"3-3——;. - THE IMPACT OF ALTERNATIVE MODES 0F THIRD PARTY INTERVENTION IN * " " RESOLNNENANGNNTNQ TNPAssEs Dissenatbn for the Degree of Ph. D. - MWTGAN STATE UNIVERSITY WILLIAM}; BIGONESS ' ‘1974 ' ' LIBRARY ' MI Michigan State University This is to certify that the thesis entitled THE IMPACT OF ALTERNATIVE MODES OF THIRD PARTY INTERVENTION IN RESOLVING BARGAINING IMPASSES presented by WILLIAM J. BIGONESS has been accepted towards fulfillment of the requirements for Pf: . W. degree in M/mfljmvwh M. 3&7, Aémm wt professor Datej—M? 93/ /1 7g 0-7639 . IIIMG & SIIIIS' BOOK BINDERY INC. JBRARV BINDERS ABSTRACT THE IMPACT OF ALTERNATIVE MODES 0F THIRD PARTY INTERVENTION IN RESOLVING BARGAINING IMPASSES BY William J. Bigoness The present research examined the impact of four alternative modes of third party intervention in resolving bargaining impasses under conditions of high and low bargaining conflict. The modes of third party intervention were (1) bargaining (no third party inter- vention), (2) mediation, (3) voluntary arbitration and (4) compulsory arbitration. The literature shows considerable disagreement regarding which style of third party intervention is most successful in facilitating bargaining settlements. For example, Northrup (1966) and Bok and Dunlop (1970) have argued that compulsory arbitration tends to under- mine meaningful collective bargaining, while Reuther (1965) and Stevens (l966) suggest that the threat of intervention may foster a greater sense of urgency for bargainers to settle their disputes. Podell and Knapp (l969) and Pruitt and Johnson (l970) found mediation enhances bargaining outcomes while Johnson and Tullar (1972) found mediation was the least effective mode of third party intervention. Johnson and Pruitt (1972) found arbitration to be a more effective style of intervention than mediation. William J. Bigoness Even greater uncertainty exists regarding what potential condi- tions either enhance or hinder alternative modes of third party inter- vention. Research by Landsburger (1955) and Vidmar (197l) suggest that stronger modes of third party intervention are more successful in high conflict situations than low conflict situations. The results of the present study shows the following: (a) (d) (e) (f) Dyads who faced no third party intervention (bargaining only) settled more contracts (ll) than dyads under any style of third party intervention. 0f the three interven- tion conditions, mediation was found to be the least successful in that only 2 of 20 contracts were settled while in the voluntary arbitration condition 7 of 20 contracts were settled and under compulsory arbitration 10 of 20 contracts were settled prior to intervention. Subjects under low conflict left significantly fewer issues unresolved under voluntary and compulsory arbi- tration than subjects negotiating under bargaining and mediation. However, under high conflict significantly fewer issues were left unresolved under bargaining than in any other condition. Dyads under mediation exchanged significantly fewer total bids than subjects under any condition. Subjects bargaining under high conflict demanded a significantly smaller total opening bid than subjects bargaining under low conflict. Subjects under high conflict conceded significantly less than subjects under low conflict. Under low conflict, subjects anticipating voluntary and compulsory arbitration had significantly higher earnings than subjects in the other two conditions (bargaining and mediation); while under high conflict, subjects in the bargaining condition earned significantly higher payoffs than subjects in the other three conditions. In general, stronger modes of third party intervention were found to facilitate bargaining settlements when subjects faced low conflict. However, under high conflict conditions, subjects bargained most suc- cessfully in the absence of outside intervention. These results tend William J. Bigoness to suggest that the superiority of a particular mode of third party intervention varies with the magnitude and intensity of the conflict between the parties. THE IMPACT OF ALTERNATIVE MODES 0F THIRD PARTY INTERVENTION IN RESOLVING BARGAINING IMPASSES By William J. Bigoness A DISSERTATION Submitted To Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Management I974 CDCopyright By William J. Bigoness 1974 ACKNOWLEDGMENTS The successful completion of any dissertation necessitates sup- port from multiple sources; this dissertation is no exception. I am particularly indebted to my dissertation chairman, W. Clay Hamner, his ‘ friendship, standards of excellence and interest in my professional growth was a major factor in the successful completion of my graduate education. The other members of my dissertation committee, Lawrence A. Messe' and Henry L. Tosi have generously contributed their time and I offered invaluable assistance. To each of these men, I express my gratitude. I also thank Richard F. Gonzalez and the Organizational Behavior faculty at Michigan State University for their continuous support. My colleagues in the doctoral program deserve a special recognition. These include particularly Lloyd S. Baird, Jay S. Kim and Randall S. Schuler, whose assistance and companionship was always most timely and appreciated. Finally, my loving wife, Leslie, has provided a special source of support. Her continuous encouragement, patience, understanding and love have contributed incalculably to the successful completion of my graduate education and my happiness during the past five years. TABLE OF CONTENTS Page LIST OF TABLES ........................ v LIST OF FIGURES ....................... vii Chapter I. INTRODUCTION AND REVIEW OF LITERATURE ......... I Literature Review .................. 5 Field Studies ................... 6 Experimental Literature .............. 10 Summary of Literature Review ............. 20 II. METHODOLOGY AND HYPOTHESES ............... 23 Statement of the Problem ............... 23 Methodology and Independent Variables ........ 24 Dependent Variables ................. 27 Hypotheses ...................... 28 Experimental Method ................. 32 Subjects and Sample Size .............. 32 Apparatus ..................... 33 Experimental Task ................. 33 The Instructions .................. 35 Data Collection and Scoring ............. 36 Pilot Study ..................... 36 III. RESULTS ........................ 39 Contracts Settled .................. 39 Issues Left Unresolved ................ 40 Total Number of Bids ................. 44 Total Opening Bid .................. 49 Total Amount Conceded ................ 54 Earnings ....................... 59 Post Experimental Questionnaire ........... 65 Mundane Realism .................. 65 Fear of Third Party Intervention .......... 7O Opponent's Bargaining Strategy ........... 74 Conflict Intensity ................. 78 Summary of Results .................. 82 iii Chapter Page IV. DISCUSSION AND CONCLUSIONS ............... 86 The Effects of Third Party Intervention ....... 87 The Effects of Initial Position ........... 89 The Effects of Third Party and Initial Position . . . 90 Role ......................... 92 Summary of the Findings ............... 93 Implications of the Findings ............. 95 Limitations of the Present Study and Directions for Future Research ................ 97 iv Table TO. II. 12. l3. 14. LIST OF TABLES Chi-Square Analysis of "Contracts Settled" ....... Means, Standard Deviations and Summary of Analysis of Variance of the "Number of Issues Left Unresolved" . . Mean Number of Issues Left Unresolved as a Function of Initial Management Position X Mode .......... Means, Standard Deviations and Summary of Analysis of Variance of the "Total Number of Bids" ........ Mean Number of Total Bids as a Function of Initial Management Position X Mode .............. Mean Number of Total Bids as a Function of Initial Union Position X Mode ................ Means, Standard Deviations and Summary of Analysis of Variance of "Total Opening Bid" ........... Mean Percentage of Total Opening Bid as a Function of Initial Management Position X Mode .......... Mean Percentage of Total Opening Bid as a Function of Initial Union Position X Role ............ Means, Standard Deviations and Summary of Analysis of Variance of "Total Amount Conceded" ......... Mean Percentage of Total Amount Conceded as a Function of Initial Union Position X Role ........... Mean Percentage of Total Amount Conceded as a Function of Initial Management Position X Role ........ Means, Standard Deviations and Summary of Analysis of Variance for "Earnings" ............... Mean Earnings as a Function of Initial Management Position X Mode ................... Page 40 42 44 46 47 48 50 53 53 55 58 58 60 63 Table Page 15. Mean Earnings as a Function of Initial Management Position X Role ................. . . . 64 16. Mean Earnings as a Function of Mode X Role ....... 64 17. Mean Earnings as a Function of Initial Management Position X Initial Union Position ........... 65 I8. Means, Standard Deviations and Summary of Analysis of Variance for "Mundane Realism" ............ 66 19. Mean Mundane Realism as a Function of Initial Management Position X Initial Union Position X Mode ....... 69 20. Means, Standard Deviations and Summary of Analysis of Variance for "Fear of Third Party Intervention" . . . . 7l 2l. Mean Fear of Third Party Intervention as a Function of Mode ........................ 74 22. Means, Standard Deviations and Summary of Analysis of Variance for "Opponents Bargaining Strategy" ..... 75 23. Means, Standard Deviations and Summary of Analysis of Variance for "Conflict Intensity" ........... 79 24. Mean Conflict Intensity as a Function of Role X Mode . . 82 vi LIST OF FIGURES Figure Page l. Diagram of Experimental Design ............ 