Labour Arbitration and Conflict Resolution: Back to Our Roots

The Honourable Warren K. Winkler, Chief Justice of Ontario

Presented at Queen’s University, 30 November 2010, Queen’s School of Policy Studies

Donald Wood Lecture

One of the people I have respected the most over the course of my career used to commence every meeting with the words, “I’ve called you together so that we can resolve your indifferences.” I have to concede that what I am about to discuss here this afternoon is not a matter of indifference to me. Quite to the contrary.

My thesis is that labour arbitration as we know it, that institution that we are all so dependent on in the field of labour relations, has lost its course, has lost its trajectory, has lost its vision. It is at risk of becoming dysfunctional and irrelevant. We all need to hearken back to what that vision was.

Labour arbitration was intended to be a procedure through which disputes could be resolved in a timely way, on the merits, in an affordable fashion, and with finality. But since the end of what I call the “golden age” of labour arbitration, which spanned the period from roughly 1944 to 1967/68, it has lost its sense, its vision, and has gone off the rails. The risk is tremendous if that happens. What I am going to do is describe what I see as the golden age of labour arbitration, and then outline what I think the three reasons are for why it is off track and at risk of becoming irrelevant. Then I want to address what I think can be done to bring it back on track, back on target, back on focus and in line with its vision—and what we ought to do to make that happen.

I begin with some history, a thumbnail sketch. In the years right after the Second World War, labour arbitration really saw its beginning. The reason for that can traced back to the labour unrest that accompanied the Second World War and the huge production requirements that were made on employees during that period. Yet there were illegal strikes and many other kinds of disruption to production.

Of course, there was reaction by governments. Aside from a short twoyear period during which there was a thing called the labour court, which was part of the high court of Ontario and really was an abject failure, two sets of labour concepts came into play legislatively. One was PC 1003, the federal statute; the other was the Labour Relations Act of Ontario. Both statutes contained a trade-off. They provided that there could be no economic sanctions during the term of a collective agreement. In other words, it was a no strike proviso. Then, to accompany that, the quid pro quo was that there was to be a labour dispute arbitration process, a tripartite process to resolve those disputes that happened during the term of the collective agreement. This was seen to be a pressure relief valve so that disputes could be resolved in an amicable way during the term of the collective agreement. The main architect of this process was Professor Jacob Finkelman.

In 1944, after a fourteen-year career as a law professor at the University of Toronto, Jacob Finkelman became the first chair of the Ontario Labour Relations Board. He had written much of the Ontario Labour Relations Act that he was to administer. He was also the person who came up with a vision of what labour arbitration was to be all about, namely, a process by which labour disputes could be resolved during the term of an agreement in a timely way that was affordable and final. The idea was that the courts should be kept out of labour relations, so there were to be no appeals or judicial reviews. Consequently, there was a final and binding clause, and the whole sense of this process was that it was to be industrial relations based.

That was the underpinning of this labour arbitration process—it was to be industrial relations based, not litigation based. Therefore at all times it had to be plugged in to a sound labour relations policy in the workplace. The reasoning for this was simple: if these disputes did not get resolved in a timely way, then when the collective agreement came up for a renewal, the parties would of course be dealing with this harboured discontent and dissatisfaction and it would be almost impossible to negotiate a collective agreement without a work stoppage. In this way the process was finely balanced. This is how the system was intended to work and that was how it started off.

This initial period of labour arbitration in the years from 1944 to 1967 is what I call the “golden age” of labour arbitration because the characteristics of the process and outcomes were closest to what had originally been envisioned. What were these characteristics?

There was a tripartite board consisting of two persons from each side— one labour, one management—and a neutral chairperson. The parties would appear and present their cases, and this was a non-adversarial, non-acrimonious type of proceeding. The usual process was that the parties would print up their submissions in a brief; it would be short, and it would contain evidence. This evidence would often conflict, but nobody worried about that. There were frequently no witnesses called, there were no cross-examinations. There was no acrimony because it was non-adversarial. People showed up and presented these cases, which were short and over with quickly.

