Employee relations

IR and HR perspectives on workplace conflict What can each learn from the other?

David Lewin*

Anderson Graduate School of Management, University of California at Los Angeles, Los Angeles, CA 90095-1481, USA

Abstract

This paper presents an analysis of industrial relations (IR) and human resource (HR) perspectives on conflict in the employment relationship. While both perspectives recognize the existence of employment relationship conflict, IR’s premises about such conflict are that it stems from an employer–employee power imbalance, is enduring, often requires institutional interventions in the forms of unionism and legislation to correct the power imbalance, and can be constructive even when the conflict is dealt with in adversarial, nonproblem-solving fashion. HR’s premises about employment relationship conflict, by contrast, are that such conflict stems from poor management, can be partially reduced by organizational and workplace innovations that build an employer–employee unity of interests, can be still further reduced through cooperative, mutual gains-oriented problem-solving techniques, and, as a consequence of improved management, will fade from the employment scene. These premises are then examined in relation to case and empirical evidence on grievance procedures, employment discrimination, and employee involvement/participation programs. The analysis finds some support for, but also important limitations on, each of these key perspectives. Recommendations are offered to enable members of the IR and HR communities to incorporate into their respective domains key lessons from each other’s dominant perspective on employment relationship conflict. D 2001 Elsevier Science Inc. All rights reserved.

1053-4822/01/$ – see front matter D 2001 Elsevier Science Inc. All rights reserved. PII: S1053-4822(01)00049-3

* Tel.: +1-310-206-7666; fax: +1-310-476-0324. E-mail address: [email protected] (D. Lewin).

www.HRmanagementreview.com

Human Resource Management Review 11 (2001) 453–485

1. Introduction

Both industrial relations (IR) and human resources (HR) focus on conflict in the employment relationship. However, IR specialists emphasize that conflicts of interest in the employment relationship are relatively widespread and enduring — that in such conflicts, employers hold a power advantage over employees, which in turn implies that there will be considerable and ongoing conflict in the employment relationship; that institutional inter- ventions, through unions or government, are required to correct the power imbalance; and that the open surfacing of conflict can have positive, not just negative, consequences for the parties to the employment relationship.

By contrast, HR specialists focus more on organizational effectiveness and emphasize that although employment relationship conflict is indeed widespread, such conflict stems largely from poor management and thus can be substantially reduced, if not eliminated, through organizational and procedural innovations and other management improvements; that a unity of interests among employers and employees can be achieved; that remaining conflicts can best be dealt with through mutual problem-solving techniques that promote cooperation and integration rather than adversarialism and fractionalization; and that power imbalances will fade from the employment scene as employers and employees come to have similar goals and objectives.

As the 20th century drew to a close, it appeared to many that relatively more cooperation-oriented HR had supplanted relatively more conflict-oriented IR as the dominant perspective on the employment relationship and as the dominant frame for research and practice. In this regard, the decline of private sector unionism in the US and elsewhere is a key datum. Nevertheless, it will be argued here that HR can still learn much from IR in analyzing and dealing with conflict in the employment relationship. By the same token, IR can learn much from HR about the management and resolution of employment relationship conflict.

To develop these arguments, Section 2 of this paper analyzes leading IR and HR premises about the employment relationship. Sections 3 and 4 examine in greater detail the workplace practices and policy implications emanating, respectively, from IR and HR perspectives on conflict in the employment relationship. Then, in Section 5, empirical evidence for and against these two perspectives on conflict in the employment relationship is presented and discussed. Section 6 summarizes key issues drawn from this inquiry and offers recommendations to members of the IR and HR communities concerning how they may incorporate these lessons into their respective perspectives on employment relation- ship conflict.

2. IR and HR premises about the employment relationship

The employment relationship can be said to be mixed motive in nature in that there are incentives for both cooperation and conflict between employer and employee. The incentive for cooperation is grounded in the combined contribution of capital and labor that is required

D. Lewin / Human Resource Management Review 11 (2001) 453–485454

for the production of goods and services to achieve economic progress, while the incentive for conflict is grounded in differences between labor and management over their respective roles and responsibilities in the production of goods and services and in the distribution of economic returns derived therefrom. From an IR perspective, the employer has long been considered to have a power advantage over the individual employee. This power imbalance stems from the considerable assets that the firm is able to amass relative to the individual, thereby giving a labor market advantage to the employer, and from the internal governance structure of the firm in which authority and responsibility for decision making and conflict resolution are vested in owners and managers, thereby placing the employer in a position of superiority over the employee (Kaufman & Lewin, 1998).

In the most extreme form of this unbalanced power relationship, a master–servant philosophy prevailed and there was a complete absence of both worker participation in the determination of terms and conditions of employment and due process in the protection of employee rights. While recognizing that the power imbalance in the employment relation- ship can vary considerably by workplace, industry, employee skill level, and economic conditions, IR specialists have nevertheless largely regarded conflict in the employment relationship as inevitable and enduring (see, e.g., Kornhauser, Dubin, & Ross, 1954). This, in turn, has led IR specialists to support institutional mechanisms, such as unions and legislation, to (attempt to) redress the power imbalance in the employment relationship (i.e., ‘‘level the playing field’’) and thereby bring employment conflict to the surface where it can be dealt with openly and fairly.

HR specialists also recognize the potential for and presence of conflict in the employment relationship but, in contrast to IR specialists, have often focused on the management of such conflict so as to achieve a unity of interest between employer and employee. The creation of personnel departments in large firms early in the 20th century can be said to have reflected this view, as did the later adoption of employee counseling programs and employee relations training programs for supervisors, the still later adoption of flexible work schedules and work–family life balance programs, and recent work redesign initiatives that feature the use of teams and increased employee participation in workplace decision making (Lewin & Mitchell, 1995). Indeed, the conversion of personnel departments into HR departments late in the 20th century (as well as the virtual disappearance of labor relations departments) underscores the view that a unity of interests between employer and employee can potentially be achieved and, thus, that win–lose type conflict in the employment relation- ship can be contained, minimized or perhaps even eliminated (see, e.g., Ulrich, Losey, & Lake, 1997).

Consistent with this HR orientation and markedly different from the IR orientation toward conflict in the employment relationship, third parties, such as unions, arbitrators, and government agencies, representing the ‘‘weaker’’ side in an adversarial struggle over terms and conditions of employment, are not especially salient and, for the most part, not required. Instead, employment relationship conflict can be managed through the use of various workplace and organizational innovations undertaken by professionally trained managers with the assistance of HR staff. This line of reasoning also helps us to understand why IR specialists have had a relatively strong external or macro-orientation toward conflict in the

D. Lewin / Human Resource Management Review 11 (2001) 453–485 455

employment relationship, in contrast to the relatively strong internal or micro-orientation toward employment relationship conflict that has prevailed among HR specialists.1

3. The IR perspective on conflict

Given the aforementioned premises about conflict in the employment relationship, it is not surprising that IR scholars and practitioners have tended to focus on certain overt manifes- tations of employment relationship conflict, such as union-organizing efforts, strikes, grievances, and turnover. Consider, for example, the premise that conflict in the employment relationship is widespread and enduring. In this regard, Bendix (1956) traced the evolution of management ideology and showed that, during the first half of the 20th century, the dominant ideology changed from social Darwinism to scientific management to human relations. Though each of these management ideologies differed substantially from the others — social Darwinism emphasized survival of the fittest in the labor market; scientific management emphasized the application of industrial engineering principles to the design of work; human relations emphasized the importance to management of understanding workers’ psychological and social needs — management practices were roughly similar under each of these ideologies in that management gave orders and workers were expected to obey.

Often, however, and as Bendix (1956) also showed, workers regarded management’s orders as unfair, did not want to obey, and sought to form unions to redress their grievances. While management opposition to unionism was expected under the ideologies of social Darwinism and scientific management, it was far less expected under the ideology of human relations, which seemed to share with workers (and unions) concern for worker well being, safer working conditions, and noneconomic aspects of work life. Yet, even the most human- relations-imbued employers and managers opposed unionism with a vengeance and went to great lengths to suppress union-organizing activity. More recently, Jacoby (1997) showed how such prominent firms as Eastman Kodak, Sears Roebuck, and Thompson Products (now TRW) spent the better part of the 20th century opposing employee unionism and variously used generous benefit plans, attitude surveys, and company unions to do so.

Despite employer opposition, however, unionism grew rapidly in the US between the mid- 1930s and mid-1950s, sparked by the Great Depression, high labor demand during World

1 That these different perspectives on conflict in the employment relationship are of long standing is perhaps best illustrated by the famous Hawthorne experiments and subsequent re-analysis of the data from those experiments. Researchers who conducted the original set of experiments attributed large productivity increases primarily to the strong bonds, or we-feeling, that developed among workers who were part of the ‘‘treatment’’ group (Roethlisberger & Dickson, 1939). Such bonding, it was said, satisfied workers’ social needs, and this interpretation became an important element in Maslow’s (1943) development and McGregors’s (1960) subsequent refinement of the concept of a need hierarchy. Some four decades later, Franke and Kaul (1978) concluded from their re-analysis of the Hawthorne data that two factors, namely, the economic environment, meaning the Great Depression, and managerial discipline, as reflected in the Hawthorne plant management’s replacement of two of the six original experimental team members (allegedly for violations of work rules), accounted for more than 90% of the variance in hourly, daily, and weekly production during the course of the experiments.

D. Lewin / Human Resource Management Review 11 (2001) 453–485456

War II, and favorable labor legislation. Strikes (including national emergency strikes) became commonplace events following World War II, labor relations were highly adversarial, and union leaders and employers further manifested their opposition to each other in political activity. In short, overt conflict was at the center of union–management relationships during this period.

With the decline in private sector unionism, which began in the mid-1950s and continues today, conflict in the employment relationship seemed to some to be on the wane. But it would be a mistake to regard union membership, per se, as a measure of conflict in the employment relationship. Consider, for example, the large amount of worker protective legislation that has been enacted in the last four decades or so, ranging from the 1964 Civil Rights Act to the 1970 Employee Retirement Income Security Act (known as the Pension Reform Law) and Occupational Safety and Health Act (OSHA) to the 1986 Immigration Reform and Control Act to the 1990 Americans With Disabilities Act to the 1993 Family and Medical Leave Act. These and other federal statutes, to say nothing of similar state and local government statutes, are aimed at protecting one or another term, condition, or right of employment for privately employed workers — and at its core, such protection is from the employer who cannot be relied upon voluntarily to ensure such terms, conditions, and rights. The existence of this legislation supports the view that conflict in the employment relationship is widespread and enduring. So, too, does the growing volume of employment discrimination cases filed under the Civil Rights Act and other anti-employment discrimina- tion legislation.

At the micro or organizational level, nonunion firms have increasingly adopted grievance and grievance-like procedures to the point where perhaps a majority of publicly traded nonunion US firms have one or another such procedure in place (Lewin, 1997). While, in some instances, these procedures are adopted to ward off employee unionization, the more dominant rationale for adopting such procedures is that employers want to detect and correct unfair workplace treatment — which, if undetected or uncorrected, may result in legal action under one of the aforementioned statutes (Feuille & Delaney, 1992). Surprisingly, perhaps about one quarter of these nonunion grievance procedures culminates in binding third party arbitration (Lewin, 1997). In any case, the growth of these conflict resolution mechanisms also supports the premise, drawn from the IR perspective, that employment relationship conflict is widespread and enduring.

