GRIEVANCE MEDIATION I. T H E COAL INDUSTRY EXPERIMENT STEPHEN B. GOLDBERG*
The promise of labor arbitration was that it would deter strikes, that it would be speedy, inexpensive, and informal, and that the arbitrator's decision would be sensitive to the labor relations problems underlying the grievance. The performance of labor arbitration has, in many respects, lived up to its promise. It has deterred strikes—strikes during the term of a collective bargaining contract are not a major national problem—and it has, on the whole, been faster and less expensive than litigation. In other respects, however, arbitration has been less successful than was anticipated. Delays of six to nine months from the request for arbitration to the receipt of the arbitrator's written decision are common. Costs are substantial, with the average arbitrator's bill in excess of $1300 per case. Adding the cost of the transcript, the briefs, and attorney's fees, each party may pay more than $5000 to arbitrate a grievance. Complaints about excessive formality are also common. Finally, arbitration today is not, on the whole, sensitive to the labor relations problems underlying a grievance. It is not, if you will, problem-oriented. Let me explain. Every grievance originates as a problem between two or more people at the workplace. Frequently, however, as the grievance moves through the grievance procedure, it is discussed less and less as a workplace problem, and more and more as a matter of contract right. Indeed, when the grievance reaches arbitration, it is generally presented exclusively in terms of contract interpretation, with little or no reference to the problem which led to the grievance. This approach frequently has a deleterious effect on the useful•Member, National Academy of Arbitrators; Professor of Law, Northwestern University, Chicago, 111.
ness of the arbitrator's decision to the parties. Indeed, all too often I have heard it said that the arbitrator's decision was a marvelous piece of scholarly writing, but that it left the parties with the problem that they had to begin with. In response to these concerns, we have developed a procedure that, by combining elements of both mediation and arbitration, is designed to resolve grievances more quickly, inexpensively, and satisfactorily than conventional arbitration. (The "we" to whom I refer consists of Rolf Valtin, whom all of you know as a former president of the Academy; Bill Hobgood, former Director of the Office of Mediation Services for FMCS, and a former Assistant Secretary of Labor; and Dr. Jeanne Brett of the Kellogg Graduate School of Management at Northwestern, who is largely responsible for the statistical evaluation of our work.) The procedure that we have devised is this: After the final step of the internal grievance procedure, the parties are given the option of going to a form of mediation rather than directly to arbitration. The mediator is an experienced arbitrator with mediatory skills. He or she first seeks to assist the parties in reaching a mutually satisfactory settlement. The focus, at this stage of the proceedings, is not exclusively on the contract, but is also on the underlying problem. The contract, of course, establishes the parameters within which a settlement can be reached. However, as all of you know, it is impossible to write a contract that covers all the problems that are going to arise during its term. Accordingly, there is ample room for creative problem-solving and dispute resolution within the parameters of the contract, and this is the process in which the mediator and the parties initially engage. If a settlement is not achieved through this initial discussion, the next task of the mediator, who, you will recall, is an experienced arbitrator, is to provide the parties with an advisory opinion as to how the grievance is likely to be decided if it goes to arbitration. This opinion is immediate, it is oral, it is nonbinding. We hope that the parties will accept this opinion and resolve the grievance accordingly, but if they do not, they are free to arbitrate. If they do, the mediator may not serve as arbitrator, and nothing that has been said at mediation for the first time, including the mediator's advisory opinion, may be used against a party at arbitration. Thus, mediation is a no-risk procedure. The parties may come out with a settlement, but if they do not,
ARBITRATION—PROMISE AND PERFORMANCE
they can go on to arbitration with neither party being any worse off as a result of having gone to mediation. The entire mediation procedure is informal. Evidence is elicited in a narrative fashion with no rules of evidence, no transcript, and no record. The proceedings are brief and to the point. Our goal is to resolve three grievances per day. The theoretical advantages of mediation are manifold. First, it should be faster than arbitration because there will be no transcript, no briefs, and no written decision. It should be less expensive than arbitration for essentially the same reasons. It will be less formal than arbitration, which should make the parties more comfortable and more capable of focusing on their problem. It will be more problem-oriented than arbitration, and that should improve the quality of the outcomes, since dealing with the real problem should lead to a real solution. Perhaps the most important theoretical advantage of mediation—its long-range advantage—is that it should serve to teach settlement skills to the parties. They will be working with a skilled neutral in attempting to resolve disputes. This will provide them with on-the-job training in dispute resolution. If they can