Introduction

     During the last two decades or so, considerable research has been conducted on non-union employment dispute resolution processes and outcomes. Earlier, the vast bulk of dispute resolution research focused on unionized settings and on the formal grievance procedures that were included in virtually all collective bargaining agreements. In that literature much attention was paid to grievance filing, processing, and settlement, with grievance arbitration singled out for particular attention. With the subsequent decline in unionism, the rapid rise of the non-union firm, and the adoption by such firms of one or another type of employment dispute resolution procedure, research attention became increasingly drawn to this non-union landscape. While much of this research has also singled out arbitration for particular attention, a substantial portion of this research has focused on postÐdispute resolution outcomes. Such research essentially attempts to answer the question, "What happens to the disputing parties after their disputes are settled?"

     In a series of studies that address this question, Lewin selected samples of non-union employees who had used their respective organizations' employment dispute resolution procedures (that is, grievance filers) and matched samples of employees in the same organizations who had not used these procedures (that is, grievance nonfilers). He then compared the job performance ratings, promotion rates, and work attendance rates of these grievance filers and nonfilers in each organization before, during, and after a specific period of grievance filing and settlement. These comparisons showed no statistically significant differences between the two employee groups before or during grievance filing and settlement, but statistically significant differences were apparent thereafter. In particular, both job performance ratings and promotion rates of grievance filers were significantly lower than the job performance ratings and promotion rates of grievance nonfilers during the period following grievance filing and settlement. The postÐdispute resolution work attendance rates of grievance filers were also lower than those of grievance nonfilers, though these differences were either marginally significant or insignificant. However, additional comparisons of postÐdispute resolution employee turnover rates showed such turnover—especially voluntary turnover—to be significantly higher among grievance filers than among grievance nonfilers.

     Going further, Lewin reported the same pattern of findings when comparing samples of supervisors of employee grievance filers with samples of supervisors of employee grievance nonfilers in these non-union organizations. That is, these samples of supervisors did not differ significantly in terms of job performance ratings, promotion rates, and work attendance rates before or during employee grievance filing and settlement, but they differed significantly thereafter. In particular, postÐdispute resolution job performance ratings, promotion rates, and work attendance rates were significantly lower among the supervisors of grievance filers than among the supervisors of grievance nonfilers. Further, and again similarly, comparisons of postÐdispute resolution turnover rates showed such turnover—especially involuntary turnover—to be significantly higher among supervisors of grievance filers than among supervisors of grievance nonfilers.

     These findings, together with those of other researchers, lend themselves to two contrasting answers to the question, "What happens to the disputing parties after their disputes are settled?" One answer is "retaliation," meaning that non-union employees who choose to avail themselves of the dispute resolution procedures established by their employers by actually filing grievances will in effect be punished for doing so, as will their supervisors. A second answer, markedly different from the first, is "revealed performance," meaning that non-union employees who file grievances under their employers' dispute resolution procedures and their supervisors are (or turn out to be) systematically poorer performers than their nonÐgrievance filing counterparts and their supervisors. From a management perspective, the first of these answers runs counter to the ostensible rationale for and objectives of non-union employment dispute resolution systems, whereas the second answer provides support for these systems and for the decisions made according to them.

Managerial Reactions to PostÐDispute
Resolution Outcome Data

        While scholarly debate continues over these contrasting answers to the question, "What happens to the disputing parties after their disputes are settled?" little research attention has been paid to how managers of non-union organizations react to the aforementioned evidence about postÐdispute resolution outcomes or to how such reactions affect changes in non-union organizations' employment dispute resolution systems. Hence, this paper aims to reduce this knowledge gap by addressing the following two questions: "How do managers react to evidence about post-dispute resolution outcomes in their own organizations once that evidence is communicated to them?" and "To what extent do such reactions lead to changes in these organizations' dispute resolution systems?"

     In seeking answers to these questions, a study of four non-union organizations was conducted in which key findings from prior analyses of these organizations' employment dispute resolution procedures were communicated by the researchers to senior-level managers of these organizations. All four organizations are publicly traded business enterprises, each of which has had a non-union dispute resolution procedure in place for at least a decade. In one of these organizations the dispute resolution procedure specifies external arbitration as the final step; in another of these organizations the chief executive officer (CEO) serves as the final step; in yet another a three-member management committee serves as the final step; and in the last a chief administrative officer serves as the final dispute resolution step. Table 1 summarizes these and selected other characteristics of the four organizations that participated in this study.

