Mrs J Bampton v Chief Constable of Nottinghamshire Police: 2601538/2016
EMPLOYMENT TRIBUNALS
Case No 2601538/2016
Between
Mrs J BamptonClaimantChief Constable of Nottinghamshire PoliceRespondent
Before
Employment Judge MilgateDate 10 July 2018
JUDGMENT
[1]When calculating the Claimant’s pension entitlement under the Police Pension Regulations 1987 the correct comparison is between the Claimant and a full-time police officer with the same actual length of service as the Claimant (PC Y) and it is appropriate to apply the pro rata principle.[2]A reference to the CJEU is unnecessary.
REASONS
[1]The Claimant, a former Police Constable in the Nottinghamshire Constabulary, brings claims arising out the decision to award her an ill-health pension of £11,840.83 per annum (subject to commutation and taxation) under the Police Pension Regulations 1987 SI 1987/257 (‘the 1987 Regulations’). Page 1 of 37[2]It is agreed that the Claimant worked on a part-time basis at various stages of her career with the Nottinghamshire Force. It is also agreed that for the majority of her service (both full and part-time) she was a member of the 1987 Police Pension Scheme (‘the 1987 Scheme’), which is operated in accordance with the 1987 Regulations. The scheme is what is commonly known as a ‘final salary’ or ‘defined benefits’ scheme whereby, broadly speaking, benefits are determined by reference to pensionable pay and reckonable service at the date of calculation.[3]The Claimant retired from her office in 2016 on the grounds of permanent disability. Neither her appointment history nor the circumstances of her ill-health retirement are material to the Claimant’s case, which relates solely to the application of the 1987 Regulations to the calculation of her pension.[4]It is common ground that the calculation of the Claimant’s pension has been carried out in accordance with the provisions of the 1987 Regulations. However the Claimant contends that the effect of the calculation is to discriminate against her because of her part-time service. She also contends that the calculation is indirectly discriminatory on grounds of sex, on the basis that female police officers are more likely than male officers to work on a part-time basis.[5]Accordingly she argues that the pension award made to her under the 1987 Regulations is:-(i) in breach of Article 5 of the Part Time Worker Regulations 2000; and/or(ii) in breach of the equality rule in Section 67 of the Equality Act 2010; and/or(iii) incompatible with Article 157 of the Treaty on the Functioning of the European Union (‘TFEU’), which she argues has direct effect and so provides the basis, if necessary, for a separate cause of action. (It should be added that the Respondent does not accept that Article 157 gives her a free-standing cause of action, but that is not an issue for determination at this hearing.)[6]The Claimant’s complaints are brought against the Chief Constable of Nottingham Police who resists the claim in its entirety. Originally there was a second Respondent – Kier Pension Unit – but all claims against that organisation were dismissed at an early stage. However the Secretary of State for the Home Department (‘the Secretary of State’) has since been joined as an interested party pursuant to Rule 35 of the Employment Tribunals (Constitution and Rules of Procedure) Regulations 2013. This is on the basis that the outcome of this case may have implications for every police force in England and Wales to which the 1987 Regulations apply. As a result the Home Office, as the government department with responsibility for the police, clearly has a legitimate interest in these proceedings. In this judgment all references to ‘the parties’ should be taken to include the Secretary of State unless otherwise stated. B. The issue for determination at this hearing[7]An agreed list of preliminary issues in this case was prepared by the parties at an early stage. Matters in dispute include (but are not limited to):-(i) whether the Claimant can rely on statistical evidence of disparate impact, rather than identifying a male comparator for the purposes of the equal pay claim and if not, whether her named comparators are employed on equal work; and(ii) whether the Claimant can rely on herself as a comparator for the part-time workers’ Page 2 of 37 claim and if not, whether the other comparators she has put forward are appropriate.[8]However those issues are not before me today. Instead, at a Preliminary Hearing on 8 September 2016, Employment Judge Heap agreed with the parties that a rather different comparator issue should be determined at the outset of this litigation, namely whether, when calculating the Claimant’s pension entitlement under the Regulations, the correct comparison is between:-(i) the Claimant and a full time Police Officer with the same reckonable pensionable service as the Claimant (PC X); or(ii) the Claimant and a full time Police Officer with the same actual length of service as the Claimant (PC Y).[9]This issue is quite separate from the question whether the male comparators the Claimant has named in her claim form are appropriate for the purpose of showing she was engaged on equal work or for showing she suffered less favourable treatment as a parttime worker. Instead the question I have to determine involves consideration of whether it is appropriate to pro rate pension benefits for part-time workers under the 1987 Regulations, and if so whether that exercise has been performed in accordance with legal principle.[10]Currently the 1987 Regulations require a pro-rating exercise to take place. This is done by reference to PC Y, who has worked as a police officer for the same length of time as the Claimant but has always done so on a full-time basis (with the inevitable result that PC Y will have worked more hours than the Claimant). The Respondent and the Secretary of State say that this is the correct approach and that the use of PC Y as a comparator accords with both legal principle and common sense. By contrast the Claimant argues that the 1987 Regulations are discriminatory and that the correct comparison is with PC X who has worked exactly the same number of hours as she has (albeit within a shorter time period because PC X has always worked full time). At the hearing before Judge Heap the parties were in agreement that a decision on this issue would, in all likelihood, narrow the issues in the case.[11]Accordingly this hearing was listed to determine whether PC X or PC Y was the correct comparator. By a letter to the Tribunal of 24 October 2016, the Claimant’s Solicitors pointed out, for the avoidance of doubt, that the issue I have to decide is relevant not only to the Part-time Workers complaint, but also to the complaints under the Equality Act and Article 157. The hearing proceeded on the basis that that was indeed the case. C. Procedure at the hearing[12]At the hearing, which took place over two days, I heard no evidence but had the benefit of oral submissions from all three parties. In addition I had before me the parties’ skeleton arguments and a bundle containing relevant legislation and authorities. I asked for the skeleton arguments to be amended after the hearing as the position taken by the Claimant changed somewhat during the course of oral argument with the effect that new ground was covered. Those supplemental skeletons were produced in timely fashion and I am grateful to the parties’ representatives for the care with which they have been prepared.[13]During the hearing I was handed a number of graphs by both the Respondent and the Secretary of State to illustrate their arguments. Some of these (which I have labelled Graphs A to C) are annexed to this judgment. Page 3 of 37 D. Agreed facts[14]The Claimant was appointed to the office of Constable on 9 December 2002. She retired her office with effect from 8 March 2016 on the grounds of ill health.[15]Between her appointment on 9 December 2002 and 31 March 2015 the Claimant was a member of the 1987 Police Pension Scheme (‘the 1987 Scheme’) which is governed by the 1987 Regulations.[16]It is agreed that the Claimant is entitled to an ill-health pension under the 1987 Regulations. (For completeness it should be added that the Claimant became a member of a new police pension scheme on 1 April 2015 and remained so until her retirement in 2016. However her pension entitlement under the new scheme is not relevant to the present claim.)[17]The Claimant’s actual length of service during which she was a member of the 1987 Scheme was 12 years and 113 days (ie the period between her appointment on 9 December 2002 to 31 March 2015, when the 1987 Scheme closed).[18]For the majority of the period in which the Claimant was a member of the 1987 Scheme she worked full time. However there were also a number of periods of part time work. As a result the Claimant’s reckonable pensionable service for the purposes of the 1987 Regulations is calculated as 11 years and 247 days. (More is said about the calculation of reckonable pensionable service below. However at this point I would simply emphasise that the Claimant’s reckonable pensionable service is clearly shorter than her actual length of service because of her periods of part-time working.)[19]The first period when the Claimant worked part-time commenced on 1 April 2009. Her final period of part time service for the purposes of the 1987 scheme was between 4 January 2015 and 31 March 2015. During that final period the Claimant worked for 25 hours per week. At the same time a full-time police constable was required to work for 40 hours per week.[20]The Claimant’s pensionable pay for the purposes of the pension calculation under the 1987 Regulations was £37,446.10 per annum at the time of her retirement. (I interject here to emphasise that this was the full time equivalent salary and not her reduced parttime salary.) Accordingly, because latterly she worked part-time, her actual aggregate salary over the period of 12 months preceding retirement was somewhat less at £23,403.81 per annum.[21]The Claimant’s pension contributions were based on her actual salary and not on her pensionable salary, so that she paid less when she earned less.[22]Had the Claimant worked solely full time during her period of actual service (ie for a period of 12 years and 113 days) then her annual pension under the 1987 Scheme would have been £12,482.03. However if the Claimant had worked solely full time for the length of her reckonable pensionable service (i.e. for a period of 11 years and 247 days) her annual pension would also have been £12,482.03. As I explain in more detail below1, this anomalous result occurs because of the design of the 1987 Regulations, whereby pension accrual for those who have only ever worked full-time remains static during the period of ten to thirteen years’ pensionable service. 1 See paras [40] and [41] below. Page 4 of 37[23]In fact, as noted above, the Claimant’s annual pension under the 1987 Scheme is £11,840.83. It is agreed that this figure has been calculated accurately in accordance with the 1987 Regulations. E. Calculation of pension under the 1987 Regulations[24]In order to understand the issue for determination at this hearing it is necessary to consider how pension benefits under the 1987 Regulations are calculated. I shall first give an overview of the relevant provisions. Secondly I will consider how pensionable service is calculated. Finally I will set out some detailed calculations for both full-time and part-time officers to illustrate how the 1987 Regulations operate. (i) Summary of the relevant provisions[25]Under Reg B3 of the 1987 Regulations, where a police officer retires on the ground of permanent disability, he or she will be entitled to an ill-health award. Provided certain conditions are satisfied that award can take the form of an ill-health pension. There is no dispute that the Claimant satisfied those conditions and so was entitled to an ill-health pension.[26]The calculation of the ill-health pension is governed by Part III of Schedule B of the 1987 Regulations. Crucially for the purposes of this case, Part III draws a distinction between those who have only ever worked full-time and those who have had one or more periods of part-time service.[27]Under paragraph 1(a) of Part III, where all of the police constable’s reckonable pensionable service has been full-time, the calculation of pension entitlement is subject to 3 distinct sets of rules:-(a) up to 5 years’ pensionable service - paragraph 2 applies;(b) more than 5 years but no more than 10 years - paragraph 3 applies;(c) more than 10 years - paragraph 4 applies.[28]By contrast, under paragraph 1(b) of Part III, where some or all of reckonable pensionable service has been part time, there is just one calculation method which is set out in paragraph 4A2.[29]Like paragraphs 2, 3 and 4, paragraph 4A refers to ‘pensionable service’ and it is therefore necessary to understand how this concept is calculated. (ii) Calculating reckonable pensionable service[30]As explained above, the Claimant’s actual length of service was 12 years and 113 days. It is agreed that had the Claimant worked full time throughout this period, then under the 1987 Regulations her ill health retirement pension would have been calculated on the basis that her reckonable pensionable service was also 12 years and 113 days3. 2 It should be added that paragraph 4A only applies ‘in a case where, if the part-time service had been full time service, pensionable service would not exceed 30 years: see Sch B Part III para 1(b)(ii). However the Claimant clearly satisfies that condition. 3 Under Regulation A9(1) of the 1987 Regulations pensionable service is calculated in complete years plus fractions of years. Page 5 of 37[31]However special rules apply where a police constable has, like the Claimant, spent one or more periods in part-time service. These rules are found in Part IV of Schedule J to the 1987 Regulations, at paragraph 8. Their effect is to take each period of part-time service and then multiply that period by ‘the appropriate factor’, to enable service to be calculated relative to an officer who has worked full time throughout the same period.[32]Under paragraph 8 the appropriate factor is A/B, where A is the total number of hours worked by the officer with part time service during the period of part-time working and B is the number of weeks in that period multiplied by 40 (the figure of 40 is used as that is the number of hours in a full-time working week): see Schedule A to the 1987 Regulations, referring to provisions in the Police Regulations 1987 (SI 1987/851). The effect, put simply, is to aggregate the number of hours of part-time work in the period and express those hours as a proportion of full-time working.