National Union of Metalworkers of South Africa v Commission for Conciliation Mediation and Arbitration and Others (PA6/19; PR03/18 ; PR50/18) [2020] ZALAC 8; (2020) 41 ILJ 1629 (LAC) (18 February 2020)

IN THE LABOUR APPEAL COURT OF SOUTH AFRICA, PORT ELIZABETH

LAC Case no: PA6/19

LC Case no: PR03/18 & PR50/18

In the matter between:

NATIONAL UNION OF METALWORKERS

OF SOUTH AFRICA

Appellant

and

COMMISSION FOR CONCILIATION

MEDIATION AND ARBITRATION

First Respondent

FEIZAL FATAAR N.O.

Second Respondent

SOLIDARITY TRADE UNION

Third Respondent

UASA

Fourth Respondent

METAL AND ENGINEERING INDUSTRIES

BARGAINING COUNCIL

Fifth Respondent

MOTOR INDUSTRY BARGAINING COUNCIL

Sixth Respondent

WIDNEY TRANSPORT COMPONENTS (PTY) LTD Seventh Respondent

RAMSAY ENGINEERING (PTY) LTD

Eighth Respondent

EURO METAL FINISHES (PTY) LTD

Ninth Respondent

AUTO INDUSTRIAL MACHINING DIVISION Tenth Respondent

ISANDO FOUNDRY DIVISION

Eleventh Respondent

HUBCO FORGINGS DIVISION

Twelfth Respondent

AUTO INDUSTRIAL GROUP (PTY) LTD Thirteenth Respondent

AUTO INDUSTRIAL FOUNDRY DIVISION Fourteenth Respondent

AUTOCAST SA (PTY) LTD

Fifteenth Respondent

AUTOCAST SA (PTY) LTD ALUMINIUM Sixteenth Respondent

BORBET SA (PTY) LTD

Seventeenth Respondent

DANA SPICER AXLE SA (PTY) LTD

Eighteenth Respondent

MW WHEELS SA (PTY) LTD

Nineteenth Respondent

SP METAL FORGINGS BOKSBURG (PTY) LTD Twentieth Respondent

SP METAL FORGINGS UITENHAGE

(PTY) LTD

Twenty-First Respondent

TORRE AUTOMOTIVE (PTY) LTD

Twenty-Second Respondent

ZF LEMFORDER SA (PTY) LTD

Twenty-Third Respondent

MALBEN ENGINEERING CC

Twenty-Fourth Respondent

Heard: 26 November 2019

Delivered: 18 February 2020

Summary: Demarcation dispute -

Coram: Davis and Sutherland JJA and Murphy AJA

JUDGMENT

SUTHERLAND JA

Introduction

[1] The controversy in this appeal is about whether the regulation of the Terms and Conditions of Employment of employees in businesses which are engaged in the manufacture of components to be used in assembling motor cars properly belongs within the jurisdiction of the Metal and Engineering Industries Bargaining Council (MEIBC) or within the jurisdiction of the Motor Industry Bargaining Council (MIBCO).

[2] Section 62 of the Labour Relations Act 66 of 1995 (LRA) regulates such disputes.[1] The two Bargaining councils and several employers cited, in the appeal, as the 7th to 25th respondents were parties to a demarcation enquiry. The employer respondents made common cause with MIBCO (the 5th respondent) that they belonged within its jurisdiction. MEIBC (the 4th respondent) contended that they belonged under its jurisdiction. NUMSA, the present appellant, likewise contended that the several

employers be regulated by MEIBC.

[3] A commissioner issued an award in which some of the employers were assigned to each of the two bargaining councils. The orders made in the award at [59] – 61] read thus:

‘[59] The applicants, Auto Industrial Foundry Division, Autocast SA (Pty) Ltd, Autocast SA (Pty) Ltd Aluminium, Borbet SA (Pty) Ltd, Dana Spicer Axle SA (Pty) Ltd, MW Wheels SA (Pty) Ltd, SP Metal Forgings Uitenhage (Pty) Ltd and ZF Lemforder SA (Pty) Ltd, SP Metal Forgings Boksburg (Pty) Ltd and Malben Engineering CC fall within the scope and registration of the …. Metal and Engineering Industries Bargaining Council.

[60] The applicants, Auto Industrial Machining Division Isando Foundry Division, Hubco Forgings Division, Widney Transport Components (Pty) Ltd, Ramsay Engineering (Pty) Ltd, and Euro Metal Finishes (Pty) Ltd, fall within the scope and registrFfitation of the fifth respondent. They are demarcated from the scope and registration of the fourth respondent to that of the ……. Motor Industry

[61] Torre Automotive falls within the scope and registration of the [MEIBC]’

[4] This award aggrieved several of the employers assigned to MEIBC and also aggrieved MIBCO who thereupon brought independent review applications. A cross-review was brought too, about the employers assigned to MIBCO. The matters were consolidated. A single judgment was given.[2] The judgment set the award aside in part and dismissed the cross-review.

[5] The order of the Review Court reads thus:

1. That part of the award issued by the second respondent [the

commissioner] on 31 July 2017 under case number ECPE 2470-15 in which he found that certain of the applicants in the proceedings under review fall within the scope of registration of the sixth respondent [MEIBC] is reviewed and set aside.

