ON APPEAL FROM THE EMPLOYMENT APPEAL TRIBUNAL
UKEAT/0492/08
Royal Courts of Justice
Strand, London, WC2A 2LL
Before :
LORD JUSTICE SEDLEY
LORD JUSTICE CARNWATH
and
LORD JUSTICE JACOB
Between :
PAUL BUCKLAND | Appellant |
- and - | |
BOURNEMOUTH UNIVERSITY HIGHER EDUCATION CORPORATION | Respondent |
Mr Antony White QC and Mr Tom Brown (instructed by UCU Legal Services) for the Appellant
Mr Jason Galbraith-Marten and Mr Edmund Williams (instructed by Messrs Martineau) for the Respondent
Hearing date: Monday, 8 February 2010
Judgment
Lord Justice Sedley :
The issues
This case, as Sir Richard Buxton noted when giving permission to appeal, raises at least two important and (surprisingly) unresolved questions of employment law:
(i) Whether the conduct of an employer who is said to have committed a fundamental breach of the contract of employment is to be judged by a unitary test or a “range of reasonable responses” test.
(ii) Whether an employer who has committed a fundamental breach of contract can cure the breach while the employee is considering whether to treat it as a dismissal.
The history
The way in which these issues have arisen is described in the decision of an employment tribunal sitting at Southampton in June 2008, to the lucidity and organisation of which I would wish to pay tribute. Taking the facts (which are also summarised in the Employment Appeal Tribunal’s decision of 8 May 2009, §2-20) very shortly, the key ones are these.
The claimant, Professor Buckland, held a chair of environmental archaeology at the defendant university. One of his annual tasks was to mark students’ examination papers, which would be routinely second-marked to ensure consistency. The 2006 examinations included 16 resits, of which the claimant failed 14. The second marker, Mr Haslam, endorsed the marks. At a meeting attended by the claimant, the board of examiners, chaired by Dr Astin, checked and confirmed the results. Note was taken of the high failure rate and of the need to address its causes.
However, the programme leader on the archaeology course, Dr Russell, very shortly afterwards took it on himself to re-mark the papers. The tribunal accepted that this was done because of legitimate concern about the throroughness and evidencing of the second markings, but they found that it was done without the authority of Dr Astin. Dr Astin, however, was concerned about possible appeals and so had the scripts re-marked by another staff member, Mr Hewitt. The marks Mr Hewitt awarded were not far away from those awarded by the claimant, but they elevated some students from a straight fail to a marginal zone in which their marks in other subjects might rescue them.
Because, however, of the claimant’s continuing protests at what had happened, the University in mid-October set up an inquiry chaired by Professor Vinney. The claimant, who considered that Professor Vinney was not of sufficiently independent status, gave written evidence to the inquiry but declined to appear before it, a decision which the tribunal considered extremely unfortunate. It had been made clear to him that the inquiry was not going to question the teaching of the course: it was concerned only with the unauthorised re-marking of the resit papers. But the tribunal judged that “unless everything happened on the claimant’s terms then he would not cooperate”.
The report, published in January 2007, vindicated the claimant. It criticised the board of examiners for not being more concerned about the failure rate, but was clear that the claimant’s marks, once accepted by the board, should have been final. It went on to criticise both the lack of a sufficient benchmark of suitability for second markers and the way in which a third marking had been undertaken. And it made constructive proposals for improving rules and practice.
The claimant was far from satisfied with the report. He was angry that Dr Astin in his evidence had impugned his integrity as an examiner. But he went on to contend – without, the tribunal found, a shred of justification – that Professor Vinney had supported both Dr Russell’s and Dr Astin’s assaults on his integrity as an examiner, and that the inquiry had set out simply to exonerate management.
Professor Darville, the head of department, sought to mollify the claimant, agreeing that the university ought never to have got itself into this position over re-marking, undertaking to ensure that it could not happen again and urging the claimant to meet Mr Hewitt as Vinney had recommended. He ended by expressing his gratitude to the claimant for his stand and assuring him that he had “won the war that will end all wars” on the subject, even if he had lost some battles on the way.
The claimant was not mollified. He considered that “Astin and his management cronies” had got away with it. By letter of 22 February 2007 he resigned with effect from the end of July, when his obligations to his students would have been fulfilled.
The proceedings
Professor Buckland brought proceedings for unfair constructive dismissal.