Each side would be somewhat loath to bring forward arguments at hearing that would inflame the other side—whether or not it was viewed as a “winning argument.” On the management side, almost all employers belonged to the Central Ontario Industrial Relations Institute; it provided surveys on salaries, and wages and benefits, as well as other relevant statistics. While managers represented the interests of employers at collective bargaining, in arbitration they were principally motivated by sound labour relations and a non-adversarial approach. This non-adversarial and non-acrimonious perspective permeated everything and made the system work as successfully as it did. There was simply none of the litigiousness that followed in later decades; it was most common to have no lawyers involved so that neither the nominee nor the presenter would be a lawyer. I characterize it as something of a “golden age” of labour arbitration, because it was.

The chair of the arbitration boards in those years was, almost invariably with only the odd exception, a county court judge. The judges did this in their spare time, and so almost all arbitration cases were heard on Friday afternoons and Saturday mornings. Fundamentally, what regulated your behaviour was that there had to be sound labour relations in the workplace. The client did not want it any other way, and so if you got offside and started becoming too adversarial or legalistic, they simply did not stand for it.

That is what labour arbitration was like in the golden era from 1944 through until 1967, and that is exactly how Professor Finkelman wanted it. Professor Finkelman, notwithstanding that he left the academy in 1944 when he started chairing the Ontario labour board, was always referred to as The Professor. He had his finger right on the pulse, he knew exactly what he was talking about, he understood human nature, and he had common sense. He knew how this system had to work and, as long as he was around, that’s how it did work. If someone got offside, The Professor would ensure that the person understood that a behavioural change was required.

In the mid-1960s, Professor Paul Weiler of Harvard prophesied that labour arbitration’s success would bring about its own failure. That was a self-fulfilling prophecy; it is my thesis that this is precisely what happened. Three factors brought it about. The first was the change in the cadre of arbitrators. The second was the change in the jurisprudence for arbitrators. The third was the culture of labour arbitration itself. The first two of these changes were, in hindsight, for the better. While in the short term they had a ripple effect and caused disruption, in the long term they turned out to be for the better. The third one turned out to be for the worse.

I begin with the first two factors. Of course, all three factors were inextricably intermingled in a joint dynamic, and not truly separable, although I consider them sequentially.

In March 1967, the Parliament of Canada, after a heated debate, decided that judges should no longer be allowed to do work other than sit as a judge. The debate in Parliament makes it clear that this was aimed, in part, at the county court judges who were hearing labour arbitration cases. The result was dramatic because they laid down their tools, by necessity, immediately. If you look back, the last reported case by a county court judge was in 1968; there was a short lag due to the cases they were seized with. But as of March 1967, they were no longer allowed to arbitrate.

This had a very sudden and direct impact because this cadre of arbitrators who were doing almost all of the arbitrating up until then were gone. Their place was taken by a group of law professors, mainly from Queen’s University and the University of Western Ontario, who picked up where the judges left off. While they did a great job, there was obviously a loss of institutional knowledge because these judges had been doing this for a long time. On top of that, there was pick-up time—a steep learning curve for the new professors because they were young, they had not worked in industrial establishments, and they had to pick up very quickly and get into the mix with these arbitrations. As the professors started to be initiated to these cases, because they came from the academy decisions tended to get longer, and so there was a bit of a delay in getting the reasons out.

That is one of the ripple effects that happened in labour arbitration as a result of those changes in the mid-1960s. This change in the cadre of arbitrators in the long run turned out to be for the better because they turned out to be tremendous arbitrators. But they also brought a little more of a legalistic approach to things, albeit with common sense. They were steeped in wisdom because they did these cases over and over again and everybody got to know who they were. They have become the new cadre of labour arbitrators for an indefinite period of time into the future.

The second factor that brought this golden age of labour arbitration to an end was the extension of jurisdiction of labour arbitrators. This was an incremental process; it did not happen all at once; nor did it happen because of any one single event or incident. Even so, there were certain highlights that one can point to that I think are indicators of what happened. These were seminal events in this process of the extension of the jurisdiction of labour arbitrators.

The first of these is the key case of Polymer;1 it involved the unlawful strike of workers at the polymer plant in Sarnia, and a claim by the employer for damages as a result of the illegal strike. The chair of the arbitration board was Professor Bora Laskin, who would later become a member of the Ontario Court of Appeal and, subsequently, the Chief Justice of Canada.