2. Paragraphs 59 and 61 of the award are substituted with a ruling that the applicants fall within the scope of the Motor Industry Bargaining Council, and are so demarcated.

3. The cross-review and conditional cross-review are dismissed.

4. There is no order as to costs.

[6] In short, the review judgment reversed the assignment of employers to the MEIBC and assigned them all to MIBCO. NUMSA appeals against that judgment, seeking, in effect, an order assigning all to MEIBC.

The issue on Appeal

[7] At the appeal hearing, the nub of the case was refined. It had become common cause that the award had to be set aside. The reason for setting it aside was that the commissioner had made a material misdirection by subordinating the enquiry to the fact that in 1962, there had been a Demarcation Determination by the Minister, later modified by a Clarification Notice in 1964 by the then Industrial Tribunal. The substance of these instruments had the effect of subjecting the respondent employers mentioned in paragraph [59] of the award, as cited above, to the jurisdiction of MEIBC. The commissioner incorrectly supposed that the Determination and the Clarification remained binding on these parties in 2017 and therefore did not evaluate afresh the question of whether these employers should be subjected to the jurisdiction of MEIBC. The balance of the employers were assigned by the commissioner to the jurisdiction of MIBCO.

[8] Accordingly, because the award was tainted by that irregularity, it had to follow that it was correctly set aside on review. The direct consequence of such a finding, logically, is that no part of the award could stand or be severed and revived because the contamination extended to the whole enquiry. Such circumstances therefore imply that the Review Court had to either remit the matter for a fresh hearing or make the decision that the commissioner had been required

to make.

[9] The parties are ad idem that the matter should not be remitted. This is a correct stance because there are no facts in contention and all the factual material necessary to reach a decision was before the commissioner and thus also before the Review Court. The Review Court was therefore correct to make a demarcation order itself. Plainly, the self-same considerations about the factual material are applicable to the appeal against the Review Court’s decision.

[10] Thus, the question of whether to differ from the Review Court’s judgment on the re-assignment of the several employers to MIBCO is straight forward: was that decision correct?

The jurisprudence of demarcation disputes

[11] It is at once apparent that demarcation disputes between Bargaining Councils (or their putative constituents) are a sui generis species of dispute. The very foundation of the idea that various economic activities can be logically defined and categorised into

silo-like realms is contrived. The exercise of drawing dividing lines between “industries” must be understood to be artificial and is necessary only because of the policy choices which underlie the effort to distinguish various (frequently closely related) industrial activities for the purpose of segmenting the responsibility of regulating terms and conditions of employment into invented convenient sectors, themselves having no objective existence and being the product of the imagination of the policymakers. A grasp of what these policy decisions were made for and what was sought to be achieved by so doing is a critical dimension of approaching the exercise correctly.

[12] The notion that, for the practical purposes of regulating employment conditions in economic activities, by assigning some enterprises to one or other bargaining council proceeds from the foundational idea that “grouping” like with more or less alike is a sensible pragmatic approach. Central thereto is the attempt, by the use of words, to describe the supposedly distinguishable economic activities in definitions which are almost always complex, wordy and often hair-splitting. The task aims at describing the characteristics or attributes of industrial activities. Then, the characteristics or attributes of a business enterprise are described and the two are compared. Just as it is not objectively possible to determine when night ends and day begins, and a practical answer depends on what you want to pinpoint that moment for, so it is with demarcation of so-called distinct “industries”.

[13] Another dimension of the exercise that warrants acknowledgement is that the exercise is as much one of creation as of adjudication. The meaning of words of the defined scope of a bargaining council can be indeed adjudicated, but that is not always enough. The management of the reality that economic activities within the invented sectors, sometimes differently described, often overlap and, cannot therefore, in logic, be truly separated, means that a pragmatic policy decision to locate a given enterprise on one or other side of an imaginary fence is an inescapable aspect of the task of demarcation. It resembles, in some respects, an Interest Arbitration.[3] What is sought is what may usefully be called the “best fit”– an idea that defies precision and is axiomatically fact–specific.

[14] The nature of the demarcation exercise was addressed by this court in SA Municipal Workers Union v Syntell (Pty) Ltd & others:[4]

[21] In the main, arbitrations under the LRA are those which address disputes of right and are adjudicative proceedings proper. In s 62, the word 'arbitration' is not used to describe the process. Indeed, if a 'demarcation' issue arises in any ordinary adjudicative proceedings, those proceedings must be stayed until the demarcation issue is decided in the distinct process provided for in s 62.

[22] The s 62 process, as is evident from its provisions, contemplates more than a conventional adversarial contest between interested parties. It presupposes a broader investigative role. In such a context, whether or not an onus in any sense exists is not obvious.

[23] These considerations which are imbedded in the provisions of the section underscore its sui generis character. The s 62 process was commented on by Francis J in Coin Security (Pty) Ltd v CCMA & others (2005) 26 ILJ 849 (LC) at paras 43 and 63:

'[43] The function of a CCMA commissioner in a demarcation dispute is a classic case of the legislature entrusting a functionary with the power to determine what facts are about the making of a decision and the power to determine whether or not they exist. It is fundamental to the effective operation of the Act that the commissioner must be a repository of such power. …

[63] The demarcation process is one entrusted to a specialist tribunal in terms of the provisions of the Act. The demarcation decision is one involving facts, law and policy considerations. In demarcation decisions, there will, more often than not, be no one absolutely correct judgment. Particularly in decisions of this sort, and given the provisions of the Act, there must of necessity be a wide range of approaches and outcomes that would be in accordance with the behests of the Act. Due deference should therefore be given to the role and functions and resultant decisions of the CCMA in achieving the objects of the Act. This approach will not only be consistent with these principles, but also consistent with the need for the Act to be administered effectively.