learn dispute resolution skills as a result of this experience, and can utilize those skills in the earlier steps of the grievance procedure, they should be able to settle more grievances without the need for a mediator or an arbitrator. That, of course, is the most desirable result of all. Finally, it is possible that mediation, in the long run, will serve to improve the relationship of the parties. The combination of a less adversarial approach, which will generate fewer hostilities than arbitration, and the increased ability of the parties to resolve disputes may, in time, lead to an improvement in their relationship. There are two central theoretical risks of mediation. One is that it just won't work—that all or most mediated grievances will go on to arbitration. The other is that it will work too well—that the parties, because of the availability of speedy and inexpensive mediation, will not settle grievances at earlier steps of the procedure. Either of these outcomes would be counterproductive, since either would add to, rather than reduce, the time and money devoted to grievance resolution. In order to test this procedure, we began in November 1980 what was, at that time, intended to be a six-month experiment with grievance mediation in the bituminous coal industry. The
six months have stretched out to two and one-half years, we have mediated more than 300 grievances in the coal industry, and another 25 or so in other industries. The results have been as follows: Of the 325 cases which have been to mediation, 275, or 85 percent, have been finally resolved without resort to arbitration. Approximately 50 percent of the settlements have been compromises, in which each side walked away with something. Another 25 percent have been noncompromise settlements—that is, either the company granted the grievance in its entirety or the union withdrew the grievance in its entirety. In about 25 percent of the grievances, the mediator has given an advisory opinion. Approximately half of those advisory opinions have been accepted and the other half of the cases have gone to arbitration. Of those which have gone to arbitration, the arbitrator's award has been that predicted by the mediator in 75 percent of the cases. We have, in fact, been able to mediate an average of three grievances per day. The cost of mediation has averaged $200 per grievance, $100 per party. This does not include travel expenses, but since those expenses are apportioned among the three cases the mediator hears in a day, they haven't amounted to much. The financial savings to the parties have been enormous. In the coal industry alone, the parties have already saved approximately a quarter of a million dollars ($250,000) over the costs of arbitrating a like number of cases. The average time from the request for mediation to the final resolution has been 15 days. There has not, on the whole, been any substantial diminution in the internal settlement rate—the rate at which the parties resolve grievances themselves. The typical pattern is that in the first six months in which mediation is available there is a slight drop, anywhere from 5 to 10 percent, in the internal settlement rate. After that it tends to hold pretty steady at the new rate. The financial consequences of a 5 to 10 percent drop in the internal settlement rate, coupled with a sharp reduction in the frequency of arbitration, is, as demonstrated in the coal industry, enormous cost savings because of the minuscule cost of mediation compared with that of arbitration. The nature of the outcomes in mediation has been impressive. Little would be accomplished by attempting to describe them here, as each must be understood in context. Many, however,
ARBITRATION—PROMISE AND PERFORMANCE
have been of a sort that simply could not be reached in arbitration because of the inherent limitations on the arbitrator's function. We are beginning to find some evidence of improvement in settlement skills. There are no statistical data on this, but both company and union representatives in some relationships report a change in the approach to grievances at the earlier steps of the grievance procedure. They report a change from an exclusive focus on who is right and who is wrong to an effort to work out a mutually acceptable settlement. Indeed, some report statements such as "When we get to mediation, the mediator is going to try to make us work out a solution we both can live with. So why don't we do it right here, and save money, and not have to go to the mediator?" That, of course, is just what we are hoping for. Perhaps the acid test of any innovative approach to grievance resolution is whether it meets the needs of the participants. One of the things that we have been doing in connection with this project is interviewing all the participants in grievance mediation. We ask them a series of questions about their experience, one of which is whether they prefer mediation or arbitration as a means of dispute resolution. Company operating personnel preferred mediation over arbitration by a ratio of six to one, and union local officers and committeemen preferred mediation over arbitration by a ratio of seven to one. The grievants themselves preferred mediation over arbitration by a ratio of two to one—less than the others because they tend to be more concerned with the outcome of their particular grievance than with the grievance resolution process, but still the grievants preferred mediation over arbitration two to one. In explaining why they preferred mediation to arbitration, each group referred to time and cost savings, but what they focused on primarily was the informality and problem-solving approach of mediation. Thus, one company representative said, "I like the informality. It creates an atmosphere of people trying to solve problems through talk rather than being enemies in a legal process. This is a much better way to approach these problems." I am sure that many of you have questions, and I will try to anticipate some of them. What's new about mediating grievances? Haven't some arbitrators been doing this, with the consent of the parties, for years? Indeed, they have, and in large measure the inspiration for our work goes back to those arbitra-
