Assignment

367

T

Industrial and Labor Relations Review, Vol. 63, No. 3 (April 2010). © by Cornell University. 0019-7939/00/6303 $05.00

REFRAMING THE REPRESENTATION DEBATE:

GOING BEYOND UNION AND NON-UNION OPTIONS

MORDEHAI MIRONI*

The author argues for replacing the traditional union/non-union dichotomy with a more dynamic, nuanced typology of employment relationships and for viewing the contemporar y workplace as an organization with multiple distinct but interacting employment models. Such a framework has clear potential advantages for the public policy debate as well as for those making a strategic choice of employment relations models. Moreover, the framework benefits researchers who may use it to refine their empirical studies of the diverse forms and combinations of representation and voice in contemporar y workplaces in order to determine which perform best in different settings for different work groups. The real alternative to union organization, the au- thor suggests, is not “no unions” or lack of representation, but some innovative form of representation and voice. Whether these alternatives are sustainable and powerful enough to do the job remains an empirical question that the field should be tackling with greater clarity and vigor.

* Mordehai Mironi is an associate professor of law at Haifa University. The author would like to thank Professor Thomas Kochan, who read earlier drafts and provided valuable comments, and to his research as- sistant, Ms. Irena Nutenko, for superb work.

he study of employment relations in unorganized or non-unionized enter- prises compared to unionized enterprises has received a great deal of scholarly attention, and much of the field of industrial relations has been implicitly or explicitly influenced by it, whether in terms of the relevant concep- tualization (Wheeler and McClendon 1998; Hyman 2001), empirical research (Freeman and Medoff 1984; Hirsch and Macpherson 2003; Booth and Byron 2004; Mishel and Walters 2003; Doucouliagos and Laroche 2003; Iverson and Currivan 2003; Kaufman and Bennett 2006), or public policy debate (Budd 2004; Howell 2005). This is particularly true in the United States, where unionism and labor-management policies tend to be especially polarized and dominated by rigid

ideological positions. This may be one of the reasons why substantial changes in U.S. labor law are so infrequent and difficult to achieve. Yet this simple union/non-union dichotomy1 (Wheeler and McClender 1998) captures neither the diversity of the models of voice and representation observed around the world nor recent innovations in voice and representation in the United States it- self. What is needed, therefore, is a richer, more nuanced approach for conceptualizing, evaluating, and choosing forms of voice and

1 Labor law jurisprudence has also suffered from be- ing premised on this dichotomous framework. American labor law is a case in point. Two examples illustrating the drawbacks that follow from this simplistic framework are the difficulty in applying the available protection for concerted activities to employees who press their own or their colleagues’ grievances in a non-unionized employment relations model (Gerson and Brightbritt 1983; Ciarrocca 1987) and the difficulty in reconciling cooperative programs with the prohibition against employer domination of labor organizations (Kochan, Katz, and McKersie 1986:234–35).

368 INDUSTRIAL AND LABOR RELATIONS REVIEW

representation that are suited to the modern work force and to contemporary employment relations. Such an approach may enable both sides of the divide to break out of the ideological constraints that are holding back labor policy reforms.

My purpose in this paper is to review the range of forms of voice and representation that exist in various settings in the United States, as well as in other parts of the world, and to propose a new, more nuanced typol- ogy for structuring theor y, research, and policy debates.

The need for a fresh approach is obvious. Policies to support or promote unionization are failing to realize their intended objec- tives, to use Budd’s (2004) framework, of providing a voice at work that improves the economic conditions of employees and enhancing efficiency and equity in econo- mies and societies. Worldwide decline in unionization and increasing workplace diversity have led to the decline of indus- trial relations as a field of study, in part because scholars of traditional industrial relations have been reluctant to broaden their analysis to consider alternative forms of voice, representation, and governance in employment relations. Moreover, many in the U.S. labor movement view collective bar- gaining as the only legitimate form of worker representation, and they chastise scholars and policy-makers who advocate consider- ing alternative forms. As a result, current policy debates end up focusing exclusively on reforms that remain within the ambit of the National Labor Relations Act, and they inevitably get mired in an instinctive business-versus-labor ideological battle— one that for at least a quarter-centur y has failed to produce results. On both research and policy grounds, therefore, we need to rethink the simple dichotomy that has historically been used to study and discuss policies relating to worker representation.

In order to open up the normative public policy debate and to enrich our conceptual- ization of both that and the strategic choice of employment relations models, I propose in this article to embark on two different routes. First, I suggest abandoning the histori- cally used dichotomous union/non-union

paradigm, and adopting instead a multidi- mensional and continuum-based typology of employment relations models that is more varied, nuanced, and dynamic. Second, I of- fer a sharper and more sensitive resolution of the unit of investigation, one that views the modern workplace not as a monolithic organization but as one that contains within itself multiple types of employment relations models.

The Terminological Problem

The terms traditionally used in the literature—”non-unionized” and “union- ized”—are inadequate to describe a model of employment relations of a particular work force or an enterprise. In the U.S. literature, “unionized” usually refers to situations in which a majority or threshold percentage2 of the workers concerned are members of a trade union, thus enabling the latter to become a collective bargaining representa- tive or agent of that work force. Sometimes the term refers to an enterprise in which the employer and its work force are bound by a collective agreement. Both references are inadequate and can be misleading when used to characterize the collective union representation employment model.

There are two contexts in which refer- ence to membership in a trade union as a distinguishing aspect of the collective union representation employment model is inac- curate. First, in industrial relations systems,

2 Different legal regimes require different degrees of support in order for the union to be recognized as a collective bargaining agent or as a party to a collective agreement. The required level of support may also var y between sole enterprise and industr y-wide collective agreements. For example, in the United States and Canada the union is required to achieve majority sup- port in order to be recognized as a bargaining agent (Getman, Pogrebinand, and Gregory 1999). In contrast, the Collective Agreements Law (1957) that regulates collective bargaining in Israel defines the “representa- tive union” for an enterprise collective agreement as the union that has the largest number of members and no fewer than 1/3 of the workers in the bargaining unit. There is no minimum level of support that is required under Israeli law for an industr y-wide or nationwide collective agreement.

REFRAMING THE REPRESENTATION DEBATE 369

unions have traditionally provided a broad range of social ser vices, such as medical care, mutual aid, loans, and pensions for their members. Second, it is inapplicable to industrial relation systems, such as those in England, the United States, continental Europe, and Israel, where formal or infor- mal craft unionism, professional association membership, or closed shop arrangements give certain membership organizations some degree of control over job opportunities, licensing, grading, certification, professional ethics, or malpractice insurance.3

In both instances one may find a signifi- cant number of people who carr y union or membership cards, but who neither seek nor benefit from collective bargaining representa- tion.4 Among them are employees who work under what is commonly referred to as a non- union employment relations model, as well as government officials and self-employed persons. Note that the opposite situation exists in industrial relations systems that espouse an exclusive representation model in collective labor relations, particularly in “right-to-work” states. There, a substantial proportion of the workers in a bargaining unit may not be members of the union although they are covered by a collective bargaining agreement.

The fact that a collective agreement applies to the enterprise and its work force cannot be used to differentiate between what we usu- ally refer to as unionized and non-unionized models of employment. This is especially true in Europe, where employers’ associations play an important role in establishing working conditions (Jacobi, Keller, and Müller-Jentsch 1998; Katz and Darbishire 2000; Budd 2004). These employers’ associations sign nation- wide or industr y-wide collective agreements

that apply directly5 and indirectly,6 through various legal instruments and doctrines, to all enterprises, including those operating ac- cording to a non-unionized employment rela- tions model, and to their workers, whether or not there is union representation in the particular enterprise.