        A single common method featuring an electronically (e-mail) transmitted report followed by a face-to-face presentation was used to communicate to the management of each organization the findings from prior research on that organization's non-union dispute resolution procedures and on the four organizations' non-union dispute resolution procedures as a whole. Through these communications it became clear to the managements of these four organizations that employee grievance filers and their supervisors had lower postÐdispute resolution job performance ratings, promotion rates, and work attendance rates and higher postÐdispute resolution job turnover rates than employee grievance nonfilers and their supervisors—and also that no such significant differences existed prior to and during grievance filing and settlement. The electronically transmitted report was accompanied by a survey that each manager was asked to complete and return within one week of receiving the report. At the conclusion of each presentation, a survey of the attending managers was conducted and was followed shortly thereafter by an in-depth interview of each manager. These data were used to address the question, "How do managers react to evidence about postÐdispute resolution outcomes in their own organizations once that evidence is communicated to them?" Approximately six months later, a second set of surveys and in-depth interviews were conducted among the same managers of these four organizations. The data obtained were used to address the question, "To what extent do such reactions lead to changes in these organizations' dispute resolution systems?"

        In terms of how managers react to evidence about the postÐdispute resolution outcomes of their own organization's employment dispute resolution procedure, the data in table 2 indicate that they are highly likely to agree with a revealed performance explanation of this evidence, especially when that evidence is communicated to them in a report compared to a presentation format. Hence, an average of between 83 and 92 percent of the managers in each of the four organizations agreed with a revealed performance explanation, while only between 8 and 9 percent agreed with a retaliation explanation, of their own organization's postÐdispute resolution outcome data in responding to a survey administered shortly after these data had been communicated to them via an electronic report. When surveyed some two weeks after a face-to-face presentation of these same data, however, an average of between 55 and 67 percent of the managers in each of the four organizations agreed with a revealed performance explanation, while between 25 and 36 percent agreed with a retaliation explanation, of their own organization's postÐdispute resolution outcome data. The differences between both pairs of responses are statistically significantly.

     Regarding the evidence about postÐdispute resolution outcomes in these four non-union organizations as a whole, the data in table 3 indicate that the managers who participated in this study agree with a revealed performance explanation of this evidence, especially when that evidence is communicated to them in a report compared to a presentation format. This seems no different from the aforementioned reactions of managers to the evidence of their own organization's postÐdispute resolution outcomes. Among these 47 managers, however, only 32 or 68 percent agreed with the revealed performance explanation of these aggregated data, while 14 or 30 percent agreed with a retaliation explanation, based on their responses to a survey administered shortly after these data had been communicated to them via an electronic report. Further, only 21 or 45 percent of these managers agreed with a revealed performance explanation, while 24 or 51 percent agreed with a retaliation explanation, based on their responses to a survey administered two weeks after a face-to-face presentation of the aggregated postÐdispute resolution outcomes data. These differences—between and across surveys—are all statistically significant.

        Certain preliminary conclusions can be drawn from these findings. First, managers of non-union organizations react differently to postÐdispute resolution outcome data when those data pertain to their own organization(s) compared to a larger set of organizations. Specifically, managers are significantly more likely to react to/interpret their own organization's postÐdispute resolution outcome data as reflecting the actual performance of grievance filers (and their supervisors) relative to grievance nonfilers (and their supervisors), rather than retaliation against grievance filers (and their supervisors), then when reacting to/interpreting multiple organizational postÐdispute resolution outcome data. Stated differently, managers are significantly more likely to assign a retaliation explanation to the postÐdispute resolution outcome data of other organizations than to their own organization. Second, managers of non-union organizations are significantly more likely to react to/interpret postÐdispute resolution outcome data as reflecting revealed performance and significantly less likely to react to/interpret such data as reflecting retaliation against grievance filers (and their supervisors) when those data are communicated to them in a relatively distant electronic report format than in a more personal face-to-face presentation format. Analytically, the presentation format can be most fundamentally distinguished from the report format in that the presentation is made to a group of managers, whereas as the report is communicated (electronically) to individual managers. During such a presentation managers can interact with and question the presenter(s) as well as each other, which aids and likely deepens the interpretation of postÐdispute resolution outcome data; no such opportunity attaches to the communication (in electronic or hard copy format) to individual managers of a report on postÐdispute resolution outcomes. Hence, in this study the face-to-face presentation of postÐdispute resolution outcome data appears to be associated with a managerial tendency to react to/interpret the data less from an initial, singular revealed performance perspective and more from a closely considered retaliation perspective.