25 vii CHAPTER I INTRODUCTION AND REVIEW OF LITERATURE Introduction The expansion of strife and conflict within numerous Spheres of contemporary society during the past decade has caused social scien- tists to devote renewed attention and energies to the study of con- flict and conflict resolution strategies (Coser, 1956; Schelling, 1960; Pondy, 1967; and Schmidt and Kochan, 1972). Conflict in inter- national relations, unionization and collective bargaining in the pub- lic sector, efforts to achieve racial and sexual equality and con- sumerism have been among the more fertile grounds for the study of conflict and its consequent resolution. Among the most fruitful strategies frequently employed in resol- ving conflict is the intervention of a third party who is not a direct participant to the immediate conflict. Blake, Shepard and Mouton (I964) have argued that conditions for a third party resolution exist when two disagreeing parties have reached an impasse, and it is assumed that no further interaction can produce a change in the dis- agreement. Authors (Kerr, 1954; Elkouri and Elkouri, 1960; Prasow and Peters, 1970; and Pruitt and Johnson, 1970) have cited numerous attri- butes which third parties introduce to a conflict situation which enhance a third party's capacity to bring about accord between the I 2 disputants. The third party has the capacity to (1) introduce origi- nal ideas; (2) be impartial; (3) act as a communication catalyst; (4) bring parties to a more rational mood; (5) clarify the intentions of the parties toward each other; and (6) relieve the sense of per- sonal weakness that is otherwise inherent in making concessions. The extent to which a third party is successful in resolving bar- gaining impasses is often determined, in part, by the power granted the third party by the principal contestants. On a continuum of in- creasing power, third party intervention is characteristically classi- fied (Bok and Dunlop, 1970; Chamberlain and Cullen, 1972) into one of the following categories; (1) conciliation; (2) mediation; (3) fact finding; (4) voluntary arbitration; and (5) compulsory arbitration. Conciliation and mediation involves those responses by a third party aimed at bringing about agreement or compromise and settling the conflict. Under conditions of conciliation and mediation the decision making responsibility, regarding mutual accommodation, remains inter- nalized within the bargaining dyad. Fact finding_involves the third party in a more investigative role because he is obligated to explore the issues causing conflict and report each party's respective posi- tion relative to a bargaining settlement. A fact finder's report is customarily made available to individuals or groups outside the con- flict, thus, often increasing the pressure upon the bargainers to reach a settlement (Hildebrand, 1967). Arbitration entails a third party's entering into the conflict and making a final determination regarding the terms of the settle- ment. Under a system of voluntarygarbitration both bargaining parties must mutually agree to the intervention of a third party regarding any specific bargaining impasse. Compulsory arbitration differs from voluntary arbitration in that a superior power, in relation to the bargaining parties, may dictate that should an impasse occur that the dispute will be submitted to compulsory arbitration. Decision making under an arbitration system, in terms of closure, is granted to the third party and thus externalized beyond the dyad (Urban, 1971). Which of the preceding modes of third party intervention is most effective and under what conditions has remained an issue of contro- versy among both practitioners and academicians. Only during recent years have researchers begun systematically to investigate the impact of various modes of third party intervention. Rosenau (l969) drawing principally from the field of political science, succinctly summarized that state of knowledge pertaining to third party intervention when he stated that despite an enormous amount of conventional and legal wisdom leading to prescriptive solu- tions and ringing affirmations not much is known about third party intervention. In reference to mediation, Podell and Knapp (l969) concluded that evidence concerning the contribution of the mediator has remained anecdotal and controversial. Vidmar (1971) concurred when he stated "that despite common assumptions that the presence of a neutral leader can reduce internegotiator conflict and thereby enhance negotiation outcomes, there is, in fact, very little experimental research on the effects of mediators on the negotiation process (p. 49)." Despite such apparent uncertainty, bountiful predictions exist. Such viewpoints often are based on descriptive studies of collective bargaining which often, unfortunately, are tinged with emotional reasoning. Numerous authors contend that the use of arbitration, voluntary or compulsory, deters the tendency of bargainers to resolve conflict. Northrup (1966) contends that when closure by arbitration is available, the incentive for the bargainers to agree sharply declines, even to the point where bargainers will force intervention. Bok and Dunlop (1970) have argued that compulsory arbitration even within a few industries tends to undermine an important ingredient in productive labor-management relations, namely, the willingness of the parties to bargain conscientiously over their differences. It should be noted, however, that these researchers have presented only limited data to support their conclusions. Informed opinion, however, is not unanimously aligned in opposi- tion to arbitration. The late Halter Reuther (1965) contended that the threat of government intervention fosters a greater sense of urgency for the bargainers to settle their disputes. Stevens (1966) has stated that the contention that the availability of a strike is an essential part of collective bargaining may not be as self-evident as most proponents seem to think. Loewenberg (1970) found that the availability of compulsory arbi- tration did not terminate collective bargaining activity among police and fire fighters in Pennsylvania. Two-thirds (132 of 198) of the municipalities examined reached negotiated settlements prior to resorting to compulsory arbitration. Loewenberg concluded that the advent of compulsory arbitration did not prove detrimental to meaning- ful collective bargaining. In light of the expanding reliance upon third parties in resolv- ing bargaining impasses it becomes increasingly crucial that we more fully understand the impact of alternative modes of third party inter- vention. The underlying area of interest in this paper is whether different modes of third party intervention hinder or facilitate mutual accommodation. In particular, the issue under investigation is whether the anticipation of different modes of intervention has an effect on the behavior of bargainers in a distributive bargaining task under simulated conditions designed to represent high and low degrees of bargaining conflict. In summary, researchers are divided on the issue of what effect, if any, alternative modes of third party intervention have upon bar- gaining behavior. This study will attempt to expand our knowledge and understanding pertaining to this issue. Prior to discussing the experimental design which will be utilized in this study, as well as specific hypotheses which will be tested, a review of relevant prior research examining third party intervention, upon which this study will build, will be presented. Literature Review Research pertaining to third party intervention originates pri- marily from two distinct sources. The first is that research based upon field studies drawn from the actual experiences of practicing mediators and arbitrators or the observations of individuals having access to such data. The second major area of research encompasses research conducted by behavioral scientists utilizing laboratory studies and experimental simulations as methods of investigation. Unfortunately, the extent of cross fertilization between these two camps has been minimal and disappointing. While this study will utilize the methodological framework employed by the latter group of researchers, an attempt will be made to utilize the findings from field studies to advance our understanding of third party interven- tion. For organizational purposes, a review of the literature and examination of the field studies is presented, followed_by a review of pertinent laboratory research relevant to third party intervention. Field Studies Among the first to present evidence related to mediation was Peters (1952, 1955). A member of the California State Conciliation Service, Peters sought to bring to the public's attention, primarily through a series of case incidents, the more common issues that fre- quently arise in collective bargaining and the conciliator's role and contribution in resolving conflict. One of Peters' principal contri- bution was answering such commonly raised questions as "Why do nego- tiations take so long?" "Why is the whole process accompanied by so much threat and bluff?" "What does a mediator do?" Peters argued that bluffing is an accepted tactic by which each party attempts to gain a little more than they are actually willing to settle for. Lengthy and heated negotiations often serve as a form of catharsis for worker hostility thereby preventing its emergence in the form of slowdowns, wildcat strikes and generally disruptive labor relations. The mediator's function, according to Peters, is not how to keep emotions, personalities and politics out of collective bar- gaining, but rather how to assist negotiators to meet the inevitable psychological as well as economic needs of their respective parties. Utilizing Bales' interaction process analysis, which he applied to recordings of actual mediation sessions, Landsburger (1955, 1956, 1960) sought to determine what personality and personal attributes distinguished "good" and "bad" mediators. Landsburger concluded that a "good" mediator, among other things, learns rapidly, has original ideas, and is able to sell these ideas vigorously to others. In addi- tion, Landsburger (1955) found that the extent to which a mediation session went through a hypothesized phase movement was of some valid- ity in indicating the extent to which Specific items in dispute be- tween the parties were resolved. Second, the ultimate success of the session could be partially predicted from the parties' state of mind when they embarked upon the session: the more hostile their expressed feelings, the less likelihood of success. Third, all mediation ses- sions progressed through this hypothesized phase movement not only those resulting in successful outcomes. Drawing from a cross cultural sample, Jackson (1952) sought to determine what factors and conditions enhanced mediation effective- ness. He suggested that mediation needs to be flexible and adaptable to the particular dispute. Proper timing of the mediation effort is crucial in all instances; public debate between the parties has a deleterious effect upon the likelihood of peaceful settlement in most situations; but public reports and recommendations by neutral media- tors may be appropriate and useful in many disputes. Lovell (1952), Rehmus (1953) and Warren (1954) have argued that a mediator's effectiveness is primarily determined by the success with which he is able to bring pressure to