VII. EMPLOYEE VOICE IN THE ANGLO-AMERICAN WORLD: CONTOURS & CONSEQUENCES

Employee Voice in the Anglo-American World: An Overview of Five National Surveys

Peter Boxall and Peter Haynes
University of Auckland, New Zealand

    Drawing on data from a linked series of large-scale worker surveys in the United States, the United Kingdom, Canada, Australia, and New Zealand, this paper identifies and analyzes key features and developments in employee voice across the Anglo-American world. These include the extent and nature of unsatisfied demand for union membership and free-riding, as well as the diversity and variation in supply of and preferences for employee voice of different types, in the different countries.

Introduction

      This paper gives a brief overview of a series of linked surveys of worker voice conducted in five English-speaking countries: the United States, the United Kingdom, Canada, Australia, and New Zealand. The survey is currently being conducted in Ireland. The objective of these surveys is to compare the voice workers have in their workplaces with the voice they want. The focus is on employee views of workplace voice, not on views about employee voice transmitted by management or union respondents. Collectively, the surveys enable us to identify and analyse key features and developments across the English-speaking world. The underpinning scholarly philosophy is one of openness to voice innovation: to trends in the way workers wish to express their interests and to institutional reforms and novel practices that enable them to do so.

    The surveys in New Zealand (Boxall, Haynes, and Macky 2005), Australia (Teicher, Holland, Pyman, and Cooper 2005) are based on the pathbreaking 1994-95 Worker Representation and Participation Survey in the United States (Freeman and Rogers 1999; Freeman 2005) and the 2001 British Worker Representation and Participation Survey (Diamond and Freeman 2002; Bryson and Freeman 2005), which followed it. The Canadian survey (Campolieti, Gomez, and Gunderson 2005) uses the 1996 Canada-US Labour Attitudes Survey by Lipset and Meltz (1997, 2004), in which many items were drawn from Freeman and Rogers (1999). To the extent necessary, questions were modified in the different countries to fit the particular institutional and demographic context.
      The method involved in these surveys has generally been the telephone interview using standard computer-assisted telephone interviewing (or CATI) techniques. Care was taken in each country to ensure representativeness. The original U.S. survey sampled 2,408 adult workers but excluded the public sector and workers in firms with fewer than 25 employees. The other surveys did not exclude any groups. In Canada, 1,495 workers were sampled, and in both Australia and New Zealand 1,000 workers were drawn randomly from residential telephone directories. The British study involved face-toface interviews with some 1,300 randomly chosen workers.

Institutional Characteristics

      The overall study contains three sets of adjacent, English-speaking countries: the United States and Canada, the United Kingdom and Ireland, and Australia and New Zealand. These countries make up three "large and small" pairs. They share much in common with their near neighbors while holding on to some cherished distinctions. In terms of commonalities, all six countries are democracies with a heritage of British concepts of law and contract. Five of the six have decentralized systems of collective bargaining (enterprise and/or workplace-based) or have evolved in this direction over the past twenty years: the United States, Canada, the United Kingdom, New Zealand, and, in a more uneven pattern, Australia. The outlier is Ireland, which has successfully developed a corporatist system over the past two decades, but one that includes a strong element of decentralised involvement through balloting of union members over potential settlements (Baccaro 2003).

     One way of identifying differences within the set is to group them by geography and look at their regulatory traditions in industrial relations. The North American pair, the United States and Canada, has a common Wagner Act heritage but demonstrates small, important differences in its operation. Very significantly, with respect to Canada, Campolieti, Gomez, and Gunderson note that "there are no prohibitions against groups of employees dealing directly with management, absent union representation, regarding their terms of employment. As a consequence of this small legal difference, there is latitude for Canadian employers to innovate with a variety of non-union mechanisms" (2005, 3).