[24] More recently, Van Niekerk J affirmed this perspective in National Bargaining Council for the Road Freight Industry v Marcus NO & others (2011) 32 ILJ 678 (LC) at para 22:

'It should also be recalled that Coin Security is also authority for the point that a demarcation involves considerations of fact, law and social policy and that in these circumstances, due deference ought to be given to a commissioner making a demarcation award (at para 63 of the judgment). As I understand the judgment, in demarcation judgments there will be, more often than not, no single correct judgment, and that a wide range of approaches and outcomes is inevitable. A reviewing court should be attuned to this reality, and recognize it by interfering only in those cases where the boundary of reasonableness is crossed. Further, Coin Security recognizes that a demarcation is provisional — s 62(9) of the LRA requires a commissioner to consult with NEDLAC before making an award. As the court in Coin Security observed, the case for judicial deference is all the more compelling in these circumstances. In short, far from encouraging an expansive approach to a demarcation, the Coin Security judgment requires this court to recognize the specific expertise of commissioners who undertake this task and to defer to that expertise.'

The factual context

[15] The employers, who initiated the review application sought to upset the decision in the award to assign them to the MEIBC, were at the time of the section 62 enquiry, subject to the jurisdiction of the MEIBC and had been since at least 1962. The point of the demarcation enquiry was to test whether, in 2017, that position was still the best fit. The employers argued it was not.

[16] At the enquiry, the sole witnesses were Mr Manners, the owner of two of the enterprises and Mr Pauw of MIBCO. Manners advanced a rationale why the status quo should not continue. There was no evidence-based rebuttal put up to sustain the status quo although the two witnesses were cross-examined.

[17] The common cause facts were recorded in the award. What these recorded facts acknowledge is that every employer is a producer of motor car components and that their exclusive or dominant enterprise is the manufacture of car components. This evidence is summarised in paragraph [7] of the award. It does not warrant repetition because the significant inference, drawn from the facts, is common cause: all the businesses are predominantly or exclusively manufacturers of motor vehicle components.

What are the “industries” that are subject the jurisdiction of MEIBC and MIBCO?

[18] Each Bargaining Council has a defined scope as issued by the Registrar of Labour Relations and promulgated in the Government Gazette. The full texts are too labyrinthine to digest and thus only the portions which are pertinent to common cause facts are addressed in this analysis.

[19] It is useful to begin with the scope of MEIBC. It is described initially as the “Iron, steel, engineering and Metallurgical industries”. The use of plural - “industries” - is important. It denotes the omnibus nature of the field sought to be regulated, a feature that becomes ever more critical in delineating the margins of the scope of its jurisdiction.

[20] There are eight major areas of activity described. The primary question is whether the employers, save for one, properly belong in (c), the “General engineering and manufacturing engineering and metallurgical industries” (GEME). It is also alleged by NUMSA that some of the employers are also covered by (b), the area described as the “production of alloys and/or the processing and/or recovery and/or refining of metals (other than precious metals) and /or alloys from dross and/or scrap and/or residues.” The Plastics Industry, (g) is the location, NUMSA alleges, where one employer, Torres, belongs, it being the sole business working in plastics.

[21] First, the ramifications of being located in GEME, (c), is dealt with and I shall return to (b) and (g).[5]

[22] GEME is further defined in a suffix to the definitions, identified as (a). That definition describes certain activities which are carried out on “machines”, “articles” and “vehicles”. However, these “vehicles” exclude “a motor vehicle”. Moreover, this definition of GEME excludes the “motor

industry”. It must follow that the activities described herein can be identical to the activities in the motor industry, but for policy reasons, are excluded from the jurisdiction of MEIBC.

[23] What constitutes the supposedly distinct “motor industry” is further defined in (i). This definition is a long list of activities related to cars in one way or another. Most are not pertinent to the controversy. In (i)(aj) it says this:

‘the business of manufacturing establishments wherein are fabricated[6] motor vehicle parts and/or spares and/or accessories and/or components thereof.’

[24] Thus, having described the GEME, and having excluded from its purview, the “motor Industry” as defined, the definition goes on to exclude from the definition of motor industry, as defined, certain types of businesses which perform those motor industry related activities, but under the specified circumstances are not to be regarded as being in the motor industry. In terms hereof, notwithstanding the nature of the activities covered, the following types of businesses are not regarded as being in the motor industry:

‘For the purpose of this definition …. “Motor Industry” as defined above shall not include the following:

(i) The manufacture of motor vehicle parts and/or accessories and/or

spares and/or components in establishment laid out for and normally producing metal and/or plastic goods of a different character on a substantial scale;

(ii) …

(iii) The manufacture and/or maintenance and/or repair of –

(aa) civil and mechanical engineering equipment and/or parts thereof whether or not mounted on wheels;

(bb) agricultural equipment or parts thereof, or

(cc) equipment designed for use in factories and/or workshops.