In the case of direct applicability, the ex- istence of collectively negotiated terms and conditions of employment in the enterprise stems solely from the employer’s membership in the employers’ association that is party to the agreement. Therefore, in industrial relation systems characterized by sectoral bargaining, collective agreement coverage may be divorced both from union member- ship and from the existence of collective labor relations.

In the case of indirect applicability, the terms and conditions of employment of col- lectively negotiated industr y-wide or nation- wide agreements may apply to the enterprise and its work force even though neither side is organized: the employer may not necessarily be a member of an employers’ association, and few if any of the employees covered may be unionized—or, indeed, have the least de- sire for any collective representation. Thus, from a workplace governance perspective,

3 The building trades in the United States and the doctors’ unions in Europe are good examples of such situations.

4 One can envision a hospital employing a few hun- dred doctors, under individual employment contracts, all clinging to their union membership and paying their union dues. Ostensibly, the hospital should be considered highly unionized, but from a workplace governance standpoint, the hospital is a typical non- unionized enterprise.

5 Direct applicability refers to a situation in which the terms of an industr y-wide or nationwide agreement that has been negotiated by an employers’ association applies by law to all the relevant workers who are employed by all the employers who are members of the employers’ association. In such cases it is necessar y to prove neither that the particular enterprise joined the employers’ association in order to be represented by it in collec- tive bargaining, nor that the particular enterprise or a requisite percentage of its employees wishes to operate pursuant to the unionized employment relation model.

6 Indirect applicability is based on the practice of government-issued extension orders. In European-style industrial relation systems, such as those existing in Belgium, Germany, Portugal, France, and Israel, the industr y-wide or nationwide collective agreements may be made applicable by the national government in whole or in part to the entire industry or to a geographic region of the countr y. The effect of such extension orders is to apply the terms of the relevant collective agreement to enterprises that are not members of the employers’ association and to their employees (who themselves are not necessarily union members). (Jacobi, Keller, and Müller-Jentsch 1998; Barreto and Nauman 1998; Katz and Darbishire 2000.)

370 INDUSTRIAL AND LABOR RELATIONS REVIEW

the result is an enterprise that, on the one hand, is legally bound by certain terms and conditions of employment that are derived from a collective agreement negotiated by an employers’ association and a union, and, on the other hand, operates on a daily basis under a typical non-unionized employment relations model without the presence of an active union or any other form of collective representation.

In sectoral systems of industrial relations, as many as 80–95% of the workers in the industr y may be covered by the collective agreement, even though only a small minority of them—29% in Germany, for instance, and 9% in France—are union members (OECD 1997; Budd 2004:121). Thus, the applicability of the industr y-wide or nationwide collective agreement or parts thereof is dependent neither on membership in a collectivity (a union or an employers’ association) nor on the employment relations regime in the enterprise. The norms stemming from an industr y-wide collective agreement or an extension order will apply to the enterprise and affect the individual worker’s employ- ment conditions much like protective labor legislation, for which extension orders ser ve as a substitute mode of regulation.7 Nonethe- less, since the enterprise may function with no collective representation and possess all the characteristics of what has commonly been referred to as a non-unionized em- ployment relations model, being subject to such industr y-wide norms does not change the nature or dynamics of the employment relationship.

Furthermore, the European model of statutor y works councils, especially those with decision-making powers (Frege 2002; Müller-Jentsch 1995; Weiler 2004; Goldman 2005), as well as recent developments in the United States where NGOs have negotiated and advocated for improved working condi- tions (Kochan 2005; Erickson et al. 2002), demonstrate that some form of collective

representation may be in effect even where none of the commonly used indicators is present. First, the employees are not union- ized; second, the employer is not a member of an employers’ association; third, there is no collective agreement that applies to the enterprise and its work force; fourth, there is no union presence. The opposite may also hold. As will be shown in the following pages, in today’s array of employment rela- tions models one may encounter collective employment relations that do not provide for representation, and individual non-union- ized employment relations or workplace governance models that are characterized by active employee representation.

In sum, both a new terminology and a new conceptual framework are urgently needed.

The Proposed New Terminology and Conceptual Framework

The preceding discussion suggests that the search for a better conceptual framework may not be limited to inventing a new terminol- ogy. Only a multidimensional framework will capture the richness and uniqueness of many developments and innovations in employ- ment relations models and forms of represen- tation. Therefore, I propose a framework that characterizes employment relations along two dimensions: individual versus collective, and direct versus represented. Combined, the two dimensions can accurately portray not only the two prototypes of employment relations models but also the different options that exist between and alongside them.

The Individual-Collective Dimension

Using terms such as “individual” and “col- lective” overcomes part of the terminological problem. These terms are both conceptually clearer and richer than “union” and “non- union” and carr y no negative connotations. For instance, the term “individual employ- ment relations model” is clearer than Budd’s “human resource management,” which he uses to describe a particular model of work- place governance (Budd 2004). And it is more accurate, since, at least in Britain, advanced human resource management policies and

7 Extension orders have the same normative effect as the collective agreement. The provisions of both types of instruments automatically affect individual employment contracts and may not be waived by the individual employee.

REFRAMING THE REPRESENTATION DEBATE 371

union sector than in the non-union sector (Sisson 1993; Beaumont and Hunter 2003). Using the term “individual employment relations model” avoids the negative con- notation associated with terms such as “non- organized,” “management control” (Weiler 1990), and “managerialism” (Clegg 1979).8

As to the definitional aspect, the new terms offer narrower and more process-oriented definitions and differentiating criteria. What we commonly refer to as “non-unionized” and “unionized” enterprises are in fact workplaces where employer–employee relations are, re- spectively, primarily individual and primarily collective in nature. The main differences between the two are in the processes by which employment rights and duties are established, modified, administered, and enforced.

Under the collective employment rela- tions model, working conditions tend to be uniform either across the board or for a particular group of employees, and they are established (and when necessar y modified) with some form of participation by employees acting as a collective entity. In the extreme form of the model, the majority of and most important working conditions are spelled out in a detailed collective agreement, which is the outcome of bilateral negotiations between management and representatives of the employees. These workers’ representatives are also usually involved in the day-to-day adjustment, administration, and enforce- ment of these negotiated working conditions. Finally, the collectivity acts as a constraining factor on many management decisions that may affect the workers concerned. What is being measured here is the scope of issues and the form and intensity of the collectivity’s participation in, for example, negotiation and day-to-day administration, as against infor- mation and consultation. Therefore, at the high collective end of the continuum, there is little room for unilateral decision-making and differential treatment by management. The result is a significant encroachment on

management discretion, as well as on the autonomy of individual employees.

By contrast, under the individual employ- ment relations model, terms and condi- tions of employment are determined on an individual basis. In the extreme form of this model, these terms and conditions of employment are established unilaterally, changed and administered by management alone, at its discretion. Save for outside le- gal constraints and the ever-present fear of unionization, these management decisions are business-driven. They are motivated by internally held beliefs regarding good human resource management, and they are intended primarily to attract, maintain, and motivate the work force. In the weaker form of the model, some of the terms and conditions of employment are established through nego- tiation, taking into account the individual employee’s bargaining power.