     The Antipodeans, Australia and New Zealand, have a shared heritage of compulsory, state-provided arbitration of labor disputes, dating back more than a hundred years. This is now completely dismantled in New Zealand, and largely so in Australia, in favor of collective bargaining and individual contracting (Boxall, Hayes, and Macky 2005; Teicher et al. 2005).
      By contrast, the United Kingdom and Ireland share a volunteerist tradition. They have, however, parted company over the past twenty years. Instead of following the U.K. path of restricting union power, including abolishing the closed shop (Bryson and Freeman 2005), Irish governments have adopted a social partnership model in which central union and employer organizations reach national wage deals (Baccaro 2003; Geary and Roche 2005). The U.K.- Ireland pairing is interesting in another sense: both countries are members of the European Union. They are thus subject to European directives and institutional diffusion, including the spread of works councils, although the scope to customize these institutions is considerable (Geary and Roche 2005).
      This method of grouping the six countries is the obvious one. A second method of categorizing them-by tractability of reform-produces a different set of strata. In terms of the politics of systemic change in industrial relations, New Zealand, the United Kingdom, and Ireland look the most tractable systems (probably in that order). All have made substantive changes to their regulatory regimes over the past twenty years. On the other hand, Canada and Australia are federalist systems, in which change is more problematic. This seems to be particularly the case in Australia, where difficulties between upper and lower houses and between federal and state parliaments almost inevitably delay and complicate intended reforms (Teicher et al. 2005).

     Nevertheless, change does occur in the Canadian and Australian systems, and the stark contrast in this second sense is with the United States. As we show below, all of the other systems seem to be evolving through complementary change in their voice institutions. By this we mean that non-union voice, such as joint consultative committees and works councils, tends to complement union voice and the two interact in ways which are generally productive, thus enlarging voice options for workers. The U.S. system, on the other hand, is predicated on competitive change: union and non-union voice are substitutes and employer opposition to unionism is a much bigger factor than anywhere else in the English-speaking world. It is this difference that stands out more than anything from a reading of these surveys.

     Our analytical framework is a simple one that considers the demand for, and supply of, workplace voice, including trade unionism (see, for example, Bryson and Gomez 2003). We note here key findings across the surveys and leading analytical work in particular studies. We begin with the opportunities for workers to join and belong to unions and then consider employee voice in a broader sense.

Unions: Demand and Supply
      Representation gaps: the extent of unsatisfied demand for unionism.The picture of declining union density in the Anglo-American world is well known. Over the period 1980-1995, union density fell by 39.4 percent in Australia, 46.1 percent in the United Kingdom, and 39.8 percent in the United States (Verma and Kochan 2004). Density levels in the studies reported here range from 13 percent in the United States to 21 percent in New Zealand, 23 percent in Australia, 29 percent in the United Kingdom, and 30 percent in Canada (Boxall, Hayes, and Macky 2005; Bryson and Freeman 2005; Campolieti, Gomez, and Gunderson 2005; Freeman 2005; Teicher et al. 2005). Although this trend continues, the voice surveys identify a substantial level of unsatisfied demand for unionism or a "representation gap." Following Freeman and Rogers (1999), this gap is calculated by asking those who have no union at their workplace that they can join whether they would be "very likely" or "fairly likely" to join one if asked.
      In New Zealand and Australia, those who would join a union if one were formed at their workplace represent around one-third of those working in non-unionized workplaces (Boxall, Hayes, and Macky 2005; Teicher et al. 2005). This finding almost exactly parallels the U.S. figure of 32 percent for 1994 (Freeman and Rogers 1999, 68) and the figure of 36 percent for British workers (Bryson and Freeman 2005, 10). More recent survey evidence suggests that the representation gap in the United States is rising and may now be as high as one in two workers in non-union workplaces (Freeman 2005).