Another key premise of the IR perspective on conflict is that the employer has a power advantage over the individual employee. Support for this premise comes in part from the continued presence, albeit a markedly smaller presence, of employee unions in some sectors of the private economy and from the much more substantial presence of employee unions in the public sector; from persistent attempts of certain worker groups, ranging from janitors to medical doctors, to organize and bargain collectively with employers; and from the aforementioned growth of worker protective legislation and of employee lawsuits charging employment discrimination and wrongful termination. But it is also and further supported by the growing efforts of employees to develop skills and competencies that are transferable from firm to firm, that is, to develop general rather than firm-specific human capital, which is in turn based on an expectation that a lifetime career with a particular

D. Lewin / Human Resource Management Review 11 (2001) 453–485 457

company is unlikely to materialize or continue. Ironically, it is employers, especially in large firms, that have contributed strongly to this changed employee expectation through their decisions of recent years to downsize or ‘‘rightsize’’ their workforces, employ more workers on short-term contracts, and outsource certain work that was previously performed by firms’ own employees.

Worker concern about the power imbalance in the employment relationship is not, of course, just a recent phenomenon. Writing about this imbalance early in the 20th century and focusing their attention on different types of unionism, Commons (1934), Hoxie (1921), and Perlman (1928) settled on ‘‘job-conscious’’ unionism as the type that best fit the US economy and society. This type of unionism emphasizes the worker’s property in the job and the use of collective power to get employers to recognize such property. Later, and on a broader scale, Bakke, Kerr, and Anrod (1967) and Dunlop (1957) developed IR systems frameworks that featured union, management, and government representatives interacting to produce a ‘‘web of rules’’ to guide labor–management relationships. Such a web of rules, it was believed, would legitimate workers’ claims to property in the job and feature open and equitable dealing with employer–employee conflict, thereby reducing the power imbalance in the employment relationship and in the internal governance of the firm. That this belief proved more or less valid for only a relatively short period, during which unionism flourished, nevertheless also supports the IR premise that conflict in the employment relationship stems in part from the power imbalance in that relationship.

Contemporary research on worker re-instatement and union and nonunion grievance procedures is also relevant to the IR premise of an imbalance in the employment relationship. For example, US and Canadian studies show that workers who are re-instated to their jobs following dismissal have significantly poorer job performance, higher absenteeism rates, and higher turnover rates than other workers, and these consequences stem in part from employer failure to re-integrate dismissed workers into the workplace. In studies of union and nonunion grievance procedures, scholars have found that grievants have significantly lower post- grievance settlement job performance, promotion, and work attendance rates and significantly higher turnover rates than nongrievants, whereas no significant differences between these two employee groups on any of the four work-related measures were found in the periods prior to and during grievance filing (Lewin, 1987, 1992; Lewin & Peterson, 1999). Other studies have found that significant portions of unionized employees who believe that they have suffered unfair workplace treatment nevertheless do not file grievances — that is, they remain silent — and, further, that fear of reprisal is the most widely cited reason by nonunion employees for their failure to file grievances despite the fact that grievance procedures are available to them (Boroff & Lewin, 1997; Lewin & Boroff, 1996). These findings about formal mechanisms that have been established to deal with workplace conflict appear to provide additional, contemporary support for the proposition that there is a fundamental power imbalance in the employment relationship.

A third IR premise about the employment relationship, one predicated on the first two premises noted above, is that institutional interventions in the form of unions and government are required to deal with employer–employee conflict. Such scholars as Commons (1934), Hoxie (1921), and Perlman (1928), among others, strongly supported unionism and bargain-

D. Lewin / Human Resource Management Review 11 (2001) 453–485458

ing rights for private sector workers and championed the Norris–LaGuardia and Wagner Acts. Similarly, Aaron (1973), Taylor (1967) and other scholars supported unionism and bargaining rights for public employees and advocated Wagner Act-type state government legislation to cover public employers and employees. More recently, the Commission on the Future of Worker Management Relations (1995) and Kochan (1995) articulated the rationale for new federal legislation that would allow the use of voluntary employee teams in unionized settings where such teams may otherwise run afoul of Taft–Hartley Act unfair labor practice prohibitions. More broadly, IR scholars and practitioners have strongly supported worker protective legislation involving anti-discrimination in employment, pensions, workplace safety and health, immigration, family leave, etc.

How effective have these institutional interventions been in correcting the employer– employee power imbalance and in mitigating conflict? To the extent that workers do have and exercise their rights to form and join unions and bargain collectively with employers, the Wagner Act, as amended by the Taft–Hartley Act, can be said to be effective. The same can be said of state-level Wagner/Taft–Hartley-type legislation for public sector employees. Moreover, the collective bargaining agreements negotiated by unions and employers under these laws, and the grievance procedures contained in most of these agreements which permit conflict to be surfaced and dealt with openly (if not always equitably), must be counted as major pluses of these institutional interventions. Indeed, the arbitration of grievances arising under collective bargaining agreements may be the single most positive contribution of institutional intervention to conflict resolution in the employment relationship. Similarly, the fact that a very large portion of the workforce is now protected from discrimination in employment and from employer abuses of pension, workplace safety, family leave, and other terms, conditions and rights of employment constitute another, far larger, achievement of institutional intervention into the employment relationship.

But institutional intervention into the employment relationship only goes so far and has negative consequences as well. For example, the Wagner/Taft–Hartley Acts are predicated on an adversarial model of the employment relationship, explicitly allowing the parties to engage in economic warfare through strikes, lockouts, and other power tactics (though some of these tactics, such as secondary boycotts, were later outlawed), and specifying sets of management and union unfair labor practices. Further, both the National Labor Relations Board (NLRB) and the courts have attempted sharply to separate union interests from management rights, which can be said to undermine the concept of joint or mutual interests. Thus, the NLRB, backed by the courts, recently ruled that certain employer-initiated voluntary employee participation plans violate the law in that they unfairly compete with labor organizations and reflect employer domination or interference with labor organizations (Estreicher, 2000). Further, the union election campaign procedures of the law, as administered by the NLRB, have long been subject to manipulation by management. In particular, Weiler (1990) has shown that employer tactics to delay union representation elections result in significantly lower union win rates in such elections. And, Weiler (1990) has also shown that most unfair labor practice charges filed under the law involve allegations of the basic right, presumably guaranteed under the law, of employees to form or join labor organizations without employer interference, threat, reprisal, or domination. Therefore, and despite its positive attributes, the

D. Lewin / Human Resource Management Review 11 (2001) 453–485 459

legislation governing union–management relations in the US may be said also to have contributed to conflict in the employment relationship and even to the power imbalance in the employment relationship.

With respect to other worker protective legislation, several studies have found that enforcement under these laws is spotty, that the probability of an employer being found in violation of any one of these laws is quite small, and that the penalties imposed on employers for violating the laws are modest at best (Weil, 1997). Perhaps more important, none of these laws, save the vesting provisions of the Employee Retirement Income Security Act, establishes an affirmative employee claim to a term, condition, or right of employment — that is, they do not establish property in the job. Rather, aggrieved employees must take action or, more accurately, decide whether or not to take action only after an employer has, for example, downsized the work force or terminated particular employees or engaged in other actions that employees believe represent unfair workplace treatment. In this sense, even the wide body of worker protective legislation beyond the Wagner/Taft–Hartley Acts appears to underscore as much or more as it corrects the fundamental power imbalance in the employment relationship.

A fourth premise of the IR perspective on the employment relationship is that workplace conflict can be healthy in that, occasionally, a little blood has to be spilled in order to obtain ‘‘voluntary’’ employer–employee agreement. In this regard, studies of strikes have occupied a prominent place in the IR literature. Early on, Hicks (1932) posited that strikes were mistakes that would never occur if the parties had adequate knowledge of settlement possibilities. Chamberlain (1951) showed, however, that the strike was one among a set of power tactics, exercisable by both workers and employers, that could influence each party’s cost of agreement and cost of disagreement with the other party. Going further, Kerr and Siegel (1954) concluded that strikes stemmed from the geographical and social isolation of workers in certain industries (e.g., mining, maritime, and construction) from the larger community, and Karsh (1956) used the case study method to emphasize the importance of employees’ perceived inequitable treatment by management in deciding whether or not to strike or continue to strike. Ashenfelter and Johnson (1969) found that strike activity was procyclical, rising with economic expansions and falling with economic contractions, but they also developed an asymmetric information bargaining model that emphasized the conflicting objectives of union leaders and union members as a key determinant of strikes. Later, and focusing on the industry rather than the economy-wide level of analysis, Kaufman (1981, 1983) showed how strikes result from differences between unions and employers in their uses of wage and price information to form expectations of bargaining settlements.

The evolution of this research on strikes has been such as to call attention to underlying conditions that bring about strikes which, in turn, implies that a problem-solving approach on the part of labor and management can reduce the incidence of strikes. A similar, if more positivist, argument in this regard was posed by Dubin (1954), who contended that strikes are merely one form of industrial conflict and that such conflict can serve certain useful functions, including bringing problems to the attention of management and union officials, providing information for problem diagnosis, allowing workers to let off steam under controlled circumstances, and potentially creating the conditions not only for settlement

D. Lewin / Human Resource Management Review 11 (2001) 453–485460

but for altered, improved relationships between the parties. In this regard, a recent study of workplace violence suggests that some portions of this violence occur because workers lack alternative mechanisms for dealing with unfair workplace treatment or work-related stress (Denenberg & Braverman, 1999).

On balance, it may be said that strikes (and other manifestations of workplace conflict) are not entirely or perhaps even largely negative in terms of their consequences for management or workers. While some strikes are political in nature, reflecting, for example, competition for election to union offices, the bulk of them reflect certain factors or conditions that have led workers to the extreme action of withholding their labor. Such withholding should serve as a signal to management and union officials that workplace problems need to be identified, diagnosed, and corrected, and this means more than merely ‘‘settling’’ the strike. The IR premise that strikes can serve positive (as well as negative) ends thus seems well founded, and is surely consistent with the view that conflict is an enduring feature of the employment relationship.

4. The HR perspective on conflict

Like IR scholars and practitioners, HR scholars and practitioners have focused attention on certain overt manifestations of employment relationship conflict, but these are relatively more microlevel manifestations, including low productivity and job satisfaction and high absent- eeism and turnover. In addition, HR specialists have drawn considerable attention to covert employment relationship conflict through their studies of informal work groups (among employees and management personnel), organization culture, and decision-making networks.

As noted earlier, a fundamental HR premise about employment relationship conflict is that such conflict stems in large part from poor management and can thus be overcome by improved management. A classic example in this regard is the scientific management movement of the early 20th century in which principles of industrial engineering were applied to the re-design of work in order to improve productivity. Taylor (1911) and his disciples (notably Gilbreth, 1911) believed that management–worker conflict was caused by poorly designed jobs and differences over the division of economic returns from a relatively low level of productive output. Taylor claimed that by applying industrial engineering principles to the re-design of work (including managerial work), productivity would increase dramatically and lead to much larger economic returns to management (capital) and labor (though not to a different division of the shares of those returns). And, in fact, this proved to be the case. But it was also the case that workers increasingly chaffed under a work system that required them unswervingly to follow the orders of supervisors (foremen), and that enabled management to tighten (or threaten to tighten) standard work times at its own behest. Moreover, this work system treated workers as unidimensional, that is, only as ‘‘economic men.’’

To overcome this narrow conception of the worker and to deal with growing worker dissatisfaction with scientific management, leaders of the human relations movement turned employers’ attention to workers’ social needs and to the importance of the informal work

D. Lewin / Human Resource Management Review 11 (2001) 453–485 461

group (Mayo, 1933; Roethlisberger & Dickson, 1939). That such work groups could have major influences on productivity and product quality was the key insight of human relations theorists.2 Accordingly, and following the prescriptions of these theorists, employers adopted a variety of employee assistance and psychological counseling programs to meet workers’ noneconomic needs, and also instituted new incentive and reward programs to motivate industrial work groups (Bendix, 1956). Personnel departments bore major responsibility for designing and assisting employers to implement these new human-relations-based programs (Jacoby, 1985). More fundamentally, these initiatives reflected the belief that employer– employee conflict could be reduced (if not eliminated) through improved management.

