SC 78/2009

IpsofactoJ.com: International Cases [2010] Part 1 Case 12 [NZSC]


SUPREME COURT OF NEW ZEALAND

Coram

Air Nelson Ltd

- vs -

New Zealand Amalgamated

Engineering, Printing &

Manufacturing Union Inc

ELIAS CJ

BLANCHARD J

McGRATH J

WILSON J

ANDERSON J

17 MAY 2010


Judgment

Anderson J

(delivered the judgment of the court in which Elias CJ, Blanchard J & McGrath J, joined)

  1. Air Nelson Ltd, the appellant, carries on business as an airline operator. Its “line maintenance” (relatively minor repairs and the servicing of aircraft between flights and overnight) is mainly carried out by its own employees. The rest, amounting to between one and two per cent, is carried out by contract engineers who each spend about five hours of their working week on line maintenance. If the amount of line maintenance work required at any particular time during the day is more than the available employed line maintenance engineers can do, Air Nelson deploys contract engineers to assist. In the words of the Employment Court, “This does not occur regularly or frequently but does occur routinely.”[1]

  2. In June 2007, when line maintenance employees were engaged in a lawful strike, contract engineers carried out a routine inspection of an aircraft and replaced some cabin and reading lights. The following day a contract engineer carried out more line maintenance by performing what is known as an engine compressor wash. Whether, but for the strike, the inspection, light replacement and compression wash would probably have been performed by employees or by contract engineers is a factual issue on which no finding has been made.

  3. The respondent union asserts and the appellant denies that by carrying out the particular line maintenance during a lawful strike Air Nelson contravened s 97 of the Employment Relations Act 2000. That section provides:

    97.

    Performance of duties of striking or locked out employees

    (1)

    This section applies if there is a lockout or lawful strike.

    (2)

    An employer may employ or engage another person to perform the work of a striking or locked out employee only in accordance with subsection (3) or subsection (4).

    (3)

    An employer may employ another person to perform the work of a striking or locked out employee if the person—

    (a)

    is already employed by the employer at the time the strike or lockout commences; and

    (b)

    is not employed principally for the purpose of performing the work of a striking or locked out employee; and

    (c)

    agrees to perform the work.

    (4)

    An employer may employ or engage another person to perform the work of a striking or locked out employee if—

    (a)

    there are reasonable grounds for believing it is necessary for the work to be performed for reasons of safety or health; and

    (b)

    the person is employed or engaged to perform the work only to the extent necessary for reasons of safety or health.

    (5)

    A person who performs the work of a striking or locked out employee in accordance with subsection (3) or subsection (4) must not perform that work for any longer than the duration of the strike or lockout.

    (6)

    An employer who fails to comply with this section is liable to a penalty imposed by the Authority under this Act in respect of each person who performs the work concerned.

  4. The dispute came before the Employment Court which found in favour of Air Nelson, holding that the limited amount of line maintenance could properly be regarded as the contract engineers’ own work rather than that of striking employees: New Zealand Amalgamated Engineering, Printing & Manufacturing Union Inc v Air Nelson Ltd (No 2) [2007] ERNZ 725 (EC) at [28].

  5. The union appealed to the Court of Appeal on a question of law as to the meaning of the words “the work of a striking or locked out employee” as used in s 97. The Court of Appeal, being of the opinion that these words mean the work a striking or locked out employee would probably have been performing had he or she not been striking or locked out, found in favour of the union: New Zealand Amalgamated Engineering, Printing & Manufacturing Union Inc v Air Nelson Ltd [2009] NZCA 349, [2009] ERNZ 178.

  6. In order to assess the reasoning in the Courts below it is necessary to examine Finau v Atlas Specialty Metals Ltd [2009] NZCA 348, [2009] 3 NZLR 771 which the Court of Appeal heard and determined simultaneously with the present case. In that case, some but not all employees at an engineering factory took lawful strike action in the course of bargaining for a collective agreement. Two machinists refused to operate a coil slitter machine and were suspended as parties to the strike. Two other employees were instructed to operate the coil slitter, which they were trained to operate and from time to time did in fact operate when the usual operators were unavailable. They refused and were similarly suspended. At issue in the litigation was the lawfulness of their suspension, in light of s 97.