The role of unilateral decisions driven by internally held considerations of employment relations rather than by grudging compliance with union pressure (Hall 2006) needs to be further emphasized as a distinguishing ele- ment in the typology of enterprises. This is so since quite often an enterprise that operates under the “individual employment relations model” is more reluctant than its “collective employment relations model” counterpart to take actions that worsen employees’ working conditions. Equally, it may be more commit- ted to job security (Osterman 1987; Foulkes 1980) and to the welfare of its employees. Yet the two types of enterprise differ completely in the reasons and motives underlying their adoption of such policies.

The same idea holds for employer- promulgated voice mechanisms, non-union employee representation programs, and dispute resolution procedures that provide due process. As Budd (2004) and others have noted, the employees’ voice and other non- union employee representation programs are intended to directly and indirectly improve productivity by harnessing and sharing the knowledge and experience of the workers, by improving coordination (thus enhancing individual and organizational performance), and by ensuring that the interests of the employees in equitable terms and conditions

8 It can be argued that the decision to use the broader and conceptually richer term “individual employment relations” rather than “management control” or “mana- gerialism” may conceal the underlying power realities that are conveyed by these terms.

372 INDUSTRIAL AND LABOR RELATIONS REVIEW

of employment are factored into manage- ment decisions (Freeman and Lazear 1995; Kaufman and Levine 2000; Kaufman and Gottlieb-Taras 2000; Budd 2004). Dispute resolution procedures that are established by employers are aimed at improving perfor- mance by fostering a sense of fairness on the part of the employees. In addition, they have the potential to save the enterprise lawyers’ fees, management time, and other monetary and non-monetary costs associated with agency or court litigation (CPR 2002; Wheeler, Klaas, and Mahony 2004). Finally, employee voice and non-union employee representation programs, as well as dispute resolution proce- dures, also act as union substitutes (Kaufman and Gottlieb-Taras 2000; Budd 2004; Wheeler, Klaas, and Mahony 2004).

The improved working conditions, advanced human resource management practices, commitment to job security, employee voice and non-union employee representation programs, and dispute reso- lution procedures that may all be found in enterprises adhering to the individual em- ployment relations model share another common characteristic: namely, they are unilaterally9 introduced and may be unilater- ally dismantled if the employer believes that they are no longer needed or are too costly or risky (Kochan 2005).

The Direct-Represented Dimension

The representation aspect in the collective model is also important, and it is often over- looked. The individual (non-unionized) em- ployment relations model emphasizes direct relations between each individual employee and management. By contrast, the collective (unionized) employment relations model stresses a strong element of institutionalized collective representation.

Since independent and institutionalized collective representation is the raison d’etre of moving from the individual to the collective employment relations model, its sources of

power and unique role within the enterprise establishment need further elaboration. The representatives in a collective representation setting are usually regular employees of the enterprise, appointed by the group they rep- resent rather than by management, typically assisted by the external professional staff of an outside union.10 Once in office, the regu- lar employees who are union representatives enjoy special status and privileges vis-à-vis management. This special status derives not from their value to the enterprise as employ- ees, but from their role as representatives of the collectivity. In addition to having a role as regular employees, they have an “institu- tional personality,” which provides them with certain immunities that are backed by law and by the power of the collectivity. They are the only important functionaries acting with formal power and authority inside the enter- prise who are independent of its hierarchy.11

The involvement of professional outside union staff emphasizes another distinguish- ing aspect of the collective representation paradigm. Save for the continual presence of legal constraints, the individual employment relations model involves the autonomous relationship between two parties—manage- ment and the employee. In the collective employment relations model there are two additional parties: the in-house employees’ representatives and the officers of the outside union. Note that by marshaling professional and other resources, the union strengthens the employees’ representation. Finally, the outside union and its officers may have in- terests and goals that are unrelated to the enterprise and its work force, and it is even more removed, insulated, and immune from management’s control.

The Matrix

The discussion of the two dimensions of the employment relations models and their implications can be better appreciated by viewing the 2x2 matrix shown in Figure 1.

9 Budd argues that the statement that human re- source management is unilateral is definitional (Budd 2004:209).

10 In unions representing the building trades in the United States, it may be a full-time business agent.

11 Workers view the independence of the representa- tion as a crucial element (Freeman and Rogers 1999:7).

REFRAMING THE REPRESENTATION DEBATE 373

Cell 2, the individual-direct (ID) employment relations model, and Cell 4, the collective- represented (CR) employment relations model, are the two prototypes traditionally used, known respectively as non-unionized and unionized employment relations.

The Two Promising Cells

Cell 1, individual-represented (IR), and Cell 3, collective-direct (CD), are at the center of a number of contemporar y public policy debates, and in some ways are at the frontier of employment relations practice. Yet researchers and policy-makers assume that there is an inevitable tendency to gravitate either to Cell 2, individual-direct (ID), or to Cell 4, collective-represented (CR), and they continue to devote most of their attention to these two cells. In fact, as will be demon- strated, not only are there stable alternatives at various points on the ID–CR continuum, there are also interesting and at times com- plementar y alternatives outside it. Looking at the present state of unions and collective bargaining, Cell 1—individual-represented (IR)—and Cell 3—collective-direct (CD)— deser ve special attention. These two models are especially important today because of the increasing de-collectivization of negotiations of working conditions (Davies and Freedland 2004) and the increasing resort to non-union work force agreements as a way to deviate

from external norms such as statutor y provi- sions or awards (Katz and Darbishire 2000; Carley and Hall 2000; Collins 2001; Barnard and Deakin 2002; Kilpatrick 2003; Barnard, Deakin, and Hobbs 2003).

Cell 1—Individual-Represented (IR)

The IR12 individual-represented cell differs from the individual-direct (ID) cell (Cell 2) in that the former entails representa- tion of the employee by an independent expert—a lawyer, an agent, or a professional negotiator. To a large degree, this form of individual representation is likely to correct the information asymmetr y between man- agement and the individual employee, and to reduce the gap in bargaining power and expertise in negotiation and advocacy. The IR model has been prevalent for many years in the performing arts (Moskow 1970) and in professional sports (Bower 1977), as well as in the higher echelons of management. Recently, it has spread to highly professional employees and itinerant experts (Barely and Kunda 2004), as well as to regular employees, through worker centers (Fine 2006).

In the arts and in sports the common pat- tern is a basic collective agreement (CR),

12 This concept is different from Troy’s individual rep- resentation. Troy uses the concept to describe the free- market form of workplace governance (Budd 2004:85).

Represented

Direct

Collective Individual

Figure 1. The New Typology of Employment Relations

374 INDUSTRIAL AND LABOR RELATIONS REVIEW

supplemented by individual negotiations con- ducted by agents or attorneys who represent these highly paid employees (IR). This is an interesting two-tier structure of workplace governance (CR + IR). The first level follows the CR model, and it establishes minimal standards and work rules for all employees in the particular profession or industr y. Lay- ered on top of the CR model is the second model, IR, which is available for those who wish to have (and who can afford) individual representation. According to Moskow, in the performing arts there is typically a third level, which is a CR-type industr y-wide agreement. This collective agreement, negotiated by the union and the agents’ association, regulates the relationship between individual actors and the agents who represent them in nego- tiations with individual studios or producers.

