Mandatory agency shop laws as an explanation of ... - Springer Link

Mandatory Agency Shop Laws As an Explanation of Canada-U.S. Union Density Divergence* D A P H N E G O T T L I E B T A R A S and A L L E N P O N A K

University of Calgary, AB CANADA T2N 1N4 I. Introduction We posit that differences in statutory treatment of union security provisions in Canada and the U.S. may significantly contribute to the union density gap between the two countries. Union density in Canada and the U.S. now stands at 30 percent and 13 percent, respectively. In 1999 Canadian public sector density was almost 71 percent, while private sector density was just over 18 percent (Akyeampong, 1999). The comparable American breakdown is about 37 percent in the public sector and 9.5 percent in the private sector (Lipset and Katchanovski, 2001). In other words, the rate of unionism in Canada is approximately double the unionism rate in the U.S. in both the public and private sectors. It was not always so. For most of the 20th century, union growth patterns were remarkably similar, and by 1966 union density in both countries stood at 30 percent. From that point forward, however, unionization patterns diverged and a union density gap between the two countries developed and persisted. II. Bi-National Comparisons The interest and value of such comparison arise from the shared features of the industrial relations systems in the two countries. Canadian labor law is largely modeled on the Wagner Act. American international unions played a major role in organizing Canadian workers throughout the 20th century, and today American-headquartered unions represent almost one-third of unionized workers in Canada. The same businesses operate on both sides of the border in industries as diverse as automobile manufacturing, energy exploration and refining, telecommunications, transportation, and retail trade. The two countries share many cultural and political values. Scholars have used the divergence in union density as a natural experiment to test the effects of economic and contextual factors (Meltz, 1985, 1986), public attitudes (Lipset and Meltz, 1997; Riddell, 1993), public policies (Taras, 1997; Wood and Godard, 1999; Kaufman and Taras, 2000; Weiler, 1983), union vitality (Widenor, 1995; Rose and Chaison, 1996; Meltz and Verma, 1996) and even irregularities in the density measures themselves (Troy, 1992).

JOURNAL OF LABOR RESEARCH Volume x x n , Number 3

Summer 2001

542

JOURNAL OF LABOR RESEARCH

Conventional wisdom had it that Canada is kinder, gentler, and more collectivist (Lipset, 1989), providing fertile ground for unions to grow. Perhaps union strength in Canada rests on "demand-side" explanations - - that Canadian workers translate their nascent collectivism into union support. Survey data, however, belie the continued reliance on this stereotype (Bowden, 1989; Baer, Grabb, and Johnston, 1993; Lipset and Meltz, 1998). Nonunion Canadian workers for many years have indicated much less likelihood of voting in favor of union representation given the opportunity than Americans (Bowden, 1989; Lipset, 1989). According to Bowden (1989, p. 734) Canadians express less approval of unions than Americans. Lipset and Meltz (1997) confirmed these findings, adding that apparently Canadian managers today say they are twice as likely as their American counterparts to threaten adverse consequences or take reprisals against employees who participate in union organizing. Furthermore, the leftof-center New Democratic Party (NDP), a creature of the labor movement, and a once influential source of union succor, is losing influence in the Canadian political landscape. The political agenda is strongly tilting towards the right, as embodied by the fiscal conservatism of the newly emerged federal Alliance Party and the provincial Progressive Conservative governments in Alberta and Ontario. Indeed, half of all Canadians feel that "Canadians have become more like Americans in the past decade" (Maclean's, 1999, p. 48). Political pundits (Simpson, 1993; Newman, 1995) and political scientists (Nevitt, 1996) have noted this attitudinal shift and its impact on unions (Panitch and Swartz, 1993). (Although there is emerging consensus of a shift to the right, few would disagree that collectivist and corporatist values described by Lipset were operative at the time when regulatory approaches to labor law were entrenched in Canada.) A 1993 study of attitude surveys rejected Lipset's well-known thesis that national character and belief systems differ between Canadians and Americans, finding instead that the map of regional cultures involves three major segments: a left-liberal Quebec, a conservative American south, and a moderate sector that includes the remainder of both Canada and the U.S. (Baer et al., 1993). Hence a Manitoba resident has more in common with a Minnesotan than a New Yorker has with an Alabaman. If the explanation for the bi-national density gap is not Canada's greater public support for unions, then economic factors may be responsible. There has been a noticeable decline in private sector union density in Canada, but it is not nearly as marked as that experienced in the U.S., and economists attribute much of it to "deindustrialization" - - structural changes in the economy as jobs are lost in traditionally heavily unionized manufacturing and resource extraction industries and diverted towards the less unionized service, financial, and high tech sectors (Reid and Meltz, 2001, p. 154; Mainville and Olineck, 1999; Akyeampong, 1999). In the U.S., deindustrializing evidence must be bolstered by a host of noneconomic variables (such as managerial opposition to unions) in order to explain density decline (Kochan et al., 1986; Kaufman and Stephan, 1995). Meltz (1986, p. 143) found little impact on density of variations in the distribution of employment by industry and argued that differences in legislation and in the presence of a labor-oriented political party in Canada have more discernible effects.

DAPHNE GOTTLIEB TARAS and ALLEN PONAK

543

There also is literature demonstrating that Canadian unions exhibit greater organizing vitality than their American counterparts. According to Rose and Chaison (1996), Canadian unions are more active in recruiting new members and assign a high priority to organizing. Canadian unions have greater organizing success (Robinson, 1992; Meltz and Verma, 1996). They have been more able to resist concession bargaining and deliver gains to their members (Widenor, 1995). They also have been more politically influential. Thus there may be more credence to the "supply-side" than the "demandside" explanations of union density. Underpinning the various supply-side explanations is a host of public policy differences between Canada and the U.S. Among the better known are differences in certification procedures and variations in the speed of certification, labor board remedies, first contract protections, protection of striking workers, and treatment of nonunion representation systems (Taras, 1997; Budd, 1996; Wood and Godard, 1999; Weiler, 1983; Kaufman and Taras, 2000). Researchers have concluded that Canada's labor laws generally are more hospitable to unions and propose that some of the union density gap can be explained by public policy. The large international literature testing public policy changes as an explanation for union decline adds support for this approach (Brown, 1997; Kenyon and Lewis, 1993). Surprisingly, union security as a comparative public policy variable has been overlooked. There are a multitude of U.S. and British studies that have addressed the impact of union security laws on union fortunes (Schiller, 1999; Addison and Siebert, 1998). A comprehensive recent review concluded that RTW laws reduce union density in the U.S. (Moore, 1998). J Within Canada, Martinello and Meng (1992) examined the effects on union coverage of both economic measures and interprovincial differences in public policy, and their only significant finding was that mandatory dues increase the probability of coverage. Thus, scholars have used union security as a variable within countries. However, in the comparative literature this variable is virtually unnoticed, which is somewhat unnerving given that Canada's mandatory agency shop policy approach is distinctly different from RTW and may well explain at least some variance in the labor movements' growth patterns in the two countries. We suggest that there are sound reasons to propose that the Canadian approach of a mandatory agency shop as a de minimus standard has protected union institutional strength and financial stability, while the American approach of either having union security subject to free collective bargaining or severely restricted by the passage of state RTW laws has produced the opposite result. In making this assertion, we do not debate whether unions and collective bargaining are good or bad - - we take it as a given that public policy has declared that the right to join and promote unions are statutory rights available to workers. Nor do we enter the debate over whether union security provisions invade individual liberties. Our goal is more limited. 2 We simply wish to review the Canadian public policy approach to union security and pose the argument that a legislative framework that promotes the agency shop might well have a great deal more impact on union density than the absence of discussion of the sub-