[33]So, to use a straightforward example, if the period of part-time service is twelve weeks and the part-time officer has worked 20 hours per week during that period (ie half the full-time hours) then the calculation of reckonable service attributable to that period is:- 12 weeks x A = 12 weeks x 20 x 12 = 6 weeks B 40 x 12 Accordingly in this example the reckonable pensionable service attributable to this period of part-time working is six weeks.[34]The net result is that, in the Claimant’s case, although she started work on 9 December 2002 and remained a member of the 1987 scheme until 31 March 2015 (a period of 12 years and 113 days) her reckonable pensionable service, applying this formula, is 11 years and 247 days. (iii) Detailed calculations under Part III of Schedule B Full timers: paragraphs 2-4[35]I explained above that the calculation of an ill-health pension under the 1987 Regulations is governed by Part III of Schedule B, which draws a distinction between full time and part-time service. I also explained that where all service has been full-time different rules apply, depending on the length of reckonable pensionable service. These rules are set out in paragraphs 2-4. It is immediately clear when these rules are examined that under the 1987 Regulations pension does not accrue at a uniform rate throughout a full time police officer’s career.[36]Under paragraph 2:- ‘Where the policeman has less than 5 years’ pensionable service, the amount of the pension shall be not less than a sixtieth of his average pensionable pay4 and… shall be of an amount equal to a sixtieth of that pay multiplied by the period in years of his pensionable service’. So for example, where a police officer has 4 years’ pensionable service then under paragraph 2 his or her pension entitlement is to 4 sixtieths of average pensionable pay. 4 For these purposes average pensionable pay can be taken to be the policeman’s full-time salary on retirement. Page 6 of 37[37]However under paragraph 3 the calculation is somewhat different:- ‘Where the policeman has 5 or more years, but not more than 10 years’ pensionable service…the pension shall be of an amount equal to 2 sixtieths of his average pensionable pay multiplied by the period in years of his pensionable service’. As a result, a full-time police officer with 7 years’ pensionable service will have a pension entitlement of 14 sixtieths of his average pensionable pay and a full-time officer with 10 years’ pensionable service will have a pension entitlement of 20 sixtieths of his average pensionable pay.[38]Under paragraph 4, the formula changes again, becoming rather more complex:- ‘Where the Policeman has more than 10 years’ pensionable service, the pension shall be not less than 20 sixtieths, nor more than 40 sixtieths, of his average pensionable pay and, subject as aforesaid and to Paragraph 5, shall be equal to 7 sixtieths of that pay with the addition –(a) of an amount equal to a sixtieth of that pay multiplied by the period in years of his pensionable service up to 20 years, and(b) of an amount equal to 2 sixtieths of that pay multiplied by the period in years by which his pensionable service exceeds 20 years.’[39]Paragraph 4 therefore provides, so far as relevant to this case, that pension entitlement calculated in accordance with its provisions shall be no less than 20 sixtieths of average pensionable pay. Subject to that minimum entitlement, the calculation is 7 sixtieths plus 1 sixtieth per year of pensionable service between 10 and 20 years.[40]As noted above, the effect of these paragraphs is that the rate of pension accrual is not uniform throughout a full-time police officer’s service. Instead the calculation differs for different blocks of service. Where the officer has less than five years’ service he or she will receive 1 sixtieth of pensionable pay per year of service. However between 5 and 10 years’ service the entitlement is more generous, amounting to 2 sixtieths per year of service so that, by the time the police officer reaches ten years pensionable service his pension entitlement is to 20 sixtieths of average pensionable pay. However at this point pension entitlement reaches a plateau. This is the result of paragraph 4, which introduces a new calculation, the effect of which would be to reduce entitlement in years 10 to 13. To avoid that result paragraph 4 provides for pension entitlement to be frozen at 20 sixtieths of pensionable pay between 10 and 13 years’ service to ensure police officers do not lose what they have already accrued (ie so that there is no drop in entitlement after 10 years’ service). The effect is that those who work full-time for 10 years and 1 day will get 20 sixtieths of pensionable pay as will those working full-time for 12 years and 364 days. Only when reckonable service exceeds 13 years does the rate of accrual begin to rise again, because at that stage the basic calculation set out in paragraph 4 takes the officer over the 20 sixtieths minimum.[41]The following examples illustrate how this works in practice. It is assumed in each case that average pensionable pay is £37,446.10 per annum (which is the full time equivalent salary for the Claimant). Example 1 Officer A has worked only full-time and has completed exactly 10 years’ Page 7 of 37 pensionable service on retirement. In this situation paragraph 3 applies and his pension calculation is £37,446.10 (pensionable pay) multiplied by 2/60 for each complete year of service. The multiplier is therefore:- 10 years x 2 sixtieths for each year = 20/60 The calculation is accordingly £37,446.10 x 20/60 = £12,482.03. Example 2 Officer B has worked only full-time and has completed 12 years and 364 days’ pensionable service on retirement. In this situation paragraph 4 applies, which means that the calculation is £37,446.10 (pensionable pay) multiplied by 7/60 plus 1/60 for each complete year of service, subject to a minimum of 20/60. In this case the minimum is applied, because without it the multiplier would only be 7/60 + 10/60 =17/60. Officer B therefore receives £37,446.10 x 20/60 = £12,482.03. This is exactly the same pension as Officer A, despite the disparity in length of service. Example 3 Officer C has worked only full-time and has completed 18 years pensionable service on retirement. In this situation paragraph 4 also applies, but this time the multiplier is 7/60 plus 1/60 for each complete year of service, as follows:- 7/60 + 18/60 = 25/60 The minimum of 20 sixtieths is therefore irrelevant and accordingly the calculation is:- £37,446.10 x 25/60 = £15,602.54[42]During submissions, various labels (such as ‘progressive’ and ‘regressive’) were used to describe parts of the 1987 Scheme. However I have not found these terms particularly helpful. The important point, whatever label is attached to the scheme, is that for some reason (and it is not clear to me why) the 1987 Scheme provides for pension to accrue at different rates for different periods of service. This has the effect of producing anomalies between full-time workers so that in the above examples Officer B will receive the same pension as Officer A, even though Officer B will have worked about 6,000 hours more than Officer A. This may, of course, be perceived by Officer B as being quite unfair. Part-timers: paragraph 4A[43]It is now necessary to consider the formula for part-time workers. As explained above, this is set out in paragraph 4A. This paragraph is much simpler to apply, adopting the same formula whatever the length of pensionable service. The aim is to pro-rate the part-timer’s pension entitlement by reference to a full-time officer with the same actual length of service. That is achieved by applying the formula N x R/Q where:-(i) N is the amount of the annual pension that would have been payable if all of the Page 8 of 37 Claimant’s service had been full-time (ie the full-time equivalent salary);(ii) R is the period in years of C’s reckonable pensionable service (calculated by applying the appropriate factor to each period of part-time service as explained in paragraphs 30 - 34 above); and(iii) Q is the period of pensionable service she would have had if the periods of parttime service were treated as if they were periods of full-time service.[44]Pausing there, it is clear that the comparison under paragraph 4A is therefore between the Claimant and a full-time officer with the same length of actual service (PC Y). Applying paragraph 4A to the Claimant[45]As noted above all parties agree that, so far as the Claimant is concerned, the calculation under paragraph 4A has been carried out in accordance with the 1987 Regulations. The relevant figures are as follows:-(i) To calculate N - the pension the Claimant would have received had she worked full time throughout - we need to know two things, namely her actual length of service (which was 12 years and 113 days) and the full-time equivalent salary (which was £37,446.10). In this situation paragraph 4 governs the pension calculation. Accordingly, as pensionable service is between ten and thirteen years, the minimum entitlement of 20/60 of pensionable pay applies. N is therefore 20/60 x £37,446.10 = £12,482.03;(ii) The figure for R (the Claimant’s reckonable pensionable service) is 11 years and 247 days (c.11.677 years)5; and(iii) Q is 12 years and 113 days (c.12.309 years), representing what the Claimant’s pensionable service would have been if she had worked full-time throughout.[46]As a result if we apply the paragraph 4A formula to the Claimant, the pension calculation is:- £12,482.03 x 11.677 = £11,840.83 12,309[47]Accordingly the Claimant receives £11,840.83, whereas an officer who had always worked full-time with the same actual length of service would receive £12,482.03. To put it another way, a pro-rating exercise has been performed such that the Claimant receives 94.863% of the pension she would have received had she worked full-time throughout her career, reflecting the fact that she actually worked 94.863% of the hours PC Y would have worked over the same period. F. The essence of the dispute[48]Paragraph 4A was introduced into the 1987 Regulations by the Police Pensions (Part-time Service) Regulations 2005 (the ‘2005 Regulations’). Ironically, given the present proceedings, it appears that the 2005 Regulations were intended to remove any disadvantage suffered by part-time workers under the previous version of the scheme by ensuring that the Claimant’s entitlement is pro-rated exactly in proportion to her part-time working. The Explanatory Notes to the 2005 Regulations provide as follows: 5 See para [34] above. Page 9 of 37 ‘The amendments [including the introduction of paragraph 4A] make provision for part-time police officers’ pension benefits to be calculated as if they had been full-time officers, and then pro-rated for periods of part-time service. This replaces the previous basis of calculation under which part-time working counted as pensionable service on the basis of the actual hours served. This disadvantaged part-time officers because under the [1987] Scheme reckonable service is accrued at a faster rate after 20 reckonable years.’[49]The clear purpose of the 2005 amendments was therefore to ensure part-time officers were not disadvantaged by the accelerated accrual afforded at various stages of a police officer’s career under the 1987 Regulations. The Respondent and the Secretary of State contend that paragraph 4A achieves this aim by comparing the Claimant’s service with an officer (PC Y) who has worked full-time throughout the period in which the Claimant was employed and so has the same period of actual service (ie 12 years and 113 days). PC Y’s pension entitlement (which is calculated by reference to the full-time salary) is then pro-rated to reflect the relative hours worked by the Claimant during the comparison period. They argue that, far from being discriminatory as the Claimant alleges, this exercise achieves parity and fairness for part-time officers because like is compared with like; the only change that is made to the Claimant’s entitlement by reason of her parttime status is a lawful pro-rata adjustment. As a result the Claimant receives a pension that is exactly proportionate, in terms of hours worked, to that received by a full-time officer and this will be true at whatever stage of her career her pension is calculated. The Respondent also points out that the part-timer’s pension will be proportionate to the contributions he or she has made.[50]The Claimant for her part maintains that whatever the intention behind the 2005 amendments, the effect in practice is discriminatory. Her claim, as advanced in her claim form, was that the correct comparison should be with a police officer (PC X) who has always worked full time but has the same period of reckonable pensionable service as she has (ie 11 years 247 days). Otherwise PC X accrues pension at a higher rate per hour than the Claimant.[51]However the Claimant’s case had been refined somewhat by the start of this preliminary hearing and although paragraph 6 of her original skeleton continues to maintain that she should be compared with PC X, this is now on the basis that PC X will have worked the same number of hours as the Claimant over their respective careers. (The change is significant because it allows reliance to be placed on the ECJ’s decision in Schonheit v Stadt Frankfurt am Main6, a case discussed in detail later in this judgment.)[52]Having made that change to the pleaded case, Ms Romney QC puts forward the following comparison to illustrate the alleged inequality that lies at the heart of the claim. Once again all calculations are based on pensionable pay of £37,446.10 (the Claimant’s full-time equivalent salary). The Claimant The Claimant’s reckonable pensionable service is 11 years and 247 days. When expressed in weeks this period equals 608 and 6/7th weeks. As the full-time working week is 40 hours, she has therefore worked for 24,354 and 2/7th hours. As demonstrated at paragraph 46 above, under paragraph 4A the Claimant’s pension entitlement is £11,840.83 per annum. This works out at £0.486 per hour worked. 6 [2006] 1 CMLR 51 Page 10 of 37[53]The Claimant’s position is then compared with PC X:- PC X PC X, who has always worked full-time, also has reckonable pensionable service of 11 years and 247 days (albeit this has been accumulated over a shorter period than in the Claimant’s case). Like the Claimant, PC X has therefore worked for 24,354 and 2/7th hours in total. Applying paragraph 4 (because PC X has only ever worked full-time) and assuming average pensionable pay is £37,446.10 per annum as in the Claimant’s case, PC X is entitled to a pension of 20 sixtieths of pensionable pay. This amounts to 20/60 x £37,446.10 = £12,482.03 per annum. PCX’s pension therefore works out at £0.513 per hour worked, somewhat higher than the hourly rate for the Claimant - even though both the Claimant and PC X have performed exactly the same hours of work.