The employment tribunal (EJ Twiss, Mr J Shah, Mr D S Robinson) held, first, that the delay between letter of resignation and its effective date was not such, in the circumstances, as to amount to an affirmation of the contract of employment. This is not now contested, but it of leaves open the question whether (a) there had been, and (b) still was, a repudiatory breach.
As to the first of these questions, the tribunal concluded that the University had indeed done something calculated to destroy the relationship of trust and confidence which is implicit in all contracts of employment: through Dr Astin, who was the chairman of the board of examiners, and without consulting the claimant or the second marker, it had procured and accepted the re-marking of the resit papers – an unequivocal affront to his integrity as to which his “sense of grievance … was fully justified”. It went, they found, “beyond what could be regarded as one of those incidents of professional life that a person must accept and move on”. There had been no need, they found, to take this course in order to ensure that students were fairly dealt with. The claimant’s belief that it was part of a much larger process of academic dumbing-down might well have a bearing on his prospective loss (it might have led him to resign in the not too distant future in any event), but it did not mitigate the insult.
Having made these careful findings, the tribunal asked themselves a question which neither party had asked: had the Vinney inquiry cured the breach? They decided that it had not. They did not accept that it lacked status or independence, but they accepted that its report did not afford the “the kind of exoneration and reinstatement that the claimant was, in the circumstances, entitled to”. They concluded:
“…. While it is understandable that the report was not more explicitly critical of those people (and it has to be said that it was quite critical of them) he was, we consider entitled to more explicit vindication and exoneration. The University had been guilty of a fundamental breach of contract and in our view it needed something very clear to rectify that breach if indeed it were possible to do that.”
The tribunal accordingly found that the claimant had been dismissed. They did not go on, however, to decide whether the dismissal, albeit constructive, had been unfair.
A further ground of unfair dismissal based on the making of protected disclosures in the course of the dispute failed because the disclosures, while they had occurred, had played no part in the termination. This ground has not been renewed.
The EAT (HH Judge Peter Clark, Mrs A Gallico, Mr B R Gibbs) allowed the University’s appeal. Taking the opportunity to untangle a thicket of authority on the correct legal test, they upheld the finding that the University had been in fundamental breach of the claimant’s contract of employment. But they overset the tribunal’s finding that the Vinney report had not cured the breach. They expressed themselves satisfied that, regarded objectively, it had done so. This alone would have been a simple disagreement, insufficient for a successful appeal; but they also held that the tribunal had erred in principle by adopting a subjective test, allowing the EAT to substitute its own view.
Having thus held that there had been no constructive dismissal because the University’s repudiatory breach had been cured before Professor Buckland purported to accept it, the EAT dealt briefly with two remaining issues, delay and fairness, which had consequently become moot. As to delay, they found no reason to differ from the tribunal (albeit they focused on the delay between the initial events and the Vinney report, whereas the tribunal had focused on the long period of notice which the claimant eventually gave). As to fairness, had they upheld the tribunal on dismissal they would have sent the claim back on the question whether the University had some substantial reason for their conduct which rendered the dismissal fair, since the tribunal appeared to have overlooked this.
This appeal and cross-appeal
In their logical order, the issues now raised, or contingently raised, are these:
(a) Is the occurrence of a fundamental breach of a contract of employment, at least on the employer’s part, to be gauged by a conventional contract test or by a ‘range of reasonable responses’ test? The University raises this issue by way of cross-appeal. If it succeeds in principle, the court is invited to find that the University’s conduct, notwithstanding the tribunal’s criticisms of it, lay within the range of reasonable responses to the problem which confronted it.
(b) Does the law permit a party which has committed a repudiatory breach of a contract of employment to preclude acceptance by curing the breach? This represents the claimant’s principal ground of appeal.
(c) If it does, was the tribunal’s decision that the University had failed cure its breach of the claimant’s contract of employment (i) legally sound and (ii) factually tenable. This represents the claimant’s fallback ground of appeal.
(d) Did the University advance a triable case that, if there had been a constructive dismissal, it had been fair? If it did, the tribunal overlooked it and if necessary the respondent seeks a remission.
(a) What is the correct test of repudiatory conduct by an employer?