     A couple of caveats should be entered here. The method used assumes that those who say they are "fairly likely" to join would actually join a specific union when, and if, the chance arose. How much of the "fairly likely" category represents soft support for unionism that would be unlikely to translate to membership is debatable (and, indeed, some of the "very likelies" may be deterred from joining in practice). Taking a more realistic view of the likely translation from hypothetical to actual union support does change the picture somewhat. In the New Zealand case, for example, 67 percent of those "fairly likely" to join also reported that, if a union was formed at their workplace, it would make no difference to them personally, and a further 8 percent reported that they would be worse off (Boxall, Hayes, and Macky 2005, 6). If we remove three-quarters of those fairly likely to join from the calculation (i.e., those who think unionism will make no difference to them or make them worse off), the realistic catchments for unions might be existing New Zealand union density (21 percent) plus the very likelies (6 percent) plus 25 percent of the fairly likelies (3 percent). This gives a more realistic potential union density in New Zealand of 30 percent (rather than 39 percent if we counted all the fairly likelies). Applying this reasoning to the Australian data produces the same figure: potential union density is around three of ten workers (Teicher et al. 2005).
      There is also the question of whether latent support for unions is spread fairly evenly across workplaces, making it hard to reach critical mass in particular workplaces, or concentrated in some workplaces which could realistically be organized by unions. There is no doubt that much of the latent support would not be converted to a majority in a U.S. workplace. On this issue, however, analysis of the U.S. data suggests that "worker needs for representation and participation have a non-normal distribution": some workplaces have high levels of needs and should be organizable if unions actually reached them Freeman (2005, 2).

      Characteristics of workers with unsatisfied union demand. Analysis of the voice surveys throws light on which workers are most likely to experience unsatisfied demand for unions. The Australian, Canadian, and New Zealand surveys indicate that young workers, those with low income, and workers in small workplaces are much more likely to have unsatisfied union demand (Boxall, Hayes, and Macky 2005; Campolieti, Gomez, and Gunderson 2005; Teicher et al. 2005). In other words, the more vulnerable groups of workers are less well covered by unions. As Campolieti, Gomez, and Gunderson put it, "Frustrated desire for unions is highest among the youngest and most economically disadvantaged" (2005, 22).
      The findings in the surveys on youth are important in terms of challenging anecdotally based accounts that young workers have become more individualistic and anti-union. Although young workers (often defined as younger than twenty-five years) have union membership levels one-third to half of older workers, the surveys paint a very positive picture of the attitudes of young workers to unions (Bryson, Gunderson, and Meltz 2004; Freeman and Diamond 2003; Gomez, Gunderson, and Meltz 2002; Haynes, Vowles, and Boxall 2005). On the whole, the differences between younger and older workers' attitudes to unions are slight or nonexistent. If anything, younger workers' attitudes and willingness to join are more positive than those of their older counterparts. Without underestimating the difficulties of recruiting young workers who also have high labor mobility, this finding suggests that unions can approach the issue of enlisting young workers with greater confidence.
      In terms of those who want unions, analysis of the British and U.S. data offers another insight. The more problems workers see in their workplace, the more they want a union voice. In the United States, "among workers with no needs, 26% say they would favor a union compared to 45% among those with 4 needs and 81% of those with 7 or more needs' (Freeman 2005, 10). In the United Kingdom, where works councils are now a factor, higher perceptions of need call forth stronger demand for the more formal and independent voice mechanisms: "the greater the number of needs, the greater the need for the stronger forms of representation: the mean number of needs is 4.71 for those wanting a combination of works council and union, compared with 3.68 for those wanting a union only, 3.01 for those wanting a works council only, and 2.11 for those with no desire for collective representation" (Bryson and Freeman 2005: 12).

      Free-riding. In the United Kingdom, Australia, and New Zealand, union membership remains voluntary even where unions have bargaining recognition. It is not surprising that there are problems with free-riding in these countries, which undermine the effectiveness of unions. The pattern is very similar: around four in ten workers in unionized British, Australian, and New Zealand workplaces are free riders (Boxall, Hayes, and Macky 2005; Bryson and Freeman 2005; Teicher et al. 2005). This contrasts with the situation in Canada, where the "Rand formula" allows agency or union dues shops (Campolieti, Gomez, and Gunderson 2005, 4). Canadian union density, at around 30 percent, is the highest in the five countries surveyed so far and may well have benefited from this specific regulation.