[54]The Claimant relies on this comparison to say that the calculation under the 1987 Regulations is skewed in favour of full time workers because the effect of Paragraph 4 is to give PC X the benefit of a ‘guaranteed minimum’ of 20 sixtieths of average pensionable pay. She contends that, by contrast, the guaranteed minimum does not apply in her case. Instead when the calculation of her pension is performed under paragraph 4A a different – and in her view less favourable – formula applies purely on the basis that she worked part time for some portion of her career.[55]Putting this into context, this means that the Claimant’s pension is some £641.20 less per annum than that granted to PC X, even though PC X has the same length of reckonable pensionable service as the Claimant and so has worked exactly the same number of hours.[56]The contrast is even more marked in a further example provided by the Claimant. PC Z PC Z has always worked full-time and has ten years’ pensionable service. He has therefore worked 20,857 1/7 hours in total (ie considerably less than the Claimant). Applying paragraph 4 (because PC Z has only ever worked full-time) and assuming once again that average pensionable pay is £37,446.10 per annum, PC Z is entitled to a pension of 20/60 of pensionable pay. This amounts to 20/60 x £37,446.10 = £12,482.03 per annum. PC Z’s pension therefore works out at £0.598 per hour worked, significantly higher than the Claimant’s hourly rate. According to the Claimant this is another example of a fundamental inequality produced by the 1987 Regulations.[57]The Claimant therefore urges me to adopt PC X as the appropriate comparator to put an end to these anomalies and the alleged discriminatory treatment she has received.[58]For completeness it should be added that the Claimant accepts that she accrues pension at the same hourly rate as a full-time police officer with the same actual length of service as she has (ie PC Y), albeit she does not accept PC Y is the appropriate Page 11 of 37 comparator. The calculation for PC Y is as follows:- PC Y PC Y has always worked full-time and has reckonable pensionable service of 12 years and 113 days (the same actual length of service as the Claimant). PC Y has therefore worked for 641 and 6/7 weeks, which amounts to 25,674 and 2/7th hours in total. Applying paragraph 4 (because PC Y has only ever worked full-time) and assuming again that average pensionable pay is £37,446.10 per annum as in the Claimant’s case, PC Y is entitled to a pension of 20/60 of pensionable pay. This amounts to 20/60 x £37,446.10 = £12,482.03 per annum. PC Y’s pension therefore works out at £0.486p per hour worked, exactly the same hourly rate as the Claimant.[59]Finally the Respondent refers to a further example which it would be useful to set out at this stage, namely officer PC ZA:- PC ZA PC ZA has worked full-time and has reckonable service of 13 years. He or she has therefore worked 27,114 hours over their career. Applying paragraph 4 (because PC ZA has only ever worked full-time) and assuming that average pensionable pay is £37,446.10 per annum as in the Claimant’s case, then PC ZA is entitled to a pension of 20/60 of pensionable pay (7/60 + 13/60). This amounts to 20/60 x £37,446.10 = £12,482.03 per annum. PC ZA’s pension therefore works out at £0.46p per hour worked (ie the lowest hourly rate out of the Claimant, PC X, PC Y and PC Z, despite the fact PC ZA has worked the greatest number of hours). This demonstrates once again7 that the 1987 scheme produces anomalies – even between full time workers. These are discussed in more detail below. G. The relevant legislative provisions[60]As noted above, the Claimant pursues her claim under three different causes of action, namely(i) Regulation 5 of the Part-time Workers (Prevention of Less Favourable Treatment) Regulations 2002,(ii)section 67 of the Equality Act 2010 and(iii) Article 157 TFEU.[61]The relevant legislative provisions, in so far as they relate to the issue for determination before me, are described below. I also set out a brief summary of the principles of interpretation where UK legislation is inconsistent with EU law. The Part-time Workers (Prevention of Less Favourable Treatment) Regulations 2000: SI 2000/1551 (the ‘PTW Regulations’)[62]Central to the Claimant’s case is the allegation that the calculation of her pension amounts to less favourable treatment on the ground that she was a part-time worker. This allegation gives rise to a claim under the PTW Regulations. However it is also the foundation for the Equal Pay and Article 157 claims which allege indirect discrimination (on the basis that the dominant number of part-time workers are female and therefore suffer 7 Similar examples are discussed in paras 41 and 42 above. Page 12 of 37 group disadvantage).[63]The right not to be discriminated against on the grounds of part-time status is derived from the European Part-time Workers Directive 97/81/EC (the ‘PTW Directive’) which was implemented into UK law by the PTW Regulations. Regulation 5 sets out the basic right of a part-time worker to be treated no less favourably than a comparable fulltime worker. It provides as follows:- ‘(1) A part time worker has the right not to be treated by his employer less favourably than the employer treats a comparable full time worker8:-(a) as regards the terms of his contract; or(b) by being subjected to any other detriment by any act, or deliberate failure to act, of his employer. (2) The right conferred by paragraph (1) applies only if:- (a) the treatment is on the ground that the worker is a part time worker, and (b) the treatment is not justified on objective grounds’.[64]Regulation 5(3) then proceeds to introduce the ‘pro-rata principle’ as the basic method of determining whether there has been less favourable treatment of part-time workers. It provides as follows:- ‘(3) In determining whether a part-time worker has been treated less favourably than a comparable full-time worker the pro-rata principle shall be applied unless it is inappropriate.’[65]What is meant by an ‘appropriate’ application of the pro-rata principle is clearly fundamental to my decision. Some guidance is given in Regulation 1(2) which defines the pro rata principle to mean:- ‘…Where a comparable full time worker receives or is entitled to receive pay or any other benefit, a part time worker is to receive or be entitled to receive not less than the proportion of that pay or other benefit that the number of his weekly hours bears to the number of weekly hours of the comparable full time worker.’ 66. ‘Weekly hours’ is then defined for the purposes of this principle by Regulation 1(3) to mean (so far as relevant to the present case) the number of hours the worker is required to work in a week without absences or overtime.[67]It is clear from these provisions that, provided it is appropriate for the pro-rata principle to be applied, then under the PTW Regulations the pro-rating exercise is to be performed by comparing the number of weekly hours performed by the part-timer as against the number performed by the full-time worker. A simple example would be where a comparable full-time worker works 40 hours per week and earns £20,000 per year. Under the pro rata principle a part-time worker who works 24 hours per week should earn no less than 24 fortieths of the full-time salary ie no less than £12,000 per year.[68]As noted above, the PTW Regulations implement the PTW Directive, which itself implements the Framework Agreement on part-time work (which is annexed to it). The principle of non-discrimination is set out at clause 4 of the Framework Agreement:- 8 The terms ‘full-time worker’ and part-time worker are defined in Regulation 2 by reference to time worked and the custom and practice of the employer. Page 13 of 37 ‘(1) In respect of employment conditions, part time workers shall not be treated in a less favourable manner than comparable full time workers solely because they work part time, unless different treatment is justified on objective grounds.[69]Clause 4 then goes on to deal with the possibility of pro-rating benefits for part-time workers:- (2) Where appropriate the principle of pro rata temporis shall apply.’ (3) The arrangements for the application of this clause shall be defined by the Member States… having regard to European legislation, national law, collective agreements and practice.’[70]In contrast to the PTW Regulations, the text of the PTW Directive does not define the phrase ‘pro rata temporis’. However considerable latitude is given to Member States in applying this principle, provided due regard is paid to relevant European jurisprudence. So much is clear from paragraph 16 of the preamble to the PTW Directive which provides as follows:- ‘Whereas, with regard to terms used in the Framework Agreement which are not specifically defined therein, this Directive leaves Member States free to define those terms in accordance with national law and practice… providing that the said definitions respect the content of the Framework Agreement.’[71]There is no rule of EU law which prevents the principle of pro rata temporis being applied to the calculation of pension benefits for part-time workers9. The Equality Act 2010[72]The Claimant also brings an equal pay claim. The domestic equal pay provisions are set out in Part 5 of Chapter 3 of the Equality Act 2010 (the ‘EqA 2010’). These provisions are designed to secure equality in pay and other contractual terms as between men and women. This is achieved by means of a ‘sex equality clause’ which is automatically read into the employee’s contract or, in the case of an office holder like the Claimant, into the terms of appointment to that office: see EqA 2010 s 66. The clause ensures parity of terms between the employee and his or her chosen comparator provided(i) they are both engaged on like work, work rated as equivalent or work of equal value and(ii) the employer cannot rely on what is known as the ‘material factor defence’ set out in EqA 2010 s 69. Provided these conditions are satisfied the equality clause has the effect of modifying contracts in so far as they incorporate discrimination (be it direct or indirect) on grounds of sex.[73]A similar provision to the sex equality clause – referred to as a ‘sex equality rule’ - is implied into the terms of occupational pension schemes. Under section 67(1) of the Equality Act 2010 (EQA 2010) a ‘sex equality rule’ is implied into every occupational pension scheme that does not already include one. This rule requires that men and women performing equal work are treated equally in relation both to the terms on which they are permitted to join the scheme and (of relevance in the present case) to the terms on which they are treated once they have become scheme members: see s67(3). As a result if such a term is less favourable to an employee than it is to a comparator of the 9 See Schonheit v Stadt Frankfurt am Main, supra note [6] Page 14 of 37 opposite sex doing equal work, then the offending term is modified so as not to be less favourable: see s67(2)(a). It is common ground that the 1987 Regulations contain such an equality rule.[74]The sex equality rule, like the sex equality clause, is subject to what is known as the ‘material factor defence’. This is set out in EqA 2010 s 69. Under that provision a sex equality rule will have no effect if the trustees or managers of the pension scheme in question can show that the difference in terms is due to ‘a material factor, which is not the difference of sex’: see EqA s69(4). As a result, where the claimant has established equal work and an inequality in terms then a trustee or manager of the scheme will have a defence if the difference can be explained by reference to a material factor and provided that there is no independent evidence of any kind to show that sex has had any influence on the inequality. This clearly imports principles in line with those of direct and indirect discrimination, as the voluminous case-law in this area makes clear. So, for example, in Glasgow City Council v Marshall [2000] IRLR 272, HL Lord Nicholls explained at paragraph 18:- ‘…[where there is a prima facie case] the burden passes to the employer to show that the explanation for the variation is not tainted with sex. In order to discharge this burden the employer must satisfy the tribunal on several matters. First that the proffered explanation or reason is genuine… Second the less favourable treatment is due to this reason…. Third that the reason is not ‘the difference of sex’. This phrase is apt to embrace any form of sex discrimination, whether direct or indirect…’10[75]In light of these principles it is the Claimant’s case that the Respondent has breached the sex equality rule in EqA 2010 s 67 on the basis that the calculation performed under paragraph 4A of Part III of Schedule B to the 1987 Regulations to determine her pension entitlement is indirectly discriminatory against female members of the 1987 Scheme (see paragraph 40 of the amended ET1). The Respondent defends the claim on a number of grounds. These include the argument that the Claimant has adopted the wrong comparator (ie PC X) and that when the correct comparison is made there is no less favourable treatment (see paragraph 20 of the response). The Respondent also relies, without prejudice to its other arguments, on the defence of material factor (see paragraph 25 of the response). Article 157 of the Treaty on the Functioning of the European Union[76]The EqA 2010 is intended to give effect in UK law to certain EU legislation including Article 157 of TFEU and the EU Equal Treatment Directive 2006/54 (the ‘Recast Directive’).[77]Article 157 of TFEU (which began life as the Treaty of Rome) sets out the principle that men and women should receive equal pay for equal work. It is drafted in very simple terms and provides as follows:- ‘(1) Each Member State shall ensure that the principle of equal pay for male and female workers for equal work or work of equal value is applied.’ 10 Although the Marshall case was decided under the Equal Pay Act 1970 the judgment of Lord Nicholls is still cited as providing a ‘road map’ through the relevant statutory provisions, now to be found in Chapter 3 of the Equality Act 2010: see Calmac Ferries Ltd v Wallace UKEATS/0004/13 per Langstaff P. Page 15 of 37 (2) For the purposes of this Article ‘pay’ means the ordinary basic or minimum wage or salary and any other consideration, whether in cash or in kind, which the worker receives directly or indirectly in respect of his employment, from his employer. Equal pay without discrimination based on sex means “(a) that pay for the same work at piece rates shall be calculated on the basis of the same unit of measurement; (b) that pay for work at time rates shall be the same for the same job’.