Recognizing the existence of IR-type (Cell 1) situations is important. They represent an interesting, sophisticated, and in certain respects promising pattern of workplace governance. The IR-type highlights inde- pendent representation of the employee as an important paradigm, one that introduces new representation options for the growing number of contract workers, such as those now staffing much of the information tech- nology industr y (Barley and Kunda 2004). Moreover, this model is free of the strong constraints associated with unions and their detailed work rules and contract administra- tion procedures. Furthermore, it also avoids the need to organize workplaces one at a time, with all the attendant risks, conflicts, and legalistic delays that are commonly pres- ent in the organizing campaign and election process. Consequently, it may be particularly attractive to professional employees (nurses, doctors, lawyers, IT professionals, and so on) who are not tied to a specific employer for a long period of time and who cherish their individual identities, but who at the same time want to be compensated fairly and to work under standards consistent with accepted professional practice.13

The independent representation property of the IR model calls into question the alleged willingness of many non-unionized compa- nies to negotiate a tailor-made individual employment contract for every employee and the resistance voiced against the CR model due to its uniformity and rigidity. One can as- sume that if a majority of employees engaged by an ID enterprise (Cell 2), not merely senior executives and talented professionals, choose to transfer to an IR model (and consequently each one of them is represented by a shrewd negotiator or lawyer14 when negotiating the individual employment contract), the stra- tegic choice of employment relations model adopted by the enterprise might change. If the enterprise were able to change, it might refuse to hire those employees who decline to sign the employer-imposed standardized employment contract and insist on being rep- resented while negotiating their own terms and conditions of employment. If not, it is likely that the enterprise would abandon its union avoidance strategy and opt instead for the collective employment relations model (CR-type or Cell 4). So far, however, compa- nies have not had to confront this choice.15

Cell 3—Collective-Direct (CD)

Cell 3—CD—is probably the territor y of most cutting-edge employment relations practices and innovations in workplace gov- ernance. In terms of industrial democracy, the CD model brings to mind the ancient form of direct democracy (as opposed to modern representative democracy). One example of the straight CD model, the Aus- tralian “Enterprise Flexibility Agreement” (also referred to as the “non-union collective agreement” (Katz and Darbishire 2000)), was

13 The important issue of individual representation within a collective labor relations regime will not be dealt with in this paper.

14 One of the wonders of our time is the fact that the vast majority of employees choose not to be represented by a lawyer or other professional when negotiating the terms of their employment, which is one of the most important transactions of their life, and one with far- reaching consequences.

15 It has been suggested that had the Internet bubble lasted, probably we would have seen more and more employees engaging professionals to negotiate their individual employment contracts.

REFRAMING THE REPRESENTATION DEBATE 375

agreement” (Katz and Darbishire 2000)), was recently integrated under Australian “work choice” legislation. Management negotiates enterprise flexibility agreements or a non- union collective agreements directly with the affected employees, without the latter being represented by a union. Such agree- ments are intended to adjust industr y-wide or national standards to the particular needs of the enterprise and its work force.16 Under the Australian regime, the CD model is not merely a collective voice or a collective beg- ging mechanism. For an employer that prefers to distant itself from the unionized CR model, it is the only available avenue for exempting its enterprise (or part of its work force) from to industr y-wide or national standards. Since the alternatives, from the employer’s point of view—a standard collective agreement or being subject to industr y-wide or national standards—are less desirable, the employer has a strong incentive to reach an agreement with its work force.

A similar phenomenon is the “Work Force Agreement” under recently enacted U.K. legislation, which regulates information and consultation procedures, maternity and parental leave, fixed-term employment, and working time (Davies 2004). These statutor y provisions establish uniform labor standards. Nonetheless, an enterprise may opt out of the uniform standards by signing a regular collective agreement negotiated with a union—establishing a CR system—or by entering into an enterprise-specific work force agreement, which is an agreement that it negotiates either with ad hoc representatives of its work force or with the majority of the employees in the enterprise—in effect a CD arrangement (Carley and Hall 2000; Collins 2001; Barnard and Deakin 2002; Kilpatrick 2003; Barnard, Deakin, and Hobbs 2003). The approach taken by the U.K. Informa- tion and Consultation of Employees (ICE) Regulations of 2005 is an extreme example

of a CD system. Not only may the agreement be negotiated directly with the employees or with their ad hoc representatives rather than with an institutional and independent employees’ representation or a union, but it can provide for direct means of information and consultation rather than via employee representatives (Hall 2006). In their co- determination or decision-making capacity, the European works councils are similarly situated. The only difference is that the works council is an institutional rather than an ad hoc form of representation.

From a workplace governance perspec- tive, the most intriguing aspect of the Aus- tralian enterprise flexibility agreement and the U.K. work force agreement is that the enterprise-specific terms and the conditions of employment are determined collectively and uniformly without a union, through a consensus-building process in which manage- ment and all affected employees take part directly or through ad hoc representatives; this is a clear-cut example of a CD employ- ment relations model. Note, however, that the concept of representation in the enterprise flexibility agreement or the work force agree- ment is ver y different from the collective representation of employees under the CR model. The former is an ad hoc representa- tion, and it is far less independent, power- ful, and resourceful than the permanent, institutionalized representation found in the CR model.

The Internal Continuums of the Matrix

Having introduced the four cells of the matrix, I now turn to the last and most im- portant point about the proposed typology of employment relations models—viewing the broad array of employment relations models and workplace governance as a con- tinuum with the two prototypes, the Cell 2 individual-direct (ID) and Cell 4 collective- represented (CR) employment relations models at its two extremes. Industrial rela- tions literature, which often addresses the different degrees of a union’s strength or involvement in the employment relations of an enterprise, confirms the view that the continuum approach is conceptually more

16 Since by definition its terms var y from the national arbitration award, an enterprise flexibility or non-union collective agreement has to be approved by the arbitra- tion tribunal. See Pittard (1996:62); Fells and Kelffington (1998:234).

376 INDUSTRIAL AND LABOR RELATIONS REVIEW

chotomous “union/non-union” framework. In addition, since an employment relations model or model of workplace governance that is located in a particular cell in the matrix might be at any of various points on each of the two dimensions, it is preferable to view each cell as a continuum. The fact that a given mechanism of workplace gover- nance (collective bargaining, for example, or works councils) can easily be located at different points within a cell demonstrates that such a mechanism means different things in different countries. Finally, due to the dynamics and the continuous nature of employment relations, the employment relations of an enterprise can be expected to move at times within a cell and from one cell to another. In other words, one can envision that in addition to the ID–CR continuum there may be additional contin- uums inside the matrix.

One such continuum exists between the ID and IR cells. The particular location on this continuum may be a function of the intensity of representation insofar as the highest can be obser ved in the day-to-day representation of performing artists by their agents and lawyers. Much lower intensity may occur in the case of top executives, who are usually represented by lawyers only at the point of entr y and exit. The lowest degree of repre- sentation is exemplified in the “companion” under recent U.K. legislation. The Employ- ment Relations Act enables an employee to be accompanied by either a trade union member, a lay trade union representative, or a co-employer. Another example is the ad hoc representation in agency or court proceedings by an individual from an NGO or another type of community-based orga- nization (Fine 2006).