     This conclusion is reinforced by the interview comments rendered by managers during the course of this study. For example, the operations director of an aerospace company commented as follows:

When I looked at the report on what happens to employees who use our "speak up" procedure, it seemed to confirm my personal view that those employees are whiners and complainers compared to the bulk of our workforce. In fact, I said as much in responding to the electronic survey that accompanied the report. But after listening to the presentation on this research and to some of the questions raised by my peers following that presentation, I was more inclined to think that retaliation against these "complaining" employees probably had occurred.

This comment pertains to the operations director's reaction to the postÐdispute resolution outcomes data for his company alone. In another example, the information technology director of an express delivery company commented as follows on how he was influenced by the postÐdispute resolution outcome data for the four companies as a whole:

For my company, I thought that the [postÐdispute resolution outcome] data were quite consistent with my belief that the employees who tend to use our grievance-like procedure are those who are at the lower end of the performance distribution. In fact, this seemed quite obvious. But after looking at the data for all the companies combined and after attending the presentation on these data, I began to reconsider. I don't think you can simply explain the much lower job performance and promotion rates and the much higher turnover rates for employees who use these types of procedures [compared to those who do not] simply by performance differences; there's got to be some retaliation going on here.

For other managers who participated in this study, the postÐdispute resolution outcome data on supervisors of grievance filers compared to the supervisors of grievance nonfilers were especially compelling. Illustrative of these managers is the director of customer service for a medical supply company, who commented as follows:

When I first read the dispute resolution report on my company, I focused on the data regarding grievance filers—much like I focus on data regarding our customer service associates when I receive customer satisfaction reports. My reaction was that these data more or less made sense in that, in my experience, employees who complain a lot or formalize their complaints tend to be poorer performers than other employees who attend to their work. But when I reviewed the data for my company concerning the supervisors of grievance filers [compared to the supervisors of nonfilers], I was floored! As it happens, I've known most of these supervisors personally and it simply doesn't make sense to say that the supervisors of employees who wind up filing grievances are worse performers than the supervisors of employees who don't file grievances. You've got to remember that our company goes to great pains to tell employees about our internal employment dispute resolution system and how to use it, just as we go to great pains to tell our customers about our customer dispute resolution system and how to use it. If the [dispute resolution] report on my company is accurate, there has to have been retribution against our supervisors whose subordinates wound up filing grievances. And that is really disturbing.

        Still other managers who participated in this study were roughly equally divided—one might say torn—between the revealed performance and the retaliation explanations of the postÐdispute resolution outcome data that were communicated to them. An example of one such manager is the director of human resources for a financial services company, who commented as follows:

I'm very familiar with our internal employee dispute resolution system—hell, I helped design it. We believe in this system, I can tell you that with certainty. So, when I saw the report on the postÐdispute resolution outcomes of employees and their supervisors, both in our company and in other companies, I first thought that the data meant that employee grievance filers are poorer performers than other employees—and that the same is roughly true for their supervisors; this made sense to me. But later, especially after the presentation that was made to us about this report, I began to think that this couldn't be just a performance story; it had to involve something else. What is that something else? As a human resources professional, I know that this something else is retribution or retaliation or whatever other word you want to use. It's very troubling, especially because I still feel that employees who complain a lot and who use our dispute resolution system to pursue those complaints are not as good performers as our other employees.

     All in all, then, the managers who participated in this study reacted to reports about their own and other non-union organizations' postÐdispute resolution outcome data by initially interpreting these data to accord with a revealed performance explanation, but later they interpreted these same data to accord with a retaliation explanation. The revealed performance explanation was more likely to be adopted when the data pertained specifically to respondent managers' own organizations' rather than to other organizations; when the data were communicated electronically through individual reports rather than through face-to-face presentations to groups of managers; and when the data were limited to comparisons of employee grievance filers and nonfilers rather than expanded to include supervisors of grievance filers and nonfilers. Stated differently, as the data from studies of these non-union organizations' postÐdispute resolution outcomes were more deeply and widely communicated to managers of these organizations, the dominant management reaction to these data shifted away from a revealed performance explanation and toward a retaliation explanation.