Modern employment law is a hybrid of contract and status. The way Parliament has done this is to graft statutory protections on to the stem of the common law contract. Thus by s.94 of the Employment Rights Act 1996, every employee is given the right not to be unfairly dismissed. By s.95 dismissal is exhaustively defined for the purposes of the statutory right as – in short – termination by the employer with or without notice, termination by effluxion of time, or termination by the employee “in circumstances in which he is entitled to terminate it without notice by reason of the employer’s conduct”. The last of these, in legal terms, is constructive dismissal – that is to say an act which is not an explicit dismissal but which in law has the same effect.
What circumstances can bring about a constructive dismissal is determined not by the Act, which is silent on the subject, but by the common law. The common law holds that they must be circumstances amounting to a fundamental or repudiatory breach of contract by the employer. The first question, raised by the cross-appeal, is whether to have that character the employer’s conduct must fall outside the range of reasonable responses to whatever situation has arisen. Jason Galbraith-Marten for the University submits to us, as he submitted to EAT, that it must do so if it is to constitute a repudiatory breach. Antony White QC for Dr Buckland contends that there is no such test: the question is an objective and unitary one.
The fundamental term of Professor Buckland’s contract of employment which the tribunal held to have been breached was the requirement of mutual trust and confidence. The University does not contend that this finding was not open to the tribunal, but by its cross-appeal it contends that it was a response to the wrong question: the tribunal should have asked whether what the University had done was within the range of reasonable responses open to it in the situation confronting it in and after July 2006.
The EAT considered this issue in great detail. The need for such consideration arose from the proliferation of recent authority and dicta on the subject, not all of it consistent. With both gratitude and relief this court can refer the reader to the detailed description and thoughtful analysis of these in the judgment of the EAT [2009] ICR 1042, §24-46. For my part I would respectfully endorse both EAT’s reasoning and its conclusion at §47:
“In summary, we commend a return to settled authority, based on the following propositions: ”
(1) In determining whether or not the employer is in fundamental breach of the implied term of trust and confidence the unvarnished Mahmud test should be applied.
(2) If, applying the Sharp principles, acceptance of that breach entitled the employee to leave, he has been constructively dismissed.
(3) It is open to the employer to show that such dismissal was for a potentially fair reason.
(4) If he does so, it will then be for the Employment Tribunal to decide whether dismissal for that reason, both substantively and procedurally (see Sainsbury v Hitt [2003] IRLR 23 ), fell within the range of reasonable responses and was fair.
To the EAT’s reasons one can now add the remark of Underhill P in Amnesty International v Ahmed [2009] ICR 1450, §70, in relation to the EAT’s decision in this case, that he was “sympathetic to the contention that it is unhelpful to introduce into the concept of constructive dismissal a conceptual tool devised for an entirely different purpose”.
For their part, both counsel before us accept the EAT’s formulation. But – and it is perhaps as well that it has occurred at this point of time– Mr Galbraith-Marten contends that the “range of reasonable responses” test forms part of the Mahmud exercise at stage (1), as well as of the fairness issue at stage (4), if that is reached.
I would unhesitatingly reject this submission. It ignores what the EAT clearly, and correctly, meant when it spoke of “the unvarnished Mahmud test”. In Mahmud v BCCI [1998] AC 20, 35, Lord Nicholls reiterated that the test of breach of a fundamental term of a contract of employment was objective: “A breach occurs when the proscribed conduct takes place”. Lord Steyn (at 47) said much the same.
Mr Galbraith-Marten accepts this without demur. But – he says – the conduct of an employer which is said to have objectively broken the contract of employment is often (as here) conduct which the claimant alleges was in fundamental breach because it was unreasonable. This must entitle the employer to show that it was not – in other words to argue that it lay within the band of reasonable responses. He accepts that this has the effect of replicating the same issue at stages (1) and (4), but that simply means that by the time stage (4) is reached, if it is reached at all, the job is done.
This approach finds support in the EAT’s decision in Abbey National plc v Fairbrother [2007] IRLR 320, and with qualification in Claridge v Daler Rowney Ltd [2008] ICR 1267, §30. But, without retracing the complex path which the EAT was compelled to take, it is an approach which cannot stand with the authority of Western Excavating v Sharp [1978] ICR 221, in which this court counterposed the objective test and the unreasonableness test of constructive dismissal and held in clear terms that the former was the correct one.