bear upon the parties. More specifically, they emphasize that the mediator's role allows him to channel and concentrate existing pressures to break deadlocks. In these authors' view, the mediator's primary role is to force the par- ties to change positions rather than resolving personality clashes or providing negotiating skills. Kerr (1954) discussed the inevitability of conflict in an indus- trial society and examined mediation as one means of keeping indus- trial conflict within controllable bounds. Principal contributions which a mediator brings to industrial conflict, according to Kerr, include (1) reduction of irrationality; (2) removal of nonrationality; (3) exploration of solution; (4) assistance in the graceful retreat and (5) raising the cost of conflict. Kerr referred to mediation designed to resolve existing conflicts as "tactical" mediation. The advancement of long-term industrial peace is beyond the scope of tactical mediation. "Strategical" media- tion, on the other hand, involves processes designed to structure the environment to accommodate long term industrial peace. Strategical mediation involves such processes as (1) the integration of workers and employees into society; (2) the advancement of stability within society; (3) the dispersion of grievances; (4) ideological compati- bility; (5) secure and responsive relationship of leaders and members; and (6) the structuring of conflict within acceptable boundaries. Strategic mediation is today referred to as preventive mediation and is practiced frequently by state and federal mediation services. Douglas (1962) attempted a psychological analysis of the inter- action of bargainers and mediators. Her work is based on observations and records of actual negotiations. Douglas maintains that meaningful negotiations go through three stages: (1) establishing the range, 9 i.e., setting the limits of the bargaining range; (2) reconnoitering the range, i.e., through interpersonal interaction and not conflict, each party searches for areas in the other's range where he may make inroads without firm commitments; and (3) confronting the decision- reaching crisis, i.e., reaching the impasse or agreement. In concluding her study, Douglas questions three areas of the conventional wisdom of mediators and negotiators about collective bar- gaining. She characterizes the following as fiction: (1) the possi- bility of a strike is essential to resolution of conflict; (2) compro- mise is an essential part of agreement making; and (3) the ultimate agreement is narrowly circumscribed by market economics. All three of these points have raised substantial controversy and are largely re- jected by those with experience in industrial relations (Rehmus, 1965). Numerous additional field researchers have documented the assets they believe third parties may contribute in resolving a bargaining impasse. Chamberlain (1965) hypothesizes that mediation or concilia- tion fares best when the deadlock has risen over some misunderstan- ding; it fares worst when the intent of both parties is simply to hold the best bargaining position. Chamberlain does, however, concede that mediation and conciliation in labor relations are the most effective devices yet developed to head off strikes and enable collective bar- gaining to function. Prassow and Peters (1970) maintain that mediators intensify nego- tiations; that they are confidants for both parties; that their exploratory conversations on disputed issues should be off the record; that lawyers (presumable well-grounded in the principle of contract 10 law and judicial proof) make the best mediators; and that the old adage about experience being the best teacher certainly plays an important role. Miner (1969) argues that the most effective conciliator or media- tor is a "communications catalyst" and that successful mediation is a function of perceived impartiality and the capacity of the third party to win trust. The mediator aids parties in determining their relative position in the bargaining range; he more fully explains positions; he points out nonapparent bases for agreement; he helps search for solu- tions; and he generally facilitates the agreement. Experimental Literature Although the preceding review of field studies and observations by practitioners pertaining to third party intervention would appear to provide ample evidence for generalizations such is unfortunately not the case. As McGrath (1966) has stated, although field studies possess the great advantage of realism their generalizability is limited. In addition to the limitation cited by McGrath, most field studies dealing with third party intervention have been written by direct participants to the collective bargaining situation and are, therefore, potentially confounded. In efforts to overcome the inherent limitations relating to the case study approach numerous researchers have turned to experimental simulations (Campbell, 1960; McGrath, 1966; Bass, 1966; Johnson and Pruitt, 1972) as a method of examining third party intervention strategies. McGrath (1966) asserts that experimental simulations, 11 unlike field experiments, offer greater experimental control, permit replication and thereby can escape the particular and advance the generalizability of findings. Vaughan and Bass (1967) argue that although simulations cannot reproduce real-life operations in every detail, "business games“ often make it possible to make measurements of potentially important vari- ables. Among the first researchers to investigate the impact of third party intervention, within the context of experimental simulations, was McGrath and his colleagues (McGrath and Julian, 1962; Julian and McGrath, 1963; Vidmar and McGrath, 1967). Employing what he termed a tri-polar model of negotiations McGrath has systematically attempted to examine what factors enhance a mediator's effectiveness. Based on several studies McGrath (1966) concluded that the principal component a mediator provides within a bargaining situation is that of providing structure. The amount of structuring activity (by the mediator) does not correlate with group success, but the timing of such activities does. Successful groups are characterized by high structuring activ- ity during the initial bargaining phases, followed by low structuring activity in the final bargaining phases, whereas unsuccessful groups have highest structuring activity during the final phase. The successful mediator focuses the group on its task suffi- ciently well and sufficiently early, so that they can grapple with the issues. A second key element contributing to mediation effectiveness is the mediator's ability to refrain from displaying negative inter- personal feelings himself, and not to permit negotiating parties to do 50. 12 Podell and Knapp (1969) examined bargainers' willingness to accept a solution offered by a mediator once a bargaining deadlock had been reached. Their study simulated a labor-management wage negotia- tion by college students in a laboratory atmosphere, with the Ss assuming the role of the labor negotiator. In order to study the pro- cess of mediation, the authors argue, it was necessary to guarantee a need for it, and this need was achieved by assuring that no agreement would occur prior to the mediator's intervention. A negotiation dead- lock was accomplished by having the "management" role played by a re- search assistant who appeared to be a peer of the subject, and who followed a prearranged program of "hard line" bargaining. Once the deadlock was achieved, social pressure from the Programmed Management Negotiator (PMN) and the experimenter induced the subject to accept mediation. The final aspect of the simulation was to offer the sub- ject exactly the same concession on two different issues, one of these offers coming directly from the subject's opponent, (the PMN), and the other coming through the mediator. The researchers found that suggested concessions attributable to the mediator were more readily accepted by the subjects than compar- able concessions offered by the subjects' opponent. They conclude that "the present results indicate that concessions which come through a mediator create less of an impression of weakening of bargaining position than identical concessions which came from one's opposite negotiator (p. 517)." Podell and Knapp's findings, however, are not so illuminating as they might first appear because of several methodological limitations. First, no attempt was made to determine whether mediation produced 13 more effective or more frequent settlements than the parties would have reached under a condition of no intervention. In essence, no control group was included in the study. Second, by having the sub- jects bargain against an opponent who rigidly adhered to a pre-pro- grammed hard-line bargaining position foreclosed the feasibility of the parties' reaching a settlement prior to the mediator's interven- tion. Such constraints may have enhanced the parties' receptiveness to the mediator's suggested settlement. Third, the Ss belief that the mediator's proposals were actually a basis for settlement may have been based on prestige. Specifically, the 55 may well have felt that management would not refuse to accept the mediated proposal any more than he could, since the negotiators were students while the experi- menter-mediator was evidently a professor. Utilizing a bilateral monopoly bargaining game, Pruitt and Johnson (1970) created a simulated situation similar to that employed by Podell and Knapp (1969), such that subjects would, by design, en- counter a bargaining impasse. The researchers hypothesized that fol- lowing the occurrence of such an impasse a suggestion by a mediator of a point of settlement halfway between the positions of the two nego- tiators would produce substantial concessions. Support was found for the conclusion that a third party can stimulate concessions by suggesting an equitable solution to both negotiators. Additional hypotheses advanced by Pruitt and Johnson suggested that mediation would be most effective under conditions of high time pressure, where the other party moved slowly toward agree- ment, and in situations where the opposing party had made few l4 concessions. None of the hypothesized interactions were found to be significant. Pruitt and Johnson's study suffers several of the limitations cited earlier when discussing Podell and Knapp's research. Briefly, subjects were programmed to bargain against a fixed schedule which was designed to discourage a negotiated settlement prior to intervention. Additionally, no alternative mode of third party intervention (i.e., arbitration) was included in this study. In generalizing Pruitt and Johnson's findings to broader collec- tive bargaining situations, it should be noted that by utilizing a bilateral monopoly bargaining game the parties were not presented the variety of negotiable issues usually present in a collective bargain- ing situation. The bilateral monopoly bargaining game compels the parties to bargain over a single issue and thus prohibits "horse-trad- ing" between alternative issues. One of the few studies designed specifically to examine the impact of alternative modes of third party intervention was Urban's (1971) research. This study utilized 150 upper-division college stu- dents as subjects. The subjects engaged in a distributive bargaining task. The objective for each bargainer was to maximize his total pro- fit for the bargaining session. The bargainers negotiated in a state of incomplete information using written communications. The subjects were randomly assigned to bargain in one of three intervention modes; no arbitration (NA), voluntary arbitration (VA) or compulsory arbitration (CA). In the NA mode, bargainers negotiated to agreement without outside intervention. In the VA mode the bargainers could request the third party to intervene and conclude negotiations 15 by rendering a decision. In the CA mode, if the bargainers could not reach agreement, the third party would automatically intervene and conclude negotiations. In the VA and CA modes, the experimenter used an arbitration scheme to allocate the fixed resource. Urban hypothesized that bargainers in the NA mode of intervention would be more conducive to mutual agreement than bargainers in the VA and CA modes respectively. The dependent variables utilized in this study were adopted from the Siegel and Fouraker (1960) model of nego- tiation. The aspects of bargaining behavior studied were: (1) ini- tial profit expectation; (2) duration of bargaining process; and (3) yielding behavior of bargainers. Urban's results failed to support his hypotheses. Bargainers in the VA mode initially requested the least initial profit, concluded negotiations in the least amount of time and exchanged the least num- ber of bids. NA bargainers initially requested the greatest profit and exchanged the greatest overall number of bids. The CA bargainers required the greatest time to conclude negotiations but were not significantly different from NA bargainers in terms of number of bids. CA bargainers requested more profit initially than VA bargainers but less than NA bargainers. In terms of yielding behavior, no significant differences were found between the three modes. NA, VA, and CA bargainers did not dif- fer significantly in regard to amounts yielded, changes of position, and concession rates. Vidmar (1971), in the context of an experimental simulation of negotiation process, sought to determine if a neutral leader or "mediator" could increase negotiation effectiveness. Vidmar desired 16 to study the impact of mediation under two distinct conditions. First, a mediator's contribution within the context of a discussion group where the group members were not given representational role obligations. The group was told their goal was to set forth a solu- tion which was constructive and acceptable to both groups. Secondly, Vidmar examined mediator's contribution in a situation characterized by participants encouraged to assume representational role obligations and were instructed that the groups task was to set forth a construc- tive solution to the problem, but each representative was to make sure that his organization's viewpoints were represented in the final writ- ten group decision. Vidmar hypothesized, based on earlier field studies by Lands- burger (1955), that a mediator's contribution to negotiation effec- tiveness is, in part, determined by the degree of conflict in any par- ticular situation. By analyzing the solutions developed by discussion and negotiation groups, Vidmar sought to determine the extent to which conflict may serve as a moderator of mediation effectiveness. The results of this study indicated that mediators tended to improve the performance of negotiation groups, and on the overall suc- cess criterion, conceptually the most important, mediated negotiation groups performed significantly better than those without a mediator. The mediators employed by the experimenter in this study it is interesting to note, were peers of the group participants and pos- sessed no power or status differential upon which they might draw. This fact led Vidmar to suggest "had the mediators had more power or status, the difference between the mediated and unmediated negotiation groups might have been greater (p. 55)." 17 In summary, Vidmar's study suggests that the presence of a media- tor can improve negotiation effectiveness, but indicates that his con- tribution may be contingent upon the magnitude of the conflict and his status. Johnson and Pruitt (1972), employing a simulated collective bar- gaining game, sought to determine the impact that the anticipation of alternative modes of third party intervention had upon concession mak- ing in negotiations. Johnson and Pruitt hypothesized that concessions would be more rapid under the threat of binding, as opposed to non- binding, third party intervention. The authors' study involved 50 male graduate students in a course on collective bargaining. Manipulation of the independent variable, anticipated mode of third party intervention, was achieved by distri- buting different instructional sheets to the subjects contingent upon the experimental treatment. Research results confirmed that negotiators faced with a binding decision (arbitration) made larger and more frequent concessions than those faced with a non-binding (mediation) decision. However, the authors make two qualifications with respect to their findings. First, their hypothesis received strong support from those bargainers representing the union position but only marginal support from bar- gainers representing the management position. Furthermore, the dif- ferent concession strategy found between the binding and non-binding conditions can be clearly seen only during the late phases of the col- lective bargaining simulation. Following an examination of data collected by means of a post experimental questionnaire, Johnson and Pruitt suggest that certain 18 factors attributable to their experimental design may explain the pre- viously unexplainable reluctance of union negotiators to be confronted by a binding decision. First, in examining the collective bargaining simulation materials the authors discerned that the structure of the offer scales was such that the management negotiator was able to move well beyond the midpoint on all important wage scales before reaching his breakeven point while union negotiators could not do so. Further- more, the hypothetical mediator was described in the simulation material as an "advanced management trainee" once again potentially encouraging the union negotiator to reach an agreement prior to third party intervention. If, as Johnson and Pruitt suggest, their study was significantly compromised for reasons noted above, their findings must certainly be considered highly tentative. Johnson and Tullar (1972), in perhaps the most ambitious study of third party intervention to date, sought to investigate four forms of third party intervention under conditions of both high and low need to save face. The four forms of third party intervention were: (1) non- binding suggestion (mediation); (2) binding decision (arbitration); (3) selection of the best of the bargainers own proposals (govplan); (4) no third party intervention (control). As in the Johnson and Pruitt study mode of third party intervention was manipulated by use of alternative instructional sheets. The second independent variable, need to save face, was manipulated by the inclusion or exclusion of a video tape and voice recorder. Subjects in the high need to save face condition were told that the equipment was there to make a training film dealing with "novice bargaining" to be used in various classes in the area. 19 Johnson and Tullar's findings were that there was a significant main effect based on style of intervention and an interaction effect of intervention by face saving. In the face-saving condition those pairs expecting the govplan form of intervention were quite far from agreement while those not expecting any intervention were close to agreement. Those anticipating mediation and arbitration were at a moderate distance from agreement. In the low face-saving condition, those expecting arbitration had all reached agreement while those anticipating mediation were farthest from agreement. After interven- tion those who received a suggestion for settlement were no more likely to reach agreement than were those who did not receive a sug- gestion. The authors concluded that when the need to save face is strong bargainers are more willing to let an outsider come in and resolve the conflict. The expectation of any outside intervention appears to decrease the likelihood of agreement prior to that intervention. Par- ties appear even willing to allow the possibility of forced compliance to a third party's demands. When there is a low need to save face, those expecting arbitration expend considerable effort in order to avoid intervention and settle things on their own. Although need to save face should not be equated as a direct corollary of conflict intensity, it does appear logical that some cor- relation between the two phenomena is plausible. If this is the case, Johnson and Tullar's findings are of particular importance. What their findings imply, at least indirectly, is that bargainers, con- fronted by the most difficult bargaining situations, bargain least effectively when confronted by binding arbitration. 