  7. The Employment Court: Finau v Southward Engineering Co Ltd [2007] ERNZ 522 (EC).

  8. The Court preferred the latter, being of the view that s 97 is not intended to cripple a business where only some of the employees are on strike. In its opinion the particular task approach would tilt the balance of interests too much in favour of unions and employees. For example, such an approach would prevent an employer from rearranging rosters efficiently to use the services of non-striking employees unless they explicitly agreed to such a change. The particular task approach would render s 97 essentially prohibitory and very largely rob it of any efficacy for employers subject to strike action. noted that while many employees will have only one specified job, task or role that they perform consistently, others may rotate around a range of tasks or have certain duties which they principally perform and other duties which they perform from time to time. Section 97 therefore had to be construed in a manner that gives effect to the relevant Part of the Act not only in relation to employees with a single consistent role but also in relation to those employees whose tasks change from time to time. It then examined “work of a striking employee” as a dichotomy: did the expression contemplate inquiry into the particular task that, but for the strike or lockout, would have been done by the striking or locked out employee; or did it contemplate an examination of the type of work usually done by a worker who is on strike or locked out?

  9. The alternative was to treat the work of the striking employee according to its general type so that the question would be whether the type of work that was being done by a striking employee was work normally done by the non-striking employees. If it is the type of work which comes within the normal duties of the non-striking employees, then those employees are not being asked to do the work of a striking employee but their own work. The Employment Court preferred this alternative, saying (Finau at [30] and [31]):

    We prefer the “type of work” approach which would enable employers to direct non-striking employees to do particular tasks within the range of work they normally perform but would require the agreement of those employees to do work they do not normally perform.

    This construction is based on the concept of what may properly be said to be work which an employee normally performs. We take the view that it comprises tasks which the employee regularly or routinely performs in the course of employment. This would not include tasks an employee might occasionally be required to do pursuant to a “catch all” provision of an employment agreement … The key is what an employee actually does as a matter of practice, rather than what may be contained in a job description or otherwise be provided in an employment agreement.

  10. The Employment Court disposed of the issue in the present case by applying its decision in Finau. It held that the extent to which the contract engineers were deployed to do line maintenance work during the strike was within the range of work which they routinely performed, was standard practice and unexceptional. The Court added:

    In this case, we wish to make it clear that whether work which is done by a contractor or non-striking employee during a strike is the work of a striking worker or that person’s own work will be a matter not only of the type of work but also the extent to which the person does that type of work when there is no strike. To a large degree this decision turns on its relatively unusual facts. It should not be taken as a licence to employers to break strikes by changing significantly the pattern of work normally performed by contractors or non-striking employees.

    (New Zealand Amalgamated Engineering, Printing & Manufacturing Union Inc v Air Nelson Ltd (No 2) [2007] ERNZ 725 (EC) at [31])

  11. There might seem to be some awkward aspects of the Employment Court’s reasons in these cases. In relation to the type of work approach, it uses the words “usually”, “normally”, “regularly”, and “routinely” as if they are synonymous, whereas they have subtle differences of meaning. In Finau the Court ultimately settled on “regularly” and also “routinely”. In its reasons in the present case the Court took account of the range of work routinely performed, and considered the extent of work and cautioned against significantly changing patterns. As we indicate later in this judgment, what is likely to be of most importance in a case of multi-faceted work is its usual pattern.

  12. Another matter is that beyond giving the reasons for rejecting the alternative, the Court was sparse in its reasons why it preferred the type of work approach. Essentially the Court considered that alternative (particular task) to be unbalanced and prohibitory in effect. Yet s 97 is intended to be prohibitory except in the specific restricted situations described in subsections (3) and (4), compliance with which exempts an employer from the general prohibition. It is clear that the section intentionally tilts the balance in favour of striking or locked out workers. It is a firm antistrike-breaking mechanism. It confers employment-related rights on employees and “constrain[s] the bargaining power of the employer for the benefit of striking or locked out employees”: Carter Holt Harvey Ltd v National Distribution Union Inc [2002] ERNZ 239 (CA) at [30].

  13. The Court of Appeal appeared to think that the Employment Court may have shifted its focus from the work of the striking or locked out worker to the work of one who is not striking or locked out. A reading of paragraph [30] of the Employment Court’s reasons may suggest that, but, as we note later in this judgment, examination of the usual work of one who is not striking or locked out as well as one who is striking or locked out is part of the method for determining whether there has been or might be a breach of s 97; and in any event, paragraph [31] and the judgments in Finau and the present case, taken together, show the type of approach the Employment Court favoured.