It is nevertheless arguable, I would accept, that reasonableness is one of the tools in the employment tribunal’s factual analysis kit for deciding whether there has been a fundamental breach. There are likely to be cases in which it is useful. But it cannot be a legal requirement. Take the simplest and commonest of fundamental breaches on an employer’s part, a failure to pay wages. If the failure is due, as it not infrequently is, to a major customer defaulting on payment, not paying the staff’s wages is arguably the most, indeed the only, reasonable response to the situation. But to hold that it is not a fundamental breach would drive a coach and four through the law of contract, of which this aspect of employment law is an integral part.
Where, if at all, the reasonableness of the employer’s conduct may enter the picture is through the statutory additions to the law of contract. Assuming, in other words, that there can be conduct which is both reasonable and a fundamental breach of contract, a constructive dismissal claim would be impossible to decide unless stage (1) was tested objectively on ordinary principles and reasonableness deferred to stage (4).
As counsel agree, the statutory provision for justification of a dismissal appears at more than one point not to fit constructive dismissal, notwithstanding the express inclusion of constructive dismissal in the statutory scheme. But, reverting to my example, if the employer could satisfy a tribunal that, albeit not a listed reason, the unexpected lack of funds amounted to some other substantial reason for dismissal within s. 98(1), and that in all the circumstances the employer had “acted reasonably in treating it as a sufficient reason for dismissing the employee” as required by s. 98(4), it might be arguable that the claim should fail. Although it will need to be revisited on the last of the issues before us, this is a conundrum we do not have to resolve in order to decide the cross-appeal.
For these reasons I would dismiss the cross-appeal.
(b) and (c) Can a repudiatory breach be cured before acceptance, and if so, was this one cured?
The tribunal, having found a repudiatory breach on the University’s part, asked and answered a question which neither party had put to it: by the time Professor Buckland purported to accept the breach by giving notice of termination, had it been cured?
The case for saying that the Vinney report had cured it was not inconsiderable. It had, as the tribunal found, vindicated Professor Buckland, whose own petulance had kept the wound festering. But, in a balanced and fair-minded judgment, the tribunal concluded that the breach had, even so, not been cured because the slur on his integrity remained.
The EAT substituted its own, contrary, view not because it considered the tribunal’s conclusion to be perverse but because it considered that the tribunal had erred in law by using a subjective test – Professor Buckland’s own wounded feelings – rather than an objective one. This criticism was in my respectful opinion unfounded. The tribunal at each stage of its material findings was careful to say that Professor Buckland not merely felt but was entitled to feel as he did about the original insult to his standing and about the limits of the Vinney report. Their conclusion that, by contrast, “it needed something very clear to rectify that breach if indeed it were possible to do that” was objectively arrived at and tenable.
If the case were to reach this point, therefore, I would hold that the ground on which the EAT allowed the University’s appeal was not well founded. But the issue of cure falls to be decided in this court on a more fundamental ground, because Mr White contends that the law of contract, employment law included, knows no such doctrine.
It is common ground that no decided case holds in terms that a repudiatory breach, once complete (that is, not a merely anticipatory breach), is capable of being remedied so as to preclude acceptance. It follows, Mr White submits, that, absent waiver or affirmation, the wronged party has an unfettered choice of whether to treat the breach as terminal, regardless of his reason or motive for so doing. There is, in other words, no way back.
I confess that, if this is the state of the law, it seems to me capable of working injustice. It may, as Sir Richard Buxton remarked in granting permission to appeal, be the point at which the law of contract meets sensible industrial relations. The reason is that something may well be said or done in haste or anger which, if left unredressed, would strike at the root of the employment relationship – an accusation of dishonesty for example – but which is regretted and retracted before the other party, who may be either employer or employee, acts upon it by ending the relationship. Mr White may be right when he says that in the great majority of cases the parties will want the relationship to survive and the retraction will be accepted. But the law is there for those cases where, for what may be less than creditable reasons, the wronged party will not climb down: does the law in such cases ignore the olive branch?
The case for saying it does is formidable. It is not simply that no case holds that a repudiatory breach can be cured unilaterally by the party in default. If that were all, I hope that this court would not be deterred from so holding for the first time. It is that employment law in this regard forms an integral part of the general law of contract, so that either a doctrine of cure has to apply across the board or good and sufficient reason has to be found for introducing it into employment law alone. I will consider these two possibilities in turn.