20 Such a conclusion tends to support Northrup's impressions drawn from extensive field experience while at the same time contradicting earlier research by Johnson and Pruitt. Additional research designed to shed light on this problem would appear to be needed, particularly in view of the fact that numerous states have recently passed laws requiring compulsory arbitration in the field of public employee bar- gaining (Weissbrodt, 1973). Johnson and Tullar's study raises several issues for future research, for example the ineffectiveness of mediation as a mode of third party intervention challenges results of several earlier studies (Podell and Knapp, 1969; Pruitt and Johnson, 1970; and Vidmar, 1971). As discussed earlier, many earlier studies were structured in such a manner that they encouraged affirmative findings relating to mediation effectiveness. The effectiveness of mediation in resolving bargaining impasses remains an unresolved issue. Summary of Literature Review As the preceding review illustrates both researchers and practi- tioners are divided as to what style of third party intervention most enhances conflict resolution. Even greater uncertainty surrounds what potential factors or conditions might improve or hinder alternative styles of intervention. Landsburger (1955), Douglas (1962), and Chamberlain (1965), all drawing from field experience, endorse mediation as an effective mode of conflict resolution. Their findings appear supported by laboratory studies conducted by Podell and Knapp (1969), Pruitt and Johnson (1970), and Vidmar (1971). However, as Vidmar himself admits, there 21 is in fact, very little experimental research on the effects of media- tors on the negotiation process. As previously noted several of the studies cited above, in support of mediation effectiveness, have suf- fered from significant methodological limitations. A recent study by Johnson and Tullar (1972), designed to overcome such limitations, found parties receiving a suggested settlement from a mediator were no more likely to reach agreement than were those who did not receive a suggestion. Arbitration as a mode of conflict resolution engenders even greater controversy. Northrup (1966) and Bok and Dunlop (1970) adamantly insist that the threat of arbitration seriously erodes the collective bargaining process. Stevens (1966) and Loewenberg (1970), on the other hand, find no support for such assertions. Laboratory research by Urban (1971), Johnson and Pruitt (1972), and Johnson and Tullar (1972) also appears equally divided on the impact of arbitra- tion on bargaining behavior. To date only two studies (Vidmar, 1971; Johnson and Tullar, 1972) have attempted to weigh the impact of moderating variables upon third party intervention. Presently no study has attempted to evaluate the impact of the three most commonly utilized styles of third party intervention (mediation, voluntary arbitration, and compulsory arbi- tration) as well as including a control group within the context of a single study. The study described in the following chapters examined the impact of four alternative modes of third party intervention upon bargaining behavior under conditions of high and low conflict. The review of literature presented above documents the controversy currently 22 existing pertaining to the effects of the anticipation of alternative modes of third party intervention. By adopting a somewhat different methodological framework than earlier studies it is hoped that a more comprehensive understanding of third party intervention will be gained through this study. In Chapter II the methodology utilized in this study and hypo- theses that were tested are presented. Chapter III presents the quan- titative findings obtained from a statistical analysis of the experi- mental data collected in this study. Discussion and conclusions are presented in Chapter IV. CHAPTER II METHODOLOGY AND HYPOTHESES Introduction In the preceding chapter, third party intervention as one strat- egy of conflict resolution and literature pertaining to third party intervention was presented. This chapter begins with a brief reitera- tion of the problem under investigation, followed by a presentation of the following: (1) the methodology utilized in this study, (2) the operational definition of the independent and dependent variables, (3) the hypotheses which were tested in this study, (4) a description of the sample selected (5) a description of the instructions relevant to the experimental task and (6) the proposed statistical treatment of these data. Statement of the Problem Third party intervention has been widely heralded as an effective way to resolving bargaining impasses (Kerr, 1954; Chamberlain, 1965; Walton, 1967; and Pruitt and Johnson, 1970). However, as Rehmus (1965), Rosenau (1969) and Vidmar (1971) have noted, we know very little regarding the extent to which third parties enhance negotiation outcomes. Two unresolved questions are: (1) what mode of third party intervention is most successful in enhancing bargaining settlements 23 24 and (2) does conflict intensity have an impact upon the effectiveness of alternative modes of third party intervention. Bok and Dunlop (1970) and Northrup (1966) have argued strongly that compulsory arbitration has a detrimental impact upon the bargain- ing process. Stevens (1966) and Reuther (1965), on the other hand, believe that this proposition is not necessarily self-evident and in fact compulsory intervention may foster a greater sense of urgency for the bargainers to settle their disputes. Evidence concerning the effectiveness of mediation is also divided (e.g. see Podell and Knapp, 1969; Johnson and Pruitt, 1972). Recent research by Vidmar (1971) and Johnson and Tullar (1972) suggests that the success of third party intervention may be contin- gent upon the magnitude of the conflict between the parties. Vidmar found that mediators improve negotiation effectiveness more under high as opposed to low conflict situations. Johnson and Tullar found that when the need to save face is strong arbitration inhibited par- ties from reaching an accord prior to third party intervention. How- ever, when there is low need to save face arbitration enhanced bar- gaining settlements. Methodology and Independent Variables A 2x2x4 factorial design with fixed effects and equal cell sizes was employed in this study (see Figure 1). This experimental design was expanded by adding a fourth factor, role within the bargaining dyad (management role or union role), in cases where the individual subject's responses served as the unit of analysis rather than the bargaining dyad. The expanded experimental design therefore was a 25 2x2x4x2 factorial design with repeated measures analysis on the last variable. There were five (5) dyads (10 subjects) in each of the 16 cells of the experimental design. FIGURE 1 EXPERIMENTAL DESIGN FOR THE STUDY OF INITIAL MANAGEMENT POSITION BY INITIAL UNION POSITION BY MODE OF THIRD PARTY INTERVENTION INITIAL UNION POSITION INITIAL MANAGEMENT POSITION Soft Voluntary Compulsory Bargaining» Mediation Arbitration Arbitration 50ft x111 x112 X113 X114 7°”9“ X211 x212 X213 X214 FIGURE 1 (Cont'd) INITIAL UNION POSITION INITIAL MANAGEMENT POSITION Tough Voluntary Compulsory Bargaining Mediation Arbitration Arbitration 50ft X121 X122 x123 x124 T°"9h X221 x222 X223 x224 Factor A was initial management position, i.e. soft or tough. A "soft" initial management position was manipulated by assigning sub- jects to a treatment condition where their initial bargaining demand 26 was low, i.e., not very demanding. A "tough" initial position was manipulated by assigning subjects to a treatment condition where their initial bargaining demand was high, i.e., very demanding. Factor B was initial union position, i.e., soft or tough, and was opera- tionalized in the same manner as initial management position (see Appendix A) Factor C represented four modes of thirdgparty interven- tion, i.e., bargaining (no third party intervention), mediation, voluntary arbitration and compulsory_arbitration. Mode of third party intervention was manipulated by means of dif- ferent instructional sheets. All sheets stated that if a negotiated settlement had not been reached fifteen (15) minutes after negotia- tions began, negotiations would stop. Under each mode of third party intervention the remainder of the message was different (see Appendix 8). Under bargaining, subjects were told they should cease discussion with their opponent and reevaluate their bargaining strategy. After they had thought things over for a few minutes it was indicated that negotiations would resume. Subjects under mediation were instructed that a person unrelated to this experiment but knowledgeable in the field of industrial relations who had earlier studied these bargaining issues would present his proposal for a settlement for their con- sideration. His proposal, it was stressed, was to be considered a suggested settlement and that subjects were free to act or not act on his suggestion. After the mediator's suggested settlement was pre- sented, subjects were informed that negotiations would resume (see Appendix C). Under voluntary arbitration subjects were informed that should a bargaining impasse arise they could, if they desired, elect to submit their impasse to a neutral third party knowledgeable in the 27 field of industrial relations who had earlier studies these bargaining issues and arrived at a proposed settlement. This individual's set- tlement would be binding upon the parties. However, such third party intervention would occur only at the joint request of the subjects. Subjects under compulsory arbitration were informed that should they be unable to reach a settlement within the designated time a third party, knowledgeable in the field of industrial relations who had ear- lier studied these issues and arrived at a proposed settlement would intervene and render a binding settlement (see Appendix D). The fourth independent variable was rglg_within the bargaining dyad; management role and union role. Role was determined by randomly assigning subjects to represent either the management position or union position. Dependent Variables Ten different dependent variables were collected during the experimental session. These included both objective (task-oriented) and self-report (post experimental questionnaire) measures. The dependent measures under examination in each treatment condition as operationally defined were: 1. Number of Contracts Settled. A contract was regarded settled only i? sUBjects reached agreement on all three issues under negotiation during the initial fifteen (15) minute bargaining session. 2. Number of Issues Left Unresolved. Issues under negotia- tion not settled during the Tnitial fifteen (15) minute bargaining session were regarded as unresolved. 3. Total Number of Bids. A bid was defined as a subject's initial bid followed by his opponent's counter bid. 28 Total Opening Bid. The amount of earnings sought by subjects in their total opening bid as a percentage of the maximum they could have demanded. Total Amount Conceded. The total amount conceded by subjects as a percentage of the maximum they could have conceded based on all three issues under negotiations during the initial fifteen (15) minute bargaining ses- sion. Earnings. The total money earned by subjects in the course of the simulated collective bargaining game. A post experimental questionnaire was given to all subjects fol- lowing the completion of the simulated collective bargaining game to measure realism, fear of third party intervention, opponent's bar- gaining strategy and conflict intensity. The post experimental questionnaire is shown in Appendix E. Some items were included in the analysis as a check on the experimental manipulation and intended only for that purpose. 