The union’s argument was that an arbitrator had no jurisdiction to award damages because it was not spelled out in the collective agreement. Professor Bora Laskin, in this very important landmark case, essentially said that there was a residual jurisdiction in the arbitrator— that if you had a right to hear the case you had the right to formulate a remedy, and that was unencumbered. That was the thrust of this case, which has affected virtually every single decision in labour arbitration where there is a remedy at issue, since that time. It pervades the entire field labour arbitration.

The next really important case was Weber v. Ontario Hydro in 1995.2 In this case, Weber had been under surveillance because he was thought to be a malingerer by Ontario Hydro, and finally he sued them. This case ultimately ended up at the Supreme Court of Canada, and the Supreme Court said “no.” You must look at the nature of the dispute and then look at the four corners of the collective agreement in order to determine whether or not this dispute fits into that employmentrelated collective agreement. If the answer to that is yes, then it goes there exclusively, and it goes to arbitration; it does not go to the civil courts.

What has been the effect of Weber? It has been enormous, first of all, because of the psychological impact of the decision. This was seen as a case that was outside the boundaries—because until this point in time everyone had argued the jurisdictional phrase in the arbitration section of the statute—that arbitrators can deal with issues involving the interpretation, administration, or application of the collective agreement including any issues as to whether it is arbitrable or an alleged violation. So they thought that formed the four corners of labour arbitration.

But, with Weber, that is no longer the four corners of a labour arbitration. Weber was about a defamation case. Weber moved cases such as those related to workplace assaults or defamation into the arbitration process. These were traditionally civil court lawsuits that, until then, no one thought would come within labour arbitration, with a possible remedy of awarding damages. This was a sea change because it altered the whole concept and notion of labour arbitration, which had been defined by the jurisdictional clause dealing with the interpretation, application, or alleged violation of a collective agreement (i.e., labour arbitration had been really confined and constrained by the language of the collective agreement). Suddenly, it was a whole new era.

The next major development came with Perry Sound Social Services Administration v. OPSEU, which was a case where a probationary employee gets dismissed. There is a probationary employee clause, so that if you are a probationary employee, you cannot seek arbitration. This case goes to the Supreme Court of Canada; they go back and read section 4812J of the Labour Relations Act that says the Human Rights Act or any other industrial relations–type statute can be arbitrated even if it is inconsistent with the collective agreement. All of a sudden, cases that used to go someplace else go to labour arbitration.

These three cases effectively created a new world because the sphere of the kinds of cases that, by subject matter, arbitrators could resolve was hugely expanded, and I would say for the better, because it meant a holistic approach. It essentially meant that everything was going to be decided in the same place by the arbitrators who knew all about it and what works and does not work, and so forth.

The third major factor was a cultural change in the conduct of labour arbitration. Starting in the mid-1960s, the whole idea of an industrial relations–based arbitration process began to shift over to a litigationbased arbitration process.

Originally, the parties worried about whether or not the labour arbitration process would enhance, foster, and benefit industrial relations in the plant. This changed so that now you had the emergence of a group of people who were not worried about that at all; instead, they were worried about winning the case. They did not care whether one party or the other was upset in the workplace. If they were upset, the attitude was “Well, let them be upset.” The arbitration process changed from one that was industrial relations based to one that was litigation based. This change drove a wedge between the arbitration process and the labour relations in the plant, and it drove a wedge between the parties in such a way that they did not relate any more. This disconnect proved to be immensely harmful.

The industrial relations practitioners on both sides, who had been such an important part of this community, were displaced by this litigation mentality as the arbitration process was inundated with lawyers. It became a litigation-based arbitration hearing in which the discourse changed from one that was based in industrial relations to one based on law. As a litigation-based process, it was about winning cases, often at any cost. It was also about making technical arguments, taking a different approach to a case, adhering strictly to the rules of evidence, and insisting on the production of documents and on particulars.

The other aspect of this development was that when winning or losing becomes the main focus, the parties then have to be careful about who is chosen as an arbitrator. Each side has to give it their best shot, so the list of who is acceptable becomes a pretty skinny list. Often, this meant choosing the busiest people, because each side felt they needed to have the people who would give them their best chance—even a 1 percent better chance, because winning meant everything. Choosing busy people led, in turn, to significant delays in hearing. This was an incremental, gradual, but steady change in process.