     Returning to quantitative analysis of managers' reactions to the evidence about postÐdispute resolution outcomes featured in this study, the influence of demographic characteristics on such reactions is summarized in table 4. These regression estimates suggest that manager age and experience are significantly associated with a perceived retaliation explanation of the evidence of postÐdispute resolution outcomes, whereas education is significantly associated with a perceived revealed performance explanation of the evidence about postÐdispute resolution outcomes. Especially notable, managers with human resource, marketing, and information technology functional responsibilities are significantly more likely to favor a retaliation explanation of the evidence about postÐdispute resolution outcomes, whereas managers with finance, accounting, and operations functional responsibilities are significantly more likely to favor a revealed performance explanation. The main reason why this particular quantitative difference is especially notable is that it directly influences how non-union organizations and their managers respond to evidence about postÐdispute resolution outcomes by changing their employment dispute resolution systems and procedures—a matter this is taken up in the next section of this paper.

Changes to Non-union Dispute Resolution Systems

     The research design for this study included a third survey and a second interview of managers in the four participating organizations; these instruments were principally aimed at answering the question, "To what extent do managers' reactions to evidence about postÐdispute resolution outcomes in their own and other organizations lead to changes in these organizations' dispute resolution systems?" Recall that, on the one hand, the initial reactions of these managers to the postÐdispute resolution outcome data that were communicated to them strongly suggest that no changes in their respective non-union organizations' employment dispute resolution systems would be contemplated, let alone forthcoming. On the other hand, the subsequent reactions of these same managers to the outcome data suggest that some changes in their respective organizations' employment dispute resolution systems would indeed be contemplated and perhaps forthcoming. Which of these two contrasting inferences/predictions turned out to be more accurate?

     The answer to this question is that the "change" prediction turned out be the more accurate, though the extent to and ways in which these non-union organizations' employment dispute resolution systems changed varied considerably among the four organizations. In the aerospace company, where the employment dispute resolution system features arbitration as the final step, relatively few cases—about 2 to 3 percent annually—actually culminate in arbitration. But the presence of arbitration has long been thought by company management to signal if not fully ensure a fair dispute resolution system free from retaliation against system users. After reviewing the data on postÐdispute resolution outcomes in this company that were communicated to company management, however, seven of the twelve managers who participated in this study concluded that these data told a retaliation story, four favored a revealed performance story, and one favored a dual retaliation/revealed performance story. Among those who favored the revealed performance story were the chief financial officer (CFO), the treasurer, and the general manager of the company's largest division. Among those who favored the retaliation story were the aforementioned operations director, the chief administrative officer (CAO), the director of human resources, and the general managers of three other company divisions.

     These two groups of managers formed what can best be characterized as political coalitions that wound up meeting on three occasions to discuss and negotiate changes in the company's employment dispute resolution system. Those managers who sought primarily to overcome retaliation against grievance filers and their supervisors initially proposed replacing the middle two steps of the company's four-step employment dispute resolution system and replacing them with a single peer review step. This group of managers also proposed allowing grievance filers to be represented by outside counsel, the excising of grievance settlement decisions from grievants' personnel records, and the formal monitoring of grievants' and grievants' supervisors' postÐgrievance settlement performance evaluations by the director of human resources and the CAO. Those managers who favored a revealed performance interpretation of the company's postÐdispute resolution outcomes data initially sought to preserve the company's existing employment dispute resolution system and opposed all of the proposed changes in the system. This group was especially adamant about retaining the traditional four-step dispute resolution system—believing that it had worked well over a long period—and about not inserting outside legal representation into a system that they believed was effective in settling employment-related disputes because it avoided excessive litigiousness.

     Following its second meeting, these two groups jointly requested a meeting with the researchers to further review and discuss the company's postÐdispute resolution outcome data. In that meeting the issue of why grievance filers' and their supervisors' job performance ratings declined markedly following the settlement of grievances was discussed at length. Even those who favored a revealed performance explanation of the company's postÐdispute resolution outcome data thought that such (supposed) performance problems should have been observed earlier, that is, prior to and during grievance filing. Recognizing this, the revealed performance management group gave additional consideration to the proposals of the anti-retaliation management group. In a third and final meeting, the two groups agreed to make two changes in the company's employment dispute resolution system, namely, replacement of the middle two steps with peer review and formal monitoring of postÐgrievance settlement performance appraisals of grievants and their supervisors.          