7. 10. Mundane Realism. Self-reported realism experienced dur- ing the course of the simulated collective bargaining game. Fear of Third Party Intervention. Self-reported impor- tance of reaching an agreement during the initial fif- teen (15) minute bargaining session prior to third party intervention. Opponent's Bargaining Strategy. Self-reported percep- tion of opponent's willingness to compromise. Conflict Intensity. Self-reported perception of the extent of difference between subjects initial bargaining positions. Hypotheses The hypotheses which follow are drawn from the review of litera- ture presented above. Mode of third party intervention and conflict intensity are the primary variables under examination in this study. Based on prior_research findings and theoretical predictions, the 29 following hypotheses are presented: Hypothesis l-A. The stronger the anticipated mode of third party intervention, the greater will be the number of contracts settled. Hypothesis 1 is based on research findings by Podell and Knapp (1969) and Pruitt and Johnson (1970). Podell and Knapp found that mediation assisted parties in reaching bargaining settlements, while Johnson and Pruitt found that arbitration was a more effective mode of third party intervention than mediation. Hypothesis 2—A. Subjects bargaining under soft initial positions will settle significantly more issues than subjects negotiating under tough initial positions. Hypothesis 2-B. Subjects anticipating stronger modes of third party intervention will settle more issues than subjects anticipating no third party intervention or weaker modes of third party intervention. Hypothesis 2-C. A significant interaction is predicted such that stronger modes of third party inter- vention will result in fewer issues remain- ing unresolved under a tough initial posi- tion while a soft initial position will result in more unresolved issues under stronger modes of third party intervention. Hypothesis 2 is based on theoretical hypotheses advanced by Komorita and Brenner (1968) suggesting that bargainers under high con- flict or confronting tough bargaining strategies would reach fewer agreements than bargainers under low conflict or confronting softer bargaining strategies. The preceding hypothesis is also supported based on Johnson and Pruitt's (1972) findings that arbitration proved to be a more effective intervention strategy than mediation. The hypothesized interaction between initial position and mode of third party intervention is based on Vidmar's (1971) finding that mediation was more successful in negotiation than discussion groups. 3O Hypothesis 3-A. Total number of bids exchanged within dyads will be significantly greater under a tough as compared to soft initial position. Hypothesis 3-8. The stronger the anticipated mode of third party intervention, the greater will be the total number of bids exchanged. Hypothesis 3—C. A significant interaction will be found between initial position and mode of third party intervention such that stronger modes of third party intervention will result in significantly more bids under a tough initial position while stronger modes of third party intervention will result in fewer total bids under a soft initial posi- tion. Hypothesis 3 is based on the theoretical prediction that bar- gainers confronted by higher degrees of conflict will make a greater number of total bids in an effort to reach a settlement than subjects confronting lower degrees of conflict. Johnson and Pruitt's (1972) findings that stronger modes of third party intervention will encour- age bargainers to reach a settlement prior to third party intervention and thereby increase the total number of bids between the parties also supports the preceding hypothesis. The predicted interaction between initial position and mode of intervention is supported by Vidmar's (1971) research cited above. Hypothesis 4-A. Total opening bid will be significantly less under a tough as compared to soft initial position. Hypothesis 4-8. The stronger the anticipated mode of third party intervention, the less subjects will demand in total opening bid. Hypothesis 4-C. A significant interaction is predicted such that stronger modes of third party inter- vention will result in smaller total open- ing bids under a tough initial position while a soft initial position will result in larger total opening bids under stronger modes of third party intervention. 31 Hypothesis 4 is based on the theoretical proposition that bar- gainers confronting large discrepancies between their respective posi- tions will be less demanding in their total opening bid and Johnson and Pruitt's (1972) finding that bargainers fear stronger modes of third party intervention and therefore will be less demanding in their total opening bid to enhance agreement prior to third party interven- tion. Hypothesis 5-A. Total amount conceded will be significantly greater under a tough as compared to soft initial position. Hypothesis 5-B. Stronger modes of third party intervention will lead to a significantly greater total concession. Hypothesis 5-C. A significant interaction between initial position and mode of third party interven- tion is predicted such that stronger modes of third party intervention will result in a greater total concession under a tough initial position while under a soft initial position stronger modes of third party intervention will result in a smaller total concession. Hypothesis 5 is based on the prediction that subjects assuming a tough initial position will be forced to make greater concessions than subjects assuming soft initial positions to enable agreement to be reached. Hypothesis 5 is also supported by Johnson and Pruitt's (1972) finding that subjects fear stronger modes of third party inter- vention and therefore will make greater concessions to avoid such intervention. Vidmar's (1971) research showing mediation to be more effective under higher levels of conflict supports the predicted interaction between initial position and mode of third party inter- vention. 32 Hypothesis 6-A. It is predicted that subjects bargaining under soft initial positions will reach more settlements and subsequently obtain higher earnings than subjects negotiating under tough initial positions. Hypothesis 6-B. The stronger the anticipated third party intervention, the greater will be subjects' earnings. Hypothesis 6-C. A significant interaction is predicted such that stronger modes of third party inter- vention will result in higher earnings under a tough initial position while stronger modes of third party intervention will result in smaller earnings under a soft initial position. Hypothesis 6 is based on theoretical premises presented earlier. First, subjects bargaining under soft initial positions will reach more settlements hence obtain higher earnings. Secondly, stronger modes of third party intervention will result in greater pressure to reach agreement and subsequently result in higher earnings. Experimental Method Subjects and Sample Size Subjects for the experiment were males recruited from under- graduate business and psychology classes at Michigan State University. A total of 196 subjects took part in this experiment, with 36 partici- pating in the pilot study and 160 taking part in the main study. At the time of recruitment subjects were told they would participate in a collective bargaining exercise where they would have the opportunity to earn some money. 33 Apparatus Two classrooms were used in this study. Subjects were randomly assigned to one of five (5) dyads in each of sixteen (16) treatment conditions. Each room contained tables and chairs arranged so that the subjects faced each other across a table. Only dyads under the same experimental treatment bargained within the same room. Clocks were also provided in each room so that subjects had knowledge of the negotiating time remaining. Five dyads bargained per negotiating ses- sion. Negotiation sessions were run on four consecutive evenings. Experimental Task A simulated collective bargaining game originally developed by Campbell (1960) and significantly modified for the purposes of this study was utilized. The task involved face-to-face bargaining between two subjects in a dyad. The simulation introduced enough background information to maintain the interest of the subjects. Subjects bargained over three (3) issues (wages, hospital and medical plan, and a cost of living clause) simultaneously. Each bar- gaining issue was presented on an individual scale indicating various potential points of settlement. Each bargaining scale in addition to indicating points of potential settlement included scales informing the bargainers of "points" they would receive should a settlement be reached at any designated location. Subjects were informed that their earnings in the collective bargaining game were contingent upon the points they earned. To encourage subjects to actively bargain, point scales were constructed in a fashion to prohibit an even division of points. Potential earnings ranged from $3.30 to zero if subjects 34 failed to reach agreement on any issue under negotiation (see Appendix A and F). All subjects received an instructional sheet providing general background information regarding the company-union relationship (see Appendix G). Management representatives also received additional information pertaining to the company position (see Appendix H). Cor- respondingly, union representatives were provided with further infor- mation explaining the union position (see Appendix 1). Subjects also received an instructional sheet outlining the procedures which would be followed in the event of a bargaining impasse. These sheets stated that if a negotiated settlement was not reached fifteen (15) minutes after negotiations had begun, negotiations would stop. The remainder of the instructions were different depending upon mode of third party intervention (see Appendix B). To encourage subjects to reach a settlement prior to intervention and thereby stimulate costs similar to those incurred when bargainers fail to reach collective bargaining settlements without strikes, sub- jects were instructed that should they be unable to reach a contract settlement within the specified time (15 minutes) their earnings, based upon points accumulated, would be reduced by 20 percent. Sub- jects received 10 cents for every point earned when a contract was settled within the Specified time. Subjects requiring more than the specified time to reach an agreement received 8 cents for every point earned until 25 minutes of bargaining time had elapsed, after which they received no payments. 