Taken together, these changes resulted in enormous bottlenecks of cases and huge delays on the other end. The original precept of labour arbitration being a timely and affordable resolution, on the merits, was gone: it was no longer timely, it was increasingly unaffordable, and it was not on merits but on technicality. The principle of finality was also gone, because if somebody was unwilling to accept the decision they went into judicial review to the court. So all four of the fundamental precepts, going back to Jacob Finkelman, had been compromised.

Professor Finkelman understood that these proceedings had to be affordable because unions did not have the money to litigate these things to a standstill, or in a war of attrition. He knew that approach could not work, which is why the process had been set up the way it was originally. That is why in the “golden age” it would take only six to eight weeks to get a hearing date, the case would be over in half a day, the parties would get the decision two weeks later, and the decision was only six or eight pages long.

The arbitration process was no longer on trajectory—the vision was gone; the notion of how arbitration was supposed to work and that it was supposed to serve industrial peace and stability was gone. Of course, by this I mean that these objectives had become secondary to the goal of litigation, which is to win, regardless of how little the case. This is not what the arbitration process was intended to accomplish.

The outcomes now also reflect the changed trajectory. Now the hearing can take a year and a half or two years, with several adjournments, and it is so technical that nobody can understand the issues, which are not decided on the merits but with decisions thirty-five pages long issued six months later. This is not labour arbitration; it is labour dysfunction.

Several other things started to happen that indicated the process was no longer working. When unions could not afford some of these cases, they would wait until the day before the cancellation fee would trigger (because arbitrators have cancellation fees), and they would pull the cases or settle them, or do something else in order to get out of the case.

As well, many of the bigger employers who really had a grasp of this process opted to design their own arbitration procedure. These employers went to the expedited arbitration process and picked a cadre of arbitrators who knew their industry. They went to them regularly and they compacted the arbitration so they could have up to perhaps ten cases in a day. In a way they went back to the golden days: they did not have evidence anymore; they filed some documents; and they would have a short, agreed statement of facts. In this way, they solved their own problems.

These employers had several advantages in that they had a large volume of cases, a matured relationship with their union, and a culture in which the unions and employers trusted each other. As a consequence, not every case was a live or die matter. They were prepared, more mature, and more philosophical, and they were turning back to a labour relations–based dispute resolution mechanism that had the goal of satisfying the parties in the workplace and making it a good place to work.

The result was better production, and more satisfied workers where people wanted to work. These employers knew that the real game was not to win a particular case; rather, it was to have a big profit at the end of the year. They knew that this depended on a more satisfied and productive workforce. They also knew that it meant no strike at the end of the collective agreement. They knew that if they showed up to negotiate the collective agreement while sitting on a large backlog of grievances, they would deal with an unhappy group of employees—and their chances of reaching a settlement easily would be somewhere between slim and none. Not surprisingly, most of these expedited cases came out of meltdowns, such as Air Canada and Ontario Hydro. We have to acknowledge that the labour relations communities have always been characterized by their savvy and common sense, and the trend toward expediting cases in the big workplaces was an indication of this.

Another development was that arbitrators saw their workload going away as these cases were getting cancelled or pulled, and they saw their numbers going down. In reaction many arbitrators started to mediate disputes, which might previously have resulted in a protracted and costly process, to a successful settlement. While they did themselves out of a long hearing, they proved to the parties two things: first, they were back on an industrial relations–type proceeding, and second, they were looking at affordability and timeliness because the case was settled, everyone was satisfied, and the cost was reasonable and affordable. In effect, these arbitrators were back in play because they had redesigned what they were doing.

The major problem, however, was that these two major developments— expedited arbitration and arbitrators changing the way they handled the cases—did not occur right across the board. It was only basically the big employers who adopted the expedited arbitration, and only some arbitrators changed their handling of cases. Many continued to take the approach of hearing the case, taking notes, and then going away to send out a decision sometime later, no matter how long.

Given that this culture has not changed—except for the big employers and the arbitrators who were doing mediation—and given that the rest of this dysfunctional trajectory is still happening, what can be done to change this?