[78]A number of points need to be made about Article 157. Firstly it is clear that the concept of ‘pay’ in Article 157 is a wide one and can, in principle, include payment of a pension11, although it should be added that there are some pensions to which Article 157 does not apply. Secondly it is clear that the prohibition on discrimination set out in Article 157(2) includes both direct and indirect sex discrimination12. Thirdly, although Article 157 does not refer expressly to the principle of pro rata temporis, it is accepted that the principle applies when considering the treatment of part-time workers in this context, just as it does under the PTW Directive13. Moreover, as noted when discussing the PTW Directive, there is no rule of EU law which precludes a pension being calculated pro-rata temporis in the case of part-time employment.14[79]The provisions of Article 157 were originally supplemented by Council Directive 75/117 (the ‘Equal Pay Directive’), the purpose of which was to facilitate the practical application of the principle of equal pay. However the Equal Pay Directive and a number of other Directives dealing with the equal treatment of men and women in matters of employment and occupation were subsequently consolidated, together with relevant caselaw of the ECJ into Directive 2006/54/EC (the ‘Recast Directive’). Like the Equal Pay Directive before it, the Recast Directive is not intended to alter the scope of Article 157.[80]So far as equal pay is concerned Article 4 of the Recast Directive provides:- ‘For the same work or for work to which equal value is attributed, direct and indirect discrimination on grounds of sex with regard to all aspects and conditions of remuneration shall be eliminated.’[81]The Claimant relies on Article 157 for two reasons. Firstly she argues that the Article should govern the construction of the relevant provisions of the EqA 2010 in relation to her equal pay claim. Secondly if her equal pay claim does not succeed then she argues that Article 157, which has direct effect15, provides her with a free-standing cause of action in the employment tribunal. The Respondent raises a number of issues with regard to this aspect of her case, including whether the Claimant was a ‘worker’ within the meaning of Article 157, and whether she is entitled to bring a free standing claim. However none of those matters is for determination at this hearing which, as noted above is concerned solely with whether it is appropriate to pro-rate the calculation of the Claimant’s pension benefits and if so on what basis. 11 Barber v Guardian Royal Exchange Assurance Group C-262/88 [1990] IRLR 240 at [12] (ECJ); Ten Oever v Stichting Bedrejfspensioenfonds voor het Glazenwassers C-109/91 [1993] IRLR 601, at [8] (ECJ) 12 Schonheit v Stadt Frankfurt am Main, supra note [6] 13 See paragraph 57 of the Claimant’s original submissions citing Osterreichischer Gewerkschaftsbund v Verband Osterreichischer Banken ubnd Bankiers C 476/12 [2015] IRLR 14 Ibid at [90-91] 15 Defrenne v Sabena C-43/75 [1981] 1 All ER 122 at [40] (ECJ) Page 16 of 37 Interpretation of UK law[82]As explained by Miss Romney QC at paragraphs 51 and 52 of her original skeleton, where UK legislation is inconsistent with a directly applicable provision of EU law (such as Article 157), the UK provision must be read as though without prejudice to the directly enforceable provision; and the UK provision must, if necessary, be dis-applied or read so as to apply the provision consistently with EU law.[83]Equally, where there is an inconsistency between domestic law and an EC Directive, the principle of conformity of construction set out in Marleasing SA v LA Comercial Internacional de Alimentacion SA16 applies, namely that national law is to be interpreted so far as possible in light of the wording and purpose of the Directive. Miss Romney also points out that the interpretative tools available when conducting such an exercise are helpfully summarised in Rowstock Ltd and another v Jessemey17. H. Applying the law to the facts of this case The Schönheit Case[84]The Claimant’s principle argument is that the decision of the ECJ (as it then was) in Schönheit v Stadt Frankfurt am Main18 determines the issue I have to decide. The Schonheit case, like the present, concerns an equal pay claim brought by a female parttime worker in respect of pension entitlements under a non-linear scheme. The Claimant argues that the ECJ’s decision establishes a general principle whereby, when considering the calculation of benefits for part-time workers in such schemes, the appropriate comparator is a full-time officer who has worked the same number of hours as the parttime worker (ie comparator X). She submits that that is the case whether the claim is brought under Article 157 or the PTW Directive.19 Furthermore she argues that the same principle must apply under both section 67 EqA 2010 and also under the PTW Regulations (if necessary by means of a conforming interpretation applying the principles in Marleasing20 and Jessemey21). The Schonheit Pension Scheme: the BeamtVG[85]The Schonheit case concerned the BeamtVG, the occupational pension scheme for the public service in Germany. To understand the rationale of Schonheit, and the arguments put forward by the parties about the case, it is necessary to understand how the BeamtVG scheme operated. The pre-1992 Scheme[86]During the period 1 August 1984 to 31 December 1991 pensions for full-time workers were calculated under paragraph 14 of the BeamtVG as follows:-(i) there was a minimum entitlement of 35 per cent of pensionable pay for the first ten 16 Case-C106/89 [1992] 1 CMLR 305, CJEU 17 [2014] IRLR 368, at [22] per Underhill LJ 18 Supra note [6] 19 She relies on Osterreichischer Gewerkschaftsbund v Verband Osterreichischer Banken und Bankiers C-476/12 [2015] IRLR 67 for this proposition. 20 Supra note [15] 21 Supra note [16] Page 17 of 37 years of pensionable service22;(ii) Under paragraph 14, after ten years’ pensionable service entitlement rose “with every further year of service by 2 per cent until completion of the twenty fifth year of service, and thereafter by 1 per cent of pensionable service subject to a maximum of 75%...”[87]Pausing there, this was clearly another ‘non-linear’ scheme where, as in the case of the 1987 Scheme, pension did not accrue at a uniform rate. However the provisions governing the calculation of benefits in the BeamtVG were somewhat different to those in the 1987 Regulations. Under the BeamtVG, the rate of accrual decreased the longer the officer’s career, thus favouring a full-time officer who left after a relatively short period of pensionable service (eg after 5 years’ pensionable service he or she would be entitled to 35 per cent of pensionable pay). Another notable feature of the scheme was that entitlement was subject to a 35 per cent floor and a 75 per cent ceiling.[88]So far as entitlement for part-time workers was concerned this was also calculated under paragraph 14 of the BeamtVG, the second part of which provided as follows:- ‘In the case of part-time work… the rate of pension that would have been attained hereunder but for these departures from full-time work and before application of the maximum rate shall be reduced in such proportion as actual pensionable service bears to the period of time which but for the departures from full-time working would have been completed but shall not be less than 35% or more than 75%’.[89]This provision for part-timers was known as the ‘pension abatement’23. Put simply, it involves calculating a notional pension for a full-time worker with the same actual length of service as the part-timer (temporarily ignoring the ceiling of 75%). Then a pro-rating exercise is performed which is identical to the one carried out under paragraph 4A of the 1987 Regulations so that pension benefits are reduced in proportion to the hours worked. However there is then a final stage – which is not replicated in the 1987 Regulations - whereby the resulting figure is subject to the minimum and maximum entitlements under the scheme (35 per cent and 75 per cent respectively).[90]To explain how the pension abatement operates in the context of the BeamtVG the Advocate General gives the following example:- ‘Let us assume an official has worked for 30 years on a half time basis. His notional pension (as if he had worked full-time for 30 years) is then calculated as follows: 35 per cent for the first 10 years; 30 per cent (15 x 2 per cent) for the 11th to the 25th year; and 5 per cent (5 x 1 per cent ) for the 26th to the 30th year, making a total of 70 per cent. If the abatement is now applied, the result is a pension of 35 per cent (70 per cent x 15/30).’24 22 Under paragraph 6 of the BeamtVG, pensionable service is ‘the period of service completed by the official from the date of appointment’. However in the case of part-time workers service only counts ‘in such proportion as the reduced working time bears to normal working time’. Paragraph 6 therefore requires the same exercise to be carried out as when determining R (reckonable pensionable service) for the purposes of paragraph 4A of the 1987 Regulations. So, for example, if an official worked half time for 30 years then, under paragraph 6 of the BeamtVG, his or her pensionable service would be 15 years. Exactly the same result is obtained when calculating a parttimer’s reckonable service under the 1987 Regulations: see para 33 above. 23 The pension abatement in paragraph 14 was introduced on 1 August 1984. 24 See AG [60] and [61] Page 18 of 37[91]As the Advocate General went on to explain, if the pension abatement was not applied then the pensionable service of this part-timer would be would be 15 years25, which would give an entitlement of 45 per cent (35 per cent for the first 10 years, 2 per cent for each of the remaining years). A worker who had worked full-time for 15 years would similarly be entitled to a rate of pension of 45 per cent.[92]The figures in the Advocate General’s example do not call for application of the 35 per cent baseline or the 75 per cent ceiling. However these are important features of the pension abatement. Their effect is that there is no consistent pro rata relationship between the pension entitlements of a part-time worker and a full-time worker with the same actual length of service under the BeamtVG. This is demonstrated by Graph A, provided by the Secretary of State. Line A (the blue line when in colour) shows the accrual of pension rights of a full-timer, measured as a percentage of full-time salary. In the period up to 10 years’ service there is an entitlement of 35 per cent so the graph begins at that point. After that, entitlement increases in a straight line up to year 25 at 2 per cent per year. It then increases in another straight line, less steeply now, at only 1 per cent per year up to year 35. Then the full-timer reaches the 75 per cent ceiling and so the line is flat.[93]This should be contrasted with Line B (the green line) which shows the accrual of pension rights by a part-time worker who works half time. (This particular proportion was chosen by the Secretary of State because it is the example discussed by the Advocate General). It can be seen, for example, that as a result of the operation of the pension abatement, the part-timer’s entitlement flat lines at 35 per cent where pensionable service is between 10 and 30 years. Then as the blue line rises the final 5 per cent between year 30 and 35, the green line can rise pro rata. However after year 35 when Line A flat lines because the full-timer has reached the maximum entitlement of 75%, Line B (the green line) still continues to rise slowly.[94]The overall effect of the pension abatement is therefore that there is no uniform prorating of benefits as between full-time and part-time workers as they progress through their careers. Quite the contrary; there is little correlation between the A (blue) and B (green) lines.[95]There is a further point to note about the BeamtVG scheme. Just as there is no prorata relationship between the part-time worker and comparator Y under this scheme, equally there is no correlation between the benefits accrued by the part-time worker and comparator X. This can be seen by considering Line C on the graph (the red line) which represents a worker who has worked full-time for half of the period worked by the parttimer and so has worked the same number of hours overall as the part-time worker.[96]To read the red line on the graph the numbers on the X axis must be halved. So at the point where Lines B (green) and C (red) meet, the green worker has been working part-time for 20 years and the red worker has been working full-time for 10 years. At that point both have pension rights of 35 per cent. However thereafter the red line moves sharply upwards as the full-timer experiences a rapid accrual of pension rights in years 10 – 25, just as happened with the full-timer at Line A (the only reason the angle of Line C is less acute than that of Line A is because the numbers on the x axis have been halved). The same is not true of the part-timer at Line B (green) and the graph is consistent with the part-time workers’ service being disproportionately penalised viz a viz comparator X. The Post-1992 scheme and the transitional arrangements 25 Calculated in accordance with footnote [22] above Page 19 of 37[97]So much for the old scheme. From the start of 1992 the BeamtVG was amended to create a ‘linear’ pension scheme. To achieve this end a new version of paragraph 14 was introduced, providing that henceforth officers were to be entitled to 1.875 per cent for each year of pensionable service subject to a maximum of 75 per cent. At the same time, the pension abatement for part-time workers set out in the latter part of paragraph 14 was abolished.[98]When the scheme was amended in 1992, complex transitional provisions were also introduced to preserve existing rights. These were contained in paragraph 85 of the scheme. The idea was that workers employed at 31 December 1991 could bank their accrued pension rights under the old version of paragraph 14 and then could add 1 per cent for each further year of pensionable service up to a maximum of 75 per cent. The calculation was as follows:-(i) Step 1: calculate pension for both full-time and part-time workers as if it all accrued under the new scheme ie 1.875 per cent for each year of pensionable service (pensionable service for these purposes was calculated, as previously, under paragraph 6 of the scheme26 so that service was reduced to take account of part-time working);(ii) Step 2: compare the figure under Step 1 with a calculation under the old paragraph 14 (without the abatement) for pensionable service up to 1991 but then adding a further 1% for every further year of pensionable service from 1992 onwards. If this yields a higher figure than under Step 1 then, if the worker is full time, his pension entitlement will be the higher figure; and(iii) Step 3: for part-time workers only, compare with a third calculation done on the basis that all service was full-time and had accrued under the pre-1992 scheme ie under the old version of paragraph 14 but this time with the pension abatement. If this yields a lower figure than in (ii) above then take this figure.