544


ject might suggest. We do not propose that union security is the only, or even the most important, explanatory variable. But is it plausible that a significant percentage of the density decline in the U.S., and a measurable contributor to the higher and more stable Canadian density figure, might arise from the impact of statutes that directly address union security? III. Mandatory Agency Shop Laws In general terms, union security refers to collective agreement provisions that require bargaining unit employees to join the union and pay union dues. Such contract provisions are subject to union-management negotiation and, depending on the legal framework, may take a variety of forms ranging from no union membership requirements or dues obligations ("open shop") to a requirement that all bargaining unit employees become and remain dues-paying union members ("union shop"). (The more common type of union security contract clauses, with definitions, are set out in Table 1.) The agency shop, our focus, imposes no union membership requirements, but compels all bargaining unit employees, union members or not, to pay union dues. It is with respect to the agency shop that American and Canadian labor policies most sharply diverge. The U.S. 1947 Taft-Harley Act eliminated the closed shop altogether and gave states the right to create laws pertaining to union security. Twenty American RTW states used their powers to effectively ban the agency shop. By contrast, most Canadian provinces, 3 as well as the federal sector, have established the agency shop as a minimum level of union security to which unions are entitled as a matter of law. For most Canadian unions, the agency shop constitutes the statutory floor from which other forms of union security may be negotiated. In a very real sense, therefore, Canada has the converse of RTW - - rather than making it illegal to force employees to pay dues or join a union, as RTW laws do, we require employees in a certified bargaining unit, whether union members or not, to pay union dues. Contemporary statutory treatment of union security in Canada is set out in Taras and Ponak (2001, Table 1). 4 No form of union security, including the closed shop, is prohibited. Four provinces, Alberta, Nova Scotia, New Brunswick, and Prince Edward Island, neither legislate nor prohibit any particular form of union security, but do require the employer to deduct appropriate dues and provide them to the union when authorized in writing by an employee. Section 27(1) of the Alberta Labour Relations Code is typical: "nothing in this Act prevents a trade union from entering into a . . . collective agreement with an e m p l o y e r . . , whereby all e m p l o y e e s . . , are required to be members of a trade union." Beyond simply permitting negotiation of the full range of union security clauses, seven jurisdictions go further. British Columbia, Saskatchewan, Manitoba, Ontario, Quebec, Newfoundland, and the federal government, which together account for 90 percent of the Canadian work force, give unions the right to at least an agency shop through statutory language on dues collection. There are minor variations in approach. For example, Manitoba law provides that "every collective agreement must contain a


545

provision requiring the employer to deduct regular union dues"; in Ontario and Newfoundland the deduction is initiated following "the request of the bargaining agent"; and British Columbia law stipulates that the agency shop must be requested at the time the first collective agreement is negotiated. Some provinces treat the construction industry separately and some statutes provide limited religious exemptions. 5 These minor differences notwithstanding, the legislative thrust is clear. In six provinces and under federal jurisdiction, unions have the legal right to insist that the employer collect union fees from all members of the bargaining unit. IV. Why Should an Agency Shop Matter? North American legislation towards the agency shop falls into three categories: (1) allow the parties to negotiate an agency shop but neither require nor prohibit it; (2) prohibit the negotiation of an agency shop; and (3) require an agency-shop provision at the union's request. In the first category lie four Canadian provinces and 30 American states. The second category comprises the 20 RTW states. Six Canadian provinces and the Canadian federal sector, together encompassing 90 percent of Canadian workers, fall into the third category. Regardless of the legislative regime, some unions successfully convince all or almost all bargaining unit employees to join the union and pay dues even without contractual compulsion. Peer pressure, industry norms, union tradition, management practices, and cultural factors all may translate into near universal union joining and dues paying behavior no different than if such obligations had been negotiated into the collective agreement. But such worksites may well be anachronisms. It is our premise that a cross-sectional view of all bargaining units will show that the legal regime with respect to union security will have a discernible impact on the proportion of bargaining unit employees who join the union and pay dues. Focusing just on dues-paying behavior, we expect that under the Canadian mandatory agency-shop laws virtually all bargaining unit members would pay union dues, a lower proportion would pay union dues under systems where union security is left to union-management negotiation, and that a substantially lower proportion would pay union dues in RTW states that prohibit agency shops. Once employees pay dues, it is not much of a stretch for them also to join the union. 6 Clearly, more research is necessary to explore the individual decision to become a union member in situations requiring dues payment and in situations allowing choice on both dues and membership. (A dated, but useful, study of British teenagers suggests various models; see Cregan and Johnston, 1990). If union security laws, by making it easier or more difficult for unions to acquire agency-shop clauses in collective agreements, affect the proportion of bargaining unit employees who pay dues, union security laws then affect union financial health. Agency-shop provisions eliminate so-called "free riders," i.e., members of the bargaining unit who choose not to pay union dues, and ensure that the union collects dues from all employees in the bargaining unit who, as a matter of law, the union is required to represent. The impact of an agency shop should be especially significant

Newfoundland Prince Edward Island Nova Scotia New Brunswick Quebec Ontario Manitoba Saskatchewan Alberta British Columbia Yukon & Northwest Territories Multiple Provinces

Disaggregated by Province

Primary Industry Manufacturing Construction Transportation, Communication and Utilities Trade, Finance and Insurance Community, Business and Personal Services Public Administration

Disaggregated by type of industry

Total

19 7 29 31 196 332 57 28 88 114 4 84

20 208 89 126 61 355 130

989

No. of Agreements

35,410 7,415 47,616 47,458 559,615 620,093 99,138 55,801 155,065 331,451 9,440 252,954

22,208 251,029 202,996 319,368 147,985 908,934 368,936

2,221,456

No. of employees

Union Shop

Modified Union Shop

Maintenance of Membership

5.3 20.7 19.4 3.1 10.5 8.8 13.6 14.9 . 10.7

5.0 1.9 94.4 4.8 1.6 0.3 -

9.8

. 10.7

42.1 20.7 6.5 17.9 26.8 14.0 17.9 23.9 33.3

20.0 43.8 2.2 14.3 49.2 15.8 15.4

22.3

.

26.3 14.3 13.8 29.0 42.9 11.1 7.0 67.9 2.3 31.6 . 1.2

10.0 22.1 18.3 23.0 26.2 18.5

20.4

1.2

5.3 12.9 3.1 1.8 1.8 5.7 0.9

1.9 2.4 3.4 4.6

2.5

As a percentage of the number of agreements

Closed Shop

(500 or more employees, as of February 2000)

Union Security Provisions in Canadian Collective Agreements

Table 1

21.13 71.4 31.0 19.4 24.5 34.3 61.4 10.7 33.0 8.8 100 70.2

65.0 23.1 55.6 8.2 33.8 53.8

32.9

Rand (Agency Shop)

2.3

0.3

0.3

Open Shop

6.0

13.8 12.9 8.7 15.1 7.0 3.6 19.3 10.5

14.3

7.7

3.4 4.8 18.0

7.2

1 1.6

No Provision

9

r" >

t" 9

z

19

Private Sector

812,209

1,409,247

832,304

36,511

110,666

305,989

78,043

45,734

.

20.7

-

-

-

-

-

-

.

31.6

14.0

12.4

5.3

20.5

23.4

-

.

.