Provided that for the purpose of (aa), (bb) and (cc) above, “equipment” shall not be taken to mean motor cars, motor lorries and/or motor trucks; and

(dd) motor vehicle or other vehicle bodies and/or superstructures and/or parts of components thereof made of steel plate of 3,175mm thickness, when carried on in establishments laid out for and normally engaged in the manufacture and/or maintenance and/or repair of civil and/or mechanical engineering equipment on a substantial scale.”

(underling supplied)

[25] Thus, it is plain that the scale of operations in relation to the fabrication of motor components is a material factor in assigning the enterprise to one or another jurisdiction. This is illustrative of the artificial and pragmatic borderlines that are drawn by the definition of scope. If an enterprise is substantially not making motor parts, it is not in the motor industry, even if some such products are made.

[26] I return to the major area described in (b) of the definition of the scope of MEIBC, alluded to and cited above. Certain of the employers indeed produce goods from “dross” “scrap” or “residues” and on a literal application that definition covers their activity. However, that cannot be enough or be properly evaluated in isolation. It is an example of an overlap between two bargaining council scope definitions. The very purpose of the demarcation enquiry is to determine the best fit. The considerations pertinent thereto are addressed hereafter.

[27] I return to the major area (g), the Plastics industry. One employer, Torres, does not operate in metal products, but in plastics. Both plastics, as a substance itself, and the plastics industry are defined in passages marked (e) and (f):

‘“Plastics Industry” means the industry concerned with the conversion of thermoplastic and/or thermosetting polymers, including the compounding or recycling thereof, or the manufacture of articles or parts wholly or mainly made of such polymers into rigid, semi-rigid or flexible form, whether blown, moulded, extruded, cast, injected, formed calendered, coated, compression moulded or rotational moulded, including in-house printing on such plastics by the manufacturers, and all operations incidental to these activities;

“Plastics” means any one of the group of materials which consist of or contains as an essential ingredient an organic substance of a large molecular mass and which, while solid in the finished state, at some stage in its manufacture has been or can be forced, i.e. cast, calendered, extruded or moulded into various shape by flow, usually through the application, singly or together, of heat and pressure including the recycling or compounding thereof, but only where such compounding and or recycling is as a result of the conversion for manufacture by the same employer, but shall extrude all extrusions into mono- and multi-filament fibres and other activities falling under the scope of the National Textile Bargaining Council;’

[28] NUMSA advances an argument that these definitions to do with plastics do not incorporate a similar proviso, as exists in the definition of GEME, to exclude the motor industry. This is plainly correct. The follow-on submission is that ipso facto, bereft of that exclusion, if what Torres does, can fit into MEIBC’s plastics’ scope, that conclusion ends the debate. In my view, this is incorrect because it ignores the fact that what Torres does also fits into the definition of the Motor Industry. MIBCO’s scope includes production of components regardless of the material from which they are manufactured.

[29] In the definition of the scope of MIBCO, the identical definition of the “motor industry” appears. MEIBC’s (i)(aj) is identical to (j) in MIBCO’s scope. It too, in “exclusions: (a),” excludes from the motor industry, businesses whose dominant enterprise is the production of non-motor vehicle parts, including processes involving the working of both metal or of plastics. It must follow that establishments in which the dominant enterprise is motor car component manufacture, those businesses are intended to be included in the motor industry.

The rationale in the judgment a quo

[30] Having set aside the award, and having not remitted it, Van Niekerk J was at large to formulate the order that was appropriate. The debate before him, as it was before this Court, was directed at the following issues:

30.1 Insofar as the scope of the jurisdiction of the two rival bargaining councils was circumscribed by definitions, what was the proper approach to the interpretative exercise in attributing meaning thereto and what weight ought the conclusions enjoy in the context of a demarcation re-evaluation;

30.2 Insofar as the nature of the enterprise could be examined through the prism of the work/ production process or through the prism of the end-product produced, which, if any, was appropriate in the context of the demarcation evaluation;

30.3 Was the concept of a “value chain” legitimate and useful in locating the businesses within the compass of an “industry” for the purposes of a demarcation exercise?

30.4 What was the relevance of the history of collective bargaining between the several employers and their workers, and in the context of the demarcation re-evaluation, what was the proper weight to be attached thereto?

[31] The rationale evinced in the judgment of Van Niekerk J, in summary, was that (1) it was a necessary dimension of the exercise to interpret the definitional scope of the rivals and apply those descriptions to the common cause descriptions of the nature of the enterprises,[7] (2) the bargaining history under the jurisdiction of MEIBC for several decades, though pertinent was not a weighty factor in addressing the question of the best –fit for those enterprises at the time of the enquiry,[8] (3) an appreciation of the end-product produced by the enterprises was a better tool of analysis in this case,[9] and (4) the value chain concept was legitimate and useful in achieving the aims of the LRA.[10]

[32] Van Niekerk ultimately held at [58]:

‘In summary: the factors disclosed by the evidence indicate that the demarcation applicants fall within the scope of MIBCO’s

registration, the history of collective bargaining in the motor and metal industries is based principally on a determination that is some 60 years old and no longer binding, and the definition of scope emphasises the outcome of the manufacturing process rather than the nature of that process. All of these factors, cumulatively considered, indicate that the only reasonable outcome of the

proceedings under review is a conclusion that all of the demarcation applicants fall outside of the MEIBC’s scope and within the registered scope of the MIBCO.’