[99]The transitional provisions in paragraph 85 of the BeamtVG therefore create the possibility of part-time status affecting the calculation twice over: firstly at Step 2 where the calculation is based on pensionable service (which has already been reduced to reflect part-time working) and secondly at Step 3 where the pension abatement – which is only applied to part-timers - could reduce benefits still further. The facts of the Schonheit case[100]Mrs Schonheit was an official of the city of Franfurt am Main and a member of the BeamtVG Scheme. She had eligible service from 1 April 1965 until July 31 1999, albeit after 1 July 1992 she worked part-time. She then took early retirement with effect from 1 August 1999 on the basis of serious invalidity. As she had been in service on 31 December 1991 she was awarded a pension under the transitional provisions. This resulted in her being given a pension of 65.80 per cent of her pensionable pay. The calculation in her case, using the three steps set out above, was as follows:-(i) Step 1: Her pensionable service was calculated under paragraph 6 as being 30.39 years. As a result the calculation was 30.39 years x 1.875 = 56.99 per cent(ii) Step 2: Her pensionable service to 31 December 1991 was calculated under 26 As explained at footnote [22] above Page 20 of 37 paragraph 6 to be 26 years and 219 days (which was rounded up to 27 years). Under the old version of paragraph 14 (without the abatement) her pension entitlement would be:- 10 years @ 35% = 35% 15 years @2% = 30% 2 years @ 2% = 2% 67% To this figure must be added a figure for her post 1992 entitlement. Her pensionable service for this period (again applying paragraph 6) was 3.79 years. Under paragraph 85 this gave her an additional entitlement of 3.79 x 1% = 3.79 per cent. Adding these figures together gives an entitlement of 70.79 per cent. This figure is clearly higher than the figure obtained under Step 1 and had she worked solely full-time the calculation would have stopped there and she would have been entitled to a pension of 70.79% of pensionable pay. However because she has worked part-time there a further comparison has to be made.(iii) Step 3: This requires calculation of the pension she would have received under the old version of the scheme, assuming all her service was full time. On this basis her service was 34.18 years. Applying the old version of paragraph 14 this results in a rate of pension of 74 per cent as follows: 10 years @ 35% = 35% 15 years @ 2% = 30% 9 years @ 1% = 9% 74% The pension abatement is then applied so as to reduce the percentage in the ratio of pensionable service to actual length of service ie 30.39/34.18 x 74% = 65.80 per cent27. As this figure is lower than that calculated under Step 2 it is the lower figure that applies. As the Advocate General pointed out28Mrs Schonheit was therefore entitled to a lower pension by some 5 per cent than that of a full-time official who had completed a similar number of years of pensionable service (albeit over a shorter period).[101]Mrs Schonheit claimed that the transitional provisions were incompatible with Article 141 (now Article 157) because they gave rise to indirect sex discrimination. This was on the basis that her pension benefits were reduced disproportionately when compared to a full-time official who had worked the same number of hours and that this was done solely because she was a part-time worker. As it was predominantly women who worked parttime she argued that she had been subject to indirect discrimination and that she should have been awarded a pension of 70.79%. The reference to the ECJ[102]Mrs Schonheit’s case was considered by the German Administrative Court (the Verwaltungsgericht) which accepted that Mrs Schonheit had been treated less favourably on the ground that she was a part-time worker and that considerably more female civil 27 The figure of 30.39 years pensionable service is calculated by adding 26 years and 219 days (her pre-1992 pensionable service) to 3.79 years (her post-1992 pensionable service) 28 At para [AG 81] Page 21 of 37 servants worked part-time than men in the region in which she worked29. However there was conflicting German case-law on the issue of whether such treatment could be justified by objective factors unrelated to any discrimination on grounds of sex. The German court therefore decided to refer a number of questions to the ECJ for a preliminary ruling on various issus relating to objective justification. (There were also two questions about the scope of Article 141 - the forerunner of Article 157 - but they are not relevant to the point I have to decide and will be ignored for the purposes of this judgment).[103]The ECJ, having considered the questions posed by the reference, set out a number of principles in relation to objective justification. These included firstly, that differential treatment could be justified by reasons which were not put forward at the time the measures introducing the difference in treatment were introduced and secondly, that restricting public expenditure was not an objective that could be relied upon to justify different treatment on grounds of sex. As far as the specific provisions of the BeamtVG were concerned the ECJ decided as follows:- ‘… national legislation, such as that deriving from para 85 of the BeamtVG in conjunction with the old version of paragraph 14 thereof, which has the effect of reducing a worker’s retirement pension by a proportion greater than that resulting when his periods of part-time work are taken into account cannot be regarded as objectively justified by the fact that the pension is in that case consideration for less work or on the ground that its aim is to prevent civil servants employed on a part-time basis from being placed at an advantage in comparison with those employed on a full-time basis’30. The Court did not make any specific findings about the appropriate comparator for the purposes of establishing disadvantage. It simply accepted that the transitional provisions in paragraph 85 had a disproportionate impact on the pension entitlement of part-time workers. Does Schonheit help to resolve the preliminary issue?[104]As noted above, the Claimant argues that the Schonheit decision establishes a general principle that comparator X - a full-time worker who has worked the same number of hours as the part-time worker - is the appropriate comparator for establishing whether part-timers suffer disadvantage in non-linear pension schemes. Although the case did not deal with the matter directly, she submits that this is implicit in the reasoning of both the Advocate General and the Court. She points in particular to paragraphs 100 – 101 of the Advocate-General’s opinion where he states as follows:- ‘That the abatement can be described as disproportionate is evident from simple calculations… the introduction of the pension abatement in the event of part-time working has led to officials who have worked part-time being treated differently from officials who have always worked part-time: their pensions are different even though they have completed the same number of hours of pensionable service. (Emphasis added). Although the pension abatement was abolished when the linear calculation system entered into force in 1992, it continues to be applicable under the transitional legislation to officials who have spent some of their service working part-time. They are still confronted with this abatement and are therefore worse off than officials who were similarly in service before December 31 1991, but have worked full-time and have the 29 See Schonheit paras [34 to 39] 30 Paragraph [97] Page 22 of 37 same number of years service.’[105]She also highlights the Court’s reasoning at paragraph 96 of the judgment:- ‘As is clear from …the Advocate General’s opinion, where, over their careers as a whole, a part-time official and a full-time official have worked the same number of hours, application to the part-time official of the pension abatement rule is liable to result in his being awarded a lower rate of pension than that awarded to the full-time official under the old version of paragraph 14 of the Beamt VG.’ (Emphasis added)[106]The Claimant argues that it is possible to extract a general principle from such statements, whereby comparator X is the appropriate comparator for non-linear schemes.31 Moreover she contends that this principle should be applied identically under both Article 157 and the PTW Directive so as to support all her claims.[107]Having considered the various submissions in relation to this issue I am not persuaded that the Schonheit case establishes the general principle for which the Claimant contends. This is for a number of reasons.[108]Firstly, the questions referred to the Court in Schonheit by the Verwaltungsgericht do not concern or refer to the comparator issue. This is unsurprising - it was never in dispute that the transitional provisions in the BeamtVG disadvantaged part-time workers. Rather, the difficulty for the German court was the issue of objective justification and it is on that issue that the ECJ are asked to provide guidance. This was recognised by the Advocate General in the following passage:- ‘The applicants in the main action point out that the pension abatement results in a lower pension, by some 5 per cent in their case, than that of a full-time official who has completed a similar number of years of pensionable service. The abatement discriminated against women in particular, because in Germany’s public service, it was predominantly women who worked on a part-time basis. That is not disputed; what is important therefore is objective justification.’32[109]The ECJ made a similar point33:- ‘…it is not disputed that where the pension abatement provisions are applied with the regressive scale, the result is likely to be for the same number of hours worked, the pension paid in respect of part-time employment will be lower that that paid in respect of full-time employment… In this instance it is apparent from the orders for reference that a considerably higher percentage of female than male civil servants works part-time and is therefore affected by the relevant provisions of the BeamtVG. In those circumstances it must be held, on the basis of the information provided in this regard by the referring court, that provisions such as those at issue in the main 31 The Claimant accepts that no such principle can be extracted from European case-law dealing with linear pension schemes. In such cases it matters not whether you use PC X or PC Y – in all cases the hourly rate of accrual will be the same. As a result, where linear schemes have been considered by the Court, the language used has been wide enough to accommodate both bases of comparison: see the Claimant’s original submissions para 59 and the cases cited therein. 32 See AG [81] 33 See paras [69] to [74] Page 23 of 37 proceedings may result in discrimination against women by comparison with men in breach of the principle of equal pay for men and women for equal work, unless the provisions are justified by objective factors unrelated to any discrimination on grounds of sex’.[110]I therefore accept the submissions of both the Respondent and the Secretary of State’s that neither the Advocate General or the Court were ever called upon to consider the respective merits of comparator X and comparator Y. The focus of the legal argument was on objective justification. To the extent that there were references to comparator X these were simply to give the objective justification arguments some context. As the Respondent puts it “The ECJ looked at justifications not comparisons’34.[111]Secondly, although the Claimant made much of the fact that the Advocate General referred to comparator X when explaining that there had been unfavourable treatment under the BeamtVG and his comments were then accepted by the Court, that does not in my view establish as a general principle that comparator X is the appropriate comparator in other non-linear pension schemes.[112]It is easy to see why comparator X might have been chosen in the Schonheit case. As noted above, under the transitional provisions Mrs Schonheit’s part-time service was effectively taken into account twice. The first occasion was under Step 2 when her pensionable service (which had been reduced to reflect her part-time working) was used as the basis of the calculation to give a figure of 70.79%. The calculation at Step 2 therefore already reflected and took account of the fact she was a part-time worker. However because she was part-time, she was then subjected to a second adjustment when the pension abatement was applied in Step 3. This took her down to 65.8%. As demonstrated by Graph A, the effect of the abatement is that the part-time worker’s service (at Line B) is disproportionately penalised as compared to Line C (comparator X).[113]However the fact that both the Advocate General and the Court accept the use of comparator X to show disadvantage under the BeamtVG scheme, does not in my view establish a general principle, particularly as neither was called upon to consider whether a comparison with comparator X would be appropriate in other non-linear schemes. The broader context was simply never considered or addressed. That is a significant omission, particularly as non-linear schemes can come in many different forms.[114]Thirdly (and this is to some extent an extension of the last point) I agree with the Respondent’s submission that the rationale for using comparator X only benefits a parttime worker in the context of a scheme like the BeamtVG, where the rate of accrual diminishes the longer the period of service35. However if the rate of accrual were to be reversed (for example accruing at 1 per cent for the first 10 years, 2 per cent for the next 15 years and 3.5 per cent for the next 5.5.years) then the Claimant’s approach would disadvantage part-time workers. The following example demonstrates this point:-(i) A full-timer with 30 years service (PC Y) would be entitled to 57 per cent of pensionable salary (10 x 1% = 2 x 15% = 5 x 3.5%);(ii) If this calculation were to be pro-rated using the formula in paragraph 4A for a parttimer who has worked half time for 30 years (ie using comparator Y), then the part-timer would receive a pension of 28.75 per cent; 34 See para 60 of the Respondent’s supplementary submissions 35 See the Respondent’s supplementary submissions at paras 58 and 59. Page 24 of 37(iii) However if you adopt the Claimant’s approach and base the part-timer’s benefits on comparator X (who would have 15 years service) then the calculation gives a pension of 20 per cent (10 x 1% = 5 x 2%), 8.75 lower than using PC Y. Given such a different result it is inherently unlikely that the comparison referred to in Schonheit was intended to apply to all non-linear pension schemes.