15.6

24.8

27.4

42.1

26.9

19.5

5.9

1.3

3.7

3.6

10.5

6.4

1.3

-

100

22.2

42.7

36.2

36.8

39.7

49.4

76.5

-

-

0.6

1.0

-

-

-

-

requires neither union membership nor dues payment of bargaining unit employees, the decision to pay dues or join the union is voluntary.

Definitions: Closed shop refers to a form of union security under which an individual must first be a member of the union in order to be eligible for hire into the bargaining unit. A union shop requires that all employees in the bargaining unit must, as a condition of employment, become and remain dues-paying union members. A modified union shop is the same as a union shop, except that certain employees in the bargaining unit, typically those hired prior to certification, may be exempted from the requirement to become union members. Under a maintenance of membership provision bargaining unit employees may choose whether or not to join the union or pay dues, but once union membership is taken, must remain union members for some specified period, usually until the contract expires. An agency shop does not require bargaining unit employees to join the union, but they must pay union dues. Finally, an open shop

8.7

14.2

19.5

5.3

6.4

6.5

17.6

Source: Workplace lrfformation Directorate, Human Resources Development Canada, Government of Canada. Information provided in response to our request, February 25, 2000.

520 469

Public S e c t o r Total

307

P h o n e , Electric and W a t e r Utilities

Education, H e a l t h and W e l f a r e

77 78

Local Administration

17

Crown Corporations

Provincial Administration

22

Federal Administration

Public Sector:

Disaggregated by Sector

L/I 4~ ...,.I

0 Z >

Z

:> t"

g~

9 -] ,.-]

Z

Z7

548


in industries marked by high turnover, and those relying on low-wage and part-time workers. As new employees flow into the bargaining unit, they automatically contribute union dues, saving the union the time consuming, expensive, and not necessarily successful task of soliciting dues payment on an individual basis. The reverse is also true. In the absence of an agency shop, unions would likely experience a loss in revenue. For example, the president of a United Food and Commercial Workers local in Idaho estimated that the abolition of the agency shop in the retail food industry, due to the passage of RTW legislation in 1986, resulted in the eventual loss of half of the local union's revenues (UFCW brief, 1995, p. 35). As labor markets become more fragmented and so-called traditional work more precarious, the ability to keep up with the flow of workers into and out of the bargaining unit can only become more critical for union financial health in the future. Financial strength in turn translates into greater resources for organizing and greater ability to service existing members, both of which should promote union membership acquisition and retention.

V. Explaining Union Density It is difficult, without extensive new data analysis that is beyond the scope of this paper, to isolate the extent to which the three approaches to union security have contributed to union density changes over time. The majority of individual state RTW laws were adopted by the early 1950s. Most Canadian jurisdictions with mandatory agency shops did not enact these laws until the 1970s. The effects of the disparate approaches may have been long term and cumulative. In Canada, we are suggesting that the financial security provided by the agency shop laws enhanced servicing and organizing activities, leading to higher union membership, which in turn generated more financial resources. In the U.S., in contrast, the weakening of financial security engendered by RTW laws reduced servicing and organizing resources, leading to loss of membership and further financial erosion. Shedding light on the "virtuous cycle" (at least from a union perspective) in Canada, should provide insight, and testable hypotheses, about the "vicious circle" in which U.S. unions may find themselves. STATISTICAL EVIDENCE

Collective Agreement Provisions in Canada and the U.S. Some insight into the role of differing legal treatment of union security is provided by an analysis of contract clauses in the two countries. Table 1 provides a breakdown of the various union security clauses in Canadian collective agreements covering 500 or more employees, which represent 57 percent of employees covered by collective agreements. 7 The most prevalent form of union security is the union (and modified union) shop, found in 43 percent of major agreements, with particular strength in manufacturing. The agency shop accounts for approximately one-third of major collective agreements and is popular in public administration and public utilities, partly because there are some prohibitions on compelling union membership in these sectors. A further ten percent of major agreements, almost all in the construction industry, contain closed shop clauses. All told,


549

close to 90 percent of major collective agreements contain clauses that require employees to pay union dues. Thus, in most collective bargaining situations, unions begin with the agency shop, and some will use their strength to achieve gains in union membership, for example, through the union shop. Furthermore, there is reason to believe that some collective agreements without any dues requirement may in practice operate as if an agency shop was in effect. Under Quebec law, for example, all employees covered by a collective agreement must pay dues, even in the absence of a specific contract clause. Some public sector statutes contain similar provisions. Thus, some contracts in the "No Provision" category of Table 1 might more properly be placed in the agency shop column. Table 1 also highlights the impact of the absence of a statutorily required agency shop. The four provinces which do not provide a mandatory agency shop, Prince Edward Island, Nova Scotia, New Brunswick, and Alberta, also contain the highest proportion of contracts with no union security provisions at all, from 14 percent to over 20 percent. 8 This suggests that the law matters - - unions are not always capable of successfully negotiating these provisions without facilitating legislation. The absence of legislation setting a minimum agency shop translates into weaker union security provisions for some unions. 9 A 1995 Bureau of National Affairs survey, based on a sample of 400 collective agreements (BNA, 1995), is currently the sole source of published data on the incidence of various forms of U.S. union security provisions (Table 2). It shows that the union shop, requiring both union membership and dues payment, is by far the most prevalent form of union security and that the agency shop is relatively rare. The survey also indicates that 18 percent of collective agreements have neither membership nor dues requirements - - they are in effect open shops. The BNA survey contains a regional breakdown, showing that open shops are most likely in regions with RTW states. In non-RTW regions of the U.S. (e.g., New England) the proportion of contracts with open shops ranged from two to 28 percent. Comparisons of the American and Canadian data must be made with great caution as the sampling methods and time frame are different and there is no guarantee

Table 2 U n i o n S e c u r i t y P r o v i s i o n s in U.S. C o l l e c t i v e A g r e e m e n t s , 1 9 9 5

No Membership Modified A g e n c y Maintenance or Dues Required Union Shop UnionShop ShopOnly of Membership in Contract Proportion of Contracts

64 percent

10 percent

4 percent

4 percent

22 percent

Source: Bureauof NationalAffairs,Basic Patterns in Union Contracts. WashingtonD.C.,BNA,1995,p. 98.

550


that the coding of the contract provisions, even using similar definitions, is the same. Nevertheless, certain general patterns are evident. First, the pattern of union security provisions differs quite substantially in the two countries. American contracts are much more likely (74 percent to 52 percent) to require union membership through the union and modified union shop than is the case in Canada (even adding in closed shop provisions). The agency shop, which accounts for one-third of union security provisions in Canada is almost invisible in the U.S. (4 percent). In the U.S., membership and dues are coupled; in Canada they are more likely to be decoupled. In Canada many unions have succeeded only in requiring dues collection from bargaining unit employees without necessarily requiring union membership; in the U.S., dues are collected through union membership. At the same time, a higher percentage of American contracts require neither membership nor dues payment - - 22 percent in the U.S. versus 15 percent in Canada (and for reasons explained above, we believe that some of the Canadian open shop provisions are de facto agency shops because of legal requirements). Thus, while U.S. unions arguably have been more successful at negotiating membership requirements than their Canadian counterparts, there is, at the same time, a higher likelihood of collective agreements without any form of union security. Although there may well be bargaining power implications for compelling membership, there should be no financial difference to union revenues between the agency shop and a stronger form of union security. Canadian unions both cover a proportionately larger number of workers and make more money in doing so. Second, the data in Tables 1 and 2 confirm that agency shop patterns reflect the legal structure. In Canada, those jurisdictions with mandatory-agency shop requirements have a higher proportion of contracts that require dues payment than those jurisdictions without such laws. The U.S. differences are much more obvious as RTW laws ban the agency shop. Accordingly, the vast majority of collective agreements in nonRTW states require dues payment from bargaining unit employees; the reverse is true in RTW states.