Cells 3 (collective-direct (CD)) and 4 (collective-represented (CR)) constitute the collective continuum. The explanator y power of this continuum can be illustrated by comparing two types of enterprises in which employment relations are governed by a collective agreement. The first type is known as a “green site” (Beaumont and Townley 1985b; Bassett 1986), which origi- nally referred to new plants built in rural areas in the United Kingdom by Japanese

multinationals.17 The second is an old and well established enterprise type characterized by job control unionism18 (Brody 1993; Ko- chan, Katz, and McKersie 1986). Both types of enterprise will have an active union and high union membership. In fact, between the two, the green site union that is party to an agreement may have a higher rate of member- ship, and it may enjoy both exclusivity and institutional security within the enterprise through a closed or union shop arrangement. Nonetheless, from a workplace governance perspective, what really counts is day-to-day employment relations, which is managed to a large extent on a direct individual or team basis. From the management’s viewpoint, the collectivity places ver y few constraints on management’s initiatives or functional flexibility. It allows management considerable latitude and discretion, and shows a high degree of tolerance for differential treatment of employees. The union usually inter venes only in the most extreme cases of usurpation of power by management. In terms of work- place governance, these peculiarities assign the green site enterprise to a high CD or to a low CR location.

In sharp contrast to the above, there are enterprises or employment relationships lo- cated at the far end of the continuum, that is, at the high end of CR. Among them are large proportions of public sector enterprises, as well as the sur viving (though now rare) old industrial plants, all of which are tightly gov- erned by an elaborate collective agreement, detailed work rules, and disciplinar y codes. They typically have a strong and entrenched union representation and restrictive work

17 These plants signed collective agreements that fea- tured the following properties: a long-term agreement, closed shop, management flexibility, union involvement through consultation over a broad array of operational and work-related issues, no strikes, and final offer arbi- tration for contract renewal.

18 Job control unionism is associated with a ver y de- tailed and legalistic collective agreement, enforced by a formal quasi-judicial grievance and arbitration proce- dure. The agreement establishes uniform seniority-based terms and conditions of employment that tie employee rights to ver y narrowly defined jobs with predictable compensation packages (Budd 2004).

REFRAMING THE REPRESENTATION DEBATE 377

union representation and restrictive work practices (Kochan, Katz, and McKersie 1986; Bluestone and Bluestone 1992).

Works councils, aimed at promoting dia- logue between management and employees, exist in many European countries, includ- ing Germany, the Netherlands, Spain, Italy, and Nor way (Kaufman and Gottlieb-Taras 2000). They may ser ve as another example of how a workplace governance mechanism may be manifested in different locations along the CD–CR continuum. Legislation and social structure var y from countr y to countr y, and as a result so does the location of a particular works council on the CD–CR continuum. Assuming there is no union pres- ence in the enterprise, the specific location of the works council is determined by three criteria: (1) the degree of the works council’s independence; (2) the amount of power and influence it has over decision-making; and (3) the scope of work- and business-related issues over which it has power and influence. The influence of the works council over decision-making in the enterprise may range from the right to receive information about and explanations for management actions to joint regulation or co-determination. Other options within the range include joint con- sultation, the right to negotiate (while man- agement retains decision-making powers), and the right to delay or veto management decisions (Clegg 1979; Boxall and Purcell 2003; Goldman 2005).

Examples of mechanisms that will prob- ably be located at the low end of the CD collective-direct cell (cell 3) are collective voice and other non-union employee rep- resentation programs that are initiated by management. These employment practices usually belong to the lower part of the CD cell for three reasons. First, they are wholly dependent on management for their exis- tence. They are established and structured to operate as part of the firm’s larger system of H.R.M., and they may be terminated at management’s discretion (Kaufman and Gottlieb-Taras 2000). Whenever a collective voice mechanism is mandated by law (Boxall and Purcell 2003), its ranking within the CD cell will be higher. Second, those who partici- pate in such programs lack the independent

status, immunities, and resources of the col- lective representation or the works councils typical of Cell 4 (CR) systems (Kaufman and Gottlieb-Taras 2000; Budd 2004). Third, these mechanisms do not provide the employees with any appreciable decision-making power in establishing, changing, interpreting, or applying the terms and conditions of their employment, or enable them to challenge management decisions that affect them. True, these mechanisms enable employees to voice their concerns and to have an input with respect to some narrowly defined issues, such as job design, cost containment policies, and quality control (Kaufman and Gottlieb- Taras 2000; Verma 2000); but the final say rests with management (Purcell and Sisson 1983; Boxall and Purcell 2003; Dundon, Wilkinson, Marchington, and Ackers 2004). These properties place management-initiated voice mechanisms in the lower part of the CD cell, as they are unable to provide a mean- ingful collective voice. Finally, as traditional unions have declined, a wide variety of groups and organizational forms have emerged to fill the void. These include NGOs advocat- ing for improved labor standards, as well as demographic- and social-based organizations or informal ethnic, racial, or gender groups, particularly in the United States (Scully and Segal 2002; Kochan 2005). These groups and organizations often represent workers’ causes on a collective or group basis, without hav- ing formal status or authorization from the workers themselves. In the suggested matrix, these forms of representation are found on the line separating CD and CR, depending on the causes they espouse and the functions they perform.

Figure 2 provides the approximate location in the matrix of different models of employ- ment relations and workplace governance. As I explain below, the exact location will always be context-specific.

The proposed matrix and the various con- tinuums may be used for analyzing a certain mode of workplace governance in the abstract or for examining the employment relations of a particular enterprise. In both cases, the application is both context-specific and dy- namic, that is, the obser ver’s conclusion as to where the workplace fits in the matrix may

378 INDUSTRIAL AND LABOR RELATIONS REVIEW

change over time. As shown in Figure 3, a movement from each point within the ma- trix is possible along three continuums. For instance, at the industrial relations system level, a change in legislation that bestows new and substantive decision-making pow- ers on works councils (supplanting joint or collective consultation rights, say, with joint regulation powers (Boxall and Purcell 2003)) may cause works councils to move higher within the CR cell. Similarly, an enterprise that is successful in ousting its union and replacing the collective agree- ment with a management-promulgated employee manual may move from CR to ID. Finally, when management introduces an employees’ voice or any other non-union representation program, the enterprise may move from ID to CD or to the bottom of

the CR cell, depending on the program’s features. The reason for the new loca- tion within the matrix is that these types of programs provide workers with some measure of indirect bargaining power, as well as benefits that exceed those available in individual-based employment relations (Kaufman and Gottlieb-Taras 2000).

A Note about the Unit of Observation

A unique feature of many modern work organizations is their propensity to put into effect multiple employment relations models simultaneously. The literature tends to differentiate between two groups—core employees and peripheral employees—ac- cording to their contractual arrangements and the employment relation models un-

Figure 2

Collective Labor Relations

Works Councils

Green Sites / No Strike Agreements

CR

Performing Arts (Talents)

Professional Sports

Executives

Worker Centers

Companion (U.K.) IR

Workforce Agreement

Flexibility Agreement (Australia)