tors—to George Taylor, to Saul Wallen, to Arnold Zack, and to many others, some of whom, I am sure, are sitting here today. What we have done is to structure the mediation process in such a way as to maximize both its likelihood of bringing about a final resolution of the grievance and its long-range value to the parties. The likelihood of final resolution is maximized by using a mediator who is also an experienced arbitrator, hence is capable of providing the parties with accurate advice as to the likely outcome in arbitration, and so discouraging resort to arbitration. The long-range value to the parties is maximized by barring the mediator from also serving as arbitrator. It is on this point that our approach diverges from that of many of our predecessors. For, we believe, to the extent that the parties know that the mediator will not also arbitrate, they will devote their energies toward achieving a negotiated settlement, rather than toward persuading the mediator, who is not a potential arbitrator, of the "rightness" of their positions. This focus on a negotiated outcome should, as I have already suggested, improve the parties' negotiating skills, and so serve the crucial long-range goal of minimizing future resort to either mediation or arbitration as a means of dispute resolution. Another question I am frequently asked is: Why not use expedited arbitration instead of mediation? For, it is suggested, expedited arbitration also reduces costs and delay, in addition to which it is final and binding. Here, too, I agree to some extent—that is, if one is seeking simply a speedy and inexpensive decision, expedited arbitration is just as good as mediation. However, there are two problems. First, because expedited arbitration is final and binding, it is available for only a narrow range of cases. As a practical matter, most parties simply will not agree to have anything other than the most routine cases go through a truncated procedure that is also final and binding. In that sense the very strength of mediation is that it is not final and binding unless the parties want it to be final and binding—unless the outcome is something that both sides are satisfied with. Because the parties have that control—because they can control the outcome of mediation—they can, and frequently do, submit their most significant grievances to mediation. Thus, an important practical advantage of mediation over expedited arbitration is that it makes the same high-speed, low-cost procedure available for all grievances, not just for some. Furthermore, expedited arbitration is still arbitration—it is not mediation, it
ARBITRATION—PROMISE AND PERFORMANCE
is not problem solving, and it has none of the long-range educational advantages of mediation. A question that has frequently been raised since the Supreme Court's decision in Bowen v. U.S. Postal Service1 is as to the likely effect of mediation on duty of fair representation suits. We think that mediation can substantially reduce the likelihood both that such suits will be brought and that they will be successful. We further think that this result can be achieved in a manner that is scrupulously protective of the legitimate interests of employees. Initially, we think that many duty of fair representation suits are filed because the grievant does not think that he or she has had a fair hearing. The union representative has said, "You don't have a case, and we're not going to take it to arbitration," but the grievant may believe either that the union representative doesn't know what he is talking about, or that the union representative is simply trying to save money for the union. If that grievance goes to mediation, and the mediator—a skilled, experienced arbitrator—after listening to the grievant and all the evidence for two and a half, three, or four hours, advises the grievant that in his honest judgment, the grievance would not be sustained in arbitration, and why not, we think—and we have had some experience with this—the grievant will agree to the union's dropping the case. He or she will have had a day in court, and in many cases that will suffice. Assume, however, that it does not suffice, and that a duty of fair representation suit is brought. A union violates its duty of fair representation only if it has been arbitrary, discriminatory, or in bad faith in declining to arbitrate. We think that if the union can show that its reason for not arbitrating was that it relied upon the opinion of a skilled and experienced arbitrator that the grievance would probably be denied at arbitration, that should go a long way toward proving that the union has not been arbitrary, discriminatory, or in bad faith in declining to arbitrate. Is the successful use of mediation in the coal industry due to the nature of that industry? Certainly not. About 25 grievances, as I earlier said, have been mediated outside of the coal industry, and not one of those grievances has gone on to arbitration. The key to the successful use of mediation is not the nature of the '459 U.S