Union Density and Collective Agreement Coverage. In addition to analyzing contract provisions, there is another way to assess the impact of union security laws. Data are available on union density and collective bargaining coverage in different provinces and states. Based on surveys, these data measure the extent to which bargaining unit employees have actually joined or not joined their union, not just the type of union security clause under which employees fall. They allow us to compute the proportion of workers covered by collective agreements who are not members of the union, in effect representing a rough proxy of "free riders," the proportion of employees who the union must represent but who choose not to become union members. The measure itself is calculated as the difference between collective bargaining coverage and union membership density expressed as a percentage of union membership density. Just as we expect that the incidence of collective agreements that require neither union membership nor dues collection to decrease with the strength of legal protection for union security, we also expect that the gap between union density and collective bargaining coverage to shrink the stronger the legal protection for union security.l~

D A P H N E G O T T L I E B T A R A S and A L L E N P O N A K

551

Table 3 sets out the results of this analysis by comparing the gap between union density and collective agreement coverage in RTW states, non-RTW states, provinces without a mandatory agency shop, and provinces with mandatory agency-shop laws. As predicted, the gap is greatest where legal restrictions exist on the negotiation of union security provisions and lowest where unions are guaranteed a minimum agency shop by law. It shows that unions in RTW states must service the needs of an additional 21 percent of bargaining unit employees who are not members of the union. By contrast, the gap in non-RTW states is 9 percent.11 The Canadian data demonstrate the same pattern. Provinces with mandatory agency shop laws have a lower gap (7.5 percent, and likely even lower because many workers are paying dues when their collective agreements are silent on union security) than provinces in which union security is subject to negotiation (11.9 percent). Taken together the data in Table 3 reinforce our argument that union security laws should be included as a variable to help explain variations in union density. All things being equal, the weaker the union security laws, the more likely it is that unions will have to represent and service bargaining unit employees who are not union members. We recognize that some of these employees will pay union dues, but based on our analysis of collective agreement provisions in the previous section, this is more probable in Canada where union membership and dues paying are not as closely linked as in the U.S. The greater the gap, the more the financial strain placed on unions and the less the resources available for organizing. Weaker financial resources should translate either directly or indirectly into lower union density, j2 Table 3

Gap between Union Densi~ and Collective Agreement Coverage

Union Density, %

Percentage of Employees Covered by Collective Agreements

Percentage of Additional Workers Covered than Are Union Members a

8.13

9.85

21.2

U.S. Non-RTW States

16.14

17.56

8.8

Canada - - Mandatory Rand Formula Provinces (Nfld, Que, Ont, Man, Sask, B.C.)

31.8

34.2

7.5

Canada - - Non-mandatory Rand Formula Provinces (PEI, NS, NB, Alta)

24.4

27.3

11.9

U.S. Right-to-Work States

Sources: U.S. data adapted from data in Barry Hirsch & David Macpherson, Union Membership and Earnings Data Book: Compilations from the Current Population Survey (1999 edition), http:l/www'bna'c~176176 Canadian data adapted from data in Ernest B. Akyeampong, Unionization: An Update, 11 Perspectives on Labour and Income 45-65 (Autumn 1999); Statistics Canada, Catalogue no. 75-001-XPE. Note: aCalculated as (column 2 - column 1) / column 1.

552


VI. The Underpinning of the Mandatory Agency Shop in Canada What accounts for the Canadian public policy approach to the agency shop? Why have Canada and the U.S. taken such different approaches to the question, despite their common Wagner Act heritage? To help answer these questions we examine, in turn, the origins of the agency shop in Canada, the strong rejection of a right to work in a province that most resembles RTW states, and compare supreme court precedents in the two countries.

Origin of the CanadianApproach:DecouplingDuesAnd Membership. The agency shop was dramatically introduced in Canadian industrial relations (and folklore) during a high-profile strike by Ford production workers in Windsor, Ontario in 1945 (Moulton, 1975). Striking workers, facing concerted opposition from the company, the provincial government, and the police, barricaded the plant with their own cars and trucks. "Union shop and checkoff' was the union's rallying cry, and the central issue in dispute. An agreement signed by the Ford Motor Company with the American UAW in August 1941 had given American Ford workers both the union shop and checkoff, but Canadian management resisted a similar settlement. The federal government eventually convinced the parties to accept a binding arbitration settlement under Justice Ivan Rand of the Supreme Court of Canada. His decision, issued on January 29, 1946, imposed an agency shop, which henceforth, in Canada at least, became known as the "Rand Formula." Justice Rand refused to grant a union shop, but he allowed compulsory collection of dues from all members of the bargaining unit. He framed his decision in the context of making operational the underlying premises of a collective bargaining regime: Certain declarations of policy of both Dominion and Provincial legislatures furnish me with the premises from which I must proceed . . . . IT]he social desirability of the organization of workers and of collective bargaining where employees seek them has been written into laws.... The corollary from it is that the labour unions should become strong in order to carry on the functions for which they are intended. (Rand, 1958). Justice Rand attached conditions, however. As a quid pro quo for compulsory payment of dues, the union became liable for penalties if it engaged in future "illegal" or "wildcat" strikes. The union would lose its checkoff for a specified amount of time scaled to meet the seriousness of the violation. There also were penalties on the individual level of "three dollars per day for every day's absence from work and to a loss of one year's seniority for every continuous absence of a calendar week or part thereof" (Rand, 1958). The Rand Formula was of historic significance because it decoupled union membership and union dues, making the latter possible without the emotionally evocative step of compelling people to join a labor organization against their will. It was a classic and highly visible Canadian compromise, made even more Canadian in approach by using union security assurances to prevent industrial unrest during the life of the collective agreement. 13


553

Two decades later, Rand's approach was endorsed and expanded on by the influential Report of the Task Force on Labour Relations (known as the Woods Report after its chair, Professor H. D. Woods of McGill University). After undertaking a comprehensive review of the state of Canadian labor policy at the behest of the federal government, the 1968 Woods Report made a series of recommendations that continue to guide Canadian labor law. One recommendation was that the agency shop be stipulated as a matter of law, as a minimum form of union security for any certified bargaining agent that requested it: An important ingredient in the Canadian collective bargaining system is that a union gains exclusive bargaining authority and with it the duty to represent all employees in the prescribed unit. . . . These policies, in our view, give the union a claim to general support from employees in the unit in the union's capacity as their collective bargaining agent, whatever other functions it may perform as an instrument of social transformation. This rationale supports the agency shop form of union security. . . . We recommend that the compulsory irrevocable check-off of regular and reasonable dues be available to a certified union as a right upon the negotiation of its initial collective agreement and thereafter, and that this right be extended to a union recognized voluntarily

by the employer (Woods Task Force, 1968, emphasis added). This recommendation was widely adopted by legislatures across the country, entrenching the mandatory agency shop and symbolizing the acceptance of two fundamental principles: (1) the social desirability of unionism and collective bargaining and (2) compelling the payment of union dues is justifiable given that a certified bargaining agent has no choice but to fairly represent all bargaining unit employees. Most Canadian jurisdictions at that time had either an NDP government in office, forming the official opposition, or posing a genuine electoral threat. Enshrining union rights was a priority for the NDR It is not surprising that Canadian public policy at the time, having endorsed the legitimacy of unionism, went a step further and sought to ensure that unions would have the tools necessary for their survival and for the advancement of their goals. One such tool was the right to collect dues from employees they represented. American public policy, in contrast, has been much more tepid in fully accepting union legitimacy and has increasingly fettered the ability of unions to advance their interests through the Taft Hartley Act, Landrum Griffin Act, and court decisions. Right-to-Work M o v e m e n t in Canada. Not only have Canadian policymakers (and, as we shall discuss below, the courts) endorsed the right of unions to negotiate a full array of union security provisions, made the agency shop a minimum requirement for most of the country's unionized workers, and placed few (if any) restrictions on the purposes to which union dues may be put, legislatures have also explicitly rejected RTW.

