Hovell v Ashford & St Peter's Hospital NHS Trust

The particular issue raised in this case is whether an employment tribunal erred in law in refusing an application by the appellant that a decision to appoint an independent expert in an equal value case should be revoked. The principal reason why the appeal has been brought, however, is to question certain observations of the employment judge which, it is alleged, are legally unsound and are improperly influencing the way in which other employment judges are approaching similar applications.

The Equal Pay Act 1970 identifies three ways in which a woman may be able to establish that she should be paid the same as her male comparator. The first is where she and her comparator are employed on like work (section 1(2)(a)). The second is where the jobs have been rated as equivalent in a non-discriminatory job evaluation study: section 1(2)(b). The third is where the work is found to be of “equal value” to that of the comparator: section 1(2)(c). The equal value route cannot be pursued, however, if either of the other two routes is applicable to the work on which a woman is employed.

Section 1(5) of the Act defines the circumstances where a woman is able to claim that her job is rated as equivalent:

Once it is established that the woman is employed on like work or work of equal value, she will be entitled to recover back pay for a period of six years provided both she and her comparator have been employed throughout that period, and that there has been no material change in the nature of the work carried out by either in the interim. In those circumstances it will be possible to say that the claimant and her comparator must at all material times have been employed on like work, or work of equal value, as the case may be.

That is not, however, an inference that can properly be drawn where the woman establishes equality under section 1(2)(b). The point has recently been established by the Court of Appeal (Mummery and Smith LJJ; Lindsay J) in Redcar and ClevelandBorough Council v Bainbridge and Ors[2009] ICR 133. The Court held that as a matter of language and logic it was not possible to give retrospective effect to a section 1(2)(b) claim.

The effect of this decision is that if a woman whose work is rated as equivalent wishes to argue that she should be entitled to equal pay with respect to a date prior to the job evaluation scheme being agreed, she will have to rely on a like work or - and in practice far more likely - on an equal value claim.

The question which then arises is this: if the woman’s job has been rated as equal under the job evaluation scheme, does this mean that the two jobs must have been of equal value before the scheme was agreed (assuming that the jobs have not changed)? Is the job evaluation rating determinative of the equal value claim? This was also an issue considered in Redcar, and the Court held that it was not determinative. Mummery LJ, giving the judgment of the Court said this (para 279):

So the equality of rating is potentially of probative value, but is far from being decisive of the claim.

Where an equal value claim is advanced, the law sets out detailed procedures designed to enable the tribunal to obtain the benefit of expert evidence when determining whether or not the jobs are of equal value. An independent expert can be appointed “to assist the tribunal” by preparing a report on the question of equal value: rule 6(2) of schedule 6 to the Employment Tribunal (Constitution and Rules etc.) Regulations 2004. Counsel accepted that the appointment could be made at any stage of the proceedings, although it will normally be at an early stage.

Rule 10(4) then provides that at any stage in the proceedings the tribunal can withdraw that requirement for the expert to report. This can even be done after an expert has been appointed, although the expert must then be given the opportunity to make representations. It was this power which the employment judge refused to exercise in this case.

With that by way of relevant legal background, I now turn to the facts, which are agreed.

The claimant was employed by the respondent Trust as a social services administrator. Under the NHS Agenda for Change, pay and conditions were reorganised in the context of a job evaluation scheme (JES). This scheme identified 9 pay bands (although band 8 was further sub-divided into four categories). The claimant was placed in band 4, which covers those who had obtained between 271 and 325 points in the JES. She scored 274, that is, just falling within that band. Her three male named comparators scored 296, 298 and 305 points respectively.

The JES was implemented with effect from 1 October 2004. From that time the claimant has received the same pay as her comparators by virtue of section 1(2)(b) of the Equal Pay Act 1970.

However, having achieved that equality, the claimant then sought to establish that she had been entitled to equal pay in the six-year period prior to the implementation of the scheme. In view of the ruling in Redcar, she pursued her case under section 1(2)(c).

On 9 August 2007 an Employment Judge, at the request of both parties and in accordance with usual practice, ordered that an independent expert should be appointed to prepare a report on the question of equal value.

Subsequently, and before any expert had in fact been appointed, the claimant made an application asking the Tribunal to withdraw that requirement. It was not suggested that the circumstances had changed in any material way.

Employment Judge Malone refused that application at a case management hearing on the 12 February 2008. That ruling was unsuccessfully appealed to the EAT before HH Judge Peter Clark, and has now been challenged before this court.

Before the Employment Judge, the principal contention of the claimant was that the fact that her job had been rated as equivalent to that of the male comparators necessarily determined that the jobs were of equal value for the purposes of section 1(2)(c). Employment Judge Malone rejected the submission as being inconsistent with the decision of the EAT in Redcar ([2007] ICR 1644). (At that stage the Court of Appeal had not given their judgment on this point but their decision followed and approved the analysis of the EAT.)