  14. When the Court of Appeal considered Finau, its reasons for allowing the appeal comprised both a preference for its own construction, namely that the words in issue meant the work that a striking or locked out employee would probably have been performing but for the strike or lockout, and its opinion that the Employment Court had erred in a number of ways. Its preference was founded on a purposive approach and a counterfactual hypothesis. As to legislative purpose, it was of the view that Parliament must have intended that employers faced with strike action and non-striking employees should easily be able to ascertain what work could or could not be done. The counterfactual element required a mental construct of the workplace on the hypothesis that there was no strike.

  15. The Court was also of the opinion that the Parliamentary history of s 97 supported its view. We note that in the Employment Court, the Court of Appeal, and the reasons of Wilson J, which we have had the advantage of reading in draft, three different views of the legislative history have been relied on. In our respectful opinion no assistance at all is derived from the history.

  16. We turn now to the Court of Appeal’s reasons for finding that the Employment Court was in error in Finau.

    • First, it thought it clear that the Employment Court’s construction requires words to be read into s 97, effectively grafting onto it another section to the effect that nothing in s 97 prevents an employer requiring an employee to perform work which the employee normally performs even if, but for the lawful strike or lockout, the employee would not have been required to perform that task on that occasion.

    • Second, it held that the Employment Court’s interpretation simply does not work in the case of subsection (4), which deals with health and safety considerations. That is because, in the Court’s view, the replacement worker may be one who does not normally perform the striker’s work, leading to different meanings in different parts of s 97, which cannot have been intended.

    • Third, the interpretation would be difficult to apply in practice. That is because a strike-breaker would have difficulty in ascertaining what a striking worker’s usual work would be.

    • Fourth, the Employment Court’s interpretation leads to potential abuse by way of wide job descriptions in employment agreements. The Employment Court’s view that the key was what the non-striking employee does in practice, rather than what might be contained in a job description, would necessitate a further qualification to the meaning of “normally” in the hypothetical further subsection discussed by the Court of Appeal, giving rise to further uncertainty. The Court of Appeal was concerned that union secretaries, when called upon to advise at short notice, could not merely look at the terms of a collective agreement but would need to be fully briefed on how a particular workplace operated in practice.

    • Fifth, the Employment Court’s interpretation leaves little scope for the application of s 97. It would apply only in cases where the potential strike-breaker is being asked to do work which is not within his or her normal range of work but which is nevertheless within the range of work in the striking employee’s employment contract.

  17. When considering the present case, in the light of its judgment in Finau, the Court of Appeal held that the Employment Court had asked the wrong question. Its focus should not have been on what the contract engineers normally did but whether they did work which, but for the strike, a striking employee would have done. It considered that if the Employment Court’s decision were allowed to stand it would effectively rob s 97 of its efficacy; it would hardly ever apply. The test favoured by the Court of Appeal meant, it held, that the person who requested the contract engineers to do the maintenance in question would have easily been able to ascertain who would have done it but for the strike, and if it would probably have been done by a striking employee then s 97 was engaged. He would then have had to assess whether the work was necessary for reasons of safety. The Court of Appeal thought that the greater certainty of this approach compared with the perceived uncertainty resulting from the Employment Court’s approach was a strong pointer that its own interpretation was the correct one.

  18. Just as we have difficulty with aspects of the Employment Court’s reasons, so we do with those of the Court of Appeal. We express them in connection with the five reasons the latter gives for disagreeing with the Employment Court. As to the first reason, what is criticised as a hypothetical, indeed “phantom” subsection, is in reality just the factual consequence of the Employment Court’s approach. As to the second reason, an available response is that the analysis of respective roles is likely to be simplified. In any event, the combined reasons of the Employment Court in the present case and Finau indicate that it was not so much attempting linguistic construction as adopting a method for determining whether s 97 may have been breached. The realisation of that diminishes semantic objections. Concerning the third reason, if there is an integration of work one ought to assume that the integrated employees and/or contractors know what their usual work is. If not, that would indicate a break in usual work patterns. As to the fourth reason, which takes an assumed factual view of union practices, we think it wrong to assume that union secretaries will act in ignorance of workplace practices and realities; that unbriefed on all relevant considerations, they would make a decision whether or not to object to a situation simply on the basis of reading a collective agreement. We do not see s 97 calling for urgent decisions of that nature. Coming now to the fifth reason, we think this states what is self-evident and thereby begs the question under consideration.

  19. The judgments in the Courts below illustrate the difficulty of trying to find a definition for what is really a contextual judgment. The essential question is concerned not with the meaning of “work” but with recognising the nature and scope of particular work in particular cases. That is, the issue is not one of construction but of application. The assumption by the Court of Appeal that the case is concerned with construction led it to mischaracterise the approach of the Employment Court as one of having asked itself the wrong question. But in our view the Employment Court did not ask itself the wrong question. Rather, it took a different view of the facts.