Of all the Canadian provinces, Alberta most closely fits the profile of a RTW state. It is dominated by the oil industry and ideologically more closely resembles Montana, Colorado, and Texas than it does its adjacent provinces. It has Canada's lowest union density with private sector unionization standing at approximately 12 percent. It has chosen a labor law niche as the province most likely to defer to employer pref-

554


erences and its legislation does not contain a mandatory agency shop. The NDP is almost vanquished from holding influence in the legislative assembly (Ponak et al., forthcoming). While there have been initiatives to introduce RTW in Canada from time to time, in the mid-1990s Alberta gave RTW its most serious consideration to date, largely due to the intensive efforts of two pro-RTW interest groups - - the National Citizens' Coalition's (NCC) and Canadians Against Forced Unionism (CAFU). 14 Their sustained lobbying campaign bore fruit when a private member's bill (Motion 503) was introduced in Alberta Legislature in 1995, urging the government "to initiate a study to examine the implementation of fight-to-work legislation in the Province of Alberta." The motion was passed by a vote of 32 to 31, with 12 Cabinet Ministers supporting it. The Minister of Labour, 15 whose stated public position was that there was no need for any significant change in labor law, asked an existing government-established consultative body, the Alberta Economic Development Authority 16 to review RTW with special consideration of its economic effect in Alberta. The populist (and popular) provincial premier, Ralph Klein, stayed clear of the debate. The Alberta Economic Development Authority committee that examined RTW included representatives of business, government, and labor. It received more than 200 submissions from individuals (mostly union supporters), unions, and employers and employer associations. Unions were predictably extremely hostile to RTW, arguing that the very term was "mischievous" and "misleading", and submitting that RTW meant "Right to Work for Less" and was union busting in intent. Summaries of American AFL-CIO studies were relied on extensively. In addition, many Alberta unions took the opportunity to highlight those sections of the Alberta Labour Relations Code that they considered heavily favored employers. Much less predictably, most unionized Alberta employers and employer associations who made submissions also opposed RTW. Major employers wrote that RTW was neither necessary nor desirable for the operations of their businesses. In its submission to the Economic Development Authority, Canada Safeway (an Americanowned company) pointed out that "the current legislation in Alberta is already more advantageous to employers than in other areas [of Canada]. We see no point in drawing attention to legislation that is working." In another submission, the Vice President of Westfair Foods (a Canadian-owned wholesale and retail food and merchandise company) agreed with Safeway, adding that RTW would favor the entry of low-wage American competitors to the detriment of existing unionized grocery chains: i

There is no doubt that in our comparative experience [operating throughout Canada], Alberta has the most reasonable and balanced approach to employment and labour law in this area of the country. The Right-to-Work proposed legislation.., will cause this delicate balance to shift . . . . Trade unions will have less resources to organize other places of employment which will limit the growth of the unionized sector. While this may be seen as a positive result by some in the employer community, it has a negative consequence for those of us with established unionized businesses with relatively high rates of pay and benefits. This is because recent entrants into the retail


555

sector have tended to be large US firms with a reputation for remaining nonunion with lower rates of pay and benefits. [W]e have invested over a hundred million dollars in our unionized businesses in this province. If this legislation is passed, the so-called "level playing field" will tip further in favor of firms that will more likely remain nonunion. This trend would have a detrimental effect on our existing investment and make us less likely to invest further in our unionized businesses in the future . . . . We encourage the [Alberta] government to stay the course with the balanced approach and not be tempted to move to the right by abandoning the [Rand] formula. 9

.

.

A construction industry employers association urged the government to recognize that the introduction of RTW would destroy labor relations stability in the industry and reverse the many positive union-management achievements of the previous decade. Both the Calgary and Edmonton Chambers of C o m m e r c e feared that RTW would unnecessarily instigate labor unrest. The Human Resources Institute of Alberta, an umbrella organization representing diverse professional groups, felt that there existed neither the rationale nor evident economic benefit for RTW. Many large and potentially influential nonunion employers stayed out of the debate. Most submissions favoring RTW came from smaller nonunion companies and free-market interest groups such as the Canadian Federation of Independent Business. The Committee reviewed the evidence, concentrating exclusively on the economic rationales presented to it (avoiding any discussion of individual and collective rights, which it considered an ideological quagmire), and concluded that there were few economic benefits to be gained by RTW and that RTW legislation would disrupt Alberta labor relations. The Committee (including the employer representatives) unanimously rejected RTW. As a result, RTW was not adopted by the Alberta legislature, although some RTW lobbying efforts persist. The Alberta case is instructive because it demonstrated little appetite for a fundamental policy shift on union security even in the province where such a shift would be most expected. It also is a case study documenting that once a union presence is taken for granted in an industry, there is a higher likelihood of a union acceptance strategy by major employers (Thompson, 2001; Cooke and Meyer, 1990). Large unionized employers opposed RTW legislation fearing its destabilizing potential. Nonunion employers remained on the sidelines. The most influential members of the most conservative government in the country at the time also displayed little enthusiasm for RTW. This rejection of RTW in Alberta emphasizes the underlying entrenchment of current Canadian policy on union security and reinforces the differences between Canada and the U.S. on this issue.

The Lavigne and Abood Cases. 17 Two landmark Supreme Court decisions, one American and one Canadian, further underscore the different directions taken in the two countries. Both cases involved a major challenge to the agency shop and the use of union dues. Of particular interest for our purposes, the Canadian court in Lavigne (1991) explicitly rejected the American court's approach in Abood (1977).

556


Louis Abood was an American public school teacher who objected to paying the union dues required by his collective agreement. He asked for declaratory and injunctive relief against the discharge of any teacher for failure to pay union dues and relief from the payment of dues in support of union activities unrelated to the union's functions in negotiating and administering the collective bargaining agreement and directly servicing the needs of members in grievances and disputes. The U.S. Supreme Court unanimously held that a union is constitutionally free to spend funds for any causes not germane to its primary duties as collective bargaining agents, but that "such expenditures be financed from charges, dues, or assessments paid by employees who do not object to advancing those ideas and who are not coerced into doing so against their will by the threat of loss of government employment" (Abood, 1977, p. 236). Union members who register their dissent would be entitled to a refund of some portion of their dues. The 1991 Canadian case is strikingly similar on its facts. An Ontario community college instructor, Mervyn Lavigne, covered by an agency shop clause, objected to his union's expenditure of dues on causes unrelated to collective bargaining. Ontario is one of the provinces with a mandatory agency shop. 18 Lavigne argued that his constitutional guarantees of freedom of association and freedom of expression under the Charter of Rights and Freedoms had been violated. The outcome of the case, however, was the opposite of Abood. The Supreme Court of Canada unanimously rejected Lavigne's appeal, with three different opinions being written by the seven-judge panel (summarized in Lynk, 2000). The Canadian court discussed the perceived deficiencies of the Abood decision, conceding that the closest comparison case from another jurisdiction on which to comment is Abood (Lavigne, 1991, p. 100). First, the Court believed that Abood invited great amounts of litigation and court intervention into discerning appropriate collectives bargaining activities. "In my view," wrote Justice Wilson on behalf of the Court, "nothing could be more incompatible with the promotion of collective bargaining. Since the ultimate aim of the system is to encourage the parties to settle their own disputes, any ruling which will encourage the parties to rush off to court is clearly counter-productive" (Lavigne, 1991, p. 101). 19 Second, the Court was concerned that following Abood would result in a "disproportionate weakening" of the union voice. Corporations are permitted to advance their agenda in spite of shareholder dissent, the Court noted, while unions would be required (on request) to provide dues refunds to dissenting members of the bargaining unit. This was an unacceptable double-standard to the Canadian court, which held that "it is naive to suggest that imposing an obligation to refund dues will not work an unfair burden on unions. Given that the success of collective bargaining in this country depends in a fundamental way on an appropriate balance being struck among the various interest groups involved in industrial relations, to upset that balance will clearly defeat the legislature's objective" (Lavigne, 1991, p. 102). Third, the Court recognized that the labor movement derived clear benefits for their members from union activities in the political and social realm that might appear


