The approach our courts are required to adopt in reviewing the awards and rulings of CCMA arbitrators was revisited in the recent Labour Appeal Court (“LAC”) decision of Goldfields Mining South Africa (Pty) Limited (Kloof Gold Mine) v CCMA, case number JA2/2012 (“Goldfields").

This issue has been the subject of debate over a number of years. At the heart of the debate was the question of to what extent the Labour Court should be able to overturn CCMA awards and rulings. It was the intention of the legislature that the powers of the Court in this regard should be limited. This is the reason why the Labour Relations Act, 66 of 1995 (“LRA”) does not make provision for an appeal against the arbitration awards or rulings. It simply makes provision for the review of such awards on limited grounds (e.g. where the arbitrator commits misconduct in relation to his/her duties or there is a gross irregularity in the arbitration). But, experience over the last few years has shown that the concept of a review has been widely or narrowly interpreted by different courts – hence the debate. The approach our courts should adopt was finally decided, or so we thought, in the widely-popularised case of Sidumo & Another v Rustenburg Platinum Mines Ltd & Others [2007] 12 BLLR 1907 (CC) (“Sidumo”).

In essence, Sidumo required a review court to ask the following question: Is the decision one that a reasonable decision maker could not reach on the evidential material available. Thus the focus was largely on the outcome of the decision, as opposed to the manner in which the arbitrator arrived at the outcome.

This approach presupposes that arbitration awards (“awards”) based on defective reasoning by an arbitrator will still pass the muster required in reviews, provided that the result is one that a reasonable arbitrator could have reached. This is generally known as an ‘outcome-based approach’.

However, in various decisions, the Labour Court did not limit itself to this relatively narrow test for review. It developed the concept of the “process related review”, which it treated as existing in addition to Sidumo’s ‘outcome-based approach’. This approach accepted that, even if the outcome of the award was one that a reasonable arbitrator could have reached, an award could still be overturned if the process through which the award was arrived at was found materially wanting – for example if the arbitrator ignored material relevant facts or misconstrued material evidence in coming to his/her decision.

This approach had been accepted by the Labour Appeal Court (“LAC”) in the recent case of Herholdt v Nedbank Ltd [2012] 9 BLLR 857 (LAC) (“Heroldt”). In this case the LAC adopted a generous approach to the scope of the test for reviews. The LAC indicated that the ground of review of gross irregularity in terms of s145 (2)(a)(ii) of the LRA included “latent irregularities” and “dialectical unreasonableness” as the basis for the review of an award. This required the reviewing court to consider the reasoning of the arbitrator.

The LAC stated that a ‘latent irregularity’ occurred where the arbitrator failed to take into account material facts or took into account immaterial facts, whereas ‘dialectical unreasonableness’ was unreasonableness stemming from the process of reasoning of the arbitrator. In this regard, the LAC held that the reviewing court must consider whether the arbitrator’s decision is supported by arguments and considerations that are valid, albeit, not necessarily conclusive. In order for an arbitrator’s decision to be reasonable in a dialectical sense, he/she is required to properly consider all the relevant and material facts indispensable to a reasonable decision.

This was a far wider interpretation than the traditional approach to the concept of gross irregularity which was largely limited to the situation where the arbitrator misconceives the whole nature of the enquiry, and as a result the arbitrator misconceives his/her mandate or duties in conducting the enquiry.

The LAC’s judgment in Heroldt went on appeal, however, and the Supreme Court of Appeal did not uphold such a generous approach. It revisited and analysed the provisions of s145 of the LRA, and stated that the legislature was deliberate in rejecting the option of an appeal of awards. It deliberately chose review, on narrow grounds, so as to serve as a deterrent to parties seeking to challenge awards. This supported the purpose of the CCMA as a dispute resolution forum that is to provide for an inexpensive and expeditious dispute resolution process.

The SCA summarised the position as follows:

A review …is permissible if the defect in the proceedings falls within one of the grounds in s145 (2) (a) of the LRA. For a defect in the conduct of the proceedings to amount to a gross irregularity as contemplated by s 145(2) (a) (ii) …the arbitrator must have misconceived the nature of the inquiry or arrived at an unreasonable result. A result will only be unreasonable if it is one that a reasonable arbitrator could not reach on all the material that was before the arbitrator. Material errors of fact, are not in and of themselves sufficient for an award to be set aside, but are only of any consequence if their effect is to render the outcome unreasonable.” (our emphasis)

The SCA thus made its position clear on the issue through its judgment in the Heroldt matter. However, since the coming into force of the Constitutional Seventeenth Amendment Act, 2012 the SCA no longer has the jurisdiction to hear appeals from the LAC and the LAC is now the final body of appeal (except for Constitutional issues) when interpreting the LRA.