[115]Fourthly the factual matrix of the Claimant’s case is very different to that in Schonheit. Under the BeamtVG, as we have seen, there was a double abatement mechanism that served to reduce the pension of part-time workers disproportionately in relation to hours worked. By comparison in the Claimant’s case there is only one pro-rating exercise by virtue of which her pension benefits are reduced in exact proportion in terms of hours worked to those received by a full-time officer. It is therefore far from obvious that the Court would have adopted comparator X in her case – particularly as the EC J accepts that the principle of pro rata temporis can apply to pension benefits36. Fifthly if the Schonheit case were to compel a comparison with comparator X then the effect on the 1987 Scheme would be to create indirect discrimination elsewhere in the scheme. So for example, if the Claimant were to receive the same pension as PC X, she would then receive considerably more per hour than say PC Y or PC ZA. It would then be open to a male PC Y or PC ZA to bring a claim of indirect discrimination on the basis that he works full-time but receives less than a part-time worker who worked fewer career hours and for a shorter period. This could then give rise to further ‘piggy-back claims’ by part-time workers. Whilst the potential for piggy back claims is no reason in itself to dismiss an equal pay claim37 it would nonetheless be surprising if the ECJ in Schonheit had endorsed the possibility of such an outcome without explicit discussion of the matter.[116]In my view therefore the Schonheit decision does not assist the Claimant. I am clear that it does not establish a general principle that comparator X applies in all nonlinear pension schemes. I therefore see no reason to make a reference to the CJEU (as the Respondent had urged me to do if there was any doubt on the matter).[117]Having made my decision in relation to the Schonheit case it is therefore necessary to consider the comparator issue in relation to each of the three claims brought by the Claimant, applying established principles. The Part-time Workers claim[118]As noted above, Regulation 5(3) of the PTW Regulations provides that when determining whether a part-time worker has been treated less favourably than a full-time comparator, the pro-rata principle ‘shall be applied unless it is inappropriate’ (emphasis added). In deciding on the correct comparator for the purposes of the part-time workers claim I therefore have to decide whether the pro rata principle (or the principle of pro rata temporis under European law) should be applied to the calculation of pension benefits in this case. The Respondent and the Secretary of State argue that application of that principle is appropriate, that it has been applied in accordance with legal principle and that it ensures that there has been no less favourable treatment of the Claimant. The Claimant for her part says that application of the principle is not appropriate, that it works unfairly and that a comparison with PC X should be preferred, thereby entitling her to a higher 36 See para [90] 37 The case law on equal pay expressly recognises the possibility of a piggy-back claim: see Hayward v Cammell Laird Shipbuilders Ltd [1988] IRLR 257 at paragraphs [38] – [39]. Page 25 of 37 pension. Consideration of general principles[119]The application of the pro-rata principle was considered by His Honour Judge Richardson in James v GNER38at paragraphs [49] – [50]. He stated as follows: ‘The starting point is that the pro rata principle is to be applied unless it is inappropriate… But there are limits to its application. In its own terms it applies only to ‘pay or any other benefit’. The [PTW Regulations] apply, of course, to discrimination over a wider field. Moreover not all forms of pay or benefit will be susceptible to its application. In Matthews and ors v Kent and Medway Towns Fire Authority39 the tribunal held that it was inappropriate to apply the pro rata principle over the whole range of a financial package for fire fighters which included pension benefit, sick pay and pay for additional duties. This conclusion was not challenged in the Appeal Tribunal. A Tribunal addressing itself to Regulation 5 should always consider whether it is appropriate to adopt the pro rata principle. A Tribunal should bear in mind the fundamental purpose of the pro rata principle is to enable a valid comparison to be made between the remuneration of a part-time worker and his full-time counterpart, so as to identify whether a part time worker is being treated less favourably and if so to what extent. It is a tool for determining whether a part time worker has been treated less favourably than a full time employee… It will generally be a useful exercise for a tribunal to consider carefully how the pro-rata principle might apply to the case before it.’[120]As the passage from James v GNER emphasises, the pro-rata principle is only to be applied to ‘pay or any other benefit’ and even then it will not be appropriate to use it in every case. That makes good sense. Obviously there are some benefits where pro-rating will simply not work. The provision of safety helmets is one obvious example. On the other hand, where a benefit is easily divisible – as will almost certainly be the case where the benefit is pay or pension entitlement – then it would appear to offer a sensible and appropriate means of establishing whether there has been less favourable treatment. Certainly there appears to be no domestic authority to the effect that, as a matter of general principle, it is ‘inappropriate’ to adopt the pro rata principle in the context of pension benefits. Indeed to do so will usually reflect the expectations of the workforce. A part-time worker would not expect to receive the same benefits as a comparable full-time worker in terms of salary, holiday pay, pension or other such divisible benefit. They will however expect to be paid - or receive the benefit – on a proportionate basis.[121]The approach adopted under the PTW Regulations accords with the European jurisprudence. So for example in the Schonheit case the Court stated:- ‘…it should be stated at the outset that, as the Advocate General has noted at point 102 of his Opinion, Community Law does not preclude a retirement pension being calculated pro rata temporis in the case of part-time employment’40. 38 UKEAT/0496/04 39 [2003] IRLR 732 40 See para [90] of the Schonheit decision. The Claimant accepts that, according to EU jurisprudence, pay under the PTW Directive is equated with pay under Article 157 (formerly Article 141) and that no distinction is made between the application of the principle of pro rata Page 26 of 37[122]The Court then explains how the pro-rata principle might be applied in a nondiscriminatory way:- ‘The fact that, in addition to the number of years spent working in the civil service, an official’s actual period of service during those years, as compared with the actual period of service of an official who has worked on a full-time basis throughout his career, is also taken onto account is an objective criterion unrelated to any discrimination on grounds of sex, allowing his pension entitlement to be reduced proportionately. Paragraph 6 of the BeamtVG, pursuant to which periods of part-time employment are pensionable only in such proportion as the reduced working time bears to normal working time, gives effect to an objective criterion of that kind.’[123]As noted above41, paragraph 6 of the BeamtVG determines pensionable service in much the same way as pensionable service is calculated under the 1987 Regulations. This passage therefore suggests that the type of formula contained in paragraph 4A of the 1987 Regulations, whereby the full time pension is pro-rated exactly in proportion to relative lengths of pensionable service, is – in principle at least - permissible under European law. Should the pro rata principle be applied in this particular case?[124]I therefore turn to consider whether it would be appropriate to apply the pro rata principle in this particular case.[125]In her original submissions Miss Romney QC appeared to concede that the answer to that question is ‘yes’. As she put it at paragraph 57:- ‘Under EC jurisprudence, pay under the PTW Directive is equated with pay under Article 157… whenever a benefit falls within ‘pay’ under Article 157 the principle of pro rata temporis is ‘appropriate’ under the PTW Directive’. 42[126]However during the course of oral submissions Miss Romney abandoned this concession and proceeded to argue that the pro rata principle is inappropriate in this case. This no doubt reflects the fact that whilst a comparison with PC Y involves a pro-rating exercise, the comparison with PC X plainly does not. If PC X is adopted as a comparator there is simply an aggregation of the number of hours worked, which is then used to calculate pension entitlement under paragraphs 2-4.[127]Miss Romney’s argument that application of the pro rata principle is inappropriate is put on two alternative grounds43. The first is that in calculating the Claimant’s pension under the 1987 Regulations the Respondent had pro rated twice. The second is that if the pension is pro-rated in accordance with the 1987 Regulations then the part-time worker’s pension would be reduced disproportionately by awarding her a lower pension than a fulltemporis to part-time workers under the two regimes (see the Claimant’s original submissions para 57). 41 See footnote [22]. 42 She cited the CJEU judgment in Osterreichischer Gewerkschaftsbund v Verband Osterreichischer Banken C-476/12 as authority for this proposition. 43 See para [16.4] of her supplementary submissions. It should be noted that Miss Romney conceded in oral submissions that this case was not one of those ‘anomalous’ situations identified by Judge Richardson in James v GNER where it might be inappropriate to apply the pro rata principle. Her argument was therefore based solely on the two arguments described. Page 27 of 37 time officer working the same number of hours overall across his career. I will consider each of these arguments in turn. Double pro-rating[128]The Claimant’s argument that there has been double pro-rating proceeds on the basis that a pro-rating exercise is first performed when pensionable service is calculated under part IV of Schedule J of the Regulations (thereby reducing the Claimant’s actual service of 12 years and 113 days to a figure of 11 years and 247 days). It is then said that a second such exercise occurs when the calculation of the Claimant’s pension benefits is performed under paragraph 4A, when the formula N x R/Q is applied.[129]However this is not an accurate analysis of what takes place. Bearing in mind that the pension calculation is based upon the full-time equivalent salary, the figure for pensionable service is simply an aggregate figure, measuring the Claimant’s service over her entire career. It involves no pro-rating whatsoever. All it does is to convert the part-time worker’s weekly hours into years and fractions of years so that a pro-rating exercise can take place at a later stage when the formula under paragraph 4A is applied. It is at that point, and at that point only, that the full timer’s pension is pro-rated in proportion to the time worked viz a viz the full-time worker.[130]I therefore do not accept the premise on which this part of the Claimant’s argument is based. Only one pro-rating exercise is carried out under the 1987 Regulations44. I am therefore not persuaded that this argument provides any basis for finding that the pro rata principle (and the adoption of comparator Y) is inappropriate in this case. Disproportionate reduction in benefits[131]I now turn to the argument that the pro rata principle is inappropriate because it reduces the Claimant’s pension benefits disproportionately, resulting in her being awarded a lower pension than a full-time officer who has worked the same number of hours overall across his career. Given my finding in relation to the Schonheit case, to the effect that that there is no general principle of EU law that compels me to accept this proposition, I therefore considered the Claimant’s argument in light of the relevant statutory provisions and the operation of the 1987 Scheme. I was not persuaded by the Claimant’s arguments for a number of reasons.[132]Firstly I do not accept that the calculation performed under paragraph 4A of the 1987 Regulations produces a disproportionate reduction in the Claimant’s benefits. On the contrary the calculation used in paragraph 4 adopts the pro rata principle exactly as it is defined in Regulation 1(2) of the PTW Regulations. That provision makes it clear that the pro-rating exercise is to be based on ‘the proportion of that pay or other benefit that the number of the [part-time worker’s] weekly hours bears to the number of weekly hours of the comparable full-time worker’. Regulation 1(2) therefore requires that the comparison between the Claimant and the full-time worker be done by reference to the weekly hours 44 As pointed out in paragraph 129, N is calculated using the equivalent full-time salary. If it were otherwise, and the calculation were to be based on the Claimant’s actual salary, then she would effectively be paying twice for being part-time: once by a reduction in pensionable salary and a second time by a reduction in reckonable pensionable service. There would then truly be a double pro-rating. However that is not what occurs under the 1987 Scheme. Page 28 of 37 that each of them has worked45.[133]This is precisely what the calculation under paragraph 4A achieves. The Claimant’s lower entitlement when compared to PC Y is exactly pro rata to the relative hours worked during the comparison period. As the Respondent puts it at paragraph 40 of its supplemental skeleton:- ‘The approach provided for in the 1987 Regulations is to take a week by week pro rating as with Regulation 1(2) but then to extrapolate across the career of a particular police officer. This works by not just comparing between a part-time worker and a full-time worker in week one but by adding up successive weeks until they take their pension. That is an aggregation of each week’.