An alternative argument was that the difference in scores in this case was so small that the only proper conclusion was that the jobs were of equal value. The Trust opposed this analysis; it is to be noted that they did not contend that a difference in points demonstrated that the jobs were not of equal value; rather they argued that there is a fundamental difference between a job evaluation study assessing a whole range of jobs and a “one on one” exercise carried out by an independent expert in an equal value case.

The Employment Judge then at para.13.2 and 13.3 identified two preliminary issues which were raised by this alternative way in which the case was put:

The Employment Judge then returned to the particular issue before him and concluded

Finally, the judge sought to lay down the following guidelines for tribunals dealing with such applications in the future (para 37):

On appeal to the Employment Appeal Tribunal, Mr White QC, who also appeared before us, modified the stance that the claimant had initially taken. He expressly disavowed the argument that the job evaluation was conclusive of the equal value claim; rather, it was suggested that the fact that the jobs had been rated as equivalent sets up a presumption that they are of equal value, and that it is for the employer to rebut that presumption. In effect it was an argument that the burden of proof had shifted once the claimant could point to the fact that the jobs had been rated as equivalent. HH Judge Peter Clark rejected that submission. The fact that the jobs were in the same band was, as Redcar made clear, some evidence that the jobs were of equal value but it was not determinative of the question and it did not displace the burden of proof, which remained with the employee.

HH Judge Peter Clark accepted a submission that the judge had erred in treating the case of Home Office v Bailey[2005] IRLR 757 as authority for the proposition that it would be an error of law for a tribunal to find two jobs to be of equal value on the basis that the difference in points in the JES was very small. However, he was satisfied that this error was not material and did not vitiate the exercise of the judge’s discretion.

Mr White contends that the Employment Judge erred in law in the exercise of his discretion. In oral submissions he narrowed the scope of his argument. His central point was that the judge appeared to have thought that if there was any difference at all in the marks given in the job evaluation study, then as a matter of law no tribunal could find that the jobs were equal value without first obtaining the evidence of an independent expert. That was a fundamental misconception which coloured his whole approach to the question of withdrawing the requirement for the expert. It is not clear how the Employment Judge would have exercised his discretion had he properly directed himself, and therefore at the very least the matter should be remitted to the same judge for reconsideration.

Furthermore, even if the court was satisfied that any error of law by the Employment Judge was immaterial and that he would have reached the same conclusion in any event, it is important for the court to indicate that the judge’s approach, and in particular the guidance he sought to give at the end of his judgment (see para 24 above), was wrong.

Mr Bowers QC, counsel for the respondent Trust, submitted that the issue was a simple one. This was a case management decision by the Employment Judge. These are classically difficult decisions to challenge: see e.g. Adams v W Sussex CC[1990] ICR 546, and Noorani v Merseyside TEC[1999] IRLR 184 The decision was plainly justified. The original order to appoint the expert was made with the approval of both parties, and it was conceded that there was no change in circumstances thereafter. It will be very rare to alter an order in such cases. Even if it be the case that the judge was wrong to say that where the claimant received fewer points than the comparator, equal value cannot be established without the support of the report of an independent expert - and Mr Bowers was not seeking to uphold that part of the decision - nonetheless that did not materially affect the way in which the judge exercised his discretion in this case. The judge stated in terms that he did not accept that the difference in the allocation of points in this case was small; he considered it to be significant and it is conceded that he was entitled to do that. Accordingly, the appeal should be dismissed.

In substance, therefore, there are really two issues. First, did the judge make the error alleged? Second, if he did, was it material to the exercise of his discretion?

The appellant put her case in two ways. The first and more limited argument was that it must in principle be open to a tribunal to find that jobs were of equal value in circumstances where the claimant’s mark had been assessed lower than her chosen comparator in the JES, even in the absence of a report from an independent expert.

The second and more far reaching argument was that not only was the judge wrong to say that no equal value could be found where the difference in marks was very small, but that on the contrary, the tribunal ought in such a case to find as a matter of law that the jobs are of equal value. Mere marginal differences demonstrate that the jobs are substantially equivalent, and as a matter of law such substantial equivalence is enough to constitute equal value, at least in the absence of some cogent evidence from the employer demonstrating otherwise. There must be what Mr White termed a “margin of tolerance”.

The differences in the marks in this case were only marginal and attracted this principle. Accordingly, since the claim was bound to succeed, no legitimate purpose could be served by appointing an independent expert, and the only proper exercise of discretion by the judge, once that order had been made, was to revoke it.

I accept that the mere fact that the claimant scores fewer points than her comparator in the JES study does not mean that a tribunal is obliged to conclude that her job is not of equal value unless there is the support of an independent expert. That does appear to have been the understanding of the employment judge in paragraph 30 of his decision (reproduced in paragraph 22 above) and it is too narrow a view. A tribunal may be persuaded that a very small difference in points, particularly in the context of a wide ranging job evaluation study which has focused on benchmark jobs and has not involved a direct comparison of the jobs in issue, does not reflect a material difference in the value of the two jobs.