Employee Voice System

CD

Individual Employment Contracts

ID

Collective

Represented

Direct

Individual

REFRAMING THE REPRESENTATION DEBATE 379

diversity of today’s work organizations goes much further than core versus peripheral employees. First, an increasing proportion of those working for the enterprise do not even have a formal employment relationship with the firm or with the firm’s outside contractor (triangular employment). They are classified as independent contractors, self-employed, freelancers, and so on. Second, a high de- gree of diversity exists even among regular employees. Any attempt to characterize an enterprise as either IR, ID, or CR runs the risk of overlooking the complexity of workplace governance systems in today’s organizations. A modern work organization may have several interacting employment relations models, as depicted in Figure 4. At the center is a small group of executives, all represented by lawyers (IR). The second layer is ID. It consists of regular managers and high-level professionals. They are represented neither individually nor collectively. The third layer is high CR—the collective union representation model. Sometimes, as is frequently the case in Japan and some European countries (Golan 2000; Morishima and Tsuru 2000), this layer may have a joint consultation committee or works council—low CD—that functions side-by-side with the collective bargaining process—CR. In some instances, especially under conditions of a tight product market and a strong union presence, this layer may be further divided between first- and second-tier collective agreements. Usually the second-tier

agreement applies to new employees, and provides for lower wages, fewer fringe ben- efits, and less union-supported job security than the first-tier agreement. The two tiers may be classified as high-CR and low-CR. The fourth layer is again ID. It is composed of the peripheral and “non-standard” em- ployees, such as contingent, seasonal, part- time, and temporar y employees. They are excluded from the coverage of the collective agreement, and cannot afford professional individual representation. The fifth layer includes those classified as non-employees, as well as those in various forms of triangular employment relationships, including those who perform outsourced functions. It lumps together non-employees and employees who perform work at or for the firm, but who nonetheless are retained by other employers such as temporary work agencies, regular and employee-only subcontractors, and related or double-breasting firms. From the perspective of the enterprise, these workers fall into the ID categor y.

Another example of a firm with a mix- ture of employment relation models is the American-based multinational, which, due to legal requirements or social/cultural norms, operates in Europe under the CR model. Since multinationals tend to emphasize the idea of a uniform organizational culture, it would be interesting to compare organiza- tional performance under the two regimes, and to see how the two models of employ-

Represented

Direct

Figure 3

Collective Individual

380 INDUSTRIAL AND LABOR RELATIONS REVIEW

Figure 4

Executives (IR)

Managers / High- Level Professionals (ID)

Regular Employees (CR)

Peripheral Employees (ID)

Non-Employees & Triangular Employment (ID)

tional performance under the two regimes, and to see how the two models of employ- ment relations interact. A similar strategic problem may arise when a firm that is primarily ID acquires or merges with one that is primarily a high-CR enterprise. The situation may be particularly complicated in industrial relation systems where, by law, collective agreements and representation sur vive the transfer of ownership or merger (Braginsky 1995).

Conclusion

This paper illustrates the value of replacing the traditional union/non-union dichotomy with a more dynamic and nuanced typol- ogy of employment relationships. It offers researchers and policymakers a richer and more accurate conceptual framework for describing and analyzing the wide variety of individual and collective representation and voice arrangements that may be obser ved in different employment settings around the

world. The paper further suggests honing the study of employment models and mak- ing them more sensitive by looking at the contemporar y workplace as an organization with multiple distinct employment models that interact with each other.

The potential advantages to be gained by using this new typology are limited neither to public policy debate nor to those making a strategic choice of employment relations models. In fact, this typology may allow re- searchers to refine their empirical studies of the diverse forms and combinations of repre- sentation and voice found in contemporar y workplaces in order to see which perform best in different settings for different work groups. It appears that the real alternative to union organization is not “no unions” or lack of representation, but some innovative form of representation and voice. Whether these alternatives are sustainable and powerful enough to do the job remains an empirical question that our field should be tackling with greater clarity and vigor.

REFRAMING THE REPRESENTATION DEBATE 381

REFERENCES

Ash, Michael, and Jean Ann Seago. 2004. “The Effect of Registered Nurses’ Unions on Heart-Attack Mortal- ity.” Industrial and Labor Relations Review, Vol. 57, No. 3, pp. 422–42.

Bamber, Greg J., and Russell D. Lansbur y, eds. 1998. International and Comparative Employment Relations, 3rd edition. London: Allen & Unwin.

Barley, Stephen R., and Gideon Kunda. 2004. Gurus, Hired Guns, and Warm Bodies: Itinerant Experts in a Knowledge Economy. Princeton, N.J.: Princeton University Press.

Barnard, Catherine, and Simon Deakin. 2002. “’Nega- tive’ and ‘Positive’ Harmonization of Labor Law in the European Union.” Columbia Journal of European Law, Vol. 8, pp. 389–413.

Barnard, Catherine, Simon Deakin, and Richard Hobbs. 2003. “Opting out of the 48-Hour Week: Employer Necessity or Individual Choice? An Empirical Study of the Operation of Article 18(1)(b) of the Working Time Directive in the UK.” Industrial Law Journal, Vol. 32, No. 4, pp. 223–52.

Barreto, José, and Reinhard Naumann. 1998. “Portugal: Industrial Relations under Democracy.” In Anthony Ferner and Richard Hyman, eds., Changing Industrial Relations in Europe, Chap. 14. Oxford: Blackwell.

Bassett, Philip. 1986. Strike-Free New Industrial Relations in Britain. London: Macmillan.

Beaumont, Philip B., and Laurence C. Hunter. 2003. “Collective Bargaining and Human Resource Manage- ment in Britain: Can Partnership Square the Circle?” In Thomas A. Kochan and David B. Lipsky, eds., Negotia- tions and Change—from Workplace to Society, pp. 161–88. Ithaca, N.Y., and London: ILR Press.

Beaumont, Philip B., and Barbara Townley. 1985a. “Non- Union American Plants in Britain—Their Employment Practices.” Relations Industrial, Vol. 40, No. 4, pp. 810–25.

____. 1985b. “Greenfield Sites, New Plants, and Work Practices.” In Valerie Hammond, ed., Current Research in Management, pp. 163–179. London: Frances Pinter.

Berger, Chris J., Craig A. Olson, and John W. Boudreau. 1983. “The Effect of Unions on Job Satisfaction: The Role of Work-Related Values and Perceived Rewards.” Organizational Behavior and Human Performance, Vol. 32, No. 3, pp. 289–324.

Bluestone, Barry, and Irving Bluestone. 1992. Negotiating the Future: A Labor Perspective on American Business. New York: Basic Books.

Booth, Alison L., and Mark L. Byren. 2004. “The Union Membership Wage Premium Puzzle: Is There a Free Rider Problem?” Industrial and Labor Relations Review, Vol. 57, No. 3, pp. 402–21.

Boxall, Peter F., and John Purcell. 2003. Strategy and Human Resource Management. Palgrave: Macmillan.

Braginsky, Leslie. 1995. “Comment: How Changes in Employer Identity Affect Employment Continuity: A Comparison of the United States and the United Kingdom.” Comparative Labor Law Journal, Vol. 16, No. 4, pp. 231–61.

Brody, David. 1993. “Workplace Contractualism in Comparative Perspective.” In Nelson Lichtenstein and Howell John Harris, eds., Industrial Democracy in America: The Ambiguous Promise, Chap. 8. Washington,

D.C.: Woodrow Wilson Center Press. Brower, Jonathan J. 1977. “Professional Sports Team

Ownership: Fun, Profit, and Ideology of Power of the Elite.” International Review of Sport Sociology, Vol. 12, No. 4, pp. 79–98.

Br yson, Alex. 2004. “Managerial Responsiveness to Union and Nonunion Worker Voice in Britain.” In- dustrial Relations, Vol. 43, No. 1, pp. 213–41.

Br yson, Alex, and Stephen McKay. 1997. “What about the Workers?” In Roger Jowell, John Curtice, Alison Park, Lindsay Brook, Katarina Thomson, and Caroline Br yson, eds., British Social Attitudes: The 14th Report— The End of Conser vative Values? pp. 23–48. Aldershot: Ashgate.