557

distant from the direct needs of collective bargaining, yet were not inconsistent with the role expected of unions in a democratic society. Indeed, the political causes of unions may indirectly affect the bargaining table (e.g., by lobbying for stronger medicare coverage, a union can affect issues at the bargaining table; financial support of the NDP might create favorable labor legislation). Meaningful democracy at the workplace entails the right of labor organizations to determine their own interests. Even if compulsory dues resulted in some infringement of an individual's personal freedom, the Court concluded that this particular violation of individual liberties can be "demonstrably justified in a free and democratic society" (Lavigne, 1991, p. 144). Unionism and collective bargaining, institutions deemed to be in the public interest, were to be protected, even at the expense of individual freedom. In short, the Canadian court upheld the right of unions to require all members of the bargaining to pay union dues and refused to fetter the discretion of unions to spend those dues as they saw fit, explicitly rejecting the American approach in the latter regard. The Lavigne decision strengthens the legal basis for union security provisions in Canada, whether negotiated or imposed through legislation. 2~The Abood decision and subsequent holdings (e.g., Communications Workers v. Beck, 487 U.S. 1988), in contrast, by imposing restrictions on how unions can spend employee dues, arguably undermines further the already precarious institutional protections afforded American unionism and collective bargaining by setting a precedent favoring individual rights over the collective. 21 An added complexity in the U.S. is that while full members of unions pay full dues, sometimes covered nonmembers may pay only partial dues, termed an "agency fee?' The agency fee consists of a contribution directly linked to the costs to the union of collective bargaining, contract administration and grievance processing and is always less than full dues. It excludes union expenses such as political activities, conventions, and publications. The partial payment of dues is highly unusual in the Canadian landscape - - indeed, the Lavigne decision holds that such a bifurcation of dues is incompatible with a collective bargaining regime. Thus, a study comparing Canadian and American union fortunes must anticipate the effects on revenues of different dues structures. A dichotomous variable (dues-paying or free riding) that describes the Canadian situation simply will not capture bi-national complexities. VII. Why Union Security Might Explain Some of the Density Gap The differences between the Canadian and American approaches to union security can be quickly summarized: (1) there are no RTW laws in Canada; (2) there are no restrictions in Canada on the negotiation of compulsory union membership and payment of dues; (3) there are no restrictions in Canada on how unions may spend the dues they collect from bargaining unit employees, union members or not; and (4) the agency shop is a minimum legal obligation for most of the unionized Canadian work force. These differences, which might have a demonstrable impact on Canadian union den-

558


sity figures and on the overall healthier condition of Canada's labor movement, have received scant attention. How might stronger legal protection for union security in Canada help the labor movement? Given the imperfect data on which we rely, we cannot do more that assert a line of reasoning and seek support from other sources. Certainly the absence of RTW legislation has spared Canadian unions from facing a potent weapon in the battle to weaken union institutional support. A recent extensive review concludes that RTW laws reduce U.S. union membership by 5 to 8 percent in those states with RTW and that free riding is approximately 6 to 10 percent higher in RTW states than in non-RTW states (Moore, 1998). Union organizing efforts and successes in NLRB certification elections are significantly reduced in RTW states. 22 On the other hand, the overall impact of RTW laws on Canadian-American union density differences may be small. RTW laws in the U.S. cover only a fraction of the private-sector work force, and many of the 20 RTW states are relatively sparse in population and less industrialized. These states are in areas where unions might have an extremely hard time organizing even without RTW, and even in those 20 states, RTW laws do not cover two heavily unionized industries subject to the Railway Labor Act, railroads and airlines. We may be overstating the impact of RTW on reducing American union density. But more important in our view is whether the converse true - - namely does the positive protection of union security increase union membership, union organizing, and union strength? There is remarkably little empirical research on this question, but the sole available study suggests that mandatory agency shop legislation increases union density. Relying on labor force survey data, Martinello and Meng (1992) examined the effects of inter-provincial labor legislation differences on the likelihood that an individual would be unionized. They focused on certification procedures, strike replacement restrictions, and union security. They found that the presence of mandatory agency shop legislation boosted the long-run probability of union coverage by 17.6 percent. The other two labor legislation variables used in the model had no impact on union coverage. We believe that there are sound theoretical reasons for Martinello and Meng's study's findings. The stronger the form of union security, the more revenues flow into union coffers. Canadian workers are paying union dues in greater numbers than their American counterparts, providing unions in Canada with a proportionately greater economic base on which to build their organizing and servicing activities and enhance their bargaining power. Perhaps this partially helps explain why, with a population base of roughly one-tenth of the U.S., Canadian unions are out-organizing their American counterparts in absolute numbers (Meltz and Verma, 1996), even despite the Canadian appetite for unions being less than that in the U.S. (Lipset and Meltz, 1997). Canadian unions were more than ten times as active in filing for certifications than American unions over a fifteen-year period from 1980 to 1995. During that time, the certification rates (i.e., certifications as a proportion of certification applications) were 69.2 percent in Canada and 47.17 percent in the U.S. (Meltz and Verma, 1995), with


559

Canadian unions enjoying greater protections during union organizing and more expeditious certification procedures and elections (Taras, 1997). A fruitful area of research would be an investigation of international unions that would allow comparison of union security provisions and dues collection in RTW states, non-RTW states, provinces with mandatory agency shops, and provinces without mandatory agency shops. We attempted to launch a first-cut empirical investigation along these lines in 2000, contacting three unions that have a strong presence on both sides of the Canada-U.S. border. Neither the United Food and Commercial Workers Union nor the United Steelworkers of America keep their financial information and reporting in such a manner that enables us to extract differences between dues collection in RTW and non-RTW states or between Canada and the U.S. in terms of agency shop security clauses. The United Autoworkers declined to discuss their finances without a written submission to the President and the approval of their Executive. The UFCW well understands the concept of the "fiduciary penalty incurred by the union in representing members that are not required to pay dues" and the gap between dues-paying members of the union versus members of the bargaining unit to whom the union owes a statutory "duty of fair representation." Neverthless, the UFCW feels no impetus to gather or analyze their data along those lines. This applies to both dues collection and to union expenditures generally. The union's focus and concern is over the absolute number of members that they represent in any given state: not surprisingly the numbers are quite small in RTW states. For example, in RTW Texas the UFCW has only organized a single employer in the retail food sector, and total membership is low. Although there are thousands of potential members associated with the "world's largest hog kill" in South Carolina, the UFCW has not made organizing inroads despite their annual organizing drive. Once the union signs up members there is very little attention regarding dues. The United Steelworkers recognized that these data would be interesting to have, but they do not currently identify any differences between states on dues collection or general expenditure patterns. Based on our experiences with three major unions, we caution that researchers interested in gathering data to test some of our propositions developed herein will need a great deal of determination. Furthermore, separating out the independent effects of union security provisions from other variables affecting organizing decisions and expenditures as well as bargaining power will be a daunting task. VIII. Other Possible Effects of the Mandatory Agency Shop There are other plausible arguments to be made about the effects of public policy on union security matters that may indirectly affect union density. We examine the impact on collective bargaining, and on unfair labor practices.