[33] The various aspects are addressed in turn.

The definitions and the comparative exercise

[34] The issue is whether, insofar as the scope of the jurisdiction of the two rival bargaining councils is circumscribed by definitions, what is the proper approach to the interpretative exercise in attributing meaning thereto and what weight ought the conclusions to enjoy in the context of the demarcation re-evaluation.

[35] The appropriate approach is well established: in Greatex Knitwear (Pty) Ltd v Viljoen[11] the following was held at 344G – 345F:

“ When the tribunal is called upon to determine whether a class of employers is engaged in a particular industry it is faced with a problem similar to that with which the Courts have often been faced, viz. to decide whether a particular employer is one of those other employers, not being parties to an agreement, engaged in a particular industry, upon which the Minister has declared an agreement to be binding (cf. sec. 48 (2) of the 1937 Act; sec. 48 (1) (b) of the 1956 Act). The cases seem to show that the matter is approached along the following lines:

(a) The meaning of 'industry', as used in the agreement, is determined. This usually requires the interpretation of some definition appearing in the agreement. It seems that a restrictive interpretation is often applied, cutting down the scope of the general words used in the definition. Although not specifically invoked, the mode of interpretation appears to be that applied in Venter v R., 1907 T.S. 915 (cf. Rex v Scapszak and Others, 1929 T.P.D. 980; Rex v Ngcobo, 1936 NPD 408; R v Goss, 1957 (2) SA 107 (T) at p. 110).

(b) The activities of the employer (personal and by means of his employees) are determined.

(c) The activities and the definition (as interpreted) are now compared. If none of the activities fall under the definition, caedit quaestio; if some of the activities fall under the definition, a further question arises: are they separate from or ancillary to his other activities? If they are separate he is engaged in the industry (unless these activities are merely casual or insignificant - Rex v C.T.C. Bazaars (S.A.) Ltd., 1943 CPD 334); if they are ancillary to his other activities, he is not engaged in the industry (unless these ancillary activities are of such magnitude that it can fairly be said that he is engaged in the industry within the meaning of the definition (A.G. Tvl v Moores (S.A.) (Pty.) Ltd., 1957 (1) SA 190 (AD)).

Inherent in this approach is the possibility that an employer may be such in more than one industry (Rex v Giesker and Giesker, 1947 (4) SA 561 (AD) at p. 566), despite the difficulties that may arise from such a situation (cf. Rex v Auto-Parts (Pty.), Ltd. and Another, 1948 (3) SA 641 (T) at p. 648).

If the aforegoing is a correct reflection of the manner in which the Courts have approached the problem whether an employer is engaged in a particular industry, it is plain that the problem is only resolved by reference inter alia to the activities of the employer. Whether one uses the word 'activities' or 'work' seems merely a question of preference of language. As in the case of an individual it cannot be determined whether he is engaged in a particular industry without reference to his work, so also it cannot be determined in the case of a class of persons whether it is engaged in a particular industry without reference to the work it does. Whether that work is to be called merely 'work' or a class of work seems, again, to depend on linguistic preference or the degree of circumscription.”

(underlining supplied)

[36] First, at the level of generality, it is the clear intention of the definition of scope of the MEIBC to exclude the motor industry as defined and vice versa for MIBCO. An overlap exists. The pragmatic division is then made between the two, dependent on the scale of operations relevant to motor industry type activities. A double-proviso delineates the borderline: ie, by excluding motor industry type activities and then re-including some businesses where the dominant enterprise is not motor industry related.

[37] The common cause evidence establishes that the manufacture of motor components is the exclusive or dominant enterprise of every respondent employer. That is also true of the businesses which use raw material scrap and dross and the one employer who works in plastics; both of which are examples of overlap.

[38] Van Niekerk J dealt directly with the definitions aspect: At [44] – [45] he held:

‘[44] …. in a case (such as the present),…. what is primarily at issue is the application of a definition of scope to an agreed set of facts. ….In the present instance, given MIBCO’s definition of scope, the commissioner was required to determine whether the demarcation applicants’ businesses were manufacturing establishments in which motor vehicle parts, spares, accessories or components were fabricated. In other words, what was at issue was the application of a definition to an agreed set of facts, not unlike a jurisdictional dispute where a commissioner is required to determine whether an applicant is an ‘employee’ as defined, or whether he or she was dismissed.

[45] Had the commissioner conducted the enquiry on that basis, he would have concluded that the definition of ‘motor industry’ in MIBCO’s scope of registration clearly extends to the business of the demarcation applicants, since they conduct the business of manufacturing establishments ‘wherein are fabricated motor vehicle parts and/or spares and/or accessories and/or components thereof’. The commissioner appears to have been alive to the inevitable outcome of the application of the definition to the undisputed facts. At paragraph 27 of the award, he says:

On face value, if we compare the common cause facts relating to the applicants, as to their business activities and that they manufacture motor components, and if we apply a literal interpretation, with the Certificate of Registration of the [MEIBC] and [MIBCO], it may be possible to interpret that the applicants fall within the scope of the [MIBCO].

Quite why the commissioner did not choose to apply a literal interpretation (or the ordinary meaning) of the definition of scope and arrive at the foreshadowed result is not apparent from the award. Viewed thus, the commissioner’s award was clearly wrong, and stands to be reviewed and set aside on that basis.”