  20. A person’s work may in fact engage many tasks, not necessarily as comprehensive in practice as may be stipulated in a particular employment contract. Other employees of the same employer may have similarly described duties in their employment contract. Some may regularly, for a greater or lesser length of time, carry out some of the tasks that another worker usually does. In such cases the duties, or work, of one is integrated with, or qualified by, or adumbrated by the work or duties of another. How, and to what extent, will establish a pattern, analysis of which will determine whether a striking or locked out worker’s work is being performed in breach of s 97, rather than whether the non-striker is performing his or her own work. Where there is a departure from a pattern of integration of work, that may indicate s 97 has been contravened.

  21. The analysis referred to in the preceding paragraph may have to consider the manner in which the work of a person who is not striking or locked out impinges on that of a striking or locked out worker. So, in a case such as the present, it may be relevant to the analysis if the striking employee’s work was always, in fact, affected in a certain way, by the contract engineer’s involvement in line maintenance, so that in fact there was no impingement upon the scope of the work of the striker. Alternatively the line maintenance involvement of contract engineers on the days in question might involve a type of line maintenance that in practice was always done by employees without assistance from contract engineers; or might exceed normal involvement in terms of duration or timing. The mutual duties of good faith stipulated by s 4 of the Act, including its specific application by subs (4)(a) to bargaining for a collective agreement, mandates a fair and reasonable approach by employers and employees to the factual issues and assessments required when s 97 may be in question.

  22. Earlier we mentioned the Employment Court’s variable use of “usually”, “normally”, “regularly” and “routinely”. We think any of these fairly, if not exactly, similar terms may be apt in a particular case, depending on the circumstances. The analysis of an employee’s work may require a consideration of what it entails, usually, or normally, or regularly, or routinely. Such considerations bear on a context of integration with or circumscription of a striking or locked out employee’s work. The nuances of these near synonyms emphasise the need for analysis of the particular or individual situation when a breach of s 97 is apprehended. We do not favour a narrow, prescriptive approach that would encourage one worker’s entitlement to work being undermined by another worker’s entitlement to strike. A broader, fact-specific analysis does not derogate from the latter’s right to withhold work; it just makes it less likely that the employee will take away a non-striker’s work as well.

  23. Generally speaking, the Employment Court sought a balance between one person’s right to strike and another person’s right to work. Its approach was not inappropriate. If the contractors were doing their own work it is obvious that they were not doing another person’s work. Whether or not they were is not a question of definition. It is a question of fact relative to an ordinary English word. As we have indicated above, s 97 does not raise issues of construction but of application.

  24. We would decline to refine upon the statutory language, which is straightforward. It calls for judgment. Another worker cannot be substituted for a striking worker in the performance of the work of the striking worker. The approach adopted by the Employment Court was not in error. Nor do we think it differs in substance from the approach suggested by the Court of Appeal, based on the position that would have applied “but for” the strike. The Employment Court was right to stress that this was a matter of practical substance not potential duties under the employment contract. Looking at the matter as a matter of substance it came to the conclusion that the use of contract engineers who themselves habitually performed some of the line maintenance work (although a small proportion of it) was not in the circumstances performance of “the work of the striking employee”. That assessment was open to the Employment Court. The Court of Appeal was not able to substitute its judgment on the facts. Its characterisation of the approach of the Employment Court as “asking the wrong question” was wrong.

  25. In accordance with the view of the majority of the members of this Court, the appeal is allowed with costs of $15,000 to the appellant, together with its reasonable disbursements to be fixed if necessary by the Registrar. Costs in the Court of Appeal should now be fixed by that Court in light of this Court’s judgment.

    Wilson J

  26. This appeal raises the single and narrow issue of whether the references in s 97 of the Employment Relations Act 2000 to “the work of a striking or locked out employee” are to the actual work of that employee or to the type of work. Which interpretation is correct is a question of law, and is therefore a question which may be the subject of an appeal from the Employment Court to the Court of Appeal under s 214 of the Act and a further appeal to this Court under s 214A. The appellant contends that, as the Employment Court found,[2] the references are to the type of work. The respondent, supporting the judgment of the Court of Appeal,[3] that they are to the actual work. The majority of this Court prefers the approach of the Employment Court. I do not. My reasons are as follows.