     In the express delivery company, where the employment dispute resolution system features the CEO as the final step, this system has been in place since the company's founding and contains specific language intended to assure employees that they will not suffer retaliation for using the system. In fact, it is also company policy to discipline any supervisor or manager who retaliates against an employee for filing a grievance. This helps to explain why the twelve company managers who participated in this study initially strongly favored the revealed performance explanation of the company's postÐdispute resolution outcome data. As with the aerospace company managers, however, upon further deliberation and consideration of the researchers' face-to-face presentation of the outcome data, several of this express delivery company's managers came to favor a retaliation explanation. When surveyed shortly following the presentation, seven of these managers indicated that they supported the retaliation explanation, three indicated that they supported the revealed performance explanation, one indicated that she supported a dual explanation, and one indicated that he could not decide between the two explanations.

     Of the three company managers who supported the revealed performance explanation, one was the CEO and another was the chief operating officer (COO). Among the managers who supported the retaliation explanation were the company president, the human resources director, the information technology director, and the general managers of the company's major business units. In several meetings of these two groups, supporters of the revealed performance explanation argued strongly in favor of retaining the company's dispute resolution system, emphasizing its long standing and record of success. Notably, one of the success criteria—probably the main one—used by these managers was keeping its employees non-union! Those managers who favored the retaliation explanation acknowledged the long standing of the company's employment dispute resolution system but argued that the postÐdispute resolution outcome data indicated that the system not only was far from being an unmitigated success but it may on balance impose a net cost on the company. The cost that these managers had in mind was what they referred to as the "human capital" loss to the company, which they believed was most clearly reflected in grievants' and grievants' supervisors' comparatively high turnover rates following grievance settlement.

     From an internal organizational politics perspective, these managers were in a difficult position in pushing the retaliation story about their company's employment dispute resolution system and in proposing changes in that system. After all, both the company CEO and COO not only favored the alternative revealed performance explanation; they were leading advocates of the company's employment dispute resolution system and continued to portray it, both externally and internally, as the leading such system in the express delivery industry (and beyond). Facing what they properly deemed to be an uphill battle in this regard, the managers who supported the retaliation story took a particularly bold step by seeking a meeting to discuss this matter with the company's former CEO (and chairman of the board), who had founded the company and who had instituted the three-step employment dispute resolution system (in which, recall, the CEO serves as the final step). That meeting was in fact held, with the attending managers laying out their retaliation explanation of the company's postÐdispute resolution outcome data and offering their recommendations for changes in the dispute resolution system.

     Shortly following this meeting, during which the former company CEO was noncommittal about the proposed changes, the retaliation story managers were invited to a meeting with the current CEO, COO, and a third manager who supported the revealed performance story. In that meeting the revealed performance story managers indicated that while they still did not fully "buy" the retaliation story (proffered by its seven manager-advocates), they were nevertheless willing to make some of the changes to the company's employment dispute resolution system advocated by the retaliation story managers. These changes included the privatization of all grievance cases, meaning that the company no longer internally publicizes grievance activity and resultant grievance settlements; the appointment of a standing general manager as the second step of the employment dispute resolution procedure, replacing the general manager rotational system that had been used previously; and the expunging of grievance filing and settlement records from the personnel files of grievants and their supervisors by no later than six months following grievance settlements.

     In the medical supply company, where the employment dispute resolution system features a three-member senior management committee as the final step, the twelve managers who participated in this study were initially about evenly split in terms of their favoring the revealed performance versus the retaliation explanations of their company's postÐdispute resolution outcome data. After these managers more fully considered the data and after they attended the researchers' face-to-face presentation, they gravitated toward the retaliation explanation. On the basis of their postpresentation survey responses, eight of these managers favored the retaliation explanation, three favored the revealed performance explanation, and one favored a dual explanation. Unlike the aerospace, express delivery, and financial service companies included in this study, the medical supply company's employment dispute resolution system includes peer review (as the second step), which this company's managers believe is the key component of this system largely because it is patterned after the type of peer review found in medical schools, hospitals, and other medical service enterprises. Given that they consider peer review to be an especially fair and meritorious process, most of these managers were both surprised and concerned by their company's postÐdispute resolution outcome data.