Budd, John. 2004. Employment with a Human Face: Balanc- ing Efficiency, Equity, and Voice. Ithaca, N.Y.: ILR Press (an imprint of Cornell University Press).

Carley, Mark, and Mark Hall. 2000. “The Implementation of the European Works Councils Directive.” Industrial Law Journal, Vol. 29, No. 2, pp. 103–24.

Ciarrocca, Lori A. 1987. “The Struggle to Define Section 7 Concerted Activity: A Literal Definition Emerges.” Washington and Lee Law Review, Vol. 44, No. 4, pp. 1277–1302.

Clegg, Hugh A. 1979. The Changing System of Industrial Relations in Great Britain. Oxford: Basil Blackwell.

Collins, Hugh. 2001. “Regulating the Employment Rela- tion for Competitiveness.” Industrial Law Journal, Vol. 30, No. 1, pp. 17–48.

Colvin, Alexander J. S. 2001. “The Relationship between Employment Arbitration and Workplace Dispute Resolution Procedures.” Ohio State Journal on Dispute Resolution, Vol. 16, No. 3, pp. 643–68.

CPR Institute for Dispute Resolution. 2002. How Com- panies Manage Employment Disputes—A Compendium of Leading Corporate Employment Programs. New York: CPR.

Davies, Anne C. L. 2004. Perspectives on Labour Law. Cambridge: Cambridge University Press.

Davies, Paul, and Mark Freedland. 2004. “Changing Perspectives upon the Employment Relationship in British Labour Law.” In Catherine Bernard, Simon Deakin, and Gilliam Morris, eds., The Future of Labour Law, pp. 129–57. Oxford: Hart.

Doucouliagos, Christos, and Patrice Laroche. 2003. “What Do Unions Do to Productivity? A Meta-Analysis.” Industrial Relations, Vol. 42, No. 4, pp. 650–91.

Dundon, Tony, Adrian Wilkinson, Mick Marchington, and Peter Ackers. 2004. “The Meanings and Purpose of Employee Voice.” International Journal of Human Resource Management, Vol. 15, No. 6, pp. 1149–70.

Dunlop, John T., and Arnold M. Zack. 1997. Mediation and Arbitration of Employment Disputes. San Francisco: JosseyBass.

Erickson, Christopher L., Catherine L. Fisk, Ruth Milk- man, Daniel J.B. Mitchell, and Kent Wong. 2002. “Jus- tice for Janitors in Los Angeles: Learning from Three Rounds of Negotiations.” British Journal of Industrial Relations, Vol. 40, No. 3, pp. 543–67.

Fells, Ray, and Robert Kelffington. 1998. “Moving beyond Adversarialism: Industrial Relations and Change in the Australian Shearing Industr y.” Industrial Relations Law,

382 INDUSTRIAL AND LABOR RELATIONS REVIEW

Vol. 29, No. 3, pp. 234–46. Ferner, Anthony, and Richard Hyman, eds. 1998. Chang-

ing Industrial Relations in Europe. Oxford: Blackwell. Fine, Janice. 2006. Worker Centers: Organizing Commu-

nities at the Edge of the Dream. Ithaca, N.Y.: Cornell University Press.

Foulkes, Fred K. 1980. Personnel Policies in Large Nonunion Companies. Englewood Cliffs, N.J.: Prentice Hall.

Freeman, Richard B., and Edward P. Lazear. 1995. “An Economic Analysis of Works Councils.” In Joel Rogers and Wolfgang Streeck, eds., Works Councils: Consulta- tion, Representation, and Cooperation in Industrial Rela- tions, Chap. 2. Chicago: University of Chicago Press for the NBER.

Freeman, Richard B., and James L. Medoff. 1984. What Do Unions Do? New York: Basic Books.

____, and Joel Rogers. 1999. What Workers Want. Ithaca, N.Y.: ILR Press (an imprint of Cornell University Press).

Frege, Carola M. 2002. “A Critical Assessment of the Theoretical and Empirical Research on German Works Councils.” British Journal of Industrial Relations, Vol. 40, No. 2, pp. 221–48.

Gerson, Herbert E., and Louis P. Britt. 1983. “The Ex- panded Rights of Unrepresented Employees.” Personnel Journal, Vol. 62, No. 5, pp. 350–56.

Getman Julius G., Bertrand B. Pogrebin, and David L. Gregor y. 1999. Labor Management Relations and the Law. New York: Foundation Press.

Golan, Paul J. 2000. “Nonunion Forms of Employee Representation in the United Kingdom and Australia.” In Bruce E. Kaufman and Daphne Gottlieb-Taras, eds., Nonunion Employee Representation Histor y: Contemporar y Practices and Policy, pp. 410–52. Armonk, N.Y.: M. E. Sharpe.

Goldman, Alvin L. 2005. “Is There a Principled Basis for Separating the Authority of a Works Council from the Authority of a Collective Bargaining Representative?” In Liber Amicorum Reinhold Fahlbeck, pp. 231ff. Lunds, Germany: Juridiska Fakulteten Lunds Universitaet.

Gordon, Michael E., and Angelo S. Denisi. 1995. “A Re-Examination of the Relationship between Union Membership and Job Satisfaction.” Industrial and Labor Relations Review, Vol. 48, No. 2, pp. 222–36.

Griffeth, Rodger W., Peter W. Hom, and Stefan Gaertner. 2000. “A Meta-Analysis of Antecedents and Correlates of Employee Turnover: Update, Moderator Tests, and Research Implications for the Next Millennium.” Jour- nal of Management, Vol. 26, No. 3, pp. 463–88.

Gundersen, Bethney. 2003. “Unions and the Well-Being of Low-Skill Workers.” Ph.D. diss., George Warren Brown School of Social Work, Washington University.

Hall, Mark. 2006. “A Cool Response to the ICE Regula- tions? Employer and Trade Union Approaches to the New Legal Framework for Information and Consulta- tion.” Industrial Law Journal, Vol. 37, No. 5, pp. 456–72.

Hammer, Tove Helland. 2000. “Nonunion Representa- tional Forms: An Organizational Behavior Perspective.” In Bruce E. Kaufman and Daphne Gottlieb Taras, eds., Nonunion Employee Representation: Histor y, Contemporar y Practice, and Policy, Chap. 8. Armonk, N.Y.: M. E. Sharpe.

Hendr y, David. 1997. “Unionization and Economic Performance: Evidence on Productivity, Profits, In- vestment, and Growth.” In Fazil Mihlar, ed., Unions

and Right-to-Work Laws, pp. 35–70. Vancouver, B.C.: Frazer Institute.

Hirsch, Barr y T. 1992. “Firm Investment Behavior and Collective Bargaining Strategy.” Industrial Relations, Vol. 31, No. 1, pp. 95–121.

____, and John T. Addison. 1986. The Economic Analysis of Unions: New Approaches and Evidence. Boston: Allen & Unwin.

____, and David A. Macpherson. 2003. Union Member- ship and Earnings Data Book: Compilations from the Cur- rent Population Sur vey. Washington, D.C.: Bureau of National Affairs.

Howell, Chris. 2005. Trade Unions and the State. Princeton, N.J.: Princeton University Press.

Hyman, Richard. 2001. Understanding European Trade Unionism: Between Market, Class and Society. London: Sage.