The idea that worker needs should be systematically assessed in order to reduce conflict, improve management, and shape HR practices and programs took firm hold among employers in the second half of the 20th century. In 1955, IBM instituted an employee opinion survey that was administered annually to rotating samples of the company’s employees in the US and abroad and which continues today. Other employers gradually followed suit and, by the 1990s, more than one of every two publicly traded US companies were conducting a regular employee opinion survey. In such companies, senior management typically required middle and lower management to ‘‘feed back’’ key findings from the surveys to employees, respond to employee queries about intended management actions based on survey findings, and provide senior management with reports about actions they had taken in conjunction with their own annual performance appraisals. Moreover, some firms and some HR departments adopted specific employee satisfaction or morale scores as new operating objectives. In essence, employee opinion surveys came to be used as early warning systems to improve management and reduce potential employer–employee conflict.

In 1973, a US Department of Health, Education, and Welfare-sponsored study concluded that the typical US worker was highly alienated from work and that, if not addressed, such alienation could lead to overt industrial conflict and even workplace violence (O’Toole, 1973). Accordingly, some HR specialists advocated programs to improve the quality of work life (QWL) (O’Toole, 1974) through job enrichment and job rotation, and many employers proceeded to adopt such programs (Davis & Cherns, 1975). Indeed, a QWL movement quickly emerged followed by numerous case studies that, on the whole, reported resounding success for QWL initiatives. Such initiatives were claimed to result in higher worker satisfaction and productivity and lower employee absenteeism, turnover, and labor costs (Lawler, 1986; Simmons & Mares, 1985). When examined more closely, however, such claims proved to be overstated, and subsequent, more rigorous studies concluded that QWL programs resulted in lower productivity and product quality and higher labor costs (Katz, Kochan, & Gobeille, 1983; Katz, Kochan, & Weber, 1985). It is therefore not

2 The human relations theorists in charge of the Hawthorne experiments did not start with the belief that the informal work group could influence productivity or product quality. Rather, they merely sought to extend Taylor’s focus on the content of work by experimentally manipulating one dimension of the work context, namely, lighting. Only after finding that productivity did not co-vary positively with the amount of lighting during the experiment did the researchers search for alternative explanations, eventually learning of the importance of work group norms for productivity and other outcomes. Hence, this discovery was serendipitous (Tannenbaum, 1966, Chapter 2).

D. Lewin / Human Resource Management Review 11 (2001) 453–485462

surprising that QWL initiatives began to wane or were discontinued and, by the 1980s, had largely disappeared from the industrial scene. But this experience, as with others before it, reflects the HR premise that employer–employee conflict can be reduced by improved management.

Another HR premise about conflict in the employment relationship is that a unity of interests, that is, an alignment of goals, among employers and employees, can be achieved through new work practices. One such practice, widely adopted by US manufacturing firms in the 1970s and early 1980s following examples set by Japanese manufacturers, was the quality circle. Under this arrangement, groups of six to eight workers meet for a specified time period at the beginning of each work shift to diagnose and devise solutions to production problems. These solutions are presented to management, which decides which ones to adopt and how to implement those that are adopted. Like QWL programs, the use of quality circles eventually declined precipitously, perhaps because they shared with QWL programs the absence of direct payments or other rewards to workers who participated in them (Lawler & Mohrman, 1987). However, by involving workers in production matters that were previously left to management alone, the quality circle experience in the US reflects an important attempt by management to forge a unity of interests with employees.

Of considerable longer standing than quality circles are employee suggestion systems, especially one such system, namely, the gain-sharing plan. Beginning with the Scanlon Plan, which was developed in the 1930s in the US steel industry (Moore & Ross, 1978), and continuing to the present, many variants of gain-sharing plans have been adopted by employers. However, all such plans aim to reduce costs for a given level of production. When cost reductions are achieved, the ‘‘savings’’ are shared between owners of capital and all employees in the plant — including managerial employees. An integral part of gain- sharing plans is employee suggestions about ways to save costs. Similar to quality circles, most gain-sharing plans call for management to decide which employee suggestions will be acted on, but unlike quality circles, there is a direct payment to workers if and when the adopted suggestions result in cost savings. Perhaps because of this incentive feature, gain- sharing programs, unlike quality circles and QWL programs, have not faded from the US employment scene, and it is estimated that about 16% of US firms makes some use of gain- sharing plans (Hewitt Association, 1989). Less is known about the prevalence of suggestion systems unconnected to gain-sharing plans, but it is likely that these are used at least as widely as gain-sharing plans. In any case, employee suggestion systems in general and gain- sharing plans in particular reflect the HR premise that a unity of interests can be forged among management and employees, with consequent reductions in employer–employee conflict. The same conclusion can be extended to employer use of profit-sharing, stock ownership, and stock option plans, all of which became considerably more widely used as the 20th century drew to a close (Jones, Kato, & Pliskin, 1997; Kruse & Blasi, 1997).

The use of employee opinion or attitude surveys is also germane to the HR premise concerning a unity of employer–employee interests. Such surveys have as their main purpose what may be termed the ‘‘construction of reality’’ or, in other words, measuring what is not directly observable from personnel records or other documents. The main reality in this regard is worker satisfaction (or morale), and employers have long been interested in gauging

D. Lewin / Human Resource Management Review 11 (2001) 453–485 463

such satisfaction for the evidence or clues it provides about actual or potential employer– employee conflict. In a typical attitude survey, employees are asked about their satisfaction with personnel-related matters (e.g., pay and benefits, job security), work-related matters (e.g., work itself, safety and health, working conditions), and management-related matters (e.g., supervision, performance appraisal) (Kleiner & Han, 1997). Areas or items which elicit relatively low satisfaction (or high dissatisfaction) scores are then singled out for attention and action by management. And, management has reason to do so because substantial evidence suggests that employee satisfaction is significantly negatively correlated with voluntary turnover, absenteeism, use of leave time, and grievance filing, and significantly positively correlated with product/service quality (Batt, 1999; Boroff & Lewin, 1997; Kleiner & Han, 1997). Further, employee dissatisfaction with pay and performance appraisal in particular has been shown to lead management to change or adopt new programs and practices in these areas (Gerhart, Milkovich, & Murray, 1992; Heneman & von Hippel, 1997). Such initiatives reflect attempts to mitigate conflict and bring about a unity of interests in the employment relationship.

Still another HR premise about conflict in the employment relationship is that such conflict can best be dealt with through mutual problem-solving techniques that promote cooperation and integration. Perhaps the leading manifestation of this premise in action is the growing and widespread use of teams in the workplace and sometimes for organizational decision making more broadly, including at senior management levels (Hambrick, 1987; Katzenbach & Smith, 1993). In the 1980s, and once again following the example set by Japanese enterprises, teams began to be proffered as essential to improving product quality and customer service, performing new knowledge-based work, and making business decisions in an increasingly competitive, global, uncertain economic environment. Specialists on team-based work contend that, for such teams to function effectively, team members must share basic norms, beliefs, and expectations, be willing to undertake certain duties and responsibilities formerly reserved to supervisors and managers, and work flexibly and interchangeably with each other (Applebaum & Batt, 1994). Stated another way, team-based work requires close cooperation among team members as well as between employers and employees.

Emphasis on team-based work and decision making was further strengthened by the total quality management (TQM) movement of the 1980s, the business process re-engineering movement of the late 1980s and early 1990s, and the movement toward virtual, electronically connected teams during the late 1990s. These initiatives involved not only the re-design of work, but of work flow, information systems, and decision-making processes. TQM may be regarded as a contemporary version of scientific management, but with primary emphasis on quality rather than quantity of output, and with the team rather than the individual as the key unit of analysis and implementation (Walton, 1986). Business process re-engineering is largely technology driven, with computing-based information systems at its core and with the organization as the primary unit of analysis (Hammer & Champy, 1993). Virtual, electron- ically connected teams are entirely technology-driven and differ fundamentally from more traditional, co-located work teams in that team members are typically dispersed across multiple locations, countries, and time zones (Joshi, 1999; Townsend, DeMarie, & Hen- drickson, 1998).

D. Lewin / Human Resource Management Review 11 (2001) 453–485464

By definition, work teams require cooperation among team members. To achieve such cooperation, employers typically provide training in mutual problem-solving techniques to team members prior to putting such teams in place. In some team work systems, teams have designated team leaders, while in others, they do not. Some teams are fully autonomous, that is, completely self-managed, while others are semi-autonomous and have a designated supervisor. Whatever particular arrangement prevails, however, work teams have in common the responsibility for tasks and decisions that were previously or more traditionally performed by management. It may be said, therefore, that team-based work reflects an element of power sharing by management with employees, not just employer– employee cooperation or mutual problem solving. It is perhaps for this reason that work teams are proving to have more staying power than, say, QWL programs, quality circles, and numerous other initiatives that have been undertaken by employers at one time or another under the label of ‘‘employee participation.’’

Of course, work teams are but one example of cooperative, mutual problem-solving initiatives that have been adopted by employers or jointly by employers and unions. Nonunion employer-initiated employee involvement and participation (EIP) programs have grown especially rapidly in recent years and vary widely in terms their formality, direct versus indirect involvement, influence in decision making, range of issues covered, and membership (Kaufman, Lewin, & Fossum, 2000). For example, a narrow-issue EIP may deal only with product quality while a broad-issue EIP may deal with customer relations and service, inventory control, plant or facility security, management succession, and employment policy. Various unionized employers have formed joint labor–management committees to study, advise, and sometimes decide such matters as workplace safety arrangements, automation of work processes, productivity improvement, employee training, and equipment and facility maintenance. Such initiatives are sometimes referred to as examples of ‘‘mutual gains’’ bargaining (Kochan & Osterman, 1995) or ‘‘dual concerns’’ negotiations (Pruitt & Rubin, 1986). But whatever their form, structure, or scope, all of these initiatives are predicated on the premise that conflict in the employment relationship can be reduced through mutual problem-solving techniques that promote cooperation and the integration of employer and employee interests.

The final HR premise on conflict is that the power imbalance in the employment relationship will fade as employers and employees come to have similar goals and objectives. This premise can be applied to several contemporary developments in HR management practice. For example, many HR specialists claim that a new role has evolved for the HR function in the business enterprise, namely, that of a strategic partner that adds value to the business (Ulrich et al., 1997). By this is meant an HR function that ‘‘relies on people as a source of competitive advantage and a management culture that embraces that belief . . . operational excellence, a focus on client service for individual employees and managers . . . and HR managers that understand the human capital implications of business problems’’ (Becker & Huselid, 1999). Some HR specialists have gone even further to suggest that the HR function should be the change agent leading organizational learning and knowledge transfer (Ulrich, 1997). These new conceptions of the HR function imply that, rather than being engaged in an internal power struggle, employers and employees have a unity of

D. Lewin / Human Resource Management Review 11 (2001) 453–485 465

interest in the enhanced economic performance of the firm. By working together to increase and improve the utilization of human capital, employers and employees can jointly gain power over competitors — and it is up to the HR function to lead and devise programs, especially programs of employee financial and nonfinancial participation in the enterprise, that will help to achieve this objective.

In a more empirical vein, several groups of researchers have shown that ‘‘bundles’’ of progressive or high-involvement HR management practices are significantly associated with improvements in business performance. These findings have been reported in plant level studies (Applebaum, Bailey, Berg, & Kalleberg, 2000; Ichniowski, Shaw, & Prennushi, 1997; MacDuffie, 1995), customer sales and service work teams (Batt, 1999), and firm- level studies (Huselid, Jackson, & Schuler, 1997; Lee & Johnson, 1998; Mitchell, Lewin, & Lawler, 1990). The dominant conclusion of these studies is that any single innovative HR management practice, such as work teams or variable pay or re-designed work or employee training or skill interchangeability, has no discernible effect on productivity, product quality, sales revenue, customer service, return on capital employed, or market value. When undertaken together, that is, as a bundle, however, these practices have significant positive effects on business performance. In a related study, Huselid (1995) showed that high- involvement HR management practices were significantly positively correlated with employee productivity and significantly negatively correlated with employee turnover, and these two HR outcomes were, in turn, significantly positively correlated with firms’ return on capital and market value. This empirical evidence can be interpreted to mean that high-involvement work practices reflect employer–employee power sharing, and that such practices, together with their positive consequences for business performance, reflect a broad unity of interests and objectives among employers and employees. This is to be contrasted with firms that follow traditional work practices and that are typically characterized by an employer–employee power imbalance and widely divergent employer and employee interests.