     What the surveys show, however, is that free-riding is complex and cannot simply be depicted as opportunistic, anticollective behavior. Some 30 percent of free-riders in Britain consider union fees too high and 28 percent consider that people doing their sort of job do not join unions (Bryson and Freeman 2005, 11). The latter group is as high as 55 percent in Australia (Teicher et al. 2005). The New Zealand data show that nonmembership in unionized workplaces is highest at both the lowest and highest levels of income and education, and lowest in the middle range (Haynes and Boxall 2004). That is to say, nonmembership is "U-shaped" by level of income and education. This pattern suggests that technical free-riding may be broadly bimodal. Lower-income workers may find the cost of union membership dif- ficult to sustain at any level. On the other hand, high-income workers may feel less need for union representation as a result of their greater job security and labor market power.

Employee Voice in a Broader Sense
      The overwhelming message from the surveys is that voice preferences are now more diverse and variable. Voice preferences vary in certain key ways:

   1. By the extent of the needs expressed The Canadian and New Zealand studies point to relatively high      levels of job satisfaction in these countries at the time of the surveys (Boxall, Hayes, and Macky 2005      Campolieti, Gomez, and Gunderson 2005). Analysis of the U.K. and U.S. studies suggests that the      majority of workers presently have few problems in the workplace and see less need for union voice      (Bryson and Freeman 2005, 2, 6, 13; Freeman 2005, 10-11). There is, however, a strong desire for      unionism among the minority of workers who see numerous problems in their relationships with their      employer.

   2. By the nature of the problems. Not all problems are created equal. The U.K. study shows that "workers      preferred solving problems relating to promotion and training and skill development on their own and      preferred to deal with problems of discrimination through a trade union" (Bryson and Freeman 2005, 10).      It suggests that, although there is a strong desire for union voice in classical conflict areas such as pay and      benefits, workers increasingly question the relevance of unions in areas to do with job design and personal      development. The New Zealand study is also interesting in this regard. Large majorities of New Zealand      workers-from 69 percent to 86 percent-report that they exercise a lot or some influence over how they      do their work but only 30 percent and 25.2 percent, respectively, say they have a lot or some influence      over pay rises and perks or bonuses. Workers seem increasingly confident in dealing directly with their      supervisors and managers over how work is done.
   3. By the characteristics of workers. The Canadian survey finds that, "generally speaking, the youngest and      oldest workers and those with the least and most tenure are more likely to desire collective representation,      as compared to middle age and middle-tenure workers" (Campolieti, Gomez, and Gunderson 2005, 22).      This life-cycle interpretation of worker needs for representation is also generally supported by the New      Zealand data (Boxall, Hayes, and Macky 2005).
   4. By the characteristics of workplaces. As noted above, the British and U.S. studies argue that demand for      union voice is much greater in workplaces with high levels of problems. They point to a "workplace effect"      in generating demand for unionism. We should not assume, however, that all small workplaces are bad      workplaces. The Canadian data is interesting on the effects of workplace size. "As establishment size      increased, workers were much more likely to want collective voice, presumably in keeping with      expectations about the value of collective voice when distance from management increases" (Campolieti,      Gomez, and Gunderson 2005, 22). This reinforces the view that unions often emerge as a response to      remote, bureaucratic management that needs to be constrained in terms of the way it exercises power.

Voice Innovation: New Voice Institutions and Practices.