112 LRRM 2281 (1983).
industry but rather the perception of the parties that not enough grievances are being settled without resort to arbitration, and their desire to do better—to settle more cases without arbitrating. If both parties honestly want to settle more grievances and are willing to enter into mediation with a desire to do that, then mediation can be successful regardless of the industry in which it is used. If an employer and a union are interested in trying grievance mediation, how should they begin? What most parties have done is to utilize a six- to 12-month experimental period, during which mediation is available only on mutual consent, so that no individual grievance goes to mediation without the consent of both parties. In order to assist those employers and unions who wish to experiment with mediation, we have formed a nonprofit corporation, Mediation Research and Education Project, Inc., which provides sample rules of mediation, joint employer-union mediation training, mediator selection, recruiting, and training, and the administration of the mediation system. One final question that I am frequently asked, and that is particularly appropriate to raise here, is: What do the arbitrators who are engaged in grievance mediation think about it? You can ask them that question during the discussion sessions since nearly all of them are members of the Academy and are here today. They are Rolf Valtin, Jim Scearce, Tom Phelan, Martin Wagner, Martin Cohen, Alex Elson, Tony Sinicropi, George Larney, Jim Martin, Barbara Doering, Bill Hobgood, and myself. Let me summarize what I think are our views as arbitrators who are mediating. First, grievance mediation is highly challenging. It is both exciting and satisfying to work with the parties and help them work out their own solutions to their own problems. It is also—and I am sure you will all appreciate this—a welcome relief from the normal decision-writing obligation. It is an awfully good feeling at the end of the day that the case is over and done with, and you don't have to go home and spend a couple of days worrying about it, and writing a decision. Finally, because the mediatory approach permits us to dispose of grievances more promptly and less expensively than conventional arbitration, I think we all feel a sense of professional pride at returning the resolution of grievances to the speedy and inexpensive process it was originally intended to be. (Incidentally, we will need more mediators in all parts of the country, so if you are interested, please let any of us know.)
ARBITRATION—PROMISE AND PERFORMANCE
In conclusion, I say that mediation will never replace arbitration. Indeed, arbitration is the motor that makes mediation run. It is the fact that arbitration is out there, that it is expensive, that it takes a long time, and that the outcome is unpredictable that impels the parties to mediate and to work out settlements at mediation. While mediation, to repeat, will never replace arbitration, it is capable of reducing the frequency of resort to arbitration. It is also capable of providing more satisfactory outcomes than arbitration, and of doing so in a manner that has the long-range educational value of teaching settlement skills. Now some of you, I anticipate, are skeptical about all this. All I can say is that skepticism before trying mediation has almost invariably been replaced by satisfaction afterwards, or, in less elegant terms, "Try it, you'll like it."2 II. A MANAGEMENT APPROACH FRANK C. SARNO*
About two and a half years ago, Professor Steve Goldberg came before a group within the coal industry and said he had a new method of handling grievances. This method was quicker, cheaper, and would render better decisions than arbitration. He went on to say that, in the process of handling these grievances, the system would teach the participants problem-solving. Steve called the process "mediation" and immediately lost half his audience. For those of us within the industry who did not tune him out or mentally leave the room, the process sounded like it had some possibilities. Those of us at AM AX Coal Company considered his proposal in the light of: (1) What benefit would "mediation" play in our overall goals and objectives in labor relations? (2) What is the level of maturity in the relationship with our employees, the union, and our mine management? (3) Are we willing to let our mine management make hard decisions? 2
For a fuller discussion of the theoretical issues posed by grievance mediation, see
Goldberg, The Mediation of Grievances Under a Collective Bargaining Contract: An Alternative
to Arbitration, 77 Nw. U. L. Rev. 270 (1982). For a more extensive empirical report on the results of our grievance mediation experiments, see Brett and Goldberg, The Mediation of Grievances in the Coal Industry: A Field Experiment, 37 Indus. & Labor Rel. Rev. 1
(October 1983). *Director, Employee Relations, AMAX Coal Company, Indianapolis, Ind.