[134]Graph B, provided by the Secretary of State, helps to illustrate this point. It seeks to demonstrate how the respective benefits of full and part-time workers are calculated under the 1987 Scheme. The horizontal axis represents years of pensionable service and the vertical axis the fraction of pensionable pay (expressed in sixtieths) to which that officer is entitled. Line A (the green line if the graph is in colour) shows the accrual of pension rights for a part-time worker working 90 per cent of full-time hours (ie a simplified version of the Claimant). Line B (the blue line) is comparator PC Y who works full-time and has the same actual length of service as the part-time worker. As the Secretary of State explains:- ‘It can be seen from this that the part-time worker’s pension follows that of comparator PC Y, perfectly in proportion. That is the case whenever they retire, whether or not retirement is during years 10–13 when rights are frozen…[135]The effect is that when the calculation under paragraph 4A is performed, the Claimant receives 94.863% of the pension awarded to PC Y. This reflects the fact that she has worked 94.863% of the hours worked by PC Y who was full-time over the same period. This means – as the Claimant concedes - that the Claimant’s pension accrues at the same hourly rate as PC Y. Moreover this will be true at whatever stage in her career the calculation under paragraph 4A is performed.[136]The same point is made in the Respondent’s supplementary submissions as follows:- ‘The Claimant argued that part-time workers lose out during the 10-13 year plateau period. That is plainly inaccurate… When the full-time worker receives an acceleration, the part-time worker also receives an acceleration in proportion to service. When the full-time worker plateaus, the part-time worker plateaus proportional to their service… They each take the rough with the smooth in proportional measure.’[137]I therefore accept the Respondent’s submission that paragraph 4A is therefore entirely faithful to Regulation 1(2)46. The Claimant receives that proportion of the benefit in 45 In adopting a week by week comparison, Regulation 1(2) appears to be entirely consistent with the Framework Agreement which gives member States considerable latitude as to how the principle of pro rata temporis should be applied: see para 70 above. 46 In coming to this conclusion I do not accept, as the Claimant, argues, that this analysis confuses the question of how a person qualifies as a full-time or a part-time worker under the PTW Regulations with the question of whether there has been less favourable treatment for the purposes of Regulation 5(3). Whilst the difference in weekly hours worked is clearly a relevant factor in determining who falls within the definition of a full-time worker within Regulation 2, that does not mean that a comparison based on weekly hours is inappropriate when determining Page 29 of 37 question (in this case pension) that the number of her weekly hours bears to the number of weekly hours worked by PC Y. To adopt the wording of Judge Richardson in James v GNER, the effect is that a valid comparison can be made between the pension entitlement of the part-time worker and her full-time counterpart.[138]By contrast the Claimant’s chosen basis of comparison (which is based on the total number of hours worked) involves no pro-rating exercise and produces quite arbitrary results. At some stages the comparison with PC X will favour the part-time worker but at others it will not. So, for example, a comparison with PC X is beneficial in the Claimant’s case because she makes the comparison within the 10-13 year window. However if the calculation is done at a different stage (say between 5 and 10 years’ or over 20 years pensionable service) then the part-time worker’s hourly rate under paragraph 4A would be higher than that of PC X. Conversely if she has less than five years’ service she tends to lose out when compared to PC X because she will not be able to take the benefit of the accelerated accrual rates built into the scheme.[139]These anomalies are demonstrated by Graph C, provided by the Respondent. The horizontal axis shows years of actual service and the vertical axis shows pension entitlement. Line A (blue) represents PC Y, Line B (green) represents a part-time worker who has worked half time throughout (to keep the example simple) and Line C represents PC X. It can be seen that if PC X is taken as the comparator then the part-time worker is disadvantaged after 5 years. That disadvantage continues, as the graph shows, until year 11. It is only at that point that the comparison with PC X provides any advantage to the part-time worker.[140]Moreover it should be noted that the length of disadvantage viz a viz PC X depends on the number of part-time hours worked in a week. Given the mathematics of the 1987 Scheme, the fewer the hours the greater the disadvantage. So for example, someone who works only 10 hours per week (ie 25% of full-time) would not gain any benefit by comparing themselves with PC X until year 20. As the Respondent puts it ‘The Claimant’s comparison creates an irrational lottery for part-time workers’47.[141]I am therefore being asked to replace a method of comparison under paragraph 4A which produces a proportionate and internally coherent calculation of benefits as between full and part-time workers and which is entirely consistent with the pro-rata principle as defined in regulation 1(2) of the PTW Regulations with an alternative comparison that produces arbitrary results and which only works in favour of the Claimant because she happens to have accrued pensionable service to a particular level. I am aware of no authority that compels me to reach this conclusion and I cannot see any good reason for doing so.[142]My second reason for finding that the pro rata principle is appropriate in this case, so as to find in favour of PCY rather than PC X, is that the Claimant’s argument appears to be based on a misconception of how the 1987 Scheme operates. Having considered the 1987 Scheme I am persuaded that the reason why the Claimant appears to be treated less generously than PC X has nothing to do with her part time status. It arises solely because the issue of less favourable treatment. As the wording of Regulation 5(3) makes plain, both the existence of a full-time comparator and the application of the pro-rata principle will normally be relevant when determining whether there has been less favourable treatment. The fact that weekly hours may be relevant to one of these concepts does not preclude its use in connection with the other. 47 Respondent’s supplementary submissions para 51(4). Page 30 of 37 of the stage in her career at which she has retired48.[143]As noted above, the effect of the calculation in paragraph 4 is that as a full-time officer progresses through the period of 10 – 13 years’ pensionable service, their pension entitlement stays flat yet the number of hours they have worked increases, thereby diluting their hourly rate of pension accrual. So for example we saw at paragraph 56 above that PC Z (who has worked full-time for 10 years) accrues pension at a rate of £0.598 per hour. That contrasts with PC ZA who has worked full time for considerably longer -13 years - and yet accrues pension at only 0.46p per hour (see paragraph 59 above).[144]In other words, the effect of the plateau created by paragraph 4 between 10 and 13 years pensionable service is that full-timers will always have a higher hourly rate at 10 years service than they do at 13 years’ service. What paragraph 4A does is to ensure that the same is also true for part time workers, albeit in proportion to the full-time worker. The result is that the hourly rate of accrual for part time workers with pensionable service of between 10 and 13 years reduces at the same rate as it does for full-time worker workers. The Secretary of State explains the point this way:- ‘The overall effect of the provisions is that in years 10-13 the full-time worker has a pension of 20/60 (no more or less) and the part-time worker has that pension reduced pro rata (under regulation 4A) for her part-time working (no more or less). Both types of worker are subject to a ceiling (not a guarantee) during those three years…’49[145]The Secretary of State goes on to explain what this means for overall hourly rates by reference to Graph B:- ‘During the ‘flat years’ 10-13, any worker who retires at a point further to the left on the graph will achieve an overall hourly rate greater than that achieved by another worker (part-time of full-time) who is further to the right (ie who has worked for longer but is still in years 10-13). During that flat period, the scheme becomes steadily less generous as those years pass. It has that effect regardless of anyone’s part-time or full-time status.’50[146]This explains why under the 1987 Regulations the Claimant has an hourly rate of £0.486, yet PC ZA (who has worked full-time for 13 years) has an hourly rate of £0.46. It is the point on the flat line that is crucial not whether the individual worker is full or part time. This means that if the Claimant takes a full-time comparator who is on the flat line and to her left (PC X) she will always establish less favourable treatment. However the same is true of a full-time officer. So for example a full-time officer who retires at the same point on the line as the Claimant would also have a lower hourly rate of accrual than a part-time officer who is situated further to the left on the flat line. That is an inevitable function of the three year ‘freeze’ on benefits.[147]I therefore accept the Secretary of State’s submission that the disparity the Claimant identifies has nothing to do with her part-time status: it is simply the result of the flat progression achieved by the minimum entitlement in years 10-13 under paragraph 4. It exists only because she makes the comparison within the 10-13 year window. So whilst it 48 Strictly speaking, in the Claimant’s case it is because of the stage at which she ceased to be a member of the scheme (see para 16 above, making the point that she was a member of a new pension scheme for a short period before she retired). However this makes no difference to the point I am making. 49 See the Secretary of State’s supplementary submissions at para 11 50 See Secretary of State’s supplementary submissions paras 15 and 16. Page 31 of 37 is true to say that the 1987 Scheme produces a number of anomalies they have nothing to do with the Claimant’s part-time service. For example, a full-time worker retiring with 13 years’ pensionable service may well feel exactly the same sense of unfairness as the Claimant when comparing his or her pension entitlement to that of an officer with ten years’ pensionable service. However that is the result of the design of the scheme – any unfairness exists regardless of full or part-time status.[148]It follows from this that I do not accept the Claimant’s analysis of the 1987 Scheme to the effect that full-time officers with more than 10 years’ reckonable service are entitled to the benefit of a pension of not less than 20 sixtieths of pensionable pay, whereas she, as a part-time worker, is not. That is to misunderstand the nature of the scheme and the effect of the calculation under paragraph 4A. On a proper analysis paragraph 4A ensures that the Claimant receives exactly the same entitlement as PC Y, subject only to prorating. As a result, at whatever stage the Claimant retires, her pension entitlement mirrors the varying accrual rates applicable to PC Y but in proportion to the hours she has worked. As the Respondent puts it:- ‘… [The formula under paragraph 4A] allows part-time workers to take the benefit of all accelerated progression periods (and likewise the flat periods) and to do so in equal proportion to a full-time worker based on their hours’ (see paragraph 21 of the Respondent’s supplemental submissions).[149]This seems to me to be fair and consistent and entirely in accordance with the pro rata principle. The Claimant and PC Y accrue pension at the same hourly rate because under paragraph 4A the Claimant’s entitlement reduces in precisely the same way as for a full-timer. This achieves equity between full-time and part-time workers.51[150]Conversely if the comparison with PC X were to be adopted throughout the scheme that would disadvantage other part-timers (such as those who retired with between 5 and 10 years’ pensionable service) who would lose out significantly when they retired. They of course have not been represented in these proceedings.[151]Finally there is a reference in the Claimant’s submissions to the fact that she has not only worked the same number of hours as PC X but also paid the same pension contributions52. However I do not consider that this adds anything to the Claimant’s case. In the first place I query whether the issue of pension contributions should carry much weight, given that the 1987 scheme is a ‘defined benefits scheme’ whereby pension entitlement is dependant upon length of service and average final salary rather than the amount of contributions made. Secondly and in any event I do not think the comparison helps the Claimant’s case. The issue of pension contributions throws up as many anomalies as the issue of hourly rates. Whilst it is true that the Claimant has made the same contributions as PC X, she has made considerably fewer contributions than both PC Y and PC ZA (significantly so when compared to PC ZA who has worked full-time for 13 years). Yet if PC X is the appropriate comparator she will receive the same pension benefits as they do. By comparison if PC Y is the comparator then, because of the application of the pro rata principle, the Claimant’s pension entitlement is proportionate to 51 There also appears to be a ‘belt and braces provision in the 1987 Regulations in that paragraph 6 of Part III of Schedule B provides that ‘If in a case where any of the policeman’s [sic] service …was part-time service, the amount of the pension calculated in accordance with paragraphs 1 to 5 of this Part would be less than it would have been if the person had become entitled to receive the pension at an earlier date, then…the pension shall be of that amount instead’. This ensures that just as with full-time workers, the part-time worker’s pension rate cannot go downwards. 52 See the Claimant’s supplementary submissions para 9. Page 32 of 37 her contributions when compared with PC Y.[152]I am therefore persuaded that it is appropriate to apply the pro-rata principle in this case and that it has been applied exactly in accordance with Regulations 1(2) of the PTW Regulations. PC Y, rather than PC X, is therefore the appropriate comparator for the purposes of the part-time worker claim. The Equal Pay claim[153]The Claimant also brings an equal pay claim. The statutory provisions governing such a claim are of course different to those which apply to a part-time workers claim. However it would be surprising if the two yielded a different result in this case because the equal pay claim rests entirely on the contention that there is discrimination against the Claimant as a part-time worker. Nonetheless I have considered the arguments relating to the equal pay claim separately to see if they point to a different conclusion.