        In deciding what changes, if any, to make to this company's employment dispute resolution system, these twelve managers held an initial meeting in which they freely debated the retaliation and revealed performance explanations of the postÐdispute resolution outcome data. Those managers who favored the revealed performance explanation nevertheless questioned why such employee and supervisor performance deficiencies had not been "revealed" prior to or during grievance filing and settlement, especially in light of the company's elaborate 360Ðdegree performance evaluation system. Those managers who favored the retaliation explanation raised the same question but, in addition, paid considerable attention to the postÐdispute resolution outcome data in the four companies as a whole, which they interpreted to mean that retaliation for grievance filing did not necessarily stem from their company's particular dispute resolution system. Consequently, and uniquely among the four organizations included in this study, the managers of this medical supply company decided to survey company employees about their views concerning the company's employment dispute resolution system and their recommendations for changes in this system. From an internal organizational political perspective, this initiative represents the coalescence of an otherwise divided leadership group around the value of obtaining potentially useful information from a key constituency—namely, the employees of this organization.

     Combining the data from this employee survey with their own views of the company's employment dispute resolution system, these managers decided to change this system by requiring that every first-step grievance filed by a company employee be reviewed by a supervisor or manager other than the employee's immediate supervisor/manager; requiring the company's director of human resources to review a grievant's and the grievant's supervisor's/manager's record of job performance immediately upon the filing of a grievance; and requiring members of the third-step senior management committee to review the job performance of the grievant and the grievant's supervisor before rendering a decision about the grievance. Notably, these changes bring this medical supply company's internal employment dispute resolution system into closer accord with the company's system for resolving external customer disputes—a system that features the analysis of customer purchasing performance.

     In the financial services company, where the employment dispute resolution system features the CAO as the final step, the reactions of managers to the company's postÐdispute resolution outcome data strongly favored the revealed performance story, both initially and later on. Specifically, of the eleven financial service company managers who participated in this study, nine favored the revealed performance explanation and two favored the retaliation explanation. In this respect, these managers differed markedly from the managers of the aerospace, express delivery, and medical supply companies. Hence, one might conclude that there was little debate or "politicking" within this company about its postÐdispute resolution outcome data and that no changes were subsequently made to its employment dispute resolution system; but these conclusions would be wrong.

     Of the two managers in this company who favored the retaliation explanation, one was the chief managing director and the other was the director of human resources. Not only had these two managers worked together for a long time, including an average of fourteen years each with this company, they were instrumental in designing the company's employment dispute resolution system as well as its performance management system. Further, when they were instituted both of these systems were claimed to be leading-edge initiatives within the financial services industry. By contrast, of the nine managers who favored the revealed performance explanation, four had been employed by the company for less than five years, three had been employed for less than three years, and two had been employed for less than two years. Following the researchers' presentation of this company's postÐdispute resolution outcome data to these managers, two meetings were scheduled to discuss the data and its implications for changes to the company's employment dispute resolution system.

     During the first of those meetings, supporters of the revealed performance story strongly pressed their views and argued that the company's employment dispute resolution system was working well, in large part because it weeded out relatively poor performers, and should therefore remain unchanged. The two proponents of the retaliation story did not press their contrasting view at this meeting. However, shortly after the meeting, these two proponents (chief managing director and the director of human resources) undertook a "political" initiative by contacting one former company employee and one former company manager who had both been involved in a grievance over client assignments while working at the company and asking them if they would be willing to attend the next meeting of the eleven company managers. These two individuals agreed to do so, and in that subsequent meeting they described in considerable detail how and why their involvement in a grievance proceeding while working at this particular financial services company resulted in "black marks" against them that were sufficiently strong to both block their future promotion within the company and motivate them to pursue employment with other financial services companies.

     Obviously impressed by the specific retaliation stories told by this company's two former employees, the managers of this financial services company met a third time to discuss the postÐdispute resolution outcome data and potential changes in the company's employment dispute resolution system. In this meeting, moreover, the postÐdispute resolution outcome data for the other companies that participated in this study were given considerable attention and were used to place this financial services company's data in a larger context. Consequently, this meeting concluded with the participating managers adopting changes to the company's employment dispute resolution system, in particular, requiring all first-step grievances to be reported to the director of human resources, requiring grievants' and grievants' supervisors' performance evaluation data to be provided to and reviewed by the chief managing director or his assignee prior to the processing of any grievance, and requiring that grievance filing and settlement records be expunged from the personnel files of all grievants (and their supervisors) within one year of grievance settlement.