There was therefore some anticipation to see if the LAC would follow the SCA decision in Heroldt when it next confronted the review test issue. The Goldfields decision, handed down on 4 November 2013, was the first LAC decision to consider the test for review after the SCA’s Heroldt decision. In its judgment, the LAC recognises that the process-related grounds of review provided for in section 145(2)(a) still pertain but finds that, once the procedural defect is established, the reviewing court must go a step further and satisfy itself that the defect resulted in the award being one that a reasonable arbitrator could not have reached. In the words of the LAC, “Whatis required is first to consider the gross irregularity that the arbitrator is said to have committed and then to apply the reasonableness test established by Sidumo. The gross irregularity is not a self-standing ground insulated or independent of the Sidumo test. That being the case it serves no purpose for the reviewing court to consider and analyse every issue raised at the arbitration and regard failure by the arbitrator to consider all or some of the issues albeit material as rendering the award liable to be set aside on the grounds of process-related review.

The LAC in Goldfields reaffirmed the purpose of an arbitrator, as set out in section 138 of the LRA, to deal with the substantial merits of the dispute between parties with the minimum of legal formalities and to do so expeditiously and fairly. The relevant enquiries to make in review applications, said the LAC are the following:

“(i) In terms of his or her duty to deal with the matter with the minimum of legal formalities, did the process that the arbitrator employed give the parties a full opportunity to have their say in respect of the dispute? (ii) Did the arbitrator identify the dispute he was required to arbitrate…? (iii) Did the arbitrator understand the nature of the dispute he or she was required to arbitrate? (iv) Did he or she deal with the substantial merits of the dispute? And (v) Is the arbitrator’s decision one that another decision – maker could reasonably have arrived at based on the evidence?”

In the Goldfields matter Mr Moreki had been dismissed for allegations of misconduct. However, the arbitrator found that Mr Moreki was in fact guilty of poor performance and that the sanction for dismissal was too harsh and ordered that the Mr Moreki be reinstated.

Applying the review test that it had articulated, the LAC found that the arbitrator had misconceived the nature of the enquiry, which had been to determine whether the dismissal of Mr Moreki, based on grounds of misconduct, was fair. The arbitrator had erroneously miscategorised Mr Moreki’s conduct as poor performance, which entailed a different enquiry than in cases involving misconduct. This amounted to a gross irregularity.

The LAC, in this regard, stated that

“…the arbitrator committed a gross irregularity in the conduct of the proceedings. The conclusion he arrived at was influenced by the wrong categorisation of the case against the Third Respondent. This however is not sufficient for the award to be reviewed and set aside. The question needs to be asked: had the categorisation of the case against the Third Respondent been misconduct as opposed to poor performance, is the arbitrator’s award nonetheless one that could be arrived at by a reasonable decision – maker? In my view it is clearly not. The Third Respondent committed a serious act of misconduct…the decision arrived at by the arbitrator is not one which a reasonable – maker could reach”

Thus, where an arbitrator commits misconduct in relation to his/her duties or there is a gross process-related irregularity in the arbitration, this is not - in and of itself - a sufficient ground to warrant interference by our courts on review. The irregularity must be of such a nature that it renders the decision reached unreasonable in the circumstances.

So practically what does all of this mean? It is no longer good enough for employers or employees wishing to review an award based on one of the procedural defects provided for in section 145(2)(a), to only establish the existence of the defect, i.e. misconduct by an arbitrator in relation to his/her duties, a gross irregularity committed by the arbitrator in the conduct of the arbitration proceedings or the arbitrator exceeding his powers. It is now also necessary to show that the defect caused the ultimate result of the award to be unreasonable. Thus, the two stage test adopted by the LAC in such instances is:

  1. Was there a section 145(2)(a) defect ?; and
  2. If so, can the defect be said to be such that resulted in the decision reached being unreasonable (in the sense that it was one that a reasonable arbitrator could not have reached)?

It thus now appears that it will, once again, be more difficult to successfully prosecute review applications in the Labour Court.