise. The second point is the use of the phrase " for which he is engaged in work". This reflects the wording of IS paragraph 5(1) [JSA paragraph 51(1)(a)] which states that "remunerative work is work in which a person is engaged" for not less than 16 hours a week. It maintains the distinction which is drawn throughout the Regulation between the hours for which a person is engaged in work and the hours for which he is actually working.

    50. For these reasons I am unable to accept Mr Drabble's argument that IS regulation 5 [JSA regulation 51] is designed only to establish how many hours a week a person is to be treated as working in any week in which he does some work and that it does not affect the position of a person who in a given week does no work at all. In my opinion this argument overlooks the fact that, although a recognisable cycle of work may include periods when the person does no work at all, he is nevertheless treated as being "engaged in work" throughout his cycle of work. Every week is part of the cycle irrespective of whether the person does or does not work during that week. If the intention had been to provide that such a person was to be entitled to benefit during periods of school holidays or similar vacations, I would have expected the regulation to provide - as it does in the case of a person who is on maternity leave or absent from work because he is ill - that the person shall not be treated as engaged in remunerative work during those periods: see IS regulation 5(3A) [JSA regulation 52(1)]. The wording of IS regulation 5(3B) [JSA regulation 51(2)(c)] appears to me to have deliberately chosen to make it clear that it was not intended that the treatment of school holidays or similar vacations for the purpose of determining the number of hours in which the person is engaged in work should have the effect that the person was to be treated as not engaged in work during those periods simply because he was not required to work during those periods.

    51. I am also unable to accept Mr Drabble's alternative argument that the extent of deeming that a person is engaged in remunerative work during periods when he does not work should be confined to the weeks which are brought into account for the purposes of the averaging exercise, and that it should not extend to those weeks which are directed to be left out of account when the average is being calculated. According to this argument, periods which are disregarded in averaging should be disregarded also in regard to the question whether the person is to be treated as engaged in remunerative work during those periods. In my opinion the concluding words of IS regulation 5(3A) [JSA regulation 51(2)(c)] make it clear that the disregard which is directed by that paragraph is for the sole purpose of the averaging exercise. A direction is needed that these periods must be disregarded for this purpose because they are periods which fall within the person's recognisable cycle of work. In the absence of the direction they would, as Commissioner Goodman held in R(IS) 15/94, fall to be taken into account together with all the other periods in the cycle for the purpose of working out the average. But I do not find any words here to indicate that the direction was intended to affect any finding as to the cycle of work to which the result was to be applied after completing the averaging exercise.

    52. It seems to me that Auld LJ stated the position correctly when he said [2000] 1 All ER 686, 693 that IS regulations 5(2)(b)(i) and (3B) [JSA regulations 51(2)(b)(i) and (c)] clearly contemplate that a recognisable yearly cycle of work may include periods in which an employee does no work, and that the averaging for which they provide is part of the means of determining whether he or she is engaged in remunerative work for the purpose of IS regulation 5(1) [JSA regulation 51(1)]. As he said, it is difficult to see what point there would be in IS regulation 5(3B) [JSA regulation 51(2)(c)] if IS regulation 5(2)(b)(i) [JSA regulation 51(2)(b)(i)] applied, contrary to its express terms, only to periods of actual work. IS regulation 5(2)(b)(i) states in terms that a person's cycle of work may involve periods when the person does not work. That was why, before the insertion of IS regulation 5(3B) [JSA regulation 51(2)(c)], the averaging exercise was done in all cases by reference to the whole period of the cycle including periods when the person was not required to work during school holidays and similar vacation periods. The effect of the insertion is simply to take these periods out of account for the purpose of working out the average.

    53. As Social Security Commissioner May observed in CSJSA/395/98, paragraph 17, if as a consequence of the calculation directed by IS regulation 5(3B) [JSA regulation 51(2)(c)] the claimant's hours are sufficient to place the claimant within the definition of remunerative work in IS regulation 5(1) [JSA regulation 51(1)], the application of that definition covers the whole period of the cycle which has been identified as the cycle of work in the claimant's contract. I would hold that that is the result of the averaging exercise which has had to be carried out in this case.