Does the provision of mandatory and universal dues collection make a difference in the ease with which unions achieve first collective agreements? Union security is often a lightening rod in many first contract negotiations, a spillover from

560


acrimonious organizing campaigns. Removing that issue from the bargaining table eliminates one of the most emotive issues on which bargaining may founder, especially in new certification situations. Moreover, unions that are provided an agency shop as a statutory minimum do not have to use up finite bargaining "capital" to achieve financial security, allowing them to pursue other desirable objectives, solidifying membership support. The only study of the effect of eliminating union security provisions in American RTW states on decertification rates found a "decertification increment" of 2.4 percent that could be attributed to RTW laws (Hunt and White, 1983). Alberta experience is suggestive as one of four provinces that does not provide a mandatory agency shop. We know of several cases in which first-contract negotiations foundered over union demands for an agency shop, and decertification resulted. One case involved unionization of a small plant of one of the major international petrochemical companies, previously running union-free in both Canada and the U.S. Bargaining over a first collective agreement reached an impasse over union demands for an agency shop. Management, under explicit instructions from its American head office located in a RTW state, refused to consider compulsory dues for any of its employees. Asked to support a strike, workers balked, not wanting to endure a strike over an issue they viewed as of no particular consequence to the local worksite. They decertifled the union, and exultant senior management joked about framing the decertification certificate and sending it as a gift to the American parent company. Two other cases, one in the newspaper industry and the other in furniture manufacturing, also illustrate that access to an agency shop as a legal union right would remove a potential obstacle in the transition from certification to first contract. 23 We hypothesize that unions will be more likely to achieve first collective agreements in provinces that provide mandatory agency shops.

Does the absence of an agency shop enable the employer to accurately gauge union strength with consequences for collective bargaining? The union can take internal measures of union support at any time. An employer is much more constrained from intruding into the internal affairs of unions. But an employer may easily assess union support by the number of dues check-off authorizations from the company payroll. Bargaining strategy may be quite different in a situation where 60 percent of bargaining unit members are paying union dues than in a situation in which 95 percent are supporting the union financially. It is our hypothesis that enabling the union to require an agency shop as a minimum negates this potential source of information on union strength, slightly reducing employer bargaining power. Of course this result would be somewhat confounded by the unions' greater financial capacity to take a strike in an agency shop.

Does the absence of at least an agency shop, obtained through negotiation or legislation, lead to a higher level of unfair labor practices by management involving reprisal against union supporters in a certified bargaining unit? In a situation where all employees must pay union dues, all employees are subject to dues check-off. In contrast, in the absence of an agency shop, union supporters who want to have dues deducted


561

must identify themselves and sign authorization cards. It is a simple matter for an employer to determine who is and who is not a union supporter. This has the potential of exposing union supporters to retaliation for their union support if the employer is so inclined. Under an agency shop, it is more difficult to identify union supporters since all employees must pay union dues. Legal requirements for an agency shop can help ensure more protection for union supporters and, by extension, the union by reducing the opportunity (and temptation) to single out union supporters. It is our hypothesis that unfair labor practices and disciplinary actions by employers against union members will be higher in the absence of agency shop provisions. IX. An A g e n d a f o r Research We have posed a series of arguments and hypothesis for further testing, accompanied by our reasoning and expected outcomes. The greatest impediment to our pursuit of more definitive findings is the lack of adequate data. In particular, this topic warrants the following: 1. Collective Agreements: Data on union security provisions in current collective agreements are a good starting point for Canada-U,S. comparisons. The U.S. data are in need of updating. The Canadian data exclude collective agreements covering small bargaining units. Greater consistency is required. 2. "Covered Nonmember" : We need accurate information on dues-paying behavior by bargaining unit members on both sides of the border. The current variable for covered nonmembers as a proxy for free riding has both validity and reliability problems. Are all people who are covered by union contracts but are not union members really free riders? In Canada, we suspect that many covered nonmembers are actually paying full dues. In the U.S., covered nonmembers may be paying no dues or partial dues. 3. Union Financials: Union statements of sources and uses of funds are required to support further analysis on the impact of dues revenues on union decision-making. Do unions differentiate among organizing opportunities - - are they less likely to sow organizing seeds in fallow ground? Is their bargaining power affected by revenue sources? Do they engage in less lobbying activity? Process fewer grievances? Have less staff to service members when dues payments are proportionally lower? 4. Individual Decisions: We must investigate the relationship between dues requirements and the propensity to join or otherwise support the union. Individuallevel data are needed to determine whether employees are more likely to actually become members of the union once they are required to pay dues; or conversely, whether they will avoid engagement with the union if they have the choice of opting out of dues payment. When better data become available, scholars will be able to test for the effects on union density of different union security regimes. The ability to break down the North American data into three categories - - agency shop (1) subject to negotiation; (2) forbidden; and (3) required - - may offer a more complete picture of the impact of

562


union security on union density than would be the case using only two categories within each country. We also will be able to use the same data to explain the percentage of the Canada-U.S. density gap attributable to this public policy variable.

X. Conclusion During a period of strength the Canadian labor movement was able to secure enduring legislation providing for their institutional protection. In the 1940s Justice Rand justified his decision by invoking the spirit underlying the operationalization of a collective bargaining regime. In the 1990s the Canadian Supreme Court continued this tradition by refusing to hollow out unions' rights to a secure financial base and autonomy in decision making. Between the 1940s and the 1990s, the unions' voice was heard in legislatures throughout the country through the NDP. Canadian unions now benefit from, among other things, first-contract arbitration, job protection for striking workers, and a card process of union certification. Thus, although current Canadian attitudes towards unions show considerable detachment, the embeddedness of legal protections helps hold such challenges as RTW in Alberta at bay. During the same years, American labor was unable to translate its strength of numbers into gains in public policy: in fact, as union density peaked, legislation such as the Taft-Hartley Act and Landrum-Griffin Act eroded union institutional protections. The Canadian labor movement's achievement of union security guarantees through legislative endorsement of the agency shop as a minimum collective agreement requirement is not well publicized. We argue herein that the protection of union security might well be another "small difference that matters" (Card and Freeman, 1993) in explaining the divergent experiences of Canadian and American unions. Toward that end, we have suggested four questions worth exploring - - whether legislative policies that make it easier for unions to obtain an agency shop as a minimum level of union security: (1) enhance union financial strength; (2) make it easier for unions to obtain first agreements after certification; (3) are a source of union bargaining power; and (4) protect union supporters from possible employer reprisal. Answering these questions can shed light on differences in union density between Canada and the U.S. and might also illuminate another variable in the search for explanations for the decline in American private sector union density.