Mr White submits that the common law has addressed the issue to the extent of deciding that an anticipatory breach of contract can be withdrawn at any time up to the moment of acceptance, but no further: see Stocznia Gdanska S A v Latvian Shipping Co Ltd (No 2) [2002] 2 Ll Rep 236, §87 per Rix LJ:
“In my judgment, there is of course a middle ground between acceptance of repudiation and affirmation of the contract, and that is the period when the innocent party is making up his mind what to do. If he does nothing for too long, there may come a time when the law will treat him as having affirmed. If he maintains the contract in being for the moment, while reserving his right to treat it as repudiated if his contract partner persists in his repudiation, then he has not yet elected. As long as the contract remains alive, the innocent party runs the risk that a merely anticipatory repudiatory breach, a thing “writ in water” until acceptance, can be overtaken by another event which prejudices the innocent party’s rights under the contract – such as frustration or even his own breach. He also runs the risk, if that is the right word, that the party in repudiation will resume performance of the contract and thus end any continuing right in the innocent party to elect to accept the former repudiation as terminating the contract.”
This account of the alternative courses which may be taken in response to a repudiatory breach leave no space for repentance by a party which has not simply threatened a fundamental breach or forewarned the other party of it but has crossed the Rubicon by committing it. From that point all the cards are in the hand of the wronged party: the defaulting party cannot choose to retreat. What it can do is invite affirmation by making amends.
To introduce into this relatively clear pattern of law an exception where amends have been made or offered for a fundamental breach is to open up case after case to an evaluation of whether the amends constituted an adequate cure of the breach. The present case is not a bad example of the factual niceties which the necessary judgment may involve. Legal niceties would also lie in wait: for example, whether a subjective or an objective valuation is called for and, if the latter, whether factors personal to the wronged party count as objective or subjective factors. I do not think we are justified in releasing the contents of this Pandora’s box into the general law of contract.
But contracts of employment are not like the generality of contracts. They are inherently personal, and they concern one of the most important things in most people’s lives. Is it desirable and feasible to introduce the doctrine into this branch of contract law? I have mentioned some of the reasons why it might well be thought desirable to do so; but Mr White has reminded us of the decision of the House of Lords in Miles v Wakefield District Council [1987] 1 AC 539, in which Lord Oliver, with whom the other members of the Committee concurred, based his analysis of the effect of a partial withdrawal of services on the contractual duty to pay salary upon the general law of contract. More specifically, Browne-Wilkinson P in W.E. Cox Toner Ltd v Crook [1981] ICR 823 held (as both counsel had agreed) that the general principles of contract law applied to the issue of repudiation and acceptance which was before the EAT.
Mr Galbraith-Marten asks why, if on those general principles an anticipatory breach can be cured up to the moment of acceptance, a completed breach cannot be. The answer has to be that the anticipatory or threatened breach has by definition not yet occurred and, if withdrawn, never will. A completed breach, even if it can be compensated for, cannot be undone. The only hint of a doctrine of cure of completed breaches that counsel have been able to show us is in the judgment in W.E.Cox Toner v Crook (above), where the EAT said, citing Farnworth Finance Facilities Ltd v Attryde [1970] 1 WLR 1053:
“…. However, if the innocent party further performs the contract to a limited extent but at the same time makes it clear that he is reserving his rights to accept the repudiation or is only continuing so as to allow the guilty party to remedy the breach, such further performance does not prejudice his right subsequently to accept the repudiation”
But this formulation explicitly leaves the option of acceptance or affirmation in the innocent party’s hands.
Albeit with some reluctance, I accept that if we were to introduce into employment law the doctrine that a fundamental breach, if curable and if cured, takes away the innocent party’s option of acceptance, it could only be on grounds that were capable of extension to other contracts, and for reasons I have given I do not consider that we would be justified in doing this. That does not mean, however, that tribunals of fact cannot take a reasonably robust approach to affirmation: a wronged party, particularly if it fails to make its position entirely clear at the outset, cannot ordinarily expect to continue with the contract for very long without losing the option of termination, at least where the other party has offered to make suitable amends. The present case, for reasons explained by Jacob LJ, may be seen as the kind of exception which proves the rule.
(d) Is there a triable case that the constructive dismissal of the claimant was fair?