Researchers have also given much recent attention to organization culture, that is, to the dominant norms, values, beliefs, and expectations of members of a business (or non- business) enterprise (O’Reilly, 1989; O’Reilly & Chatman, 1996). The main analytical point made by these researchers is that culture is a mechanism of social control, more powerful perhaps than other, more formal control systems or practices. By shaping and attempting to ensure that all organizational members — employees — share these norms, values, beliefs, and expectations, leaders of enterprises can harness organization culture to achieve key goals and objectives. By positing that organizational members can come to share a common culture, these researchers are in effect, saying that there is a unity of interest among employers and employees; it just has to be realized. But even a fully shared organizational culture does not obviate the employer–employee power imbalance or necessarily protect employee rights or guarantee that employees will share in the success of a strong culture enterprise. Therefore, such IR-based initiatives as progressive discipline and just cause-based dismissal, and such HR-based initiatives as gain sharing and high- involvement work practices can, and some say should, be practiced by a strong culture enterprise (O’Reilly, 1989).

D. Lewin / Human Resource Management Review 11 (2001) 453–485466

In light of this research, it is no surprise or accident that leaders of contemporary business enterprises have come to cite organizational culture as a key success factor (Weber, 1992), much as earlier generations of business leaders cited capital investment, cost control, technological investment, and other characteristics as key success factors. Similarly, it is not surprising that business strategy specialists have come to regard organizational culture as a key element of internal organizational alignment and strategy execution (Hambrick, 1994), or that organizational behavior specialists have come to regard organizational culture as key to successful organizational change. By the same token, however, these researchers also are aware of and have documented instances in which organizational culture builds resistance to change and thwarts even the most determined management initiatives to effectuate such change (Tushman & O’Reilly, 1996). This suggests that employee norms, values, beliefs, and expectations are not always aligned and cannot easily be aligned with those of management, just as an HR function in a firm may not be able singlemindedly to fulfill the business partner or change agent roles that some advocate for it. In any case, contemporary research and practice in the areas of HR strategy, high-involvement work systems, and organization culture together appear to reflect the HR premise that conflict in the employment relationship will fade as employers and employees come to share similar goals and objectives.

5. Empirical evidence about IR and HR perspectives on conflict

What does empirical evidence tell us about the IR and HR perspectives on conflict in the employment relationship? In my judgment, there is some evidence to support each of these perspectives, but since these perspectives differ in certain important respects, this means that there is also evidence that fails to support or goes against one or the other perspective.

5.1. Grievance procedures

Grievance procedures were established in labor agreements between industrial unions and employers and quickly became widespread. With the continuing percentage decline in private sector union membership, proportionately fewer unionized workers are covered by such procedures than previously. Nevertheless, a sample of 400 collective bargaining agreements recently obtained by the Bureau of National Affairs (1995) found that all of these agreements contained a formal grievance procedure, 99.5% of them had arbitration as the final grievance step, and 97% specified the scope of issues covered by the grievance procedure (Eaton & Keefe, 1999). As noted earlier, among nonunion private sector employers about half have formal grievance procedures in place and roughly one quarter of these procedures has arbitration as the final step (Delaney, Lewin, & Ichniowski, 1989; McCabe & Lewin, 1992). While not much is known about the scope of issues covered by these nonunion grievance procedures, it appears that substantial proportions of employees, including first-line super- visors and mid-level managers, are eligible to use them. Thus, it is likely that grievance procedures now cover considerably more nonunion than unionized private sector employees.

D. Lewin / Human Resource Management Review 11 (2001) 453–485 467

The existence of these procedures can be interpreted to support the IR perspective that conflict in the employment relationship is widespread and enduring, but also to support the HR perspective that conflict can be substantially reduced by organizational and procedural innovations, that is, improved management. Following this reasoning, the grievance proced- ure can be viewed as high-involvement or problem-solving work practice. Indeed, economists Freeman and Medoff (1984) contend that by serving as a mechanism through which employees can exercise voice in the employment relationship, grievance procedures result in lower employee quit rates, longer job tenure, increased human capital (through greater employer investment in employee training), and thus higher productivity.

A different story emerges, however, from other grievance procedure research. In a study of four unionized enterprises over two separate 3-year periods, Lewin and Peterson (1999) found that grievance filers had significantly lower job performance ratings, promotion rates, and work attendance rates, and significantly higher turnover rates than grievance nonfilers in the periods following grievance settlement. There were no significant differences between grievance filers and nonfilers on any of these dimensions prior to and during the period of grievance settlement. Similar findings were reported for samples of supervisors of grievance filers and nonfilers in the four enterprises, that is, the supervisors of grievance filers had significantly lower job performance ratings, promotion rates, and work attendance rates, and significantly higher turnover rates (especially termination rates) than the supervisors of grievance nonfilers in the periods following (but not prior to or during) grievance settlement.

In another study, set in a large unionized telecommunications firm and limited to employees who believed that they had been unfairly treated at work, Boroff and Lewin (1997) found that, contrary to the predictions of exit-voice-loyalty theory (Hirschman 1970), employee loyalty was significantly negatively related to the probability of filing a grievance (exercising voice), and grievance filing was significantly positively related to intent to exit the firm. These researchers also found that fear of reprisal was significantly negatively related to grievance filing, and thus concluded that unionized workers who believe that they have been unfairly treated at work nevertheless may choose to respond with silence even when they have a contractual grievance procedure and grievance handling representatives available to them. This suggests, in turn, that unionized workers do not view the grievance procedure as a high-involvement (mutual gains) work practice or as an effective tool for redressing the employer–employee power imbalance.

Turning to nonunion grievance procedure research, Lewin (1987, 1992) analyzed job performance ratings, promotion rates, work attendance rates, and turnover rates among samples of grievance filers and nonfilers in five nonunion companies in which employees are covered by grievance procedures (two of these company procedures provided for arbitration as the final step). He found no significant differences between grievance filers and nonfilers on any of these measures prior to or during the periods of grievance filing and settlement. In the 1- and 2-year periods following grievance settlement, however, job performance ratings and promotion rates were significantly lower and turnover rates were significantly higher for grievance filers than for nonfilers (work attendance rates were insignificantly higher for filers than nonfilers). A comparable analysis of samples of supervisors of grievance filers and nonfilers produced very similar findings. In a related laboratory study, Olson-Buchanan

D. Lewin / Human Resource Management Review 11 (2001) 453–485468

(1996) found that grievance filers had significantly lower job performance than nonfilers in the post-grievance filing and settlement period.

Further, in a study of the grievance procedure in a large, nonunion express delivery company, Lewin and Boroff (1996) found that employee loyalty was significantly negatively related to the probability of grievance filing, and that grievance filing was significantly positively related to intent to leave (exit) the company. In this nonunion company, moreover, fear of reprisal had the strongest negative correlation with grievance filing. This finding is perhaps especially important because the company’s grievance procedure covers supervisors and mid-level managers, and because fear of reprisal was significantly positively correlated with job/occupational level. Virtually the same findings were reported by Lewin (2000a) in his study of a nonunion aerospace company that has had a grievance procedure in place for over a half century. That is, fear of reprisal for filing grievances was strongest among the company’s first-line supervisors and middle managers, who are eligible to use this company’s grievance procedure, and fear of reprisal was significantly negatively correlated with grievance filing by nonmanagement and management employees of this company.

Taken as a whole, recent research on unionized and nonunion grievance procedures alike suggests that while these formal mechanisms can help to resolve employer–employee conflict, they also have some important limitations. Some employees who use these procedures as well as their supervisors appear subsequently to suffer (further) deterioration, even the severing, of their employment relationships. That this is more generally known by employees is reflected in the failure of some employees to file grievances — that is, to be silent — even when they have been unfairly treated in the workplace, and in the fear of reprisal expressed by employees (and, in nonunion enterprises, by supervisors and managers) as a key factor in deciding not to file grievances. Thus, rather than serving only as effective, participatory, problem-solving devices or improved management practices of the type envisioned by the HR perspective on employment relationship conflict, grievance procedures appear in certain circumstances to reinforce the power imbalance in the employment relationship that is central to the IR perspective on conflict.

5.2. Employment discrimination

The number and size of work force groups protected under the law, together with work force reductions and other job dislocations resulting from business restructuring, re-engin- eering, and merger and acquisition activity, have produced a large volume of cases in which employees (and former employees) allege discrimination by employers in hiring, staffing, compensation, promotion, demotion, layoff, and other aspects of employment. Early on, the bulk of the cases was brought by racial minorities, notably African–Americans, and women. More recently, the bulk of the cases has been brought by older workers (i.e., those 40 years old and above) and by the growing Latino segment of the work force, although there has also been a rise in sexual harassment cases, with charges typically (but not always) brought by women against men.

In addition, and beginning in the early 1980s, the doctrine of ‘‘wrongful termination’’ (or ‘‘unjust dismissal’’) was adopted by the judiciary in various states under which certain actions

D. Lewin / Human Resource Management Review 11 (2001) 453–485 469

of employers (e.g., firing whistleblowers) were declared to violate public policy or to constitute the breach of an explicit or implicit employment contract. In particular, a record of long-term employment with the same employer, the use by an employer of a personnel manual or employee handbook describing the benefits provided to employees and/or suggesting employment continuity, and the presence of performance appraisals skewed toward the high end of the rating scales that are typically used to make such appraisals were taken by the judiciary as evidence of an implied contract of employment between employer and employee. On one or more of these grounds, some work force reduction, layoff, and other employment termination decisions made by employers have been declared illegal and overturned by the courts.

More fundamentally, the legal protections afforded to employees under standing laws, combined with the use of wrongful termination in adjudicating employment cases, have produced an important challenge to the idea and practice of employment-at-will. This idea is, of course, grounded in the concept of private property rights; under it, the owners of business enterprises or their agents (i.e., managers) can freely hire, fire, promote, discipline, and utilize employees much as they would any other resource, such as (physical) capital. Correspond- ingly, an employee is free to quit a job and seek employment elsewhere and to do so on short notice or no notice at all. However, both anti-discrimination in employment legislation and the wrongful termination doctrine in effect create and have been interpreted by the courts to uphold employee claims to property in the job, and this property right must be weighed against the property right of the business enterprise/employer in determining the legality of employer actions concerning employees (Edwards, 1997).

These efforts bear a striking similarity to earlier union efforts to define and protect worker property in the job. They are also closely similar to ‘‘classic’’ union–management relations in that the conflicts that arise under one of the aforementioned anti-discrimination laws or as wrongful termination claims are basically adversarial in nature and typically proceed toward settlement in a win–lose context. While some of these contested employer actions are resolved internally without a formal charge ever being filed, and while relatively few charges that are filed proceed all the way through a court trial, most employment-related cases involve the hiring of legal counsel by both sides to the dispute and are marked by adversarialism, win–lose negotiations, and the use of power tactics to reach settlement. Moreover, in cases involving termination and other separations from employment, settlements rarely result in the re-instatement of an employee to a job. Instead, monetary settlements of the disputes are reached and former employees search for jobs elsewhere; the original employment relation- ships have deteriorated past the point of ‘‘no return.’’

In this regard, employers are increasingly mandating arbitration for the settlement of certain employment disputes, in particular, disputes over employment discrimination and wrongful termination. A recent US General Accounting Office (1995) report found that about 10% of 1500 surveyed companies used nonunion arbitration to resolve discrimination complaints, and that another 8.5% was considering instituting such a procedure. While, as noted above, many nonunion employers have adopted grievance-like procedures, and many have adopted other alternative dispute resolution (ADR) techniques, including open-door policies, peer review panels, management appeal boards, and ombudsmen, the rise of

D. Lewin / Human Resource Management Review 11 (2001) 453–485470

mandated arbitration among nonunion employers is due largely to two factors: ‘‘employers’ desire to limit liability for employment discrimination lawsuits . . . and unjust dismissal lawsuits . . . and the courts’ newfound willingness to compel arbitration of statutory claims as a result of the [US Supreme Court’s 1991] Gilmer decision’’ (Stone, 1999, p. 28).

Once again, from an HR perspective on employment relationship conflict, the use of arbitration can be viewed as a participatory or problem-solving device that helps to improve management practice. In theory, such arbitration allows both parties to exercise voice in the settlement of disputes, promotes cooperation in the employment relationship, and reduces the employer–employee power imbalance. But in practice, mandated arbitration has few, if any, of these attributes and, instead, appears to lack basic elements of due process. As examples, nonunion arbitration systems often provide that the employer alone selects the arbitrator, does not allow employees to be represented by counsel or other parties, requires employees to pay large fees in order to get their cases heard, imposes large ‘‘burdens of proof’’ on employees, and strictly limits the remedies available to an employee who wins his/her case. Further, under nonunion arbitration systems, there is wide variation in adherence to the requirement of written opinions by arbitrators. Moreover, with judicial deferral to arbitration to settle nonunion employer–employee disputes, the employee has no subsequent recourse to the courts (or other government agencies, such as the Equal Employment Opportunity Commis- sion or comparable state government commissions) to challenge the decision of an arbitrator. For these reasons, and invoking organizational justice theory and research (Sheppard, Lewicki, & Minton, 1992), mandatory employment arbitration systems may be regarded as procedurally unjust.

That these systems may also be distributively unjust can be inferred from recent data on trends in employment discrimination litigation in the federal courts (Stone, 1999). Between 1987 and 1997, the number of employment discrimination cases that went to judgment roughly doubled (from 2592 to 5199). During the same period, however, the plaintiff (employee) win rate declined from 22% to 12%. A major factor in this regard was the large increase in cases decided at the pre-trial stage rather than at trial. In 1987, 28% of all cases was decided on pre-trial motion and plaintiffs won more than 11% of those motions, whereas by 1997, 50% of all cases was decided at the pre-trial stage and plaintiffs won only about 3% of those motions. Therefore, even though plaintiff win rates at trial in employment discrimination cases increased between 1987 and 1997, the overall plaintiff win rate in such cases declined during the same period. Most notable, a substantial portion of these cases decided at the pre-trial motion stage and which plaintiffs lost were motions to dismiss on the ground that employment discrimination and unjust dismissal claims were subject to a mandatory arbitration procedure.

Therefore, rather than largely promoting employer–employee cooperation and problem solving, improving management practice, or reducing the power imbalance in the employ- ment relationship, mandated arbitration systems for nonunion employees seem to have some of the opposite effects. When added to the evidence about negative consequences for some nonunion and unionized employees that result from their use of formal grievance procedures, a picture of contemporary employment relationship conflict emerges that appears to support the IR perspective more than the HR perspective on such conflict. It may be the case,

D. Lewin / Human Resource Management Review 11 (2001) 453–485 471

however, that grievance systems and mandated arbitration systems are essentially reactive or negative forms of employment conflict resolution that essentially worsen already deteriorated employer–employee relationships (Kaminski, 1999; Lewin, 1999). If so, then perhaps there are other, more proactive or positive forms of conflict resolution that more closely fit HR premises about and perspectives on employment relationship conflict.

5.3. EIP

More than any work force management initiative, EIP programs reflect the HR-based perspective that employment relationship conflict stems from poor management and can thus be reduced, that a unity of interests among employers and employees can be achieved, and that remaining conflicts can be dealt with through mutual problem-solving techniques. And, some of the research on EIP and new forms of work organization supports this view.

To illustrate, there have been many case and some empirical studies of nonunion companies that have undertaken such EIP initiatives as quality circles, QWL enhancement programs, TQM programs, problem-solving groups and committees, and semi-autonomous and fully autonomous (or self-directed) work teams. Most of the early and some of the recent accounts of these initiatives conclude that they have highly positive effects on individual, team, and/or organizational performance (Davis & Cherns, 1975; Dessler, 1997; Hoerr, 1990; Lawler, 1986; Simmons & Mares, 1985). For example, Corning introduced team-based production and multiskilling into its Blacksburg, VA, plant in 1989—arrangements, which contrasted markedly with the narrow, specialized, individual-based work that prevailed in other Corning plants. The company found that work teams at Blacksburg could retool a line to produce a different type of filter in about 10 min, or approximately six times faster than at any of its other plants in which work was performed individually (Hoerr, 1990).

In another example, Toyota Motor Manufacturing USA has used team-based production in its Kentucky-based automobile manufacturing plant since the plant opened in 1988. Toyota’s ‘‘Team Member Handbook’’ explicitly states the company’s ‘‘commitment to work as a team with mutual respect and equal opportunities for all . . . our abilities are maximized when we work together in a cooperative manner toward common goals . . . mutual trust and respect for each other are the basis on which team spirit is developed’’ (Toyota Motor Manufacturing USA, 1988, p. 11). Not only Toyota’s management but some researchers attribute the high productivity and high product quality of this plant to team-based work as well as to such related EIP practices as formal teamwork training and permitting work teams to select their own new members (Wellins & George, 1991). These examples tend to support the view that EIP initiatives can reduce workplace conflict and bring about a unity of employer–employee interests.

Beyond team-based manufacturing plants, such nonunion companies as Procter&Gamble, Texas Instruments, and Hewlett-Packard claim to have long relied on team-based decision making at all levels of their respective organizations, including the top management level. Senior executives of these companies and certain organizational behavior researchers conclude that such team-based decision making is especially characteristic of strong culture enterprises, and that such a culture, in turn, has positive effects on the long-term financial

D. Lewin / Human Resource Management Review 11 (2001) 453–485472

performance of these enterprises (Pfeffer, 1998). These examples further suggest that broadly conceived EIP programs can not only establish a unity of employer–employee interests, but that such unity can be sustained over long periods.

Of particular relevance in this regard is the work of Cotton (1993), who conducted a comprehensive review and assessment of EIP research and practice involving self-directed work teams, gain-sharing plans, QWL plans, job enrichment programs, employee ownership programs, quality circles, and representative participation arrangements (largely works councils and co-determination of the type practiced by European enterprises). The bulk of the research and case examples reviewed by Cotton dealt with nonunion establishments and employees. His key conclusions are that self-directed work teams and gain-sharing programs have strong positive effects on productivity and employee attitudes; that QWL plans, job enrichment programs, and employee ownership have intermediate or modest positive effects on productivity and employee attitudes; and that quality circles and representative participa- tion have weak or no effects on productivity and employee attitudes. These important conclusions appear largely to support the HR-based claims that employment relationship conflict can be reduced through improved management, organizational and workplace innovations, and cooperative problem-solving, mutual gains techniques.3

Further in this vein, consider a recent study of EIP programs in eight nonunion US-based companies (Kaufman et al., 2000). This study, which covered units or plants of businesses ranging from soap, paper, janitorial maintenance equipment, and missile manufacturing to automobile assembly, photocopier service and repair, airline transportation, and express mail and package delivery, found widespread use of self-managed work teams, project teams, employee committees, information-sharing programs, review boards, employee attitude surveys, and gain-sharing and profit-sharing plans. All of these initiatives were undertaken at various times and places by company management, occasionally to ward off employee unionization but more typically to improve organizational performance. What is especially notable about these EIP initiatives is that the companies pursued them despite the fact that they appear in whole or in part to violate US labor law or, as the researchers put it, ‘‘the only company of the eight considered here that [in its EIP initiatives] appears clearly to fall within the permissible boundaries established by the National Labor Relations Act (NLRA) is Company E (photocopier service)’’ (Kaufman et al., 2000, p. 276). That these companies are willing to risk such potential illegality suggests that they are motivated mainly by a desire to

3 The same may be said of empirical research on relationships between human resource management practices and business performance. Much of this research focuses on the effects of ‘‘bundles’’ of complementary HR practices, including but not limited to EIP practices, on return on capital, market value, revenue per employee, employee productivity, and employee turnover (see, e.g., Huselid, 1995; Huselid et al., 1997; Lee & Johnson, 1998; Mitchell et al., 1990). Other of this research examines the effects of individual ‘‘innovative’’ HR practices, such as employee information sharing, incentive compensation, and performance management programs, on business performance (see, e.g., Asch & Warner, 1997; Gerhart et al., 1992; Heneman & von Hippel, 1997; Jones et al., 1997; Kleiner & Bouillon, 1988; Kruse & Blasi, 1997; Lawler, 1990; McDonald & Smith, 1995; Morishima, 1991, 1992).

D. Lewin / Human Resource Management Review 11 (2001) 453–485 473

achieve mutual gains with their employees by forging an employer–employee unity of interest, thereby further supporting the HR perspective on employment relationship conflict.

It is necessary, however, to consider certain important limitations on EIP study and practice. For example, Cotton (1993) observes that the process by which an EIP program is adopted is at least as important as the outcomes or effectiveness of such a program. In this respect, he contends that many EIP programs are solely employer-initiated (and sometimes mandated) so that the principle of employee involvement in program design is often violated — a conclusion virtually identical to that reached by Eaton, Voos, and Kim (1997) in their study of voluntary and involuntary aspects of EIP programs. This view also accords closely with that of organizational justice researchers who contend that the processes through which decisions are made, including whether or not to have an EIP program, are as or more important to employees than the results of such decisions and programs (Sheppard et al., 1992). In addition, Cotton warns that empirical studies or case examples of unsuccessful EIP initiatives are few and far between, and he attributes this to the under-reporting of such examples and the ‘‘selection bias’’ of researchers. More recently, Godard and Delaney (2000) assessed the claim that the so-called high-performance work systems paradigm, which is heavily grounded in EIP, has replaced unions and collective bargaining as the dominant force in IR. Contrary to this claim, they find that the HR innovations underlying this paradigm have only weak empirical effects on business performance, and that numerous North American and European case studies report quite limited life spans and even outright failures of HR innovations. Taken together, these observations and conclusions tend to support the IR-based claims that power in the employment relationship is unbalanced, with the employer holding sway over employees, and that conflict is an enduring characteristic of nonunion employment relationships, including those that feature EIP programs.4

While it is sometimes thought that EIP programs began in the nonunion sector and that such programs are considerably more prevalent in the nonunion than the unionized sector of the US economy, neither of these views is supported by empirical evidence (see Eaton & Voos, 1994). Further, some authorities believe that EIP programs work better in unionized than nonunionized companies (Freeman & Rogers, 1999). In fact, extant empirical evidence suggests that the record of EIP performance in unionized companies is mixed, ranging from highly successful to highly unsuccessful experiences.

At Xerox, for example, management and representatives of the Amalgamated Clothing Workers Union formed joint committees and workplace teams whose collaborative efforts resulted in improved plant safety, work flow and production, reduced grievance rates, and the preservation of jobs which would otherwise have been eliminated (Cutcher-Gershenfeld, 1991). Similarly, in a unionized paper mill with a long record of adversarial union– management relations, very high grievance rates and poor economic performance, a joint union–management EIP program was instituted, which resulted in more flexible job

4 Industrial relations and labor/employment law scholars have also questioned the rights, protections, due process, and lack thereof for nonunion employees involved in EIP programs (see Eaton et al., 1997; Finkin, 1992; Levine & Tyson, 1990; Weiler, 1990).

D. Lewin / Human Resource Management Review 11 (2001) 453–485474

classifications and work arrangements, substantial increases in tons of paper produced, lower grievance rates, and reduced bargaining disputes (Ichniowski, 1992).

Probably the best know EIP program in a unionized setting is the New United Motor Manufacturing (NUMMI) venture involving General Motors, Toyota Motor, and the United Automobile Workers’ Union (UAW). Founded in 1984 and located in a previously closed General Motors-owned plant, NUMMI featured replacement of multiple job classifications by a single ‘‘production worker’’ classification, replacement of individual work with autonomous work teams to produce automobiles and trucks and monitor production, freedom for individual workers to slow or stop the production line, use of just-in-time materials delivery, major reductions in parts inventories, increased supervisory and mana- gerial spans of control, and substantial use of profit sharing. Researchers and practitioners alike have concluded that these practices are largely responsible for NUMMI’s very high levels of productivity, product quality, and worker job satisfaction, and for greatly reduced grievance rates compared to those that existed in the old plant (Brown & Reich 1989; Childs, 1989).

Further, the team concept, which is central to the manufacturing operation at NUMMI, has been taken much further in Saturn, a wholly owned subsidiary of General Motors that began operation in 1990. At Saturn, work teams are used not only to manufacture automobiles but throughout the organization more broadly. These teams make all major business decisions, including about strategy, product design and development, inventory control, work flow, technological change, and customer relations and service (Horvath & Syvantek, 1998; Rubenstein, Bennett, & Kochan, 1993). In other words, Saturn is a team-based organization, whereas the use of teams at NUMMI is largely limited to manufacturing. In both cases, however, the use of team-based work is widely regarded as key to the strong performance of these enterprises. These several examples show that an employer–employee unity of interests can indeed be forged, and that employment relationship conflict can be reduced through improved management, organizational and workplace innovations, and cooperative problem- solving techniques.

But not all EIP initiatives in unionized enterprises are as successful as those recounted above. For example, a well-documented EIP program at Rushton Mining, whose workers were represented by the United Mine Workers of America (UMWA), was instituted in 1973 (Goodman, 1979). The initiative, which enrolled the union as a full partner, called for the re- organization of work; job security provisions; joint committees for training and development, coordination of work crew activities, and plant safety; an intershift communication program; job rotation; and foreman planning and organizational goal setting done in consultation with workers. Early on, this ‘‘experiment’’ resulted in higher productivity, product quality, employee compensation, and job satisfaction, and in lower scrap rates, work injuries, and employee grievances. As time passed, however, Rushton’s workers became increasingly dissatisfied with increasing work loads, excessive job rotation, foremen performing work prohibited by the labor agreement, lack of management communication with union leaders, and declining ‘‘real’’ earnings. Consequently, in 1977, following a 10-day strike at Rushton and a longer national coal strike, the union’s membership voted to rescind the UMWA’s participation in the EIP program at Rushton Mining.

D. Lewin / Human Resource Management Review 11 (2001) 453–485 475

In the airline industry, a variety of EIP programs have been instituted jointly by the major carriers and unionized employees. These programs feature decentralized decision making that allows reservations agents, in-flight attendants, and baggage handlers to (attempt to) resolve customer complaints themselves rather than passing them on to management; limited job rotation; formal training and development programs; joint committees in the areas of safety, work scheduling, and hiring requirements; and department-based team meetings. However, while the pilots’, attendants’ and baggage handlers’ unions participate in these programs, skilled craft employees and operatives represented by the International Asso- ciation of Machinists (IAM) do not. This union has consistently opposed EIP programs, which IAM leaders regard as a sellout to management. In an especially notable devel- opment, United Airlines became an employee-owned airline in the early 1990s. However, the attendants’ union refused to participate in this arrangement, and growing dissatisfaction with United Airlines’ financial performance has caused the participating unions’ leaders and members to question the longevity of this ownership arrangement.

In another study set in the automobile industry, Bruno and Jordan (1999) traced the evolution of worker attitudes at the Diamond Star plant, which began in 1989 as a joint venture between Mitsubishi Motors and Chrysler but which is now largely owned and operated by Mitsubishi. As with NUMMI and Saturn, workers at this plant are represented by the UAW and automobile manufacturing is based on the principles of lean production, substantial employee participation in decision making, and widespread use of work teams. Early on, Diamond Star’s workers had highly positive attitudes about virtually all aspects of this system and plant performance was very high. Since then, however, worker attitudes have deteriorated significantly. In particular, the work force has become highly dissatisfied with their lack of control over work, the faster pace of work, management’s decision to outsource some work and failure to rectify the use of substandard parts, reduced customer satisfaction, and declining plant performance. Bruno and Jordan thus conclude that the once promising EIP initiative at Diamond Star has reached its nadir.5

Consistent with the IR perspective on employment relationship conflict, these particular examples suggest that such conflict is enduring, that management holds a power advantage over employees, and that even a very strong unity of employer–employee interests formed in the early stages of an EIP initiative will dissipate over time. When these ‘‘cases’’ are combined with prior examples of positive conflict reduction and other favorable consequen- ces stemming from EIP initiatives in both unionized and nonunion establishments, it can be concluded that neither proponents of the IR perspective nor proponents of the HR perspective have got it ‘‘exactly right’’ in so far as the analysis and resolution of employment relationship conflict are concerned.

Finally, consider that in both unionized and nonunion settings, EIP programs are practices that are usually reserved for full-time core employees, while being largely absent among part- time, temporary, and contract employees and others in contingent employment relationships. Most firms that have adopted EIP practices have also shrunk their core work forces and

5 For a similar analysis of deteriorating employee attitudes at Saturn, see Horvath and Svyantek (1998).

D. Lewin / Human Resource Management Review 11 (2001) 453–485476

expanded their contingent work forces (Lewin & Mitchell, 1995). The core work force is the one that employers have in mind when they refer to their firms’ human assets or human capital, as is reflected in the investments these employers make in EIP programs for their core employees. Contingent employees, by contrast, are treated and managed by employers largely as a labor expense — an expense to be minimized.

If a firm has a relatively small proportion of contingent employees in its work force, then conflict between contingent and core employees may be minimal and thus not costly enough to warrant employer attention. But if a firm has a relatively large proportion of contingent employees in its work force, then conflict between the ‘‘haves’’ and the ‘‘have nots’’ may become substantial, as the latter make coercive comparisons with the former. Contingent employees may complain or even demand that EIP programs be extended to them, especially when labor markets are tight. Here, an analogy can be drawn with industrial workers of the 1920s and 1930s who were not represented by unions and did not bargain collectively with employers, in contrast to more skilled and highly paid craft workers. The considerably larger group of industrial workers eventually came to demand unionization and bargaining rights and literally fought bloody battles to achieve such rights. One major result of those conflicts was passage of the Wagner Act. This is not to say that the demands of today’s contingent workers will result in legislation extending EIP programs and practices to them. Rather, it is to recognize that for the sizable segment of the US work force that is not ‘‘empowered’’ by EIP programs (or, for that matter, by union representation or most protective legislation), there remains a substantial employer–employee power imbalance and a potential for covert conflict to become overt conflict.

Recognizing the caveats about EIP programs issued by Cotton (1993), Godard and Delaney (2000), and Kaufman et al. (2000), as well as the limitation of EIP programs to core employees, there nevertheless is impressive evidence from research on nonunion EIP programs to support the HR perspective on employment relationship conflict. In particular, this evidence suggests that innovative EIP programs and practices can forge a unity of interests among employers and employees, promote mutual problem solving and cooperation, positively affect employee attitudes and performance, and improve management and organizational effectiveness. Further, this evidence, together with the growing use of nonunion EIP programs, can be interpreted to mean that, as the HR perspective also posits, the power imbalance and conflict in the employment relationship are on the wane.

But these conclusions must be tempered by the EIP experience in unionized settings, which demonstrates a more mixed record of success and failure. Apart from NUMMI, there are few examples of enduring EIP programs, and therefore of conflict reduction, in unionized workplaces. Rather, these programs come and go and are subject to the vicissitudes of external, environmental forces (such as competition, law and technology), management and operational changes, and, especially from an HR perspective, the conflict often generated by unions for political and organizational reasons. Through formal representation and collective voice, unionized workers can participate, and indeed help employers to design EIP programs (Cutcher-Gershenfeld, 1991; Ichniowski, 1992; Rubenstein et al., 1993). However, as political organizations that seek increased membership, have diverse membership interests, and often feature competing internal factions, unions also can and have opposed the

D. Lewin / Human Resource Management Review 11 (2001) 453–485 477

introduction and continuation of EIP initiatives (Brown & Reich, 1989; Bruno & Jordan, 1999; Goodman, 1979). In nonunion settings, workers do not have independent representa- tives of their own choosing and it is therefore more difficult for them to voice their concerns about or opposition to EIP programs. Further, and as Cotton (1993) notes, it is quite likely that unsuccessful experiences with nonunion EIP programs also exist despite the fact that they are rarely reported in empirical or case research. In sum, it can be concluded that the power imbalance in the employment relationship remains in place, that conflict is an enduring (if often covert) characteristic of this relationship, and that institutional interven- tions are more likely than employer-initiated EIP programs to redress this imbalance. Thus, EIP research and practice also offer some support for the IR perspective on employment relationship conflict.6

6. Lessons from IR and HR perspectives on conflict

IR and HR offer important, different, and relevant perspectives on employment relation- ship conflict. At its core, IR emphasizes the enduring nature of the employer–employee power imbalance and thus of conflict in the employment relationship, the need for institutional interventions to correct that imbalance, and the positive consequences of openly surfacing and attempting to resolve conflict, including through adversarial, nonproblem- solving approaches. By contrast, HR emphasizes that conflict stems from poor management and can be reduced by improved management, that organizational and workplace innovations can be used to develop a unity of interests among employers and employees, that problem- solving techniques and cooperation can be used to further reduce conflict, and that power imbalances will fade as a result of employer–employee goal sharing.

There is evidence to support each of these perspectives and also evidence that calls each of these perspectives into question. Empirical research and case studies in the area of grievance procedures and employment discrimination lend weight to the IR perspective on employment relationship conflict. In both unionized and nonunion enterprises, employees who feel unfairly treated at work nevertheless suffer in silence, fear reprisals for filing grievances, or, if they proceed to file grievances, experience further deterioration of their employment relationships. Employees who claim discrimination or wrongful termination have certain legal recourse under protective statutes, but are increasingly confronted with mandatory arbitration that reduces their chances of winning their claims and, in any case under any procedure, are highly unlikely to be re-instated to work. Because both grievance procedures and legal procedures are reactive (or negative) forms of employment conflict resolution, however, it is likely that conflict will be more effectively dealt with and reduced through proactive (or positive) initiatives, such as EIP programs.

6 Similar conclusions about IR and HR perspectives on employment relationship conflict apply to public sector employment relations, which are not considered here due to space limitations. See, as examples, Aaron, Grodin, and Stern (1979), Kochan (1973), Lewin (2000b), and Lewin et al. (1988).

D. Lewin / Human Resource Management Review 11 (2001) 453–485478

Here the evidence is relatively more supportive of the HR perspective on employment relationship conflict. Numerous empirical studies and case examples find that EIP programs can have positive consequences for individuals and organizations and thereby contribute to the reduction of conflict in the employment relationship. These consequences include improved attitudes, compensation, productivity, product quality, and return on capital as well as reduced grievance rates, down time, and possibly even workplace violence. EIP initiatives also often enhance worker skill, freedom at work, and worker health and safety. There is more than one side to this picture, of course. In unionized settings, some EIP initiatives have deteriorated, been voted out, or abandoned. In nonunion settings, there has been inattention to and under-reporting of EIP failures. And, in both settings, contingent workers are unlikely to be included in EIP programs. Nevertheless, the bulk of the evidence shows that, as the HR perspective would have it, organizational and workplace innovations and problem-solving approaches can improve management, build (for a time) a unity of employer–employee interests, enhance worker well being, and reduce employment relation- ship conflict.7

What HR specialists can learn from all this is that, in general, power in employment relationships is unbalanced, with the employer holding sway over the employee; that this power imbalance is lessened considerably when the economy is at full employment or where high-involvement work systems are practiced; that conflict in employment relationships does not stem from poor management alone; that, yes, a unity of interests among employers and employees can be built for a while but, except for the rare case, cannot be counted on to prevail indefinitely; that there is a place for institutional mechanisms to deal with conflict in the employment relationship; that the open surfacing of conflict can have certain positive consequences for individuals and organizations even when resolution is pursued through adversarial, nonproblem-solving approaches; and that conflict is an enduring feature of employment relationships whenever and wherever there are managers and the managed.

What IR specialists can learn from all this is that poor management is indeed one of the ‘‘drivers’’ of employment relationship conflict; that organizational and workplace innovations can improve employee attitudes and performance, enhance organizational performance, and reduce employment relationship conflict; that a unity of interests among employers and employees can be formed and can continue in place, at least for a while; that institutional mechanisms such as unions and legislation for dealing with workplace conflict sometimes exacerbate that conflict or result in the fomenting of conflict where little or none would have existed otherwise; and that, yes, it is possible to improve management and thereby contribute to the reduction of conflict in the employment relationship.

In order for IR and HR scholars to learn more from each other, especially about their perspectives on employment relationship conflict, it would be helpful if they paid more attention to research published in journals and other outlets outside of their respective narrow areas of interest. It would be even more helpful if these scholars chose to publish in each

7 The declining incidence of union representation elections can also be taken as evidence supporting the HR perspective on employment relationship conflict.

D. Lewin / Human Resource Management Review 11 (2001) 453–485 479

other’s journals and, as well, if they incorporated into their teaching articles, cases, and related material from the ‘‘other’’ perspective. For IR and HR practitioners to learn more from each other, if would be helpful if they used their professional associations and contacts to share ‘‘best’’ but also ‘‘worst’’ and ‘‘average’’ practices. IR practitioners would do well to participate in occasional management-oriented HR conferences and seminars, and similarly the other way round for HR practitioners. Perhaps by heeding this advice, IR and HR scholars and practitioners will come to learn that they do in fact share certain premises and perspectives on conflict in the employment relationship and, more fundamentally, that each dominant perspective has something valuable to offer the other. Such a potential increase in IR and HR knowledge transfer provides the proper note on which to conclude this inquiry into two important alternative perspectives on conflict in the employment relationship.

Acknowledgments

Appreciation is expressed to Professor Bruce E. Kaufman for his comments on earlier versions of this paper, and to participants in the UCLA Anderson School Faculty Lecture Series for their comments on a presentation in December 1998, which contained the core ideas elaborated in this paper.

References

Aaron B. (1973). Final report of the California Assembly Advisory Council on Public Employee Relations. Report and proposed statute. Sacramento, CA: California Assembly Advisory Council, March 15.

Aaron B., Grodin J. R., & Stern J. L. (Eds.) (1979). Public sector bargaining. Madison, WI: Industrial Relations Research Association.

Applebaum, E., Bailey, T., Berg, P., & Kalleberg, A. (2000). Manufacturing advantage: why high-performance performance work systems pay off. Washington, DC: Economic Policy Institute.

Applebaum, E., & Batt, R. (1994). The new American workplace: transforming work, systems in the United States. Ithaca, NY: ILR Press.

Asch, B. J., & Warner, J. T. (1997). Incentive systems: theory and evidence. In: D. Lewin, D. J. B. Mitchell, & M. A. Zaidi (Eds.), The human resource management handbook, Part I (pp. 175–215). Greenwich, CT: JAI Press.

Ashenfelter, O., & Johnson, G. E. (1969). Bargaining theory, trade unions, and industrial strike activity. American Economic Review, 59, 35–49.

Bakke E. W., Kerr C., & Anrod C. W. (Eds.) (1967). Unions, management and the public (3rd ed.). E. New York: Harcourt Brace Jovanovich.

Batt, R. (1999). Work organization, technology, and performance in customer service and sales. Industrial and Labor Relations Review, 52, 539–564.

Becker, B., & Huselid, M. (1999). Overview: strategic human resource management in five firms. Human Resource Management, 38, 287–301.

Bendix, R. (1956). Work and authority in industry. New York: Wiley. Boroff, K. E., & Lewin, D. (1997). Loyalty, voice and intent to exit a union firm: a conceptual and empirical

analysis. Industrial and Labor Relations Review, 51, 50–63. Brown, C., & Reich, M. (1989). When does union–management cooperation work? A look at NUMMI and GM-

Van Nuys. California Management Review, 31, 26–37. Bruno, R., & Jordan, L. (1999). From high hopes to disillusionment: the evolution of worker attitudes at

D. Lewin / Human Resource Management Review 11 (2001) 453–485480

Mitsubishi Motors. In: D. Lewin, & B. E. Kaufman (Eds.), Advances in industrial and labor relations, (9, pp. 153–182). Greenwich, CT: JAI Press.

Bureau of National Affairs. (1995). Basic patterns of union contracts. Washington, DC: BNA. Chamberlain, N. W. (1951). Collective bargaining. New York: McGraw-Hill. Childs, B. (1989). New United Motor: an American success story. In: Proceedings of the spring meeting, Industrial

Relations Research Association (pp. 453–460). Madison, WI: Industrial Relations Research Association. Commission on the Future of Worker Management Relations. (1995). Final report and recommendations.

Washington, DC: US Government Printing Office. Commons, J. R. (1934). Institutional economics vols. 1/2. New York: Macmillan. Cotton, J. L. (1993). Employee involvement: methods for improving performance and work attitudes. Newbury

Park, CA: Sage. Cutcher-Gershenfeld, J. (1991). The impact of economic performance of a transformation in workplace relations.

Industrial and Labor Relations Review, 44, 241–260. Davis L. E., & Cherns A. B. (Eds.) (1975). The quality of working life vols. 1/2. New York: Free Press. Delaney, J. T., Lewin, D., & Ichniowski, C. (1989). Human resource policies and practices in American firms.

Washington, DC: Bureau of Labor–Management Relations and Cooperative Programs, US Department of Labor (BLMR no. 137).

Denenberg, R. V., & Braverman, M. (1999). The violence-prone workplace. Ithaca, NY: Cornell University Press. Dessler, G. (1997). Human resource management (7th ed.). Upper Saddle River, NJ: Prentice-Hall. Dubin, R. (1954). Constructive aspects of industrial conflict. In: A. Kornhauser, R. Dubin, & A. M. Ross (Eds.),

Industrial conflict (pp. 37–47). New York: McGraw-Hill. Dunlop, J. T. (1957). Industrial relations systems. New York: Holt, Rinehart, and Winston. Eaton, A. E., & Keefe, J. H. (1999). Introduction and overview. In: A. E. Eaton, & J. H. Keefe (Eds.), Employment

dispute resolution and worker rights in the changing workplace (pp. 1–26). Champaign, IL: Industrial Relations Research Association.

Eaton, A. E., & Voos, P. B. (1994). Productivity-enhancing innovations in work organization, compensation, and employee participation in the union versus the nonunion sectors. In: D. Lewin, & D. Sockell (Eds.), Advances in industrial and labor relations, (vol. 6, pp. 63–109). Greenwich, CT: JAI Press.

Eaton, A. E., Voos, P. B., & Kim, D. (1997). Voluntary and involuntary aspects of employee participation in decision making. In: D. Lewin, D. J. B. Mitchell, & M. A. Zaidi (Eds.), The human resource management handbook, Part I (pp. 63–85). Greenwich, CT: JAI Press.

Edwards, R. (1997). Alternative regulatory approaches to protecting employees’ workplace rights. In: B. E. Kaufman (Ed.), Government regulation of the employment relationship (pp. 403–427). Madison, WI: Indus- trial Relations Research Association.

Estreicher, S. (2000). Nonunion employee representation: a legal/policy perspective. In: B. E. Kaufman, & D. G. Taras (Eds.), Nonunion employee representation: history, contemporary practice, and policy (pp. 196–220). Armonk, NY: M.E. Sharpe.

Feuille, P., & Delaney, J. T. (1992). The individual pursuit of organizational justice: grievance procedures in nonunion workplaces. In: G. R. Ferris, & K. M. Rowland (Eds.), Research in personnel and human resource management, (vol. 10, pp. 187–232). Greenwich, CT: JAI Press.

Finkin, M. (1992). Labor law scholarship: a critical survey. In: D. Lewin, O. S. Mitchell, & P. D. Sherer (Eds.), Research frontiers in industrial relations and human resources (pp. 525–556). Madison, WI: Industrial Relations Research Association.

Franke, R., & Kaul, J. (1978). The Hawthorne experiments: first statistical interpretation. American Sociological Review, 43, 623–643.

Freeman, R. B., & Medoff, J. L. (1984). What do unions do? New York: Basic Books. Freeman, R. B., & Rogers, J. (1999). What workers want. Ithaca, NY: Cornell University Press. Gerhart, B., Milkovich, G. T., & Murray, B. (1992). Pay, performance, and participation. In: D. Lewin, O. S.

Mitchell, & P. D. Sherer (Eds.), Research frontiers in industrial relations and human resources (pp. 193–238). Madison, WI: Industrial Relations Research Association.

D. Lewin / Human Resource Management Review 11 (2001) 453–485 481

Gilbreth, F. B. (1911). Motion study. New York: Van Nostrand-Reinhold. Godard, J., & Delaney, J. T. (2000). Reflections on the ‘‘high performance’’ paradigm’s implications for industrial

relations as a field. Industrial and Labor Relations Review, 53, 482–502. Goodman, P. A. (1979). Assessing organizational change: the Rushton quality of work experiment. New York:

Wiley. Hambrick, D. C. (1987). Top management team: key to strategic success. California Management Review, 30,

88–108. Hambrick, D. C. (1994). Top management groups: a conceptual integration and reconsideration of the ‘team’

label. In: L. L. Cummings, & B. M. Staw (Eds.), Research in organizational behavior, (vol. 16, pp. 171–213). Greenwich, CT: JAI Press.

Hammer, M., & Champy, J. (1993). Reengineering the corporation. New York: Harper Business. Heneman, R. L., & von Hippel, C. (1997). The assessment of job performance: focusing attention on context,

process, and group issues. In: D. Lewin, D. J. B. Mitchell, & M. A. Zaidi (Eds.), The human resource management handbook, Part III (pp. 79–109). Greenwich, CT: JAI Press.

Hewitt Association. (1989). Compensation research and practices. Lincolnshire, IL: Hewitt Associates. Hicks, J. R. (1932). A theory of wages. New York: Macmillan. Hirschman, A. O. (1970). Exit, voice and loyalty. Cambridge, MA: Harvard University Press. Hoerr, J. (1990). Sharpening minds for a competitive edge. Business Week, 44, 2 (December 17). Horvath, W. L. II, & Svyantek, D. J. (1998). Participative management in union settings: lessons from Saturn. In:

D. Lewin, & B. E. Kaufman (Eds.), Advances in industrial and labor relations, (vol. 8, pp. 119–138). Greenwich, CT: JAI Press.

Hoxie, R. F. (1921). Trade unionism in the United States. New York: Appleton-Century-Crofts. Huselid, M. A. (1995). The impact of human resource management practices on turnover, productivity, and

corporate financial performance. Academy of Management Journal, 38, 635–672. Huselid, M. A., Jackson, S. E., & Schuler, R. S. (1997). Technical and strategic human resource management

effectiveness as determinants of firm performance. Academy of Management Journal, 40, 171–188. Ichniowski, C. (1992). Human resource practices and productive labor–management relations. In: D. Lewin,

O. S. Mitchell, & P. D. Sherer (Eds.), Research frontiers in industrial relations and human resources (pp. 239–271). Madison, WI: Industrial Relations Research Association.

Ichniowski, C., Shaw, K., & Prennushi, G. (1997). The effects of human resource management practices on productivity: a study of steel finishing lines. American Economic Review, 87, 291–313.

Jacoby, S. M. (1985). Employing bureaucracy: managers, unions and the transformation of work in American industry, 1900–1945. New York: Columbia University Press.

Jacoby, S. M. (1997). Modern manors: welfare capitalism since the new deal. Princeton, NJ: Princeton Univer- sity Press.

Jones, D. C., Kato, T., & Pliskin, J. (1997). Profit sharing and gain sharing: a review of theory, incidence, and effects. In: D. Lewin, D. J. B. Mitchell, & M. A. Zaidi (Eds.), The human resource management handbook, Part I (pp. 153–173). Greenwich, CT: JAI Press.

Joshi, A. (1999). Networking employees for success in global teams: a case study. International Human Resource Information Management Journal, 3, 34–39.

Kaminski, M. (1999). Impact of teams on the grievance procedure: issues of voice and union commitment. In: A. E. Eaton, & J. H. Keefe (Ed.), Employment dispute resolution and worker rights in the changing workplace (pp. 219–246). Champaign, IL: Industrial Relations Research Association.

Karsh, B. (1956). Diary of a strike. Urbana, IL: University of Illinois Press. Katz, H. C., Kochan, T. A., & Gobeille, K. R. (1983). Industrial relations performance, economic performance,

and QWL programs: an interplant analysis. Industrial and Labor Relations Review, 37, 3–17. Katz, H. C., Kochan, T. A., & Weber, M. R. (1985). Assessing the effects of industrial relations and quality of

work life efforts on organizational effectiveness. Academy of Management Journal, 28, 509–527. Katzenbach, J. R., & Smith, D. K. (1993). The wisdom of teams: creating the high-performance organization.

Boston, MA: Harvard Business School Press.

D. Lewin / Human Resource Management Review 11 (2001) 453–485482

Kaufman, B. E. (1981). Bargaining theory, inflation, and cyclical strike activity in manufacturing. Industrial and Labor Relations Review, 34, 333–355.

Kaufman, B. E. (1983). The determinants of strikes over time and across industries. Journal of Labor Research, 4, 159–174.

Kaufman, B. E., & Lewin, D. (1998). Is the NLRA still relevant to today’s economy and workplace? Labor Law Journal, 49, 1113–1126.

Kaufman, B. E., Lewin, D., & Fossum, J. A. (2000). Nonunion employee involvement and participation programs: the role of employee representation and the impact of the NLRA. In: B. E. Kaufman, & D. G. Taras (Eds.), Nonunion employee representation: history, contemporary practice, and policy (pp. 259–286). Armonk, NY: M.E. Sharpe.

Kerr, C., & Siegel, A. (1954). The interindustry propensity to strike: an international comparison. In: A. Korn- hauser, A. Siegel, R. Dubin, & A. Ross (Eds.), Industrial conflict (pp. 189–212). New York: McGraw-Hill.

Kleiner, M. M., & Bouillon, M. (1988). Providing business information to production workers: correlates of compensation and profitability. Industrial and Labor Relations Review, 41, 605–617.

Kleiner, M. M., & Han, T. S. (1997). Employee communications: an overview. In: D. Lewin, D. J. B. Mitchell, & M. A. Zaidi (Eds.), The human resource management handbook, Part I (pp. 87–110). Greenwich, CT: JAI Press.

Kochan, T. A. (1973). Correlates of state public employee bargaining laws. Industrial Relations, 12, 322–337. Kochan, T. A. (1995). Using the Dunlop report to achieve mutual gains. Industrial Relations, 34, 50–366. Kochan, T. A., & Osterman, P. (1995). The mutual gains enterprise. Boston, MA: Harvard Business School Press. Kornhauser A., Dubin R., & Ross A. M. (Eds.) (1954). Industrial conflict. New York: McGraw-Hill. Kruse, D. L., & Blasi, J. R. (1997). Employee ownership, employee attitudes, and firm performance. In: D. Lewin,

D. J. B. Mitchell, & M. A. Zaidi (Eds.), The human resource management handbook, Part I (pp. 113–151). Greenwich, CT: JAI Press.

Lawler, E. E. III (1986). High-involvement management. San Francisco: Josey-Bass. Lawler, E. E. III (1990). Strategic pay. San Francisco: Josey-Bass. Lawler, E. E. III, & Mohrman, S. (1987). Quality circles: after the honeymoon. Organizational Dynamics, 15, 42–54. Lee, M. B., & Johnson, N. B. (1998). Business environment, high-involvement management, and firm performance

in Korea. In: D. Lewin, & B. E. Kaufman (Eds.), Advances in industrial and labor relations, (vol. 8, pp. 67–87). Greenwich, CT: JAI Press.

Levine, D., & Tyson, L. (1990). Participation, productivity, and the firm’s environment. In: A. S. Blinder (Ed.), Paying for productivity (pp. 183–224). Washington, DC: Brookings.

Lewin, D. (1987). Dispute resolution in the nonunion firm: a theoretical and empirical analysis. Journal of Conflict Resolution, 31, 465–502.

Lewin, D. (1992). Grievance procedures in nonunion workplaces: an empirical analysis of usage dynamics, and outcomes. Chicago-Kent Law Review, 66, 823–844.

Lewin, D. (1997). Workplace dispute resolution. In: D. Lewin, D. J. B. Mitchell, & M. A. Zaidi (Eds.), The human resource management handbook, Part II (pp. 197–218). Greenwich, CT: JAI Press.

Lewin, D. (1999). Theoretical and empirical research on the grievance procedure and arbitration: a critical review. In: A. E. Eaton, & J. H. Keefe (Eds.), Employment dispute resolution and worker rights in the changing workplace (pp. 137–186). Champaign, IL: Industrial Relations Research Association.

Lewin, D. (2000, May). Dispute resolution in nonunion organizations: key empirical findings. Paper presented to the 53rd NYU Labor Law Conference, New York City.

Lewin, D. (2000, April). Incentive compensation in the public sector. Paper presented to the Public Sector Labor– Management Relations Conference, Chicago, IL.

Lewin, D., & Boroff, K. E. (1996). The role of loyalty in exit and voice: a conceptual and empirical analysis. In: D. Lewin, B. Kaufman, & D. Sockell (Eds.), Advances in industrial and labor relations, (7, pp. 69–96). Greenwich, CT: JAI Press.

Lewin, D., Feuille, P., Kochan, T. A., & Delaney, J. T. (1988). Public sector labor relations: analysis and readings (3rd ed.). Lexington, MA: Lexington Books.

D. Lewin / Human Resource Management Review 11 (2001) 453–485 483

Lewin, D., & Mitchell, D. J. B. (1995). Human resource management: an economic approach (2nd ed.). Cincinnati, OH: South-Western.

Lewin, D., & Peterson, R. B. (1999). Behavioral outcomes of grievance activity. Industrial Relations, 38, 554–576.

MacDuffie, J. P. (1995). Human resource bundles and manufacturing performance: organizational logic and flexible production systems in the world auto industry. Industrial and Labor Relations Review, 48, 197–221.

Maslow, A. H. (1943). A theory of human motivation. Psychological Review, 50, 370–396. Mayo, E. (1933). The human problems of an industrial civilization. New York: Macmillan. McCabe, D. M., & Lewin, D. (1992). Employee voice: a human resource management perspective. California

Management Review, 34, 112–123. McDonald, D., & Smith, A. (1995). A proven connection: performance management and business results.

Compensation and Benefits Review, 27, 59–64. McGregor, D. (1960). The human side of enterprise. New York: McGraw-Hill. Mitchell, D. J. B., Lewin, D., & Lawler, E. E. III (1990). Alternative pay systems, firm performance, and

productivity. In: A. S. Blinder (Ed.), Paying for productivity (pp. 15–88). Washington, DC: Brookings. Moore, B., & Ross, T. (1978). The Scanlon way to improved productivity. New York: Wiley. Morishima, M. (1991). Information sharing and firm performance in Japan. Industrial Relations, 30, 37–61. Morishima, M. (1992). Information-sharing and collective bargaining in Japan: effects on wage negotiation.

Industrial and Labor Relations Review, 45, 469–486. Olson-Buchanan, J. B. (1996). Voicing discontent: what happens to the grievance filer after the grievance? Journal

of Applied Psychology, 81, 52–63. O’Reilly, C. A. (1989). Corporations, culture, and commitment: motivation and social control in organizations.

California Management Review, 31, 9–25. O’Reilly, C. A., & Chatman, J. (1996). Culture as social control: corporations, cults, and commitment. In: B. Staw,

& L. L. Cummings (Eds.), Research in organizational behavior, (vol. 18, pp. 263–287). Greenwich, CT: JAI Press.

O’Toole J. (1973). Work in America. Report of a special task force to the Secretary of Health, Education, and Welfare. Cambridge, MA: MIT Press.

O’Toole J. (Ed.) (1974). Work and the quality of life. Cambridge, MA: MIT Press. Perlman, S. A. (1928). A theory of the labor movement. New York: Macmillan. Pfeffer, J. (1998). The human equation. Boston, MA: Harvard Business School Press. Pruitt, D., & Rubin, J. (1986). Social conflict. New York: Random House. Roethlisberger, F. J., & Dickson, W. J. (1939). Management and the worker. Cambridge, MA: Harvard

University Press. Rubenstein, S., Bennett, M., & Kochan, T. A. (1993). The Saturn partnership: co-management and the reinvention

of the local union. In: B. E. Kaufman, & M. M. Kleiner (Eds.), Employee representation: alternatives and future directions (pp. 339–370). Madison, WI: Industrial Relations Research Association.

Sheppard, B. H., Lewicki, R. J., & Minton, J. W. (1992). Organizational justice. Lexington, MA: Lexington Books. Simmons, J., & Mares, W. (1985). Working together: employee participation in action. New York: New York

University Press. Stone, K. V. W. (1999). Employment arbitration under the federal arbitration act. In: A. E. Eaton, & J. H. Keefe

(Eds.), Employment dispute resolution and worker rights in the changing workplace (pp. 27–65). Champaign, IL: Industrial Relations Research Association.

Tannenbaum, A. S. (1966). Social psychology of the work organization. Belmont, CA: Wadsworth. Taylor, F. W. (1911). The principles of scientific management. New York: Harper and Brothers. Taylor, G. W. (1967). Public employment: strikes or procedures? Industrial and Labor Relations Review, 20,

617–625. Townsend, A. M., DeMarie, S. M., & Hendrickson, A. R. (1998). Virtual teams: technology and the workplace of

the future. Academy of Management Executive, 12, 17–29. Toyota Motor Manufacturing USA. (1988). Team member handbook. Fremont, CA: Toyota Motor Corporation.

D. Lewin / Human Resource Management Review 11 (2001) 453–485484

Tushman, M. L., & O’Reilly, C. A. (1996). Ambidextrous organizations: managing evolutionary and revo- lutionary change. California Management Review, 38, 8–30.

Ulrich, D. (1997). Human resource champions: the next agenda for adding value and delivering results. Boston, MA: Harvard Business School Press.

Ulrich D., Losey M. R., & Lake G. (Eds.) (1997). Tomorrow’s HR management. New York: Wiley. US General Accounting Office. (1995). Employment discrimination: most private sector employers use alternative

dispute resolution. Washington, DC: GAO. Report no. HEHS-95-150 employment discrimination. Walton, M. (1986). The Deming management method. New York: Perigree Books, Putnam. Weber, J. (1992, May 4). A big company that works. Business Week, 46, 37–38. Weil, D. (1997). Implementing employment regulation: insights on the determinants of regulatory performance.

In: B. E. Kaufman (Ed.), Government regulation of the employment relationship (pp. 429–474). Madison, WI: Industrial Relations Research Association.

Weiler, P. C. (1990). Governing the workplace: the future of labor and employment law. Cambridge, MA: Harvard University Press.

Wellins, P., & George, J. (1991). The key to self-directed teams. Training and Development Journal, 26–31.

D. Lewin / Human Resource Management Review 11 (2001) 453–485 485