563

NOTES *An earlier version of this article appears as "Union Security in Canada" in The Internal Governance and Organizational Effectiveness of Labor Unions, edited by S. Estreicher, H. C. Katz, and B. E. Kaufman, Kluwer Press. We are grateful for their permission to reproduce portions of the chapter. Comments from Dennis Nolan (University of South Carolina) and John Godard (University of Manitoba) are greatly appreciated. The article benefited from their excellent suggestions. Research assistance from Fred Jacques is acknowledged. We follow the Canadian spelling of "labour" for accuracy when it is used in citations, quotes, and titles. JMoore recants his earlier published review of the RTW literature that "suggest that the effects of RTW laws are more symbolic than real," admitting instead that this conclusion was "premature" and that evidence is mounting that RTW laws clearly seem to make a difference (1998, p. 464). 2Our personal view of this matter may be simply stated. Under the current Wagner Act model of union certification, the union is obliged as a matter of law to represent all employees in the bargaining unit, both those for and those against the union, fairly and equally. Since the union must represent all bargaining unit employees, it is reasonable that all employees at least pay union dues. In our view, if there ought to be "no taxation without representation" the converse also should be true - - "no representation without taxation." 3Under the Canadian constitutional division of powers, the ten individual provinces have jurisdiction over the making and enforcing of most labor legislation. The federal government has a small but influential jurisdiction that is restricted to a handful of industries, such as telecommunications and banking (MacNeil et al., 1994). 4The table also is available from the authors by request. It should be noted that most jurisdictions have separate statutes for certain groups of public employees that may also contain language pertinent to union security (Ponak and Thompson, 2001). 5In jurisdictions with legislation permitting exemptions on religious grounds, labor boards have applied a stringent four-part test that in practice allows few employees to evade dues payments: (i) the employee must object to all unions; (ii) the case is stronger if the employee can point to a specific tenet of a religion; (iii) the belief must be objectively related to a belief in the Divine; and (iv) the religious belief must be sincere (requiring evidence of devotion to the practice of the religion). The Canada Labour Board also required a fifth element, that the exemption should be a "compelling necessity" for the employee. See Lynk (2000). For a practical example of the extent of religious exemptions, in our University of Calgary faculty union, only three members are objectors in a bargaining unit of 1,700 members (as of April 2000). Usually an equivalent amount must be donated to a charity. 61t is conventionally assumed that attachment to the union begins with a decision to join the union, and dues payment naturally follows. Where dues are mandated by law, the decision-making process may be reversed. The process of attachment to the union needs further investigation. 7The data were provided by Human Resources Development Canada, Information Services Directorate. Across-jurisdiction comparisons are only available for collective agreements covering more than 500 employees, and the results are likely different than would be the case if all collective agreements regardless of size were included. For example, in our own study of Alberta's collective agreement provisions in 1995, the figures were closed shop - - 27.7%, Union or Modified Union Shop - - 39.5%, Rand formula - - 16.3%, Maintenance of Membership - - 2.1%, and Open Shop - - 14.6%. Much of the difference between the major collective agreement calculations in Table 1 and the Alberta data in this footnote can be accounted for by the prevalence of the closed shop in construction industry union certifications involving smaller bargaining units. See Ponak and Taras (1995), Appendix 2. Table 1's 90 percent figure should not be generalized to all Canadian collective agreements, both major and minor, because there may be a size effect - - perhaps small bargaining units have different bargaining outcomes in union security provisions. Further caution should be exercised with respect to the data in Table 1. The figures may understate the extent to which union certification ensures dues payment.

564


8Ontario, which has a mandatory agency shop at union request, is an exception to this pattern. We speculate that most of the no-security provision contracts fall under public sector statutes that mandate automatic dues collection from all bargaining unit employees. 9Some might suggest the absence of any union security clause might be interpreted as a sign of uniol"~strength the union enjoys such complete bargaining unit support that it feels no need to negotiate dues and membership requirements. While theoretically possible, we believe that the opposite is more plausible - - strong unions have the bargaining power to negotiate strong security provisions and do so. -

-

l~ "free-rider" measure is only a proxy because it is fraught with difficulties. First, as we have pointed out, not all Canadians who appear in this measure are free-riders. Rather, they may be paying dues as part of their collective agreement, or even when the collective agreement is silent on union security. The measure is likely to be more accurate in American RTW states than in nonRTW states (leading Schumacher to study the free-rider phenomenon only in RTW states in his 1999 study of wage effects). There may also be misspecification and faulty reporting in the category of employees who claim they are "covered" by union contracts but are not members of unions. HBudd and Na (2000) use the American Current Population Survey between 1983 and 1993 to estimate that 10 percent of private sector employees covered by collective bargaining agreements are not union members. The gap is 11.08 over these years. Using the CPS data for RTW states, the gap is 19.43. These figures are not comparable to the Canadian data as the data collection methods and time frames vary substantially. 12We recognize that we do not know the marginal cost of representing an average extra employee - - as Dennis Nolan has pointed out to us, the bargaining expenses per bargaining unit may be fixed costs, and the grievance handling costs may not increase proportionately with the size of the bargaining unit or the proportion of employees who choose not to be union members. 13One of the hallmarks of the Canadian approach is a statutory ban on strikes and lockouts during the life of the collective agreement. The grievance mechanism is intended to handle mid-contract conflict. Parties are free to strike or lockout only after the expiration of the collective agreement, and after fulfilling a number of steps outlined in the relevant legislation. The quid pro quo of the Rand solution to the Ford dispute that dues would cease if industrial unrest ensued - - was dropped in most subsequent legislation, it being left to labor boards and courts to determine remedies for illegal strikes. However, under Quebec and Alberta legislation the suspension of union dues for specified periods is one of the specific penalties that may be imposed for illegal strikes by health care workers. See Alberta Labour Relations Code, and "An Act to ensure that essential services are maintained in the health and social services sector," SQ 1986, c. 74, now RSQ, c. M-I.1.

-

-

14The National Citizens' Coalition (NCC) in Canada was formed in 1975 to foster "free markets and individual freedoms and responsibilities under limited government." Information about the NCC can be found in various submissions to the Alberta Economic Development Authority report on RTW. Its webpage is http://www.morefreedom.org.The NCC also uses a web page to describe its positions on unionism and other issues (www.citizenscoalition.org). CAFU is an affiliate of the NCC. It was headed by Rob Anders, who went on to become an Alliance Party Member of Parliament for Alberta. According to the NCC, CAFU is now in "limbo" in Alberta (telephone interview, Oct. 2000). NCC sponsored the Lavigne case before Canada's Supreme Court. 15The provincial Minister of Labour was Stockwell Day. He is now leader of the federal Alliance Party, the Official Opposition to Canada's governing Liberal Party. 16The Alberta Economic Development Authority is a consultative body first established in the 1990s by the current governing Conservative party. Co-chaired by the Premier and a private sector business leader, it gathers prominent Albertans together to comment on important matters of the day.

lTAbood v. Detroit Board of Education, 431 U.S. 209 (1977). See also Railway Employees'Department v. Hanson, 351 U.S. 225 (1956); Machinists v. Street, 367 U.S. 740 (1961); and Lavigne v. Ontario Public Service Employees Union [1991] 2 S.C.R. 211. Lavigne is at: http://www.droit.umontreal.ca/doc/csc-


565

scc/en/pub/199 l/vol2/html/1001scr2 0211.html. Justice Wilson's reference to Abood appears on page 100102 of the internet word version. Justice La Forest's discussion of Abood appears on page 135. 18In September 2000 the Ontario government, a conservative one often accused of being anti-union, introduced legislation to amend the Ