England and Wales High Court (Chancery Division) Decisions

You are here: BAILII >> Databases >> England and Wales High Court (Chancery Division) Decisions >> Freshasia Foods Ltd v Jing Lu [2018] EWHC 3644 (Ch) (04 January 2019)
URL: http://www.bailii.org/ew/cases/EWHC/Ch/2019/3644.html
Cite as: [2018] EWHC 3644 (Ch)

[New search] [Printable PDF version] [Help]

		Neutral Citation Number: [2018] EWHC 3644 (Ch)
		CLAIM NUMBER: IL-2018-000206

IN THE HIGH COURT OF JUSTICE
BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES
INTELLECTUAL PROPERTY LIST (ChD)

Rolls Building
Fetter Lane, London EC4

4 January 2019

B e f o r e :

MR DANIEL ALEXANDER QC
Sitting as a Deputy Judge of the Chancery Division
____________________

Between:

	FRESHASIA FOODS LIMITED	Claimant
	-and-
	JING LU	Defendant

____________________

Mr Douglas Campbell QC (instructed by Myerson LLP) for the Claimant
Ms Stephanie Thompson (instructed by Virtuoso Legal Limited) for the Defendant

Hearing date: 18 December 2018
Judgment with reasons to follow in writing: 19 December 2018
____________________

HTML VERSION OF JUDGMENT APPROVED
____________________

MR DANIEL ALEXANDER QC

INTRODUCTION AND SUMMARY

On 19 December 2018, I made an order inter alia granting a limited injunction until trial in the following form:

"Until judgment following trial herein or further order in the meantime the Defendant must not (whether acting by himself, his servants or agents or by others acting on his behalf) do (or authorise, cause, assist, or enable others to do) the following acts of any of them without the consent of the Claimant:

…

In respect of food products sold to Chinese retail shops and restaurants in the United Kingdom or elsewhere in the European Union, solicit the custom of, or sell or deliver to any private individual, firm, company or other person who at 28 September 2018 was (a) a customer of the Claimant to whom the Defendant had personally sold and/or delivered the Claimant's products on behalf of the Claimant, or (b) whom the Defendant had introduced to the Claimant or approached on behalf of the Claimant".

The injunction was based on a non-solicitation clause in the defendant's ("Mr Lu's") contract of employment and was made on the basis that the claim to relief in this form at trial was sufficiently arguable and that the balance of convenience favoured the grant of an injunction. The other matters in the order, including restrictions on the use of confidential information and directions to trial, were largely agreed.

The non-solicitation injunction was made in more limited form than originally sought. It left out a number of provisions including those relating to attempts to solicit, accepting work for private gain for a particular company, and in so far as it extended restrictions in respect of persons with whom the Defendant had any business dealings or knowledge in the leaving period (alleged to be 12 months) immediately prior to the end of his employment. I declined to make an order in respect of a separate non-compete provision which was sought to prevent Mr Lu from doing the following acts until trial "Own, manage, operate, consult for, or be employed in Kung Fu [his new employer] or any other business substantially similar to or competitive with the Claimant". These were refused on the footing that this wider relief was unlikely to be enforceable for various reasons and that the balance of convenience did not favour the grant of an injunction I these terms.

At the beginning of the hearing, the parties were far apart. The claimant ("Freshasia") was trying to stop the defendant from working for his new employer ("Kung Fu") at all pending trial. That was ultimately not advanced with great conviction, for good reason because the non-compete clause seemed too broad to be justifiable in the particular context. Mr Lu was contending that even the non-solicitation clause was unenforceable both for lack of clarity and because it was oo broad with the invalid parts being non-severable. However, towards the end of the hearing, he offered an undertaking, without prejudice to his contentions, not directly to contact any persons or entities who at the time of the termination of his employment were customers of the Claimant and who were set out in a list to be agreed. Mr Lu's position was therefore that he was prepared to offer some specific non-solicitation protection pending trial, which was in some respects narrower and in some respects broader than that provided by the contractual provisions. It seemed to me that it would be preferable so far as possible to keep to the wording of the contract and that an injunction framed in the terms of the contract, while not ideal, was not so uncertain as to be objectionable. Because of illness, it was not possible to give full oral reasons at the time the order was made. These are therefore the reasons for the order.

The case overall is a relatively straightforward ex-employee dispute, applying conventional principles, with two exceptions. It raises the question of (i) the extent to which the merits should be taken into account in a situation where the length of the covenant and whether it runs beyond trial is itself an issue and (ii) the approach to severance of provisions (to use a neutral term) in the light of the relatively recent Egon Zehnder case in the Court of Appeal.

Procedural history

The application first came before Snowden J on 29 November 2018 in the applications court. He provided a timetable for service of evidence and indicated that the claim was suitable for expedition. The trial has now been fixed for the week commencing 4^th March 2019. The period in question with which the court is now concerned is therefore some 3-4 months. Snowden J declined to grant an interim injunction in the meantime because Mr Lu had been given only 2 days' notice and only had a CLIPS representative. He was also concerned that the injunction sought would prevent Mr Lu from working for the new employer at all, that the prejudice to Freshasia was "not great" and that Freshasia had not moved quickly enough to justify relief without the proper notice. It was common ground that I should look at the matter afresh and consider what relief, if any, should be granted pending trial in the light of all the evidence and full argument.

FACTS

There are witness statements from Mr Jian Lan, a director of Freshasia and from the defendant Mr Lu. Pleadings have progressed to a defence and a reply was due to be served on 21 December 2018. The key facts can be summarised relatively briefly, based on the witness statements and skeletons.

Parties

Freshasia

Freshasia is a UK registered company which supplies food products, particularly dumplings and sliced meats, to Chinese retail shops and restaurants. It has between 100 and 200 employees and a turnover in the millions of pounds. It has about 500 customers in total in the UK and the EU. There are only a few employers in this industry, which is a very specialist market of frozen and packaged food sold mainly to international student customers.

Mr Lu

Mr Lu has a Masters in Marketing and, before working for Freshasia, was with Nestle China. He started work for Freshasia in a relatively junior role of marketing assistant on 12 January 2015 but shortly thereafter progressed to Marketing Advertising Manager. This role involved promoting Freshasia to its target market, running the social media platforms and conducting analysis on the marketing strategy. In his most up to date c.v. of 2018, he refers to this employment with Freshasia as being that of "marketing manager" and says that he was head of marketing, during which time he claims to have been responsible for increasing the profit significantly.

Mr Lu handed in his notice on 11 September 2018. He should have given 1 month and 3 weeks' notice under his contract of employment, but he gave less than 1 months' notice. He told his manager, Mr Lan, that he was leaving in order to work in Hong Kong and later told Freshasia's Area Sales Manager that he was leaving to work for an IT company in the UK. However in fact he left to work for Freshasia's competitor, which is a company trading as Kung Fu. Freshasia claims that he did so to do virtually the same job as he had done for them but Mr Lu has said that his role with Kung Fu is "slightly different", since he is employed as a business development manager rather than in a marketing role.

Kung Fu

Kung Fu is the trading name of a company (Oriental Food Express Limited) which is also in the business of supplying Asian foods and which is somewhat smaller than Freshasia. Apparently Mr Lu was offered a higher salary (£40,000) by Kung Fu than he had at Freshasia. Freshasia contends that this is because Kung Fu knew Mr Lu would bring customers or confidential information with him and that he has done so with Freshasia losing about £200,000 worth of business, and 55 customers since Mr Lu's departure. It is also said that he has persuaded several other employees to leave and made disparaging remarks about Freshasia. All of this is strongly disputed by Mr Lu, especially as to whether any loss of business has been due to any of his actions. Mr Lu contends that he was not in regular direct contact with customers while at Freshasia and that sales was mainly the responsibility of account managers and/or the directors. He contends that the difficulties Freshasia is experiencing are, in part, due to their failure to recruit a replacement marketing manager and are in any event not a result of any wrongful actions on his part.

Post termination restrictions

Mr Lu is subject to post-termination restrictions in his contract of employment which were incorporated in an Employee Handbook which he signed. There is no dispute, for the purpose of this application, that it is these terms, rather than earlier, much broader, terms which are relevant and that he is bound by them. Those are set out in the section on the terms below.

LAW

I shall outline the applicable principles, dealing more fully with the points where there was dispute.

(a) Approach to interim injunction

The first issue is what test should be applied on this application.

Parties' arguments

Freshasia contends the approach in American Cyanamid [1975] AC 39 is applicable to all of the covenants on the basis that they will not have expired at trial. Accordingly, it submits that it need only show that there is a serious issue to be tried as to enforceability and breach of the covenants.

Mr Lu contends that this approach is inappropriate, given that the covenants may only be held to have a 6 month duration in which case they would probably have expired by the time judgment was given, and that Freshasia should demonstrate at this stage, on the balance of probabilities, that it will make out its case for an injunction at trial. He submits that the strength of the parties' cases should be taken into account, in any event, referring inter alia to NWL Ltd v Woods [1979] 1 WLR 1294 (HL) and Business Seating (Renovations) Ltd v Broad [1989] 1 ICR 729, where Millett J held that, where half the 12 month period of the restraint would have elapsed by the time of trial that he could "not avoid taking into account the likely prospects of success at the trial". Mr Lu also submits that there is an additional and special factor making it particularly appropriate to take the merits into account, namely the potential impact of the injunction sought on his immigration status in the light of his Tier 2 visa. He submits, in effect, that where a defendant's right to remain in the UK depends to a significant degree on his not being enjoined from continuing to be employed by his current employer, the court needs to be particularly sure at the interim stage that an injunction is justified.

Discussion

In NWL Ltd v Woods [1979] 1 WLR 1294, Lord Diplock said at 1306 that a Court ought to "give full weight to all the practical realities of the situation to which the injunction will apply" and explained that the American Cyanamid approach

"was not dealing with a case in which the grant or refusal of an injunction at that stage would, in effect, dispose of the action finally in favour of whichever party was successful on the application, because there would be nothing left on which it was in the unsuccessful party's interest to proceed to trial. Cases of this kind are exceptional but when they do appear they bring into the balance of convenience an important additional element."

In Lansing Linde v Kerr [1991] 1 WLR 251, Staughton LJ said at 258:

"If it will not be possible to hold a trial before the period for which the plaintiff claims to be entitled to an injunction has expired, or substantially expired, it seems to me that justice requires some consideration as to whether the plaintiff would be likely to succeed at a trial. In those circumstances it is not enough to decide merely that there is a serious issue to be tried."

In Films Rover International Limited and Ors v Cannon Film Sales Limited [1987] 1 WLR 670, Hoffmann J said:

"In the forefront of his argument counsel for Thorn-EMI submitted that the court should not grant an interlocutory mandatory injunction, amounting to specific performance of one of Thorn-EMI's alleged contractual obligations, unless there appeared a high probability that Films Rover would succeed in establishing its legal right at the trial. In this case the Court of Appeal has gone no further than to say that Films Rover has an arguable case and, as I have already said, I propose to treat that as meaning that Films Rover is at least as likely to fail as to succeed. Counsel said that fell well short of the standard of persuasion necessary for the grant of an interlocutory mandatory injunction.

…

The principal dilemma about the grant of interlocutory injunctions, whether prohibitory or mandatory, is that there is by definition a risk that the court may make the "wrong" decision, in the sense of granting an injunction to a party who fails to establish his right at the trial (or would fail if there was a trial) or alternatively, in failing to grant an injunction to a party who succeeds (or would succeed) at trial. A fundamental principle is therefore that the court should take whichever course appears to carry the lower risk of injustice if it should turn out to have been "wrong" in the sense I have described. The guidelines for the grant of both kinds of interlocutory injunctions are derived from this principle."

That approach was recently applied in the context of a shareholder's restrictive covenant in Ideal Standard International SA & Anor v Herbert [2018] EWHC 3326 (Comm) and, in my view, it is equally apposite to employee restrictive covenants.

Three points follow from this case law.

First, the court should always take account of the practical realities in cases of this kind. If a preliminary injunction is likely to be final in effect, the court should be particularly convinced that the course it adopts at the interim stage is likely to be justified. It seems to me, in principle, that Lord Diplock's approach suggests that the merits should be taken into account in cases where, even if the covenant would or may still have a period to run at trial, the impact of a preliminary injunction would be effectively final for other reasons. Such a case may be one in which an injunction would mean that a defendant would lose his or her new job altogether and would, as a result lose a home or residence status meaning that no trial would take place because the defendant would not be in a position to pursue an arguable defence at all.

Second, there is no fixed way in which the merits must be taken into account. In some cases, where an injunction would have a particularly severe impact, it may only be appropriate to grant it if reasonably satisfied that the claim will succeed. In other situations, the merits may be taken into account as a factor in attempting to take the course least likely to lead to injustice.

Third, in considering the approach to be taken, it is necessary to take account of the nature of the injunction sought. An injunction preventing employment altogether with a particular undertaking or kind of undertaking for a lengthy period may require more intensive justification at the interim stage than an injunction merely restraining dealing with certain specific individuals with whom the ex-employee has previously dealt but which leaves the employee free to work in the industry more generally. Intensity of scrutiny of the merits should be proportional to the impact of the order sought.

This case is illustrative. Even if there is a serious issue to be tried as to its validity, the impact of enforcing the non-compete clause would, it is said by Mr Lu, be very severe. A serious issue to be tried is not enough. In contrast, the cut down non-solicitation clause would have a much more modest impact on Mr Lu and third parties, only limiting specific dealings with specific former customers/contacts. In such a case, even if the court is not satisfied that the case is a winning one, it may be appropriate to grant protection for a few months to trial, if the case is reasonably arguable and the balance of convenience otherwise favours such a course.

(b) Construction

There is no dispute as to the general principles of contractual construction and no need to set these out. It suffices to refer to the judgment of the Court of Appeal in Egon Zehnder, where Longmore LJ said:

"12. I do not accept that the principle of preference of enforceability to unenforceability can be invoked in favour of any alternative legitimate construction since the ingenuity of business and their lawyers may assert that many a plausible construction is a "legitimate" construction. This principle used to be expressed in the maxim "verba ita sunt intelligenda ut res magis valeat quam pereat" and is now called "Saving the document" (see Chitty 32^nd edition para 13-084). But the principle can only go so far since one has to start with the language used by the parties; there must, in my view, be a genuine ambiguity before the principle can be invoked and I respectfully doubt whether Waller LJ meant, in the paragraph of Turner to which we were referred, to imply that ambiguity was not required. As Lord Hoffmann observed in BCCI v Ali [2002] 1 AC 251 para 39, when referring to the principle that parties are unlikely to have intended to agree to something unlawful or legally ineffective:-

"But the primary source for understanding what the parties meant is their language interpreted in accordance with conventional usage.""

This boils down to saying that it is not permissible to come up with some construction that is vaguely plausible if it is reasonably clear that this is not what the parties intended and take that as the construction if to do so would save an otherwise void clause.

(c) Enforceability

Again, there was no dispute as to the principles. These can be taken from the judgment of Sir Bernard Rix for the Court of Appeal in Coppage & Anor v Safety Net Security Ltd [2013] EWCA Civ 1176 at [9]:

"The law relating to restrictive covenants in the employment context was not in dispute, although at trial and again on appeal the parties were able to point to different aspects of the jurisprudence in support of their respective cases. I take the following general principles from the discussion of the subject to be found in Chitty on Contracts, 31^st ed, at paras 16-105ff and in the jurisprudence there cited.

(i) Post-termination restraints are enforceable, if reasonable, but covenants in employment contracts are viewed more jealously than in other more commercial contracts, such as those between a seller and a buyer.

(ii) It is for the employer to show that a restraint is reasonable in the interests of the parties and in particular that it is designed for the protection of some proprietary interest of the employer for which the restraint is reasonably necessary.

(iii) Customer lists and other such information about customers fall within such proprietary interests.

(iv) Non-solicitation clauses are therefore more favourably looked upon than non-competition clauses, for an employer is not entitled to protect himself against mere competition on the part of a former employee.

(v) The question of reasonableness has to be asked as of the outset of the contract, looking forwards, as a matter of the covenant's meaning, and not in the light of matters that have subsequently taken place (save to the extent that those throw any general light on what might have been fairly contemplated on a reasonable view of the clause's meaning).

(vi) In that context, the validity of a clause is not to be tested by hypothetical matters which could fall within the clause's meaning as a matter of language, if such matters would be improbable or fall outside the parties' contemplation.

(vii) Because of the difficulties of testing in the case of each customer, past or current, whether such a customer is likely to do business with the employer in the future, a clause which is reasonable in terms of space or time will be likely to be enforced. Moreover, it has been said that it is the customer whose future custom is uncertain that is "the very class of case against which the covenant is designed to give protection…the plaintiff does not need protection against customers who are faithful to him" (John Michael Design Plc v. Cooke [1987] 2 All ER 332, 334).

(viii) On the whole, cases in this area turn so much on their own facts that the citation of precedent is not of assistance."

The approach to determining the enforceability of PTRs was summarised by Foskett J, in Tradition Financial Services Ltd v Gamberoni [2017] EWHC 768 (QB). It involves a three-stage process:

a. construe the covenants;

b. consider whether Freshasia has shown it has legitimate business interests requiring protection, as to which it is well established that there is a legitimate interest in maintaining a connection with both past and existing customers although that is not so in relation to merely potential customers (see Business Seating v. Broad [1989] ICR 79) as well as, of course, protection of confidential information;

c. consider whether the covenant is wider than is reasonably necessary for the protection of these interests.

(d) Severance

The issue which gave rise to greatest dispute as to the law was the approach to severance. It arises because Mr Lu contends that even if part of the non-solicitation clause is reasonable and enforceable, the parts that are not are not separate covenants but are part and parcel of the same covenant. They are therefore not severable and the whole clause is irreparably void. Freshasia contends that the law of severance does not set as strict a test of severability as Mr Lu suggests and that, even if it does, the provisions in question arguably meet that test, since they are in substance separate obligations even if not expressed as separate covenants.

Parties' arguments on the law

Freshasia contends that the law as to severance is clear and is summarised in Beckett (discussed below) and that it supported severance of any aspect in which the covenants were too wide in the present case.

Mr Lu contends that for severance to be permissible it is necessary that the unenforceable provision is capable of being removed without the necessity of adding to or modifying the wording of what remains and the proposed severance must meet the "blue pencil" test (Norbrook Laboratories v. Adair [2008] EWHC 978 at [46], citing TFS Derivatives Ltd v Morgan [2004] EWHC 3181 (QB)). That is uncontroversial. Mr Lu adds that a clause (such as clause B(2)(f) in the Employment Handbook) which purports to agree that the Court will "read down" an otherwise void restraint of trade clause to turn it into a restraint which would have been enforceable, is ineffective for the reasons given by Briggs J in Townends Grove [2004] EWHC 3432 at [60]–[62]. Again, that was not disputed.

However, he further contends that severance can only occur where there are distinct covenants and parts of a single covenant cannot be severed, referring to the line of case law and the fate of the covenant in the Egon Zehnder case where the impermissibly wide words "or interested" could not be deleted from a covenant preventing an employee from being "engaged, concerned or interested in" a competing business.

Discussion

For present purposes, discussion can focus on two key authorities, Beckett and Egon Zehnder.

(i) Beckett

Prior to Egon Zehnder, the most comprehensive recent summary of the law of severance was undertaken by the Court of Appeal in Beckett Investment Management Group Ltd & Ors v Hall & Ors [2007] EWCA Civ 613. A unanimous court (The Master of the Rolls, Carnwath, Maurice Kay LJJ) approved the following approach:

40. In a number of more recent first instance decisions, a threefold test has been applied. In the employment context, its origin is to be found in Sadler v Imperial Life Assurance Company of Canada Ltd [1988] IRLR 388. Mr P J Crawford QC, sitting as a Deputy Judge of the High Court said (at paragraph 19):

"… a contract which contains an unenforceable provision nevertheless remains effective after the removal or severance of that provision if the following conditions are satisfied:

(1) the unenforceable provision is capable of being removed without the necessity of adding to or modifying the wording of what remains;

(2) the remaining terms continue to be supported by adequate consideration;

(3) the removal of the unenforceable provision does not so change the character of the contract that it becomes 'not the sort of contract that the parties entered into at all'."

41. The final words were derived from the judgment of Buckley LJ in Chemidus Wavin Ltd v Societe pour la Transformation et l'Exploitation des Resines Industrielles SA (1978) 3 CMLR 514, 520. That concerned a restrictive trading agreement in a commercial context. The Sadler threefold test has been applied in a number of subsequent employment cases including Marshall v N M Financial Management Ltd [1996] IRLR 20 and T F S Derivatives v Morgan [2005] IRLR 246. Mr Oldham seeks to question these decisions on the basis that they are either an aberrant extension of Attwood v Lamont or that they apply only within a subset, the distinguishing characteristic of which is that they involved the future payment of money under the terminated contract.

42. In my judgment these more recent cases are neither aberrant nor merely applicable within a subset. In T Lucas & Co v Mitchell, above, at page 135, Russell LJ referred to the necessary approach, based on Attwood v Lamont, as being:

"a matter of construction, including the question whether one obligation can be removed or severed without altering the nature of the contract and without having to add or modify the wording otherwise than by excision."

43. I see no difference in principle between those words of Russell LJ and the third stage of the threefold test propounded in Sadler. It seems to me that the threefold test as there formulated is a useful way of approaching these cases and should be adopted."

Three aspects of those observations merit comment.

First, the court in Beckett was dealing with a covenant which prohibited the ex-employee from dealing with a "Relevant Client". That term was then defined in a comprehensive but unitary definitions clause where "Relevant Client" was said to include persons with various characteristics, including a provision that deemed certain undertakings to be a "Relevant Client". There was therefore a single covenant to which was applied a two-fold definition of "Relevant Client" provided in a single clause. Despite that, the Court of Appeal found no difficulty in treating the more extended deemed definition as severable (see at [8]-[10] and [30]-[44]). That conclusion was reached by the Court of Appeal, applying the same test as the judge below derived from Attwood v. Lamont [1920] 3 KB 571, who had reached a different conclusion that the clause was not severable.

Second, the Court of Appeal referred to Attwood as the "appropriate starting point". It cited the parts of the judgments to the effect that a contract can be severed "if the severed parts are independent of one another and can be severed without the severance affecting the meaning of the part remaining" and "where the two parts of a covenant are expressed in such a way as to amount to a clear severance by the parties themselves and to be substantially equivalent to two separate covenants" (Lord Sterndale MR) and "permissible in a case where the covenant is not really a single covenant but is in effect a combination of several distinct covenants" (Younger LJ). In Attwood, the plaintiff was a draper, tailor and general outfitter. The covenant required the defendant not to "carry on or be in any way directly or indirectly concerned in any of the following trades or businesses: that is to say, the trade or business of a tailor, dressmaker, general draper, milliner, hatter, haberdasher, gentleman's, ladies or children's outfitter" within a specified area. The covenant was found to be reasonable only in relation to tailoring. The Court of Appeal held that the remaining words were not severable, because this was properly understood as a case of removing words from within a single covenant, rather than a case of separating out distinct covenants.

Third, the Court of Appeal referred to Mason v. Provident Clothing Company Limited [1913] AC 724 in connection with an observation that there was "at one stage" an assumption in the authorities that the courts should be reluctant to sever in favour of an employer in a restraint of trade case. Lord Moulton there said that a covenant should only be severed

"…in cases where the part so enforceable is clearly severable, and even so only in cases where the excess is of trivial importance or merely technical, and not a part of the main purport and substance of the clause."

Mason concerned a covenant not to carry on a specified business within a 25 mile radius of the plaintiff's premises. The House of Lords held that this was too broad, and that the covenant could not be saved by reading it down to a reasonable area. In context, Lord Moulton was there considering severance of individual words or phrases as a means of reading down a single covenant which would otherwise be unenforceable.

However, the implication from the Court of Appeal in Beckett was that the historical reluctance to sever was something of a relic, notwithstanding the fact that it was authority from the House of Lords and had been subsequently adopted by the Court of Appeal. The Court of Appeal in Beckett in its language, the test it approved and its reference to authority were implicitly rejecting an unduly technical or narrow approach to the question of severability and did not focus on whether the provision to be served fell, as a matter of form, into a completely separate sub-clause.

It has not been possible on this interim application to consider all of the decisions which have sought to apply Attwood and Mason. For example, one case not discussed at the hearing was T Lucas & Co v. Mitchell [1974] 1 Ch 129, referred to in Beckett, where the Court of Appeal treated the non-dealing part of a covenant as severable where the covenant as a whole provided that an ex-employee should not "deal in any goods [of certain kinds] or solicit orders for or supply any such goods to [certain persons]", having concluded that the remaining non-solicitation and supply clauses were valid. The Court of Appeal said that the restraint which was held not severable in Attwood was "wholly different in character" and gave no guidance to what it should do. In so doing, the Court of Appeal in Lucas differed from Pennycuick V-C below who considered that Attwood prohibited severance in such a case, noting however that there was an inconsistency between Attwood and Scorer v. Seymour Jones [1966] 1 WLR 1419.

(ii) Egon Zehnder

It is against that background that the Court of Appeal came to the Egon Zehnder case ([2017] EWCA Civ 1054). The passage from the Court of Appeal relied on by Mr Lu bears setting out in full:

"Severance

26. Mr Laddie [Counsel for the ex-employer] submitted that the words "or interested" could be deleted or severed from clause 13.2.3. The judge dealt with this point shortly by saying that he did not find it appealing. Since he held that on its true construction the clause was not impermissibly wide, he did not need to say anything more.

…

29. The second reason for not allowing severance is that it is well settled that parts of a single covenant cannot be severed; it is a requirement of severance that it can only take place where there are distinct covenants (as, for example, in 13.2 which enumerates separate covenants) and, perhaps, not even then. Clause 13.2.3 is, however, a single covenant preventing Ms Tillman from engaging or being concerned in a competing business in any one of several capacities, has to be read as a whole and cannot be severed, see Chitty para 16-215.

30. That is the effect of Attwood v Lamont [1920] 3 KB 571 where severance was refused in relation to specified trades, with respect to which the employer had no legitimate reason for preventing his ex-employee from carrying on or being concerned in. Younger LJ said at page 593:-

"The learned judges of the Divisional Court, I think, took the view that such severance always was permissible when it could be effectively accomplished by the action of a blue pencil. I do not agree. The doctrine of severance has not, I think, gone further than to make it permissible in a case where the covenant is not really a single covenant but is in effect a combination of several distinct covenants. In that case and where the severance can be carried out without the addition or alteration of a word, it is permissible. But in that case only."

This court also relied on Lord Moulton's speech in Mason v Provident [1913] AC 724, 745 in which he said that it was no business of the courts to come to the assistance of an employer who had exacted an unreasonably wide covenant and, by applying their ingenuity and knowledge of the law, to carve out of the void covenant the maximum of what the employer might validly have required. Severance was also refused in the British Reinforced Concrete case for the reasons given in Attwood v Lamont.

31. This requirement that severance can only be applied to separate covenants and not to parts of a single covenant remains the law, since Attwood v Lamont was expressly stated in Beckett Investment Management Group Ltd v Hall [2007] ICR 1539 at paras 38-39 to have been adopted in later decisions.

32. Beckett also approved a threefold test propounded by Mr Peter Crawford QC sitting as a Deputy Judge of the High Court in Sadler v Imperial Life Assurance Company of Canada [1988] IRLR 388 at paras 391-2:-

"a contract which contains an unenforceable provision nevertheless remains effective after the removal or severance of that provision if the following conditions are satisfied: 1. The unenforceable provision is capable of being removed without the necessity of adding to or modifying the wording of what remains. 2. The remaining terms continued to be supported by adequate consideration. 3. The removal of the unenforceable provision does not so change the character of the contract that it becomes "not the sort of contract that the parties entered into at all"."

33. Mr Laddie submitted that this test had replaced the requirement that severance only applied to separate and not single covenants. But Beckett shows that that is not correct. The requirement is reflected in the third of the three tests adopted by Mr Crawford because it must always be doubtful whether parts of a single covenant can be deleted without the contract becoming "not the sort of contract that the parties entered into at all". It is, as Lord Moulton said, no business of the courts to create a valid covenant in order to replace an impermissibly wide covenant which an employer has sought to impose on the employee.

34. For those reasons, I do not think the words "or interested" can be deleted in order to save the covenant."

The consequence was that the whole covenant was void.

It is therefore submitted by Mr Lu that Egon Zehnder precludes, as a matter of law "internal", severance in a covenant by deletion of (say) one of a series of alternatives and that, in so doing, it is a conventional application of well-established law dating back over 100 years, reflected in Atwood and Mason.

I am not satisfied at this stage of the case that Egon Zehnder goes that far. First, it refers to and applies the approach in Beckett. Secondly, the Court of Appeal only says that it must always be "doubtful" whether parts of a single covenant can be deleted without the contract becoming not the sort of contract the parties entered into at all. That does not suggest that it is impossible. It is true that severance was refused in that case but that is better seen as an application of the general test in Beckett to the specific contract in that case, where the Court of Appeal took the view that the removal of the term did effect a significant change.

General considerations

It is also necessary to bear some wider considerations in mind.

First, enforcement of suitably drawn covenants is in the public interest, protecting the legitimate interests of an employer in intellectual and related proprietary rights which can be easily transferred and lost when an important employee moves. That interest, it might be thought, should not be unduly frustrated by too narrow an approach to severability.

Equally though, there is much to be said for the thinking behind Lord Moulton's observations in Mason and a stricter approach. It is undesirable to permit a system by which an employer might "cow an employee into submission to unenforceable restraints and at the same time, if accused of too wide a claim, successfully assert a narrower", as the Court of Appeal put it in Lucas. Overly broad covenants can put existing employers into an unfair position with respect to their employees, giving them undue power to intimidate employees with litigation into not joining a competitor, because employees cannot afford the large transaction costs and risk of doing so. As Lord Moulton said:

"…the real sanction at the back of these covenants is the terror and expense of litigation, in which the servant is usually at a great disadvantage, in view of the longer purse of his master."

That problem is as live today as it was over 100 years ago. The potential for causing disadvantage may be greater where, as in this case, other important aspects of life, including residence status and mortgage payments are dependent upon employment. The House of Lords in Mason focussed on the right of freedom of trade which post-termination covenants could limit. There are other related problems which unduly broad covenants cause. They can act as general deterrents to labour market mobility (particularly in the aggregate) and may have a knock on effect on the terms and conditions that an employee is able to secure with his or her existing employer. Although some of the reported employee covenant cases concern very well-paid employees (such as City dealers, of which Tradition Financial v Gamberoni [2017] EWHC 768 is an example) who are quite well placed to play potential employers off against each other, others do not. Now, just as in Lord Moulton's time, many employees may have limited individual or collective bargaining power. Threatening to require an employer to compete for an employee's services may be the best - or only - way of negotiating for better terms and conditions with that employer. Broad and unjustifiable restrictive covenants coupled with a threat of litigation can act as an impediment to a fair market for labour which can, in turn, result in employers being unjustly enriched by paying too little for key contributors to their success and valuable employees unjustly disadvantaged by being paid too little. That suggests that employers who maintain and assert post-termination covenants of undue width should face some penalty for doing so and not simply be able to cut them down to size, free of consequence, when challenged on them in expensive proceedings in court.

There is no doubt that the concern repeatedly expressed in this context is legitimate. However, it is for question as to how an approach to severance which makes the validity of a given prohibition depend on whether it is a "separate" or "single" covenant addresses the issue. As Russell LJ noted in Lucas, that problem is not solved by permitting severance of two restraints which are indubitably separate and distinct while refusing severance of items of a list in a single clause, if they are genuinely separate.

That point is well illustrated by the present case: by this conventional theory, the beam of the broadest non-compete clause can be readily cast out by severance because it is clearly a separate and distinct covenant, even though it is that covenant which is the main source of the problem of over-breadth of the covenants taken as a whole. Meanwhile, with that problem "solved" with no penalty to the employer for maintaining and enforcing it, intense scrutiny then focuses on the motes in the non-solicitation clause (in which no-one has shown particular interest and which are probably casual surplussage of somewhat risk-averse drafting). A rather scholastic debate is then forced upon the court as to whether these provisions should be regarded as involving separate, and therefore excisable, provisions or a single, irretrievably contaminated, whole. That can involve further meta-debate as to whether the distinction between separate and single is one of form or substance as which both sides are able to draw on ammunition from diverse courts whose actual approaches in individual cases are divergent. At its reductio, there is scope for this sort of argument to descend into debate over whether a formally unitary provision containing words or punctuation of disjunctive connotation should be regarded as separate or single in nature. Employers and employees are likely to view this with dismay, as costs rack up over commas. The stakes of such debate are high because, on this approach to the law, reflected in Egon Zehnder, the enforceability of even parts of a clause which are admitted to be perfectly reasonable and infringed depends on whether parts of the clause which no-one may have noticed much before the litigation started are severable.

Counsel for Freshasia submits that one must look to the substance not form and that has much to commend it. One way of doing so may be to recognise that, if relatively minor and reasonably isolable surgery can be performed on an otherwise legitimate set of post termination restrictions, that should be done and the remainder enforced, regardless of whether it is formally-speaking a "separate" covenant or equivalent to a separate covenant but in which whether a provision was a separate covenant was treated as a factor in evaluating whether or not the surgery was radical or major. In contrast, if a great deal of slimming is required for enforceability relative to the restrictions taken as a whole, or the effect is a radical change, that would be regarded as not the sort of contract that the parties entered into and should not be done.

However, whatever approach is adopted, if too strict an approach is taken to severance, rendering un-salvageable all covenants which, internally to a clause or separately, contain one or other stray restriction that goes too far but which are otherwise reasonable, that may lead courts to find that broader restrictions are valid, so as to avoid providing an ex-employer with no protection. That could have the paradoxical effect of making the position worse for employees since the power of surgical severance is, in part, a means by which a court is not forced into a binary decision between all-or-nothing protection (where nothing is often an unattractive option – as Counsel for Freshasia puts it, Mr Lu contended for no valid restraints at all).

Employment covenants are not the only area in which beneficiaries of rights seek to obtain maximum protection from competition for as long as possible, only cutting down to something reasonable when forced to the wire. Patent law, as a creature of statute, has specific provisions limiting remedies where a patent is only held partially valid and claims are limited (see e.g. Patents Act 1977 s. 63) and historically has taken rather a tough line on permitting amendment where protection of unjustified width has been claimed and asserted (see e.g. Terrell The Law of Patents 13^th edition (1982), chapter 8, discussing discretion to refuse amendment under earlier acts). However, in such contexts, the presence of specific statutory provisions addressing the mischief of a proprietor asserting claims of unjustified width only to retreat to narrower protection when required to do so, means that the courts do not have to rest their remedy for such conduct on the Procrustean bed of limiting severability.

Having regard to these considerations, necessarily somewhat superficially analysed on this interim application, the case law raises serious questions of approach meriting fuller argument and means that the point of law is not suitable for resolution on this application. I am reinforced in that view by the following.

First, counsel for Mr Lu submitted that the impact of Egon Zehnder is that decisions such as those in Norbrook were wrong. It has not been possible to undertake a full analysis of the many authorities, but the argument would probably suggest that other cases too were wrong, including cases decided by the Court of Appeal, such as Lucas, and recent cases which have relied on the jurisdiction to sever such as Lusona Consultancy (Accountancy & Finance) Ltd v McEwan & Anor [2017] ScotCS CSOH 52. She is not alone in taking that line. In that latter Scottish case, counsel suggested, with commendable boldness, albeit without success, that Russell LJ's approach in Lucas was anomalous, reflecting his failure to understand the earlier case law. If a court is to find that prima facie sound decisions of some of the most experienced commercial and employment law judges in the country are wrong, the optimum venue for that is not an application for an interim injunction where the central issue should be the rougher and readier question of whether limited precautionary relief should be granted over a period of a few months.

Second, permission to appeal was granted by the Supreme Court in Egon Zehnder and a hearing has been fixed for January 2019. The Supreme Court will have the various points touched upon above well in mind and numerous others.

There is a prospect that the Supreme Court will take the view that severance is permissible where a post-termination restriction goes too far to be valid but where the respect in which it does is sufficiently isolable not to affect the remaining restrictions if deleted, regardless of whether it may be described as a separate covenant, provided that deletion does not materially change the employment contract as a whole. It may consider that where an employer has extracted and/or unjustifiably relied on a much wider covenant to the detriment of the employee or others which is later cut down, that is better reflected in limitation or refusal of discretionary remedies, such as injunctions, interim or final, or in costs than in a limitation on severability.

Were the Supreme Court to take an approach along those lines, that might be consistent with the motivation or policy behind the authorities and may amount to little more than a synthesis of the practical effect of the existing case law, taken as a whole. If so, there would be a good case that, subject to the other points, the covenants in question in this case would, as limited, be upheld and an arguable case that injunctive relief would be granted in respect of them.

Third, directions have been made for a trial of this claim at which this point will be in issue possibly in the light of a decision by the Supreme Court. The approach the law should take is, in part, policy-driven, requiring more extensive analysis of the impact of the rival contentions and the full range of authorities. Moreover, it is preferable to consider these issues against a background of found facts, partly because that may shed light on the perceived importance of individual elements of the covenants.

THE POST-TERMINATION RESTRICTIONS IN ISSUE

By the time of the hearing, two covenants were relied on, a non-solicitation clause and a non-compete clause. I deal with these in turn.

Non-solicitation clause

The relevant clause is Clause B(1)(b) in the Employee Handbook and it is only necessary to consider this clause, even though there are others in substantially the same terms. It provides:

"For this reason [confidentiality] you hereby agree that you will not during your employment with the Company or any associated or subsidiary companies, for a period of six months (for non-senior employees) or twelve months (for senior employees) (hereinafter referred to as the "leaving period") in respect of any aspect of the business which the Company undertakes, solicit, or attempt to solicit the custom of, or sell, or deliver to or accept work for private gain and/or for any third party, from any private individual, firm or company or otherwise deal with any person who at the date of termination of this contract is a customer or potential customer of the Company to whom you have personally sold and/or delivered the Company's products on behalf of the Company or whom you had introduced to the Company, or approached on behalf of the Company, or with whom you had any business dealings or knowledge in the leaving period immediately prior to the date of termination of your employment."

Scope of clause – construction

Mr Lu rightly submits that the clause is not well drafted but the obligations are reasonably clear with certain exceptions. As to construction, there is one main issue namely whether the scope of the last phrase ("in the leaving period immediately prior to the date of termination of your employment") applies only to the last of the alternatives or governs all of the alternatives. Mu Lu contends that it applies only to the last whereas Freshasia contends that it applies to all of the alternatives (thereby limiting the restriction to soliciting those only in the latter part of the employment period).

In my view, the better construction is that it applies only to the latter clause. First, the structure and punctuation of the clause is more consistent with that interpretation. Second, it would be rational to provide a non-solicitation clause which prevented solicitation of customers whenever these were contacted, not merely in the final 12 months of employment since that would protect goodwill built up at any time. I do not consider the contrary to be unarguable though since it might be said that more limited protection is also justifiable. However, I do not think the enforceability of the clause turns on this point.

Protectable interest

In my judgment the non-solicitation provisions do protect a genuine and legitimate interest of Freshasia. Mr Lu contends otherwise on the footing that he did not have contact with customers (which is an issue in dispute) but that does not seem to be the right test, because the clause protects the customer connection and goodwill prospectively at the time the contract is made if he were to have such contact. The question of legitimate interest must be considered at the time that the agreement was made and, at that time, he may have had or been expected to have some contact with at least some customers. The evidence as to whether Mr Lu had significant contact with customers is also conflicting. Mr Lan says that Mr Lu's role included working closely with the sales team to carry out promotions at retail shops and that he was the person at Freshasia who was responsible for contacting customers in relation to promotions and store decorations such as posters. He also says that he contacted customers by telephone and face to face in relation to details of promotions, cookery demonstrations and promotional posters. It is also said that during his employment, Mr Lu built up good relationships with customers and was in touch with some 60% of Freshasia's main customers.

Mr Lu's evidence is to the contrary and that he had more limited contact with customers with a more general marketing role. It is to my mind more likely than not likely on the evidence at this stage that the truth lies somewhere between: Mr Lu was not significantly "customer facing" (contrast the position in Coppage), even though he may have had some contact with some customers. I am also concerned about the quality of the Freshasia evidence at this stage concerning Mr Lu's customer contact, particularly since I was not shown any document actually recording detailed contact with customers. Moreover, Freshasia had a series of account managers who were employed to have that contact. So this is an issue for trial. I am prepared to assume for the purpose of this application that at the time of the covenants it was anticipated that he might have some customer contact with some customers and that the clause was therefore included to ensure that, for a limited period, he did not exploit the specific goodwill he had thereby generated. However, for that reason, it seems to me appropriate that protection is granted only to the extent that goodwill is built up with actual customers.

Enforceability

In my judgment, the scope of prohibited activities is greater than reasonably necessary to protect Freshaisa's legitimate interest. First, as Mr Lu submits and Freshasia does not dispute it would preclude him from working in (say) a Chinese supermarket who is currently a customer of Freshasia with which he had had contact which decided to manufacture its own brand dumplings because of the provision prohibiting him from accepting work for private gain and/or for any third party, from any private individual, firm or company or otherwise dealing with any person who at the date of termination of this contract is a customer or potential customer of the Company with whom he had the relevant contact. That work would not make use of any goodwill of Freshasia. In my view, it is one thing to prevent an employee in the position of Mr Lu from soliciting custom from his previous customers for a new employer seeking to supply to those customers. It is another thing for him to be prevented from working at all for one of the customers with which he has dealt in the given ways. In my view it is likely that this aspect of the clause will be held to be too broad to be enforceable.

Second, the clause applies to "potential customers" as at the termination date which includes those who are not presently purchasing products from Freshasia but who might in the future (such as Chinese supermarkets who are in Freshasia's area of operation). If Mr Lu knew of the existence of a potential or actual customer and had approached them at any time during his employment he would be prohibited from soliciting them for 12 months after termination. Although there are different views expressed in the cases, a restriction on dealings with former "prospective customers" was found to be too wide in Norbrook (see at [104]) albeit that it could be severed, as was a restriction on dealing with former actual customers with whom the defendant had had no personal contact. In my view, and notwithstanding some indications in the discussion of the earlier case law in Coppage and elsewhere which might be said to tell in favour of such a provision, the present clause is more likely than not to be found to be too wide for similar reasons.

Third, the last aspect of the clause, which applies to those with whom Mr Lu had "any business dealings" or "knowledge" in the leaving period immediately prior to the date of termination of his employment, is potentially unrelated to those with whom he might be expected to have any goodwill and could include a wide range of trades people. In my view, this aspect of the clause is unlikely to survive scrutiny at trial. The clause also suffers from a lack of clarity as to what the scope of "business dealings or knowledge" is to be. This is potentially a very broad term and there is no clear way in which it may be interpreted to relate directly to the protection of a legitimate interest. Either it is, as I consider, very broad or if it is narrower, its scope is in the context of this particular contract unclear.

Finally, it is said by Mr Lu that the whole clause is void because it is unclear whether Mr Lu is a senior or junior employee, its duration is indeterminate. There is considerable force in this point but I do not consider that it automatically renders the whole clause void. It is true that Mr Lu is not told anywhere in his Contract or in the Employee Handbook whether he is a senior employee or a non-senior employee, and there is no contractual mechanism for determining that from the contract alone. Moreover, the word "manager" in his job title suggests that he might be senior; on the other, his low starting salary (£20,956.60 gross) and the fact that he is described as answering to a "director and senior marketing manager" suggests that he is "non-senior". On the other hand, he has described himself as the "marketing manager" of FreshAsia in a recent c.v., which may be true in effect today but not necessarily at the time the contract was made.

Those are powerful submissions but I do not accept that they are conclusive at this stage of the proceedings. First, the question of whether Mr Lu was a senior or junior employee at the time the contract was made is not, in my view, purely a question of construction of the contract. It is in part a question of fact determined by matters such as the nature of his employment as compared with other employees. Second, it is possible that, at trial, Freshasia may be able to establish that there was not in fact any serious doubt that Mr Lu was a senior employee nor any serious doubt as to which category each employee with a similar clause fell. It is too early to tell whether this will be possible and it seems to me at present that Freshasia has an arguable, albeit not strong, case that the clause is not too uncertain in this respect.

Accordingly, there is an arguable case that parts of this clause are in principle appropriate to protect Freshasia's legitimate interests.

Severability

In my judgment, in the light of the approach discussed above, it is arguable that the respects in which this clause is likely to be invalid are severable from the clause as a whole on the footing that applying the principles from the existing case law they are to be treated as sufficiently separate to satisfy the requirements in Beckett and that there is a real prospect that, in any event, the Supreme Court will reach a decision in Egon Zehnder which leads to covenants of this kind being treated as severable.

Conclusion on non-solicitation clause

In my view, had the remainder of the clause with the offending parts excised featured in Mr Lu's contract at the outset, it would arguably be sufficiently certain and sufficiently reasonable to be enforced. It is not possible to say at this stage that Freshasia is more likely than not to succeed on the claim based on this clause but it is at least well arguable: the case that an injunction will be held at trial to have been justified (i.e. as cut down) is evenly balanced.

Non-compete clause

The matter is different with the non-compete clause. This is Clause C(a)(a). It provides:

You agree not to directly or indirectly compete with the business of the Company and its associated companies during the period of employment and for the leaving period and notwithstanding the cause or reason for termination. […]

The term "not compete" as used herein shall mean that you shall not own, manage, operate, consult or be employed in a business substantially similar to or competitive with, the present business of the Company.