Mr Galbraith-Marten asks that, if all else fails, the s.98(4) issue should be remitted, which is what the EAT would if necessary have done.
I have mentioned the mismatch between constructive dismissal and the statutory unfairness test. One only reaches s.98(4) through the gateway of s.98(1) and (2). The latter of these includes in the qualifying reasons for dismissal a reason which relates to the capability or conduct of the employee. It may legitimately be said that the reason for the repudiatory conduct of the University in undermining Professor Buckland’s position as an examiner related to his capability and conduct in the role.
But how does one decide, pursuant to s.98(4), whether the University acted reasonably in treating this as a sufficient reason for dismissal? Since the University did not consciously either dismiss Professor Buckland or therefore treat anything as a sufficient reason for doing so, the question makes little sense. One has to make sense of it by translating it into the question whether the University behaved reasonably in undermining his status. So posed, the question answers itself, for the University could not intelligibly seek to justify something it said it had not done.
I do not say that this will always be the outcome in a constructive dismissal case – one can construct cases where such a logical block would not necessarily be present - but it is the inexorable outcome in this one.
Conclusion
For these reasons I would allow the appeal and dismiss the cross-appeal. The effect will be to restore the determination of the employment tribunal.
Lord Justice Carnwath:
I agree that the appeal should be allowed and the cross-appeal be dismissed for the reasons given by Sedley LJ. I also agree with Jacob LJ’s concluding comments on the desirability of avoiding unnecessary remitter to the tribunal when the EAT is in as good a position to decide the matter itself.
Lord Justice Jacob:
I agree that, for the reasons given by Sedley LJ this appeal should be allowed and the cross-appeal dismissed. I only add two short comments.
The first is that I do not share Sedley LJ’s regret in holding that a repudiatory breach of contract, once it has happened, cannot be “cured” by the contract breaker. Once he has committed a breach of contract which is so serious that it entitles the innocent party to walk away from it, I see no reason for the law to take away the innocent party’s right to go. He should have a clear choice: affirm or go. Of course the wrongdoer can try to make amends – to persuade the wronged party to affirm the contract. But the option ought to be entirely at the wronged party’s choice.
That has been the common law rule for all kinds of contract for centuries. It works. It spells out clearly to parties to contracts that if they actually commit a repudiatory breach, then whether the contract continues is completely out of their hands. The rule itself discourages repudiatory breach. In the context of employment law it means that employers know that if they treat an employee so badly as to commit a repudiatory breach, then they cannot hang on to the employee unless they can persuade him or her to decide to stay.
Next, a word about affirmation in the context of employment contracts. When an employer commits a repudiatory breach there is naturally enormous pressure put on the employee. If he or she just ups and goes they have no job and the uncomfortable prospect of having to claim damages and unfair dismissal. If he or she stays there is a risk that they will be taken to have affirmed. Ideally a wronged employee who stays on for a bit whilst he or she considered their position would say so expressly. But even that would be difficult and it is not realistic to suppose it will happen very often. For that reason the law looks carefully at the facts before deciding whether there has really been an affirmation.
This case provides a very good example. The repudiatory breach occurred in September – a time when the academic year was just about to start and a particularly difficult time for an academic to leave instantly. The Vinney inquiry was instigated shortly thereafter and it was entirely reasonable (even though he did not think much of the choice of Prof. Vinney) for Prof. Buckland to wait and see what it said before exercising his right to accept the repudiation. And it was also entirely proper for him to exercise that right by a long period of notice given the fact that his students would otherwise have been adversely affected mid-academic year. That is why the tribunals below (now unchallenged) held there had been no affirmation, either before or after the Vinney report.
Thirdly, the fact that it takes rather a lot to find affirmation on the facts in an employment contract is itself another good reason for refusing to recognise any doctrine of “cure” in that context. Once an employer has committed a repudiatory breach there will generally be some time to make for him to try make amends, for tempers to cool and for the employee to make a rational decision as whether he or she should to stay on.
Finally a word about the EAT’s “academic” decision that if it had upheld the finding of constructive dismissal it would have remitted the question of fairness to the ET. With respect I cannot see why. Even though the ET had not made a finding about this, no more evidence was required. So the EAT could have decided the point itself
Quite generally sending a case back to a tribunal or court below should be used only as a last resort. “Ping pong”, as some call it, generally serves litigants badly – prolonging things and increasing costs.