35 The Instructions Subjects were instructed to report to designated classrooms at a Specified time. They were scheduled in groups of twenty, five dyads to each of two adjoining classrooms. Appropriate instructional material was distributed to subjects dependent upon experimental treatment condition. After subjects had finished reading their instructions, the experimenter verbally went over the instructions with them answering any questions they might have. Great care was taken to ensure that each subject was totally familiar with the process before negotiations began. When the instructions were completely understood the negotia- tions began. The experimenter would exit the room leaving the door open so that he could observe the bargaining process. At the end of the initial 15 minute bargaining session, all sub- jects were requested to cease bargaining and the appropriate mode of third party intervention was introduced. Subjects who had reached a settlement, who were bargaining under compulsory arbitration, or who decided to utilize voluntary arbitration were then asked to complete the post experimental questionnaire and then leave the room. Subjects under mediation who had not reached a settlement were given the media- tor's suggested settlement at this time. Subjects who had not reached a settlement and were neither bargaining under compulsory arbitration nor utilized voluntary arbitration were given an additional 10 minute bargaining period to reach a settlement. Following this second bar- gaining session the remaining subjects were asked to complete the post experimental questionnaire. 36 After all post experimental questionnaires were completed, all subjects were reassembled in their appropriate classrooms and paid. Subjects were debriefed and then dismissed. Data Collection and Scoring Measures of the various dependent variables were taken during or after the collective bargaining game depending on the particular vari- able in question. Subjects were requested to record each of their bids on sheets provided for each issue under negotiation and to indi- cate when an agreement was reached on each issue under negotiation (see Appendix J). Analysis of these sheets provided data relating to the following dependent variables: (1) contracts settled; (2) issues left unresolved; (3) total number of bids; (4) total opening bid; (5) total amount conceded; and (6) earnings. A post experimental questionnaire (see Appendix C) containing 10 forced choice items was used to collect data on subjects' perceptions of realism, fear of third party intervention, opponent's bargaining strategy and conflict intensity. Pilot Study Prior to conducting the main experiment a pilot study was carried out to answer the following questions: 1. Were the instructions clear and comprehendible? 2. Did the subjects become involved in the bargaining process? 3. Were time constraints sufficient to allow settlements to be reached and at the same time create adequate pressure to reach agreement? 4. Were data collection techniques adequate to insure tabu- lation and analysis? 37 5. Did the manipulation of independent variables appear to be effective? 6. Could five dyads carry on negotiations within the con- fines of a single classroom at the same time? 7. How much time should be allotted to conduct the experi— ment? 8. Was the post experimental questionnaire effective in measuring the perceptions of the subjects? As a result of the pilot study: (a) The instructions were found to be clear; (b) Subjects appeared to become highly involved in the collective bargaining game; (c) Post experimental interviews with sub- jects revealed that 15 minutes was judged to be an adequate amount of time prior to third party intervention; (d) Data collection techniques were found adequate to insure data tabulation, although it was found essential to remind subjects to record each of their bids during the course of negotiations; (e) Measures designed to manipulate indepen- dent variables appeared to be operating effectively; (f) If tables were arranged in a proper fashion, five dyads could carry on negotia- tions at the same time free from undue interference; (g) It was deter- mined that approximately one hour was necessary to carry out the experiment; and (h) For the purposes of this study, the post experi- mental questionnaire was judged as adequate to measure subjects' per- ceptions. This chapter has reiterated the problem under investigation in this study: (1) what mode of third party intervention is most suc- cessful in enhancing bargaining settlements and (2) does conflict intensity have an impact upon the effectiveness of alternative modes of third party intervention. 38 The methodology which was utilized in this study as well as the independent variables (initial management position, initial union position, mode of third party intervention and role) were presented and discussed. Dependent variables under examination including number of contracts settled, number of issues left unresolved, total number of bids, total opening bid, total amount conceded and earnings were identified. Questions contained within the post-experimental ques- tionnaire were also discussed. Based on earlier research and theoretical predictions hypotheses relating to the effect of initial position and mode of third party intervention upon bargaining behavior were presented. Information relating to sample size, apparatus, instructions, data collection and scoring and finally results of an earlier pilot were discussed. Chap- ter III reports the quantitative findings obtained from statistical analysis of the experimental data. CHAPTER III RESULTS Introduction As stated in the preceding chapter, a basic 2x2x4 factorial design with equal cell Sizes was employed in this study. Dependent measures capable of being analyzed on the basis of individual responses such as total opening bid, concession rate and earnings were analyzed by adding an additional variable, role played within the bar- gaining dyad (management or union), to the factorial design thereby creating a 2x2x4x2 factorial design. This chapter first examines in general the six previously iden- tified dependent variables. Concurrence between the experimental findings and hypotheses earlier stated are discussed. Also, partici- pants' responses to four post-experimental questionnaire items designed to measure participant's perception of realism, fear of third party intervention, bargaining opponent's strategy and conflict inten- sity are presented. Contracts Settled Contracts settled served as one of the primary dependent vari- ables under study. A contract was regarded settled only when agree- ment was reached on all three issues under negotiation within the first fifteen minute bargaining session. Twenty dyads bargained under 39 40 each of four anticipated modes of intervention (bargaining, mediation, voluntary arbitration and compulsory arbitration). It was predicted that the stronger the anticipated mode of third party intervention the greater would be the number of contracts settled. Table 1 presents the results of the analysis of contracts settled. TABLE 1 CHI-SQUARE ANALYSIS OF "CONTRACTS SETTLED" Voluntary Compulsory Contracts Bargaining, Mediation Arbitration Arbitration Settled 11 2 7 10 Unsettled 9 18 13 10 x7= 6.52, p < .10 These results did not confirm the hypothesis. Dyads bargaining under voluntary arbitration reached more contract settlements than those negotiating under mediation. Further, dyads bargaining under compulsory arbitration reached more settlements than subjects under either mediation or voluntary arbitration. Surprisingly, the greatest number of contract settlements occurred under bargaining and compul- sory arbitration. A second unexpected finding was the small number of contracts settled under mediation. Issues Left Unresolved To examine further the impact of alternative modes of third party intervention, in addition to including initial management position (Factor A) and initial union position (Factor 8) within the analysis, 41 a 2x2x4 ANOVA was conducted to analyze the dependent variable, number of issues left unresolved. Significant main effects based on initial position and mode of intervention were expected. Specifically, it was hypothesized that more issues would be left unresolved under a tough rather than soft initial position. It was predicted further that fewer issues would remain unresolved the stronger the anticipated mode of third party intervention. Finally, a significant interaction effect between initial position and mode of intervention was predicted such that stronger modes of third party intervention would result in fewer unre- solved issues under a tough initial position while stronger modes of third party intervention would result in more unresolved issues under a soft initial position. Table 2 presents an analysis of variance table of issues left unresolved. 42 ¢M.P ~¢.— oF.— m¢. om._ oo.m om._ om. qm.~ oo.P mm. ¢_.P om. oo.F o¢.P om.~ mm., om._ oo._ om.e oo.N om._ oo.~ om. me. me. mm._ om._ ON. ON. oo.N om. .amqm .<.>;m3mmMz .qmmm .<.u .«t> .owz .mam uwom .a.m smack z cowpwmom ucmemmmcmz .o.m meowcH peom z onHHmoa onz: 440mmmzs Ham; mmammH no mumzaz= mxh mo muz do mHm>4mwo omcm:mmMz .m .002 .mqm pwom .0.m smack z coepwmoa pcmEmmmcmz .0.m mepwcH “mom 2 onpHmom onza 4 00 mHm>4m<2200 oz< monH00 00<0z “mom cowpwmoa pcmemmmcez meuwcH onkaoa onzz 4 00 mHm>4m00 00<0z 00:0» .0.0 00:0H z cowpwmoa acmsmmmcmz .0.m _m_pwc0 p000 z onhHmom 20020 4 uwom .0.m smack z cowowmoa ucmsm0mcmz .0.m FowuwcH 0000 2 «=00000200 Hznoz< 4<50H= 00 002 onPHmom 20020 44m00 00<0z<~m .mz<0z 0P 040 smack .0.0 z .0.0 2 00:0» 9000 cowuwmom pcmsmmocoz meuwcH onhHmoa onzz 4 umom .0.m smack z cowp_moe acmEmmocoz .0.m FowgwcH “wow 2 20HH0000 20020 4 00 mHm>4<2< 00 >m<2200 0z< 0200H00 00<02 .mmsoe .0.m 20:00 2 cowoemoa gcmemmmcmz .0.m meuwcfi “wow 2 20Hh0000 20H20 4wmugma osu gmummgm as» some PFmo one LmFFoEm «she 20.0 om.P 0_. 