[39] Contrary to the contention advanced on behalf of NUMSA that this outcome exaggerates the literal meaning of the text, in our view, a proper textual assessment indeed leads to the results articulated by Van Niekerk J.

[40] The fact that some businesses can be understood to be covered by (b) the utilisation of scrap must be weighed in the context of the whole. Similarly, the utilisation of plastics in (g) must be weighed in the whole. The best fit is the determining factor.

The emphasis on End-product or on Work Process

[41] In principle, there is no logical reason why, in vaccuo, the one emphasis is superior to the other as a tool of analysis. A selection of one over the other could be rational in given

circumstances. It may be that in some cases both are appropriate to be employed in the analysis and accordingly, one may be preferable to the other in given circumstances.

[42] This debate in this case illustrates a straight contest between process or end-product as an appropriate tool of analysis in these given circumstances.

[43] It is contended on behalf of NUMSA that work process is the only proper tool for the analysis and the notion of end-product is illegitimate. The rationale advanced is that the holy grail in demarcation exercises is to promote fairness by producing a result

where workers who do the same or materially similar work are treated identically. In my view, this ideal need not be questioned as a legitimate aspiration. However, the literal accomplishment of that aim must yield to the pragmatism of regulatory oversight in the real world. Not all floor-sweepers or office cleaners are in the Cleaning Industry. There are fitters in several industries. Deciding which type of woodworkers are in the building industry or in the furniture manufacturing industry is among the epic contests in the history of demarcations in our law. Plainly, a given work process can be utilised in many different enterprises, which on a holistic evaluation, have little or nothing in common with one another.

[44] Van Niekerk J dealt with this notion as follows at [54] – [55]:

‘[54] NUMSA ….submits that the end product of a process is not definitive of the essential character of the operation. Rather, it contends that if an operation involves the processing or shaping of metal, then the operations are to be treated as part of the metal industry. If metalworking processes are absent, NUMSA submits that the character of the operations may in principle be something else, and that the business may potentially fall within the ambit of the motor industry, provided that its operations fall within the definition of that industry. This approach would be consistent with the approach taken historically (as demonstrated by the 1962 demarcation) - it allows for metalworkers to be treated uniformly within a single centralised bargaining structure, it is consistent with the approach taken in prior decisions and it avoids the anomalies associated with employers switching industries when switching production as between automotive and non-automotive products.

[55] What this approach ignores is the definition of scope of the MEIBC and the MIBCO respectively. Excluded from the jurisdiction of the MEIBC is the motor industry, defined in paragraph (j) as ‘the business of manufacturing establishments wherein are fabricated motor vehicle parts and/or spares and/or accessories and/or components thereof’. The definition makes no reference to the form of the manufacturing process – it is confined specifically to outcomes in the form of parts, spares, accessories and components, regardless of the mode of manufacture, engineering or otherwise. For the commissioner to disregard the outcomes of the manufacturing process in favour of a determination based solely on the nature of the process, constituted a disregard for the applicable definition and contributed to an unreasonable result.’ (underling supplied)

[45] The rationale relied on by Van Niekerk J for this conclusion is, in our view, correct. The two industries are in their conception distinguished by what they produce. To overlay upon that edifice a description of the technical work process in order to distinguish them is alien to the invented foundations upon which they are distinguished. To determine the nature of the enterprise, in such a context, end-product is an appropriate tool of analysis.

[46] This approach is consistent with the dictum in R v Sidersky 1928 TPD 109 at 112-113:

‘Dr. Reitz, in favour of the appeal, argued that the character of an industry is determined, not by the kind of occupation in which the employers are engaged, but by the nature of the enterprise in which both employer and employees are associated for a common purpose. Once the character of the industry is determined all the employees are engaged in that industry, whatever the actual work may be which the employer allots to them. The accused confessedly conducts the industry of chemical manufacture, and it would be absurd to call him at the same time an employer in the building industry because he employs two bricklayers, also an employer in the engineering industry because he employs two engineers, also an employer in the printing industry because he employs two printers; and then force him as a multiple employer; to comply with all the rules that might be binding upon these various industries, though wholly inapplicable to the main industry of chemical manufacture.

I think this argument is sound.’

[47] It was contended that a critical factor was the ability of an establishment that forges steel in a mould for a motor car component, to switch to another mould not connected with the production of motor components. This notion is true, but is of little weight on the facts adduced in evidence. On the evidence, all of the businesses are in committed relationships to supply motor car components, some according to particular specification, to their customers. If they cease to pursue that enterprise as their dominant business, that fact can be investigated, at that time, to determine if they ought to be reassigned to another Bargaining Council.

[48] Reliance was also placed by NUMSA on the decision in CWIU v Smith & Nephew 1997 (9) BLLR 1240. (CCMA) to argue that technical process trumps end-product as a tool of analysis. In my view, this contention is not borne out by an examination of that award. Relying on a dictum in Food and Allied Workers Union v Ferucci t/a Rosendal Poultry Farm (1992) 13 ILJ 1271 (IC) at 1276 (a decision in the old Industrial Court) which held:

‘It is now trite law that, in determining whether employees are engaged in a particular kind of trade or industry, or, indeed, in farming operations, regard must be had not to the special nature of the work which they do, but, rather to the nature of the enterprise in which they and their employer are associated for a common purpose. Further, once the nature of that enterprise is determined, all the employees must be regarded as being engaged therein irrespective of the actual work which the employer may allot them.”,

the commissioner, thereupon, held;

“The argument on behalf of CWIU that regard must rather be had to the nature of the product and the nature of the market cannot be

accepted. Mr Buthelezi was constrained to concede that the main business of Smith & Nephew was in the textile industry.”’