     The surveys generally reveal that the extent of direct, informal, and management- driven voice is high. In New Zealand, for example, a large proportion of workers reports that their workplaces have one or more procedures for resolving workplace problems and/or for involvement in decision making. Most (86 percent) report that their organization has an "open-door" policy so employees can tell senior management about problems with their supervisors and around half report that managers in their organizations hold regular meetings with staff (58 percent) or that their organization has a personnel/ human resources department or person (50 percent). Half of workers surveyed (50 percent) reports that a committee of management and employees or staff forum meets regularly to consult over workplace issues in their workplace. Some 35 percent report an "employee involvement program such as quality circles" in their workplace. Only a very small minority (5 percent) reports that they do not have access to any of these procedures. These findings broadly mirror those in the United Kingdom (Diamond and Freeman 2002) and Australia (Teicher et al. 2005).
     A key trend is the rise of works councils and joint consultative committees and the generally positive worker attitudes to these forms of voice. These voice institutions are important because it is not possible anymore to dismiss them as driven by a management-dominated, productivity-obsessed agenda. They are important because of their potential to provide a workeroriented voice mechanism. In both Australia and New Zealand, consultative committees cover roughly twice as many workers as collective bargaining, and they are generally seen by workers as effective forms of voice (Boxall, Hayes, and Macky 2005; Teicher et al. 2005).
     The overall trend outside the United States can be summed up as a growing complementarity of union and non-union voice. The Australian study concludes that "unions maintain a significant voice in the day-to-day decisionmaking process within organisations despite the parallel development of non-union procedures" (Teicher et al. 2005, 15). This is also shown in Canada (Campolieti, Gomez, and Gunderson 2005, 21) and is very evident in the United Kingdom, arguably the most interesting case given the infusion of European voice practices: "Most workers see works councils as complementary to unions, which is how councils and unions have operated in much of Continental Europe. Three-quarters of workers believe their workplace would work better with some form of collective representative worker voice. Two-thirds think the workplace would be better with some works council. And in two-thirds of these cases (nearly half of all workers) workers want a combination of works council and union" (Bryson and Freeman 2005, 12). The British analysis suggests that "the advent of works councils creates an opportunity for unions to expand their influence in the workforce" (Bryson and Freeman 2005, 12).
     Innovative, non-union voice practices are, of course, very evident in the United States, but U.S. labor law "restricts companies in what they can do to substitute non-union collective voice for independent unionism" (Freeman 2005, 5). This makes the United States the odd case in the set and works to the disadvantage of workers. The strong preference of U.S. workers for some cooperative, union-like committee of workers and management is an "unfilled demand" (Freeman 2005, 10).

Conclusion

     These comparable worker voice surveys have been conducted in an era of declining union density across the Anglo-American world. Despite this trend, they all point to a substantial minority of workers in non-unionized workplaces who have an unsatisfied demand for unionism. The condition of unsatisfied demand for union voice-or union-like voice-does seem to be greatest in the United States, where efforts to secure reform are clearly the most difficult.
      Many of the workers who experience a representation gap are young or have low income. They work in small, private sector workplaces that are not well addressed or easily reached by current union structures, whereas others, particularly in the United Kingdom and Australia, have simply been overlooked by union representatives. The surveys confirm that union representatives should not stereotype young workers as disinterested in collectivism.
     Free-riding is prevalent in countries that allow for voluntary unionism irrespective of workplace union recognition but is more complex than it first appears: low-income workers may free-ride for cost reasons, whereas highincome workers may free-ride for relevance (lack of benefit) reasons.
      The surveys indicate that employee voice needs are complex and that there is a market for innovative voice practices. Outside the United States, non-union and union forms of voice are increasingly complementary, to the benefit of workers. Worker responses in the studies reported here imply that the caricature of non-union voice practices as toothless or ineffective is misleading and should be avoided.
      The surveys suggest that contemporary workers see unions as relevant in traditional areas of conflict-such as pay bargaining-and very relevant in seriously dysfunctional workplaces or when they personally feel vulnerable in the labour market. They suggest, however, that workers are sceptical about the relevance of unions to the developmental agenda of skill acquisition and personal growth, which may now be dominating employee thinking about their working lives.