Summary, Conclusions, and Caveats

     Considerable research evidence has emerged showing that employee users of non-union organizations' employment dispute resolution systems experience lower job performance ratings, promotion rates, and work attendance rates and higher turnover rates following grievance settlement than counterpart employees who do not use these systems. These findings apply as well to the supervisors of grievance filers compared to supervisors of grievance nonfilers in non-union organizations. Two main explanations of this evidence have been proffered and debated by scholars and employment dispute resolution specialists, namely, a retaliation explanation and a revealed performance explanation. Absent from this debate, however, have been the views of managers of non-union organizations that maintain employment dispute resolution systems. Consequently, this study sought answers to two questions about non-union organizations' employment dispute resolution systems: (1) How do managers react to evidence about postÐdispute resolution outcomes in their own and other organizations once that evidence has been communicated to them? and (2) To what extent do such management reactions lead to changes in these organizations' employment dispute resolution systems?

     The research design for this study featured an initial electronic communication of postÐdispute resolution outcome data to high-level managers in each of four non-union organizations, followed soon thereafter by a face-to-face presentation and discussion of the same data with the participating managers in each organization; surveys of the participating managers conducted in conjunction with both the electronic communication and face-to-face presentations; an in-depth interview of each manager conducted shortly after completion of the face-to-face presentations; and an additional survey and interview of each manager conducted six months after completion of the face-to-face presentations.

     Regarding the first focal question, quantitative analysis of the survey data indicated that these managers initially reacted to the electronically communicated evidence about postÐdispute resolution outcomes in their own and other organizations by favoring a revealed performance explanation. According to this explanation, employee grievance filers and their supervisors are or turn out to be systematically poorer performers than grievance nonfilers and their supervisors. Following face-to-face presentation and discussion of this same evidence, however, these managers shifted considerably in favor of a retaliation explanation. According to this explanation, employees who actually use their non-union organizations' employment dispute resolution systems suffer retaliation in the forms of lower job performance ratings and promotion rates and higher turnover rates than grievance nonfilers in the period following grievance settlement. For some of these managers in particular, their shift away from a revealed performance explanation toward a retaliation explanation was most strongly influenced by the evidence about postÐdispute resolution outcomes for supervisors of grievance filers compared to the supervisors of grievance nonfilers. For some other managers, this shift was most strongly influenced by consideration of the postÐdispute resolution outcome evidence from non-union organizations other than (or in addition to) their own.

     Regarding the second focal question, qualitative analysis of the interview data indicated that each of the four organizations eventually changed their respective employment dispute resolution systems in one or more respects, largely because a substantial proportion of managers in each organization came to favor the aforementioned retaliation explanation. From an analytical perspective, however, the process by which such changes were determined was fundamentally a political process in which those managers who favored the retaliation story constituted one coalition, those managers who favored the revealed performance story constituted another coalition, and the two coalitions met and negotiated with each other on several occasions to thrash out whether and to what extent dispute resolution system changes would be made. This political process occurred in each of the four participating organizations, and the membership of each coalition in each organization represented a horizontal mix of management functions and a vertical mix of management levels. In other words, managers formed alliances with other managers irrespective of intraorganizational function and level based on their dominant reaction to the evidence of postÐdispute resolution outcomes in their own and other organizations, and then converted these alliances into clearly demarcated coalitions for the purpose of negotiating changes in their respective employment dispute resolution systems. This particular experience comports closely with other well-known examples of political behavior among managers in business enterprises.