  27. The focus of the wording of s 97 is on what employees would have been doing if they had not been striking or locked out. This is apparent from the heading of the section: “Performance of duties of striking or locked out employees”. Under s 5(2) and (3) of the Interpretation Act 1999, that heading may be considered in interpreting the section. The focus on what the striking or locked out employees would have been doing is confirmed by the five references in the section to the performance of “the work of a striking or locked out employee”.[4]

  28. On the appellant’s argument, even if work would in all probability have in fact been performed by striking or locked out employees, that work is not for the purposes of s 97 the work of the striking or locked out employee if performed by other employees who from time to time carry out that type of work. Such a conclusion is contrary to the plain and ordinary meaning of the phrase “the work of a striking or locked out employee”. The work which such an employee would actually have been performing, if not on strike or locked out, is therefore the foundation of the section. The statutory test is whether those on strike would have been doing the work, not whether others could have done it as an extension of their usual duties.

  29. That point is well-illustrated by the evidence in these proceedings. The employed engineers who were engaged in a lawful strike performed 98 to 99 percent of line maintenance for the appellant.[5]

  30. Section 97(4) permits an employer to employ or to engage the services of employees or contractors not previously employed, if necessary for reasons of safety or health. Such employees or contractors cannot previously have been performing the work of those striking or locked out. The test for determining whether they can be engaged must therefore be whether they are carrying out the actual work of those striking or locked out, not whether they were previously performing work of the same type. There cannot be any justification for applying different tests to employees and contractors who are already employed or engaged and those who are not.

  31. The legislative history of s 97 provides some, albeit modest, support for the proposition that it is directed to the actual work which would have been performed by the striking or locked out employees. In the Bill as introduced, the clause[6] which corresponded to s 97 as enacted referred to the work “normally performed by an employee who is participating in the strike or affected by the lockout”. The change to “the work” of such an employee justifies the inference that “the work” is the actual work which would have been performed, not work of the type normally performed.

  32. More importantly, policy considerations provide strong support for interpreting s 97 widely rather than narrowly. The section is intended to be a code to regulate the use of employees or contractors as strike-breakers or in support of a lockout. Parliament has carefully legislated to provide a balance between the interests of employers in continuing to carry on their business during lawful strikes and lockouts and the interests of employees and their unions in ensuring that strikes are effective and the consequences of lockouts are controlled. To the extent that the use of employees or contractors is held to be outside the ambit of s 97, the legislative balancing of interests in that section is rendered irrelevant. It is therefore desirable to construe the section as having the widest application which its wording permits. More specifically, the protection which the section confers on employees by directing that they cannot be required to act as strike-breakers

  33. I would therefore dismiss the appeal, with costs to the respondent. should not be undermined by adopting a restrictive rather than an expansive interpretation of the application of the section.


[1] New Zealand Amalgamated Engineering, Printing & Manufacturing Union Inc v Air Nelson Ltd (No 2) [2007] ERNZ 725 (EC) at [14].

[2] New Zealand Amalgamated Engineering, Printers & Manufacturers Union Inc v Air Nelson Ltd (No 2) [2007] ERNZ 725 (EC) and the related judgment in Finau v Southward Engineering Co Ltd [2007] ERNZ 522. submits

[3] New Zealand Amalgamated Engineering, Printers & Manufacturers Union Inc v Air Nelson Ltd [2009] NZCA 349, [2009] ERNZ 178 and the related judgment in Finau v Atlas Specialty Metals Ltd [2009] NZCA 348, [2009] 3 NZLR 771.

[4] In subs (2), subs (3), para 3(b), subs (4) and subs (5). In the absence of any direct evidence as to who, but for the strike, would have carried out the work in issue, the chances of that work being performed by the non-striking contractors were therefore between one in 50 and one in 100. It is unrealistic to say that the contractors were performing their usual work. To the contrary, the work in question would in all probability have been carried out by the striking employees if they had not been on strike.

[5] Because the focus of the section is on the work which would have been performed by the striking employees, and because most of the engineers were employees rather than contractors, this is a much more significant figure than the fact (relied on by the appellant) that the contractors worked on average for five hours a week on line maintenance.

[6] Employment Relations Bill 2000 (8–1), cl 111.


Representations

C H Toogood QC and D J France (instructed by Messrs Kiely Thompson Caisley, Auckland) for Appellant.

R E Harrison QC and G Lloyd (instructed by Messrs Gregory Lloyd, Wellington) for Respondent.


all rights reserved