Part of the answer lies in restoring proportionality. This term is very significant outside labour relations, in the civil court. Proportionality means that when you hear a case, you have to keep in mind, when you determine how much work and money will be spent on the case, that it be proportionate in relation to the value or importance of the case to the party. In other words, you don’t take a $20,000 lawsuit and spend $50,000 in legal bills. It doesn’t make sense and it is an access to justice issue, because if you to do that, people cannot afford to go to court. The cost in time and money has to reflect how important the case is. In a labour arbitration, a seven-day hearing that costs $30,000 for someone’s warning letter would not pass the test of proportionality.

Much of the dysfunctional aspect of arbitration in labour relations has been caused by a lack of understanding of proportionality. In the “golden age” the process was proportionate: the unions could afford it, as could the company, and everyone was satisfied. Consequently the process served its correct purpose as a relief valve that took the pressure off, and there was no backlog of cases and no delay.

The new system, as it evolved incrementally, has become increasingly disproportionate. The parties are spending more time and money, attempting to win at a cost that is more than the case is worth, and everyone on the ground level—the people in the plants and the people in the companies—knows this. Who didn’t know this? The lawyers — the people that were sending the bills. In effect, the lawyers had this wedge driven between them and the production people in the plants and the union people in the field. The result was another disconnect.

So how do we correct this?

First, starting with the process, take the long and unwieldy grievance procedures and cut them in half—cut whatever the number of steps we need to “accordion” the grievance procedures.

Second, the entire discovery and all the production of documents and particulars should be done in the grievance procedure, and should not be permitted to happen at arbitration hearings. When the parties get to the arbitration hearing, they need to “get down to business and get it over with” without all of the manoeuvring and positioning.

Third, use arbitrators that are more available. The parties should not strategically try to obtain every ounce of advantage. The parties should take arbitrators who are not quite so busy and are more available, thereby cutting down on delays, getting the cases through, and ridding the system of the backlogs. If the parties can, they should have two processes: one standard procedure and one expedited.

Fourth, hearings must be run according to the principle of proportionality. This is a shared responsibility, including whoever is presenting the case at arbitration, as well as the arbitrators. If the participants cannot get it right and be businesslike and get things done, they should be “called out” and made to get on side.

What is at stake is the integrity of an institution that is fundamentally important to sound labour relations in Ontario. There is too much at stake to continue to tolerate the excessive litigation and lawyering at arbitration. Hearings need to be focused, on track, and run efficiently. Cases need to be compacted, with even more than one done in a day. We also need to find ways to conduct hearings better: put in an agreed-to written statement of facts, and use admissions, but do not put in written evidence in Chief so you only cross-examine on that, and so you do not have to have witnesses in Chief at all. There are examples of ways to make this process work faster and more affordably in terms of time and money. This is all about justice—justice that is affordable and efficient and that works.

There are other things that could be done as well. The parties need to look at cost. We are a long way from the days when we used to have hearings in the union hall. If you do not want to go there, you could go to the company’s board room; but the parties do not need to go to an expensive hotel for a room, which is what happens now. There are also alternatives to travel, especially in far away cases, such as video conferencing.

Paul Weiler prophetically noted that the success of the system will bring about its failure. The process became litigation based instead of industrial relations based. The shift toward a culture to win is the fundamental reason for the failure.

We have to think back to how it worked when the arbitration process was about workplaces. It is not about court, and it was never intended to be about court. Jacob Finkelman never for a second intended this to be like court, because the labour court in 1943 was an abject failure and everybody knew it. The idea was to move away from that model, and yet what did we do?

We have to get back to the basics, reformulate, and recalibrate the system to get it back on track. The changes can be made so that we return to the notion of a decision on the merits and not on some esoteric technicality—to a process that is timely, and that is affordable.


1 Re Oil, Chemical and Atomic Workers International Union, Local 16-14 and Polymer Corporation Ltd. (1959), 10 L.A.C. 51 (per Laskin, Dubin, O’Brien), aff’d 26 D.L.R. (2d) 609 (Ont. H.C.J.).

2 (1995), 2 S.C.R. 929, 125 D.L.R. (4th) 583 (S.C.C.).


Reprinted with the kind permission of Queen’s University. Copyright © 2010 School of Policy Studies, Queen’s University at Kingston, Canada

This website is maintained by the Judges' Library. Website Policies