    54. I would dismiss the appeal.

LORD MILLETT

My Lords,

    55. I have had the advantage of reading in draft the speeches of my noble and learned friends, Lord Hope of Craighead and Lord Cooke of Thorndon. I am in full agreement with them that, were it not for paragraph (3B) of regulation 5 of the Income Support (General) Regulations 1987 (corresponding to paragraph (2)(c) of regulation 51 of the Jobseeker's Allowance Regulations 1996), a school ancillary worker who does no work during school holidays would nevertheless be treated as being engaged in work throughout the year. The question which has divided their Lordships is whether paragraph (3B) of regulation 5 of the former Regulations and the corresponding provision of the latter alter this. I agree with Lord Hope that they do not. My reasons are as follows.

(1)

The 1987 Regulations

    56. Section 124(1) of the Social Security Contributions and Benefits Act 1992 makes benefit conditional on the claimant (inter alia) not being "engaged in remunerative work". This involves two concepts: (i) "being engaged" in work and (ii) "remunerative work". Where necessary the language of the Regulations distinguishes between the two concepts.

    57. As its heading indicates, regulation 5 of the 1987 Regulations is concerned with prescribing the circumstances in which a person is to be treated as being "engaged in remunerative work". This in turn involves defining the expression "remunerative work".

    58. Where a person's hours of work fluctuate, regulation 5(1) defines "remunerative work" as work "in which he is engaged" on average for not less than 16 hours a week. Regulation 5(2)(b)(i) provides that, where there is a recognisable cycle of work, the number of hours for which a person "is engaged in work" is to be determined by reference to the period of one complete cycle including periods in which the person does no work. This, as Lord Cooke observes, is for the purpose of averaging, which in turn is for the purpose of determining whether the claimant is engaged in remunerative work. For these purposes, a person is treated as being engaged in work even during periods when he does no work. The contrast, as Oliver LJ observed in R v Ebbw Vale and Merthyr Tydfil Supplementary Benefits Appeal Tribunal, Ex p Lewis [1982] 1 WLR 420, is between being "in work" and being "at work" (where however there was language which indicated that a person was not to be treated as being "in work" unless he was actually "at work").

    59. The tribunal found that the work in which the claimant was engaged had a recognisable cycle of one year. Under regulation 5 (2)(b)(i), therefore, he must be treated as being engaged in work throughout the year, including the school holidays when he does no work. Whether the work in which he is to be treated as engaged is remunerative work depends upon the average number of hours per week spread over the whole cycle.

    60. This is not, with respect to my noble and learned friend, Lord Scott of Foscote, to imply a statutory fiction into the Regulations. It merely gives effect to the distinction which is drawn in the Regulations between being engaged in work and doing work. It recognises that, if the number of hours for which a person is "engaged in work" includes periods during which he does no work, it follows that a person can be "engaged in work" even if he is not working.

    61. There is nothing particularly surprising in this, for were it otherwise the Regulations would be open to manipulation. Suppose an employer wished to put his workforce on short time during a recession. He might ask them to work 20 hours a week instead of the usual 40 hours. The unions would ask him to introduce instead a cycle of 80 hours every four weeks, leaving it to the men to decide when to come to work. If they worked 20 hours a week, they would not be entitled to benefit. If they worked 40 hours every other week, they would prima facie be entitled to benefit during the weeks when they did not work. Paragraph (2)(b)(i) prevents this. It does not merely spread the hours when the worker is engaged in work over the four weeks; it also treats him as being engaged in work during the two weeks when he is not working.