References

Baccaro, Lucio. 2003. "What Is Alive and What Is Dead in the Theory of Corporatism?" British Journal of Industrial Relations, Vol. 41, no. 4, pp. 683-706.

Boxall, Peter, Peter Haynes, and Keith Macky. 2005. "Employee Voice and Voicelessness in New Zealand." Unpublished paper, University of Auckland.

Bryson, Alex. 2003. Employee Desire for Unionisation in Britain and its Implications for Union Organising. PSI Research Discussion Paper No. 12. London: Policy Studies Institute.

Bryson, Alex, and Richard B. Freeman. 2005. "What Voice Do British Workers Want?" Unpublished paper, Centre for Economic Performance, London School of Economics.

Bryson, Alex, and Raphael Gomez. 2003. "Buying into Union Membership." In Howard Gospel and Stephen Wood, eds., Representing Workers: Union Recognition and Membership in Britain. London: Routledge, pp. 72-91.

Bryson, Alex, Rafael Gomez, Morley Gunderson, and Noah Meltz. 2004. "Youth-Adult Differences in the Demand for Unionization: Are American, British, and Canadian Workers All That Different?" Journal of Labor Research, Vol. 26, no. 1, pp. 155-67.

Campolieti, Michele, Rafael Gomez, and Morley Gunderson. 2005. "Say What? . . . . Employee Voice in Canada." Unpublished paper, University of Toronto.

Diamond Wayne, and Richard B. Freeman. 2002. What Workers Want from Workplace Organisations: A Report to the TUC's Promoting Trade Unionism Task Group. London: Trades Union Congress.

Freeman, Richard B. 2005. "Can Unions Close the Unfilled Demand for Representation and Participation in the US?" Unpublished paper, Harvard University and National Bureau for Economic Research.

Freeman, Richard B., and Wayne Diamond. 2003. "Young Workers and Trade Unions." In Howard Gospel and Stephen Wood, eds., Representing Workers: Union Recognition and Membership in Britain. London: Routledge, pp. 29-50.

Freeman, Richard B., and Joel Rogers. 1999. What Workers Want. Ithaca, N.Y.: Cornell University Press.

Geary, John F., and William K. Roche. 2005. "Anticipating the Likely Implications of the EU Employee Information and Consultation Directive in Ireland." In John Storey, ed., Adding Value through Information and Consultation. Basingstoke: Palgrave Macmillan.

Gomez, Rafael, M. Gunderson, and Noah M. Meltz. 2002. "Comparing Youth and Adult Desire for Unionization in Canada." British Journal of Industrial Relations, Vol. 40, no. 3, pp. 521-42.

Haynes, Peter, and Peter Boxall. 2004. "Free-Riding in New Zealand: Incidence, Motives and Policy Implications." Labour & Industry, Vol. 15, no. 2.

Haynes, Peter, Jack Vowles, and Peter Boxall. 2005. "Explaining the Younger-Older Worker Union Density Gap: Evidence from New Zealand." British Journal of Industrial Relations, Vol. 43, no. 1, pp. 91-114.

Lipset, Seymour Martin, and Noah M. Meltz. 1997. "Canadian and American Attitudes toward Work and Institutions." Perspectives on Work, Vol. 1, no. 3, pp. 14-20.

Lipset, Seymour Martin, and Noah M. Meltz. 2004. The Paradox of American Unionism: Why Americans Like Unions More than Canadians Do But Join Much Less. Ithaca, N.Y: Cornell University Press.

Teicher, Julian, Peter Holland, Amanda Pyman, and Brian Cooper. 2005. "Employee Voice in Australia." Unpublished paper, Monash University.

Verma, Anil, and Thomas A. Kochan. 2004. "Unions in the 21st Century: Prospects for Renewal." In Anil Verma and Thomas A. Kochan, eds., Unions in the 21st Century: An International Perspective. New York: Palgrave Macmillan, pp. 1-16.

.