[49] Upon the premise that this statement in the award correctly described the facts, the outcome that the employer was assigned to the textile industry was sound, on the facts. The efforts of that employer to escape that assignment when about 1% of the workforce was engaged in producing cotton goods specifically for medical use were rightly in vain.

[50] Plainly the nature of that enterprise was not characterised by its marginal participation in the production of cotton goods for medical use. No principle was sought to be articulated that end-product is an illegitimate tool of analysis.

The value chain argument

[51] The debate over the usefulness or legitimacy of the value chain argument is an important one. The resort to this rationale, as best we are aware, is novel in demarcation disputes.

[52] To understand its import, it is appropriate to be clear about what the phrase can be marshalled to perform. First, it is not a normative concept; it is merely a handy shorthand description of certain facts. What it describes is the fact of a process of linked manufacture of several parts of an ultimate product which process is, thus, highly decentralised, and is carried out by a multitude of independent manufacturers who source from and supply to one another material or components, or indeed services, and are reliant and dependent on one another to maintain the “chain” of supply.

[53] In the case of the motor industry being described as a value chain, the unrebutted evidence discloses that the famous brand names, known as Original Equipment Manufacturers (OEMs) whose vehicles are sold to the public have, in recent decades, moved from being totally involved in fabricating all the bits and pieces that go together to make a vehicle and have outsourced the manufacture of distinct parts to independent manufacturers. The metaphor of a chain is apt because the vehicle manufacturers are, in reality, end-user assemblers of components bought in from what are called first tier manufacturers who in turn subcontract the manufacture of the parts that make up what they supply to the OEMs to second tier manufacturers. The production of the components made by the respondent employers lies in the second tier as described by this scheme of analysis.

[54] Van Niekerk addressed this aspect thus at [56] – [57]:

‘[56] A related issue is that of the significance of the value chain of which the demarcation applicants form part. The definition of scope aside, the demarcation applicants contend, as I have mentioned above, that the commissioner committed a gross irregularity and came to an unreasonable conclusion when he rejected the applicant’s value chain argument. In essence, that argument is that but for the eighth and twelfth applicants, the percentage of each demarcation applicant’s business that relates to the manufacture and supply of automotive components exceeds 90%. In the case of the eighth applicant that figure is 80%, and 86% in respect of the twelfth applicant. It follows, so say the demarcation applicants, that for all intents and purposes the entire business enterprise of each applicant is dedicated to the value chain relevant to the motor industry. Further, the automotive components engineered or manufactured must meet automotive industry specific specifications. They are not intended or made for use in other industries. None of the applicants’ enterprises have discrete portions dedicated to the manufacturer engineering of nonautomotive components will. Put another way, of the more than 3800 employees engaged by the demarcation applicants, less than 1% are engaged in work activities not related to the production of automotive components. The demarcation applicants contend that to the extent that the commissioner failed to acknowledge these facts and ignored the nature of the end product in making his demarcation, he committed a reviewable irregularity.

[57] There is considerable merit in this argument, for it is one that aligns the business activities of the demarcation applicants with the sector as a whole. The demarcation applicants are an integral link in the chain or value system between the conception and delivery of a motor vehicle. Manners’ evidence that the value chain or system is the mode in which production is conceptualized and actualized in the motor assembly industry was not challenged. The existence of a value chain or system locates the demarcation applicants within a set of activities in the motor industry in which they receive raw materials, add value through the manufacturing process and sell the finished product to the customer located in the next highest tier. To ignore this evidence had the result of an unreasonable award.’

[55] In my view, these findings are wholly appropriate.

[56] It was argued that in certain instances the “end-product” of some of the respondent businesses were not yet recognisable as motor components and had to be finished by the next tier of enterprises. This may be correct but it does not, in this context, mean that the particular business is alienated from the value chain. The evidence tendered is that the parts so supplied are intended for incorporation into the chain and are not universal bits that are intended for or can be utilised in a wider market.

The Collective bargaining history

[57] As foreshadowed above, the fact that 60 years ago, or, for that matter, last week, a Ministerial Determination was made, is not dispositive of the question whether a particular business should be subject to the jurisdiction of a particular bargaining council. The enquiry always starts with a clean slate.

[58] Similarly, the collective bargaining practices cannot be regarded as conclusive of where a business ought to belong. The collective bargaining practice is relevant for what an account of it can contribute to whether the status quo is appropriate and ought to be continued.

[59] In this case, the collective bargaining practice was dictated by the 1962 Determination. The evidence of Mr Manners was that the relationship, thus dictated, had become dysfunctional because of the evolution of the motor industry over decades. This is a proper premise upon which to re-evaluate the practices in place. His unrebutted evidence was that the motor component manufacturers are isolated and marginalised within MEIBC by that practice. They bargained alongside other businesses with no commonality of product and no common interest. Unlike other segments of the iron steel and engineering industries that had distinctive chambers within which to engage, the motor component businesses were unrecognised. In our view, these dynamics do point towards a dysfunctionality in the established practice and brings into question the appropriateness of locating the businesses under the jurisdiction of that Bargaining Council.