02. em. mm. «0.2 mm. .o.m 20:00 05 0 ON 0 oo _ 00 F om _ as 2 oo 0 00 _ z copowmoa pcmEmmmcm2 mm. mm. Sm. oo._ 5m.F _N. o_._ om.P .o.m _8080=H “com oo._ oo._ 05., 00.0 00.. oo.0 om._ ON.N z a z a z a z a z cowoacpwne< cowpmzoene< cowom_ooz .Tmc_c_mmumm xeomrsaeou xeoucaro> “cow onHHmoa 20020 4 00 m~m>4<2< 00 >m<2200 02< 020HF00 00<02wmucma m0» gmammg0 m0» some rbmo mcp gmhFoEm msew 00. 00. 00. 00. 0¢._ 02.0 00. 00. .0.0 00.0 00._ 00.? 00.0 00.0 00.0 00.0 00._ 2 00. 00.0 F0. 00.0 00. 0F. 00. 00. .0.0 00._ 00.0 00.0 00.0 00.0 00.? 00.0 00.~ 2 0 2 0 2 0 2 0 2 cowpmcpwpc< covpmgpwng< cowummuwz mcwcwomcmm xgomfizasou xemucs_o> 20:00 smack cowpwmom pcmsm0mcm2 mepwca H000 20HH0000 20020 0empcm magma.0gmcp mo com» 0:» Lmuomg0 m0» coma Pqu on» Lofiposm make 00.0 00.0 0_. 0p. 00., 0¢.F 00.0 00.0 .0.0 2030» 00 m 00 m 00 P 00 P 00 0 00 0 00 0 00 m 2 mcowuwmoa pcmEm0ocm2 00.2 om., om.2 0m.F om._ 0m.2 mm.~ 00._ .a.m _mwoPcH 0000 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 2 0 2 0 2 0 2 0 2 newumcuwac< covpoeuwng< coprwumz 0:0:080280 ngszaeou acmpcaro> “wow 20Hk0000 20020 000H20 >Hm<0 omep 00 m<00= 000 002 00 mHm>0<2< 00 >m<220m 02< 0200H00 00<02emucw 00080 0002» we comw,msp gmaomgm mcp come meo m2» gmrmem make 00.0 00._ 00.0 00.0 00.0 00.0 00.0 00.— 00.0 00.0 00.¢ 00.0 00.0 00.0 00.0 02.0 00._ 00.0 00. 00.— m0.— 0_.0 00.0 00.0 00.2 00.2 00.— 00.0 00.0 00.0 02.2 00.0 0 2 0 2 0 2 0 2 cowpmgpwne< covuoguwne< cowpmwum2 .mcwcwomgom xgoszasou xgmucspo> Lunch cowpwmom ucmEmmocmz Fowuwcm onhHmom 20020 0_mugmn mm: m_n_xm_m ages asp coma Ppmu m2» chFmsm osww mN.~ NF.~ ma. mm. ~—.~ mm.~ No.~ mm. .o.m smack om.¢ oo.¢ oo.¢ oo.m om.m oo.m o¢.~ oo.~ 2 cowpwmoa ucmswmmcm2 40., mm._ m¢.N mm.~ NF.N 40., mm.~ em._ .o.m FawpwcH “mom om.m o¢.¢ om.~ om.m om.¢ o~.m oo.m o¢.m 2 a 2 a 2 a 2 2 2 covumguwnc< cowpmcpwng< cowpmwnm2 .mcwcwmmcmm xgomfisaeou mgmucapo> “mom onHHmoa onz: 4omk no mHm>4mmo omPmucmq ucmcoaao mewcwmmcmnm.uummnam mm: mwnwxmpv mgos mcwzcmme Ffimu ms» LmPFmEm mcew mm.~ w¢.F ¢_.~ No.~ mm. ~m.~ Fo.~ so.m .o.m smack oo.m cu m om a on m oc m om.m o¢.m o¢.m 2 cow¢_moa acmemmmcoz mm.F mm._ mm._ 4m. mm.F mo.~ mN.F om.P .o.m _anP=H pmom om.m oo.m om.m om.~ om.~ om.~ om.m om.~ 2 2 2 a 2 2 2 2 2 cowumcuwnc< :o_pmcgwnc< coTOvam2 wcwcwmmcom 22cmrzaeou 22mgc:_o> smack onHHmom onz: 4_mugma m2» Lmummgm msp came Fqu mg» LmerEm msww mm. mm. me. ~_. mm. o—.P Pu. oo.~ .o.m gmaoh om.~ oo._ oo.¢ oo.F oo._ ou.~ oo.~ oo.m 2 cowupmom ucmemmmcoz mm._ om., o_._ o_._ om.F 4F._ oo.N om.. .o.m _awpwcfi pmom o¢.N om.~ o~.~ om.N om._ o¢.~ oo.m om.m 2 2 2 a 2 a 2 a 2 coppmcpwnc< cowumguwng< cowamvumz limcwcwmmcmm xcomrquop 22mpcsro> “wow onhHmom onz: 4kazuth HUHAmzou: mom moz no mHm>4mmo om_mocma mgp cmummcm m2» some P_mu ms» gmppmsm mcha ¢P.P ¢N. mm. ¢_._ ¢~._ om.~ om.P ¢P.F o¢.~ oo.~ om.m oo.~ o¢.~ om.m om.P oo.m oo._ «m._ mm. am. qp.— mm.~ mm. em.~ oo.m om.~ oo._ om.m o¢.m oo.m ow.P oo.m 2 2 2 2 2 2 2 2 cowumcpwngq cowpmgpwnc< co_um_cm2 .mcwcwmmgmm Acomraasou 22mpczpo> smack .o.m smack z cowbwmoa pcmsmmmcmz .o.m meuwcfi “mom 2 onHHmoa onz: 4Hs do pmou oz< zwp mo umou mcmwmo pawEmmmcmz mmampu ch>PF 2o “moo mncmsmo co_c= mm ow Nu co om w¢ ov Nm 2N op pumcpcou >9 umcm>ou mmmmcucH mcv>w4 we “moo yo wmmpcmucma PF OP m m m o m w m N mm3<4u ozH>H4 mo Hmou NP mucwoa ucmEmmocmz mpcwoa cows: APPENDIX G BACKGROUND INFORMATION TO COLLECTIVE BARGAINING GAME ITII’III ...... APPENDIX G BACKGROUND INFORMATION TO COLLECTIVE BARGAINING GAME BACKGROUND INFORMATION The Townsford Company is a sma11 texti1e company 1ocated in a 1arge northern city in the USA. Townsford is high1y respected for its qua1ity work in the dyeing and finishing of raw woven fabrics. It emp1oys approximate1y 100 men. Townsford's men are among the most ski11ed to be found in the area. The genera1 business conditions of the country are good and the financia1 conditions of Townsford are stab1e. Townsford is operating at fu11 capacity and has a six month back1og of orders. Profits are not as high as at previous times, however, since the company has not raised the prices in severa1 years in order to maintain a good compe- titive position with other sections of the country. The company has been ab1e to maintain a 6 percent sharehoners' dividend and has made recent purchases of more modern equipment. The personne1 poIicies at Townsford are not the most modern but are better than those of most p1ants the same size. The past presi- dent of the company, who retired three months ago, knew most of the men persona11y and was we11 1iked. He is 1arge1y responsib1e for the reputation of Townsford as a "good p1ace to work." His successor is viewed with some suspicion by the workers, due main1y to his statements about changing some of the work procedures to achieve greater efficiency. For the 1ast 25 years, a majority of the emp1oyees have been mem- bers of the union. Re1ations of the union with the company, for the most part, have been quite good with grievances prompt1y discussed and sett1ed. The first strike occurred, however, three years ago and 1asted 15 days. The workers 1ost the fight for a s1iding-sca1e wage based on increases in the cost of 1iving index, but did get severa1 other minor fringe benefits. A1though Townsford's wage sca1e, $1.94 per hour, compares favor- ab1y with most other texti1e firms in the area, it is 3 percent be1ow those texti1e firms which emp1oy workers of equiva1ent high ski11 and produce a simi1ar high qua1ity product. Wages in the industry have not increased in proportion to increases in the cost of 1iving or in- creases in other industries. Despite occasiona1 sma11 wage increases, over a period of years Townsford's workers have s1ipped from a re1ative1y high pay sca1e to a position rough1y equiva1ent to that of 1ow1y ski11ed workers in other industries. This has caused some unrest among the workers, and there is some danger of the workers shifting into these other higher paying industries. UnempToyment is be1ow norma1 in the area, and it has been difficu1t to obtain rep1acements at Townsford. The three year contract has now expired. Negotiations broke down in the fina1 week with both sides adamant in their positions. The on1y agreement reached was that each side wou1d se1ect a new bargain- ing agent to represent it, schedu1ed to meet today in an attempt to reach a quick so1ution and avoid a strike. APPENDIX H INSTRUCTIONS TO MANAGEMENT REPRESENTATIVE APPENDIX H INSTRUCTIONS TO MANAGEMENT REPRESENTATIVE ASSIGNMENT AS COMPANY NEGOTIATOR You have been se1ected by the Townsford Company to represent it in its negotiations with the Union. Negotiations for a new two-year contract broke down 1ast week. The Union demands for genera1 wage and benefit increases are comp1ete1y unreasonab1e. If 1abor costs are in- creased it might necessitate price increases which cou1d serious1y hamper the company's competitive standing. A1though no compromises were reached in either side's position, it was decided that each side shou1d appoint new negotiators in an effort to sett1e the contract. You are to do the best possib1e job you can to get a good sett1e- ment of the contract for the company. A1though the company now has a back1og of orders, it is in danger of 1osing severa1 major customers if increased 1abor costs necessitate a significant price increase. It is essentia1 to the company, however, that the contract be sett1ed in this bargaining period. We rea1ize that this invo1ves compromises on both sides, and you are appointed to carry out binding negotiations for us. Remember, your job is to reach a sett1ement, one that is good for the company, in this negotiating period. Indicated be10w each bargaining issue you wi11 find the 1ast offer made by each party prior to the termination of bargaining. As you wi11 notice there are numerous positions for potentia1 agreement regarding each issue under negotiation. Agreements must, however, be reached on1y at positions designated. Above and be10w each bargaining sca1e you wi11 notice additiona1 sca1es identified as union points and management points. These are the points which you wi11 earn when an agreement is reached at each designated point. As you wi11 notice the more favorab1e an agree- ment you reach for your respective party the greater your point tota1. At the conc1usion of bargaining you wi11 receive ten (10) cents for every point you have earned. APPENDIX I INSTRUCTIONS TO UNION REPRESENTATIVE APPENDIX I INSTRUCTIONS TO UNION REPRESENTATIVE ASSIGNMENT AS UNION NEGOTIATOR You have been se1ected by the Union to represent it in its nego- tiations with the Townsford Company. Negotiations for a new two-year contract broke down 1ast week. The Union is thorough1y irritated with the Company's refusa1 to grant the workers bad1y needed wage and bene- fit increases. A1though no compromises were reached in either side's position, it was decided that each side shou1d appoint new negotiating agents in an effort to sett1e the contract. You are to do the best job you can to get a good sett1ement of the contract for 1abor. Union members were dissatisfied with the 1ast contract three years ago and there is serious danger of division in the ranks of the Union if a more satisfactory contract is not achieved in these negotiations. It is essentia1 to 1abor, however, that the contract be sett1ed in this bargaining period. We rea1ize that this invo1ves compromises on both sides, and you are appointed to carry out binding negotiations for us. Remember, your job is to reach a sett1e- ment, one that is good for 1abor, in this negotiating period. Indicated be10w each bargaining issue you wi11 find the 1ast offer made by each party prior to the termination of bargaining. As you wi11 notice there are numerous positions for potentia1 agreement regarding each issue under negotiation. Agreement must, however, be reached on1y at positions designated. Above and be10w each bargaining sca1e you wi11 notice additiona1 sca1es identified as union points and management points. These are the points which you wi11 earn when an agreement is reached at each designated point. As you wi11 notice the more favorab1e an agreement you reach for your respective party the greater your point tota1. At the conc1usion of bargaining you wi11 receive ten (10) cents for every point you have earned. APPENDIX J SUBJECT'S COLLECTIVE BARGAINING RESPONSE SHEET APPENDIX J SUBJECT'S COLLECTIVE BARGAINING RESPONSE SHEET Union Initia1 Position Demands Management Initia1 Position Offers APPENDIX K SUBJECT'S COLLECTIVE BARGAINING SETTLEMENT FORM APPENDIX K SUBJECT'S COLLECTIVE BARGAINING SETTLEMENT FORM WE THE UNDERSIGNED HAVE REACHED AN AGREEMENT ON ALL THREE ISSUES AND OUR CONTRACT IS THEREFORE SETTLED. MANAGEMENT REPRESENTATIVE UNION REPRESENTATIVE n 1 Illr IIIlnlvlll (‘11.!lllillllifl‘.1:l|| . Ill '11 RN STA TE IIIHIIWI INNI 22E) £5163 ii UNIV. LIBRARIES INIWIHWII(WIN 0695331