     It is also notable that revisions to these non-union employment dispute resolution systems occurred irrespective of important differences in the characteristics of these systems. As examples, the aerospace company's employment dispute resolution system has always featured external arbitration as the final step, which company management believed ensured fairness and a lack of retaliation against system users; the express delivery company's employment dispute resolution system, which was instituted at the company's founding, featured the CEO as the final step and explicit antiretaliation policy statements that company management believed insured system fairness; the medical supply company's employment dispute resolution system featured peer review as an intermediate step and a three-member senior management committee as the final step, both of which were regarded by company management as ensuring fairness and nonretaliation against system users; and the financial services company's employment dispute resolution system featured the CAO as the final step and provisions calling for especially swift settlement of employee grievances, which company management believed fit especially well with a fast-paced transactions-oriented business and industry. Nevertheless, high-level managers in each of these non-union organizations behaved very similarly in reacting to evidence about postÐemployment dispute resolution outcomes by forming internal political coalitions of like-minded managers and by subsequently negotiating dispute resolution system changes with political coalitions of differentially minded managers. Whether and to what extent these changes in the employment dispute resolution systems of the non-union companies included in this study will result in postÐdispute resolution outcomes different from those that have dominated to date remains to be seen, but this is obviously an important issue for the next stage of research on non-union dispute resolution.

Works Consulted

Bingham, L. B. 2007. "Evaluating Disputer Resolution Programs: Traps for the Unwary." Paper presented to the 59th Annual Meeting of the Labor and Employment Relations Association, Chicago, IL, January.

Ichniowski, C., and Lewin, D. 1987. "Grievance Procedures and Firm Performance." In M. M. Kleiner, R. N. Block R. Roomkin, and S. W. Salsburg, eds., Human Resources and the Performance of the Firm. Washington, DC: Bureau of National Affairs, pp. 159Ð193.

Kaminski, M. 1999. "New Forms of Work Organization and Their Impact on the Grievance Procedure." In A. E. Eaton and J. H. Keefe, eds., Employment Dispute Resolution and Worker Rights in the Changing Workplace. Champaign, IL: Industrial Relations Research Association, pp. 219Ð246.

Klaas, B. 1989. "Determinants of Grievance Activity and the Grievance System's Impact on Employee Behavior: An Integrative Perspective." Academy of Management Review, Vol. 14, no 3, pp.445Ð458.

Lewin, D. 1987. "Dispute Resolution in the Non-union Firm: A Theoretical and Empirical Analysis." Journal of Conflict Resolution, Vol. 31, no. 3, pp. 465Ð502.

Lewin, D. 1992. "Grievance Procedures in Non-union Workplaces: An Empirical Analysis of Usage, Dynamics, and Outcomes." Chicago-Kent Law Review, Vol. 66, no. 3, pp. 823Ð844.

Lewin, D. 1999. "Theoretical and Empirical Research on the Grievance Procedure and Arbitration: A Critical Review." In A. E. Eaton and J. H. Keefe, eds., Employment Dispute Resolution and Worker Rights in the Changing Workplace. Champaign, IL: Industrial Relations Research Association, pp. 137Ð186.

Lewin, D. 2001. "IR and HR Perspectives on Workplace Conflict: What Can Each Learn from the Other?" Human Resource Management Review, Vol. 11 (Fall), pp. 453Ð485.

Lewin, D. 2004. "Dispute Resolution in Non-union Organizations: Key Empirical Findings." In S. Estreicher and D. Sherwyn, eds., Alternative Dispute Resolution in the Employment Arena. New York: Kluwer, pp. 397Ð403.

Lewin, D. 2005. "Unionism and Employment Conflict Resolution: Rethinking Collective Voice and Its Consequences." Journal of Labor Research, Vol. 26, no. 2, pp. 209Ð239.

Lewin, D. Forthcoming. "Resolving Conflict." In N. Bacon, P. Blayton, J. Fiorito, and E. Herry, eds., Handbook of Industrial and Employment Relations. London: Sage.

Lipsky, D. B., R. L. Seeber, and R. D. Fincher. 2003. Emerging Systems for Managing Workplace Conflict. San Francisco, CA: Jossey-Bass.

Olson-Buchanan, J. 1997. "To Grieve or Not to Grieve: Factors Relating to Voicing Discontent in an Organizational Simulation." International Journal of Conflict Management, Vol. 8, no. 2, pp. 132Ð147.

Olson-Buchanan, J. 1996. "Voicing Discontent: What Happens to the Grievance Filer After the Grievance?" Journal of Applied Psychology, Vol. 81, no. 1, 52Ð63.

Seeber, R. L., and D. B. Lipsky. 2006. "The Ascendancy of Employment Arbitrators in US Employment Relations: A New Actor in the American Scene?" British Journal of Industrial Relations, December, in press.