I agree with the Employment Judge, however, that equal value does not mean nearly equal value, and I reject Mr White’s submission that jobs are equal if they are substantially equal. But the analysis of job value is not a science, and a slavish attachment to the marks scored suggests a degree of precision which the assessment of job value cannot bear. Jobs may be equal in value even though not precisely equal in the points scored. It may be, for example, that a particular factor is given a weighting in the job evaluation study which would not normally be provided for in a simple one-on-one job comparison. Or the employee might be able to show that her tasks have in the past differed, perhaps only in minor ways, from the typical job profile which is measured in the JES, but nonetheless sufficient to bridge any small gap. Perhaps in the JES certain jobs were not the subject of very detailed scrutiny because it was clear that they would fall within a particular grade boundary, and no-one was too concerned precisely where they fell. I do not suggest, and am not in a position to say, that any of these arguments is necessarily available in relation to the NHS claimants, but they demonstrate that there may be a proper basis for a tribunal reaching the decision that the jobs are of equal value even without the benefit of an independent expert’s report.

The converse may also be true; as Mr White conceded, it is conceivable that even where the job of the claimant has been weighted higher than the comparator in the JES, the employer is able to show that the job is not of equal value. I should emphasise, however, that the circumstances would have to be very exceptional before that could be established.

Like the Employment Judge, I do not accept the more extreme submission advanced in the alternative argument. In my judgment, the judge was correct in stating that the fact that there was only a small difference in the points assessed was not of itself capable of establishing that the jobs were of equal value. The claimant who simply says no more than “my job has been rated in the JES within a mark or two of the comparator” cannot by that assertion alone expect to succeed in her equal value claim. At the very least she must set the difference in the context of the particular job evaluation study and explain why the difference should be ignored, and she risks the tribunal not being persuaded by her submission.

Nor do I think that the analysis is assisted by introducing the language of presumptions or shifting the burden of proof. The onus of establishing equal pay always rests on the claimant. The Tribunal will have to determine the issue by considering all the evidence before it. Of course there may be cases where the claimant has adduced evidence making a strong case which will succeed in the absence of any contradictory evidence. But that is frequently the case in any trial and does not involve any formal shifting in the burden of proof.

There are other problems with this more extravagant submission. First, how small must the difference be in order for this principle of equality, or even a presumption of equality, to operate? It is an extremely imprecise principle.

Second, it cannot be assumed that the margin of tolerance, as it was termed, is necessarily applicable at all. The assumption seems to be that because the assessment is imprecise, both jobs could easily have been differently graded, the claimant’s a little higher and the comparator’s job a little lower, such that equality would have been formally established. But that does not follow. The claimant’s job might have been fixed at the highest level it could properly be given (perhaps, for example, the mark given to that job was raised following an appeal), and the comparator’s job graded at the lowest point at which it could sensibly be placed. Or both may have been placed twenty points apart where they properly belong. Even a small difference in points awarded may reflect what is perceived to be a genuine difference in the nature and quality of the jobs being performed, albeit a small one. For example, it may be that the collective parties specifically considered whether the points should be the same and deliberately chose not to make them so in order to mark what they accepted was a material difference between the jobs.

Third, the logic of the appellant’s case would be to obliterate potentially very significant differences in marking. For example, job A may be within 3 marks of job B, job B within 3 of job C, job C within 3 of D, and so forth. On the appellant’s analysis A should be treated as of equal value to B, and B to C, and C to D, and in that way all the jobs in the same grade boundary end up being of equal value to each other, notwithstanding potentially obvious differences between the value of the jobs at the lower and higher ends of the particular grade boundary.

It follows that, in my judgment, the Employment Judge went too far in asserting that a tribunal must have the benefit of an independent expert before it can find equality where the claimant’s job has been marked lower than the comparator’s job in the JES. However, he was right to say that the fact that difference in the points given by the JES is small does not of itself establish that the jobs are of equal value.

The relevant question is whether the error was material in this case. I agree with the EAT that it was not. First, it must be remembered that the judge was faced with a submission to the effect that he should discharge the expert because the case was bound to succeed. It was not suggested that he should revoke the order and allow the claimant to take her chance to pursue her case with the risk of losing. He was right to conclude that it was far from clear that her case would necessarily succeed, and that there was therefore no basis to withdraw the reference to the expert.

Furthermore, he made it plain that he did not accept that the scores were marginal in any event. In his view they were significant and Mr White expressly conceded that he could not challenge that finding. This therefore undermined the premise on which the alternative argument was advanced, namely that the marks were very close. On that basis he was plainly justified in taking the view that the tribunal would benefit from an expert’s report and that accordingly he should not change the original order. Indeed, it may have been difficult for the claimant to make good her case without one in her favour. It should also be remembered that this was an attempt to vary an order already made by consent, absent any change of circumstance. In those circumstances it is particularly difficult to mount a successful challenge.