[60] The consequent question to be posed is whether there is a better fit elsewhere. There is such a better fit: MIBCO - where all the competitor business in the integrated chain of manufacture can bargain together. Mr Pauw, the representative of MIBCO confirmed the presence of competitors of the respondents being subject to the jurisdiction of MIBCO.

The view of NEDLAC

[61] As alluded to above, section 62(9) obliges the commissioner to consult NEDLAC about the demarcation decision being made before it is issued.

[62] In this case, the view of NEDLAC was obtained. NEDLAC wrote to the CCMA on 21 November 2017. It was a scathing rejection of the award. Of relevance at this stage of the proceedings is the rejection of the rationale that the end- product analysis and the value chain thesis were illegitimate. In as much as the commissioner ought to have weighed the views of NEDLAC, the judgment of the review court is consistent with its views.

Conclusions

[63] In the result:

63.1 It is appropriate to examine the profile and structure of motor manufacture as it presents in 2017 and to give weight to the transformation of that industry over the past half century.

63.2 The concept of a value chain and the location of a business in an integrated process of manufacture is a legitimate tool of analysis.

63.3 An evaluation premised on end-product rather than work process was appropriate in the given circumstances.

63.4 A textual evaluation of the definitions of the engineering realm and the motor manufacturing realm yields a result that demonstrates a functional overlap and the need for a policy decision to draw the line of demarcation; on the facts, in favour of the jurisdiction of MIBCO.

The Order

(1) The appeal dismissed.

(2) The order of the Labour Court is confirmed.

__________________

Sutherland JA

(Sutherland JA with whom Davis JA and Murphy AJA concur)

APPEARANCES:

FOR THE APPELLANT: Adv TMG Euijen SC, with him Adv F Le Roux,

Instructed by Gray Moodliar Inc.

FOR THE 6TH AND THE 8TH TO

25TH RESPONDENTS: Adv C Bosch,

Instructed by Van Zyls Inc.

[1] The relevant text is thus: 62 Disputes about demarcation between sectors and areas (1) Any registered trade union, employer, employee, registered employers' organisation or council that has a direct or indirect interest in the application contemplated in this section may apply to the Commission in the prescribed form and manner for a determination as to- (a) whether any employee, employer, class of employees or class of employers, is or was employed or engaged in a sector or area; (b) …. (2) If two or more councils settle a dispute about a question contemplated in subsection (1) (a) or (b), the councils must inform the Minister of the provisions of their agreement and the Minister may publish a notice in the Government Gazette stating the particulars of the agreement. (3) …. (4) When the Commission receives an application in terms of subsection (1) …. it must appoint a commissioner to hear the application or determine the question, and the provisions of section 138 apply, read with the changes required by the context. (5) …..(8) (9) Before making an award, the commissioner must consider any written representations that are made, and must consult NEDLAC. (10) – (11) (12) The registrar must amend the certificate of registration of a council in so far as is necessary in light of the award.

[1] The relevant text is thus:

62 Disputes about demarcation between sectors and areas

(1) Any registered trade union, employer, employee, registered employers' organisation or council that has a direct or indirect interest in the application contemplated in this section may apply to the Commission in the prescribed form and manner for a determination as to-

(a) whether any employee, employer, class of employees or class of employers, is or was employed or engaged in a sector or area;

(b) ….

(2) If two or more councils settle a dispute about a question contemplated in subsection (1) (a) or (b), the councils must inform the Minister of the provisions of their agreement and the Minister may publish a notice in the Government Gazette stating the particulars of the agreement.

(3) ….

(4) When the Commission receives an application in terms of subsection (1) …. it must appoint a commissioner to hear the application or determine the question, and the provisions of section 138 apply, read with the changes required by the context.

(5) …..(8)

(9) Before making an award, the commissioner must consider any written representations that are made, and must consult NEDLAC.

(10) – (11)

(12) The registrar must amend the certificate of registration of a council in so far as is necessary in light of the award.

[2] The several parties were described differently as respondents in the arbitration, the two reviews and the appeal and as a result to identify any one as respondent X is not possible or useful.

[3] The phrase “Interest Arbitration” refers to a seemingly adjudicative process which is in truth not adjudicative. Typically, where collective bargaining does not yield an agreement, Management and Labour might agree to delegate the determination of a wage to a third party “arbitrator”. In truth, the arbitrator is not seized with a justiciable issue, rather the function is to create a new right, premised on a balancing of legitimate considerations within fact-specific circumstances.

[4] (2014) 35 ILJ 3059 (LAC).

[5] The text of the scope definitions is a plethora of a maddening sub-texts with no clearly distinguishable numbering. The reader must constantly refer back to what sub-division yet another alphabetical smorgasbord is presented to unearth what the provisions qualify.

[6] An argument was advanced that the term “fabricate” should be understood to be distinct from “manufacture”.

This notion is without merit in this context. The use of the synonym is purely stylistic.

[7] Judgment at [55]; see too [44]- [45].

[8] Judgment at [47] – [51].

[9] Judgment at [52] – [55].

[10] Judgment at [56] – [57].

[11] 1960 (3) SA 338 (T).