[154]Clearly the main plank of the Claimant’s case on equal pay – reliance on the Schonheit decision - has failed. Equally clearly, although Article 157 does not refer to the principle of pro rata temporis expressly, EU jurisprudence accepts that the principle can apply in the context of equal pay. The question is therefore whether it should do so in this case, so that PC Y is the correct comparator in this context as well as under the PTW Regulations, or whether legal principle points to a different comparator – PC X - in the context of equal pay. I have decided that PC X is not the correct comparator for the reasons set out below. Is the Claimant allowed to select her own comparator?[155]The Claimant’s first argument in favour of PC X relies on well-established case-law in the field of equal pay, to the effect that a woman claiming equal pay is entitled to select the comparator of her choice. The authority for this proposition is Ainsworth v Glass Tubes & Components Ltd53, cited with approval by Lord Templeman in Pickstone v Freemans plc54. As a result when a claimant wishes to show that she has been doing equal work (be it like work, work rated as equivalent or work of equal value) then it is for her to choose her own comparator and it is not for the employment tribunal to usurp that choice. On this basis Miss Romney argues that the Claimant has the right to rely on PC X and the Tribunal has to respect that choice.[156]I am not persuaded by this argument. In my view it confuses the choice of comparator for the purposes of establishing whether there is equal work (which is not an issue for this preliminary hearing) with the question that I have to determine, which is essentially whether the Claimant has suffered indirect discrimination. As far as the former question is concerned there is no question that she is free to compare herself with whomsoever she chooses. However as regards the latter exercise she does not have that freedom. To establish indirect discrimination there must be, amongst other things, an apparently neutral PCP that has a disproportionate impact on the protected class. The identification of the correct comparison for the purpose of showing group disadvantage is clearly a matter for the Tribunal as the voluminous case-law in this area demonstrates55. 53 [1977] IRLR 74 54 [1988] IRLR 357 at [26] 55 See for example Homer v Chief Constable of Yorkshire Police [2012] UKSC 15 and more recently the combined appeal in Essop and others v Home Office; Naeem v SOS for Justice [2017] UKSC 27 dealing with this issue. Page 33 of 37 ‘The term by term argument’[157]Secondly the Claimant argues that the approach adopted by the Respondent and the Secretary of State is at odds with another well-established principle of equal pay law, namely that the sex equality clause (or in this case the sex equality rule) operates on a term by term basis.[158]The leading authority on this point is Hayward v Cammell Laird Shipbuilders Ltd56. In that case the House of Lords concluded that, for the purposes of equal pay, a woman was entitled to be treated no less favourably than a comparable man under each individual provision of her contract, regardless of the position under the contract as a whole. As a result it was not open to the employer to argue that although Miss Hayward (a canteen cook) received lower basic pay and overtime rates than her male comparators (who were painters, engineers and joiners) she nonetheless was better off overall because of more favourable terms relating to meal breaks, holidays and sickness benefits. Under the equal pay provisions a term by term approach had to be applied.[159]Miss Romney argues that analysis of the 1987 Scheme by the Respondent and the Secretary of State offends against the principle enunciated in Hayward because they are looking at the pension scheme in the round. In particular she points to the graphs they have provided to show that at some points under the 1987 Scheme a part-time officer will get a higher hourly rate of pension than PC X (eg where the part-time worker has worked more than 20 years). She argues that this is irrelevant. I should consider only that portion of the scheme that is relevant to the Claimant’s claim.[160]However I do not accept this argument. This is not a case where the Respondent is arguing that a less favourable term viz a viz the Claimant can be counterbalanced by another more favourable term elsewhere. The question is whether there is less favourable treatment in the first place. In determining that question I do not accept that it is inappropriate to consider the scheme as a whole, particularly as in my view it is necessary to do so to understand how the scheme operates and to establish whether it is appropriate to apply the pro rata principle. I therefore do not accept that the principle in the Hayward case undermines the choice of comparator Y as the correct comparator for the equal pay claim57. Should there be a like for like comparison?[161]The Respondent and the Secretary of State for their part argue that the adoption of PC Y as the comparator accords with legal principle in the area of equal pay and in particular with the need for a ‘like for like’ comparison. This concept is found in section 23(1) EqA 2010, which provides as follows:- ‘On a comparison of cases for the purposes of section 13 [direct discrimination], 14 [combined discrimination] or 19 [indirect discrimination] there must be no material difference between the circumstances relating to each case.’[162]The purpose of this section is ‘simply to emphasise what would be necessarily 56 [1988] ICR 464, HL 57 In any event the comparison the Claimant is asking for is rather different to that in Hayward. The Claimant wants the operation of paragraph 4A to be considered, but only at a particular, rather limited, stage of pensionable service. That seems to me to be a different situation to the one in Hayward where a straightforward term by term comparison was made. Page 34 of 37 implicit in section 19, namely that like must be compared with like’58. According to the Respondent and the Secretary of State that is what is called for in this case: a like for like comparison with no material difference in circumstances, namely PC Y.[163]The Claimant for her part argues that this argument is misconceived and at odds with the equal pay provisions in the EqA 2010. In the first place comparisons are frequently drawn in equal pay cases between those doing different jobs in different places, for example when establishing if there is work of equal value. There is therefore no justification for a like for like comparison. However in my view once again this argument confuses the comparison to be drawn for the purposes of establishing equal work and the comparison that I am now being asked to consider. They are not one and the same.[164]Secondly the Claimant argues that section 23 has no relevance to the equal treatment provisions in Chapter 3 of Part 5 of the EqA 2010. It is found in a completely different section of the Act and in any event the equal pay provisions have their own section dealing with the issue of comparators, at section 79. (That is the section that requires a comparator to be employed by the same employer or associated employers at the same establishment, or at different establishments at which common terms and conditions of employment are observed either generally or for employees of the relevant classes.) She argues that section 79 does not contain any wording analogous to that found in section 23. Moreover section 23 is expressly limited to comparators under section 13, 14 (not in force) and 19 of the EqA 2010 and there is nothing to suggest that its provisions should be read across into the equal pay provisions. It therefore has no part to play in an equal pay claim.[165]I have decided that section 79 does not preclude a like for like comparison in this context. In my view that section is concerned solely with the ‘gateway’ question of whether there is equal work. It has no bearing on the question whether there has been indirect discrimination - which is the issue that gives rise to the comparator issue I have to determine.[166]Moreover whilst I accept that section 23(1) is not referred to expressly within the equal pay provisions of the EqA 2010, it is clear that principles of direct and indirect discrimination are imported into the material factor defence under section 69(4) EqA 201059. As a result, if this importation is to make any sense, the principle of comparison set out in section 23 must apply. The Respondent explains the point in this way:- ‘Indirect discrimination is relevant to the question of material factor under s69. …It is to that end that the Claimant alleges indirect discrimination. This requires there to be no material difference in circumstance. This can be seen by analogy with s23 EqA 2010. Whilst s 23 applies expressly to ss 13, 14 and 19, the Claimant’s reliance upon indirect discrimination requires her to demonstrate a particular disadvantage (as with s 19). This begs the question, disadvantaged to whom? It would be impossible to establish disparate impact by reference to the s79 criteria (same establishment, common terms etc). It is however the like for like comparison as with s23 that enables the question to be answered.’[167]That being the case I accept that there is no reason in principle why a like for like comparison cannot be drawn for the purposes of the equal pay claim. I agree with the Respondent’s submission60 that a comparator who is like for like with no material 58 Per Underhill LJ in the Court of Appeal in Naeem v Secretary of State for Justice 59 See para 74 above. 60 See the Respondent’s supplementary submission at para 30. Page 35 of 37 difference in circumstance (save for part-time working) will be a full-time worker with the same actual service as the Claimant (PC Y). This is a much truer comparison than PC X who is an artificial construct based on the calculation of reckonable service.[168]The authorities cited by the Claimant do not appear to be at odds with the approach I have adopted. So for example in Kuratorium (the training session case) both full-time and part-time workers attended the same training session and therefore one could be compared with the other. As they received different pay for that same session, the claim succeeded: like was being compared with like.[169]However in Helmig the part-time workers lost because a comparison with the fulltime worker’s hours in excess of 38.5 was not comparing like with like. When the proper like for like comparison was carried out it could be seen that they were paid the same as full-time workers for hours 20 to 38.5.[170]I have therefore decided that comparator PC Y is also the appropriate comparator for the purposes of the equal pay/Article 157 claims61.
Conclusion
[171]If, as I have found, PC Y is the correct comparator then there is no less favourable treatment of the Claimant as benefits are pro-rated precisely in accordance with hours worked. Prior to this hearing both the Respondent and the Secretary of State had anticipated that if I were to find in their favour on this issue that this would inevitably lead to dismissal of the claim. That had also been my understanding when reading into the case.[172]However during oral argument at the hearing Miss Romney argued that this would not necessarily be the case. She argued that there is an alternative basis for establishing less favourable treatment under the pleaded statutory provisions, namely that:- ‘… PC Y is afforded under Paragraph 4 of the 1987 Scheme an entitlement (to which part-time officers are deprived under paragraph 4A) to a pension of a minimum of 20/60’s average pensionable pay to reward 10 to 13 years reckonable service’.[173]I confess that I struggle to see how this is a different point to the one argued before me and indeed it is referred to by the Claimant at several points in her original submissions62 and responded to by both the Respondent and the Secretary of State.[174]In any event, as explained in paragraphs 149 to 150 above, I have found the argument to be misconceived. The 1987 Scheme does not confer a benefit on full-time workers with more than ten years’ service, which is then denied to part-time workers. On the contrary the part-time worker receives exactly what a full-timer with the same actual length of service receives, subject only to pro-rating. The Secretary of State puts it this way:-63 61 There is a reference in the Claimant’s original submissions to Article 157(2)(b) which provides that ‘pay for work at time rates shall be the same for the same job’. I do not consider that this compels an hour for hour comparison as it is clearly dealing with a discrete topic and has no direct bearing on calculation of pension entitlement. 62 The issue is dealt with by the Claimant in paragraphs 23 to 25, 69 and 72 of her original skeleton and illustrated by reference to the worked examples in paragraphs 28 to 31. The Respondent puts its case on the matter at paragraphs 23 of its supplementary submissions and the Secretary of State at paragraphs 9 to 11 of the supplementary submissions. 63 See paragraphs 8 to 11 of the supplementary submissions. Page 36 of 37 ‘The overall effect of the provisions is that in years 10-13 the full-time worker has a pension of 20/60 (no more or less) and the part-time worker has that pension reduced pro rata (under regulation 4A) for her part-time working (no more or less). Both types of worker are subject to a ceiling (not a guarantee) during those three years.’ I agree with that proposition. The Scheme does not treat the part-time worker less favourably in this respect.[175]There is one further point to add. I have made a finding in paragraph 147 above that the reason for the Claimant’s treatment under paragraph 4A has absolutely nothing to do with her part-time status. If I am right on this it must follow that this point cannot be relied upon as a material factor defence to the Claimant’s equal pay claim under section 69(4) or as a ‘reason why’ defence under regulation 5(2) of the PTW Regulations.[176]As a result it may be that my findings are in practice sufficient to dispose of this claim. However that is a matter on which the parties are invited to make submissions in accordance with the attached order so that if issues in this matter remain they can be identified and case management decisions taken. If no such issues are identified, I am proposing to dismiss the claim.