    62. Thus far I am in respectful agreement with Lord Cooke. Paragraph (2) is, however, expressly made subject to paragraph (3B), which applies where a person's cycle of work at a school, other educational establishment or other place of employment is one year and includes periods of school holidays during which he does not work. It provides that, for the purpose of regulation 5(2)(b)(i), those periods are to be disregarded in establishing the average hours for which he is engaged in work. Lord Cooke interprets this to mean that, contrary to the general rule for cyclical workers, he is not to be treated as being engaged in remunerative work during the school holidays.

    63. It is at this point that I respectfully part company with him. To my mind the critical point is that paragraph (2)(b)(i) is not replaced by paragraph (3B) but merely made subject to it. Paragraph 2(b) remains in full force save only to the extent to which it is modified by paragraph (3B). The two paragraphs must be read together. They are both concerned with persons who have a regular cycle of work. Neither of them is concerned with the question when a person is to be treated as being engaged in work, but only with the determination of the average number of hours for which he is engaged in work for the purpose of determining whether the work (in which ex hypothesi he must be treated as being engaged) is remunerative work. Paragraph 2(b)(i) directs that the average number of hours is to be determined by reference to a complete cycle including periods during which he does no work, and there is nothing in paragraph (3B) which modifies this. It does not affect the duration of the cycle or exclude periods during which he does no work. In the case of the present claimant, therefore, the cycle of work remains the complete year.

    64. This is confirmed by the opening words of paragraph (3B), which applies only where the cycle of work is one year and includes periods of school holidays during which he does no work. Reading paragraph 2(b)(i) as modified by paragraph (3B) and omitting immaterial words produces the following:

    "Where a person's recognisable cycle of work at a school … is one year and includes periods of school holidays … during which he does not work, the number of hours for which he is engaged in work shall be determined by reference to the average of hours worked over one complete cycle including periods during which he does not work (ie one year) but so that the school holidays shall be disregarded in establishing the average hours for which he is engaged in work."

    65. In my opinion the cycle remains constant at one complete year, and the claimant continues to be treated (by paragraph (2)(b)(i)) as engaged in work throughout the cycle including periods when he does no work. Paragraph (3B) directs that the school holidays are to be disregarded, not for the purpose of determining when a person is to be treated as engaged in work, but merely for the purpose of establishing the average number of hours worked during the cycle.

    66. This construction is confirmed by the entirely different approach taken by paragraphs (3) and (3A), which unlike paragraph (3B) are concerned with the question when a person is to be treated as being engaged in remunerative work. Paragraph (3A), for example, provides that a person shall not be treated as engaged in remunerative work on any day when the person concerned is on maternity leave or is absent from work because he is ill. This displaces paragraph (2) altogether. A claimant is not to be treated as engaged in remunerative work while on maternity leave or absent from work because of illness. But paragraph (3B) does not provide that a claimant to whom it applies is not to be treated as engaged in remunerative work during the school holidays. It does not supplant paragraph (2)(b)(i) which directs that he is to be so treated. All it does is direct that the school holidays are to be left out of the calculations which establish the average number of hours worked during the year.

    (2) The 1996 Regulations.

    67. Although the drafting of regulation 51 is different, and at first sight critically different, the result is in my opinion the same. The difference is that paragraph (2)(b)(i) (which provides for averaging over the whole cycle including periods when the claimant does not work) is not merely subjected to paragraph (2)(c) (which deals with school holidays) but applies only where paragraph 2(c) does not apply. At first sight, therefore, the two paragraphs are alternatives. But the drafting and the layout of the regulation are inelegant, to say the least. Paragraph (2)(c) cannot stand as an independent paragraph; it ought to have been included as (2)(b)(iii) and expanded since the sense requires the opening words of paragraph (2)(b) and part at least of paragraph (2)(b)(i) to be read into it. Omitting immaterial words, and with as little as possible read in, the regulation reads as follows:

    "Where a claimant works at a school and the cycle of work consists of one year but with school holidays during which he does no work, the number of hours in which he is engaged in work shall be determined by reference to the average hours worked over the period of one complete cycle but disregarding the periods during which he does no work."

 
continue previous