Iverson, D. Roderick, and Currivan B. Douglas. 2003. “Union Participation, Job Satisfaction, and Employee Turnover: An Event-Histor y Analysis of the Exit-Voice Hypothesis.” Industrial Relations, Vol. 42, No. 1, pp. 101–5.

Jacobi, Otto, Berndt Keller, and Walther Müller-Jentsch. 1998. “Germany: Facing New Challenges.” In Anthony Ferner and Richard Hyman, eds., Changing Industrial Relations in Europe, Chap. 7. Oxford: Blackwell.

Katz, Harr y C., and Owen Darbishire. 2000. Converging Divergences: Worldwide Changes in Employment Systems. Ithaca, N.Y.: ILR Press (an imprint of Cornell Uni- versity Press).

Kaufman, Bruce E., and James Bennett, eds. 2006. What Do Unions Do: A Twenty-Year Perspective. New Brunswick, N.J.: Transaction Press.

____, and Daphne Gottlieb-Taras, eds. 2000. Nonunion Employee Representation: Histor y, Contemporar y Practice, and Policy. Armonk, N.Y.: M. E. Sharpe.

____, and David I. Levine. 2000. “An Economic Analysis of Employee Representation.” In Bruce E. Kaufman and Daphne Gottlieb Taras, eds., Nonunion Employee Representation: Histor y, Contemporar y Practice, and Policy, Chap. 7. Armonk, N.Y.: M. E. Sharpe.

Kilpatrick, Claire. 2003. “Has New Labour Reconfigured Employment Legislation?” Industrial Law Journal, Vol. 32, No. 3, pp. 135–63.

Kochan, Thomas A. 2005. Restoring the American Dream— A Working Families’ Agenda for America. Cambridge, Mass.: MIT Press.

____, Harr y C. Katz, and Robert B. McKersie. 1986. The Transformation of America Industrial Relations. New York: Basic Books.

Labour Law Casebook Group. 1998. Labour and Employ- ment Law. Kingston, Ont.: IRC Press.

Lecher, Hans, Wolfgang Platzer, and Stefan Rub. 2001. European Works Councils: Developments, Types and Network- ing. Aldershot, England: Gower.

Lipsky, David, Ronald L. Seeber, and Richard D. Fincher. 2003. Emerging Systems for Managing Workplace Conflict: Lessons from American Corporations for Managers and Dis- pute Resolution Professionals. San Francisco: Jossey-Bass.

Mefford, Robert N. 1986. “The Effect of Unions on Productivity in a Multinational Manufacturing Firm.” Industrial and Labor Relations Review, Vol. 40, No. 1, pp. 105–14.

REFRAMING THE REPRESENTATION DEBATE 383

Mills, Daniel Quinn. 1982. Labor-Management Relations. New York: McGraw-Hill.

Mishel, Lawrence, and Walters Matthew. 2003. How Unions Help All Workers. Economic Policy Institute Briefing Paper.

Morishima, Motohiro, and Tsuyoshi Tsuru. 2000. “Non- union Employee Representation in Japan.” In Bruce E. Kaufman and Daphne Gottlieb-Taras, eds., Nonunion Employee Representation: Histor y, Contemporar y Practices, and Policy, pp. 386–408. New York: M. E. Sharpe.

Moskow, Michael H. 1970. “The Performing Arts.” In Seymour L. Wolfbein, ed., Emerging Sectors of Collective Bargaining. Morristown, N.J.: General Learning.

Müller-Jentsch, Walther. 1995 “Germany: From Collec- tive Voice to Co-Management.” In Joel Rogers and Wolfgang Streeck, eds., Works Councils: Consultation, Representation, and Cooperation in Industrial Relations, pp. 53–78. Chicago: University of Chicago Press.

OECD. 1997. OECD Employment Outlook, Table 3.3. Paris: Organization for Economic Cooperation and Development.

Osterman, Paul. 1987. “Choice of Employment Systems in Internal Labor Markets.” Industrial Relations, Vol. 26, No. 1, pp. 46–67.

Pierce, Brooks. 1999. “Compensation Inequality.” Office of Compensation and Working Conditions, Depart- ment of Labor, Working Paper No. 323.

Pittard, Marilyn. 1996. “Bargaining and Enterprise Flex- ibility Agreements.” Australian Business Law Review, Vol. 24, No. 1 pp. 62–70.

Purcell, John, and Keith Sisson. 1983. “Strategies and Practice in the Management of Industrial Relations.” In George Sayers Bain, ed., Industrial Relations in Britain. Oxford: Basil Blackwell.

Rogers, Joel, and Wolfgang Streeck. 1995. “The Study of Works Councils: Concepts and Problems.” In Joel Rogers and Wolfgang Streeck, eds., Works Councils: Consultation, Representation, and Cooperation in Indus- trial Relations, pp. 326. Chicago/London: University of Chicago Press.

Rubinstein, Saul R., and Thomas A. Kochan. 2001. Learn- ing from Saturn. Ithaca, N.Y.: ILR Press (an imprint of

Cornell University Press). Schwochau, Susan. 1987. “Union Effects on Job At-

titudes.” Industrial and Labor Relations Review, Vol. 40, No. 2, pp. 209–24.

Scully, Maureen, and Ami Segal. 2002 “Passion with an Umbrella: Grassroots Activities in the Workplace.” In Social Structure and Organizations Revisited, Vol. 19 (Research in the Sociology of Organizations, edited by Mi- chael Lounsbur y and Marc J. Ventresca), pp. 125–68.

London: JAI Press. Sisson, Keith. 1993. “In Search of HRM.” British Journal

of Industrial Relations, Vol. 31, No. 2 (June), pp. 201–11. Stone, Katherine Van Wezel. 2004. From Widgets to Digits:

Employment Regulation for the Changing Workplace. Cam- bridge, U.K.: Cambridge University Press.

Terr y, Michael. 1999. “Systems of Collective Employee Representation in Non-Union Firms in the UK.” Industrial Relations Journal, Vol. 30, No. 1, pp. 16–30.

Verma, Anil. 2000. “Employee Involvement and Repre- sentation in Nonunion Firms: What Canadian Employ- ers Do and Why?” In Bruce E. Kaufman and Daphne Gottlieb-Taras, eds., Nonunion Employee Representation: Histor y, Contemporar y Practices, and Policy, pp. 307–328. Armonk, N.Y.: M. E. Sharpe.

Voos, Paula B., and Lawrence R. Mishel. 1986. “The Union Impact on Profits: Evidence from Industr y Price-Cost Margin Data.” Journal of Labor Economics, Vol. 4 (Januar y), pp. 105–33.

Weiler, Anni. 2004. European Works Councils in Practice. Ireland: European Foundation for the Improvement of Living and Working Conditions.

Weiler, Paul C. 1990. Governing the Workplace: The Future of Labor and Employment Law. Cambridge, Mass.: Har vard University Press.

Wheeler, Hoyt N., Brian S. Klaas, and Douglas M. Mahony. 2004. Workplace Justice without Unions. Kalamazoo: W. E. Upjohn Institute for Employment Research.

____., and John A. McClendon. 1998 “Non-Union and Unionized Employment Relations in the United States.” In Greg J. Bamber and Russell D. Lansbur y, eds., In- ternational and Comparative Employment Relations, 3rd edition, pp. 63--88. London: Allen & Unwin.

Copyright of Industrial & Labor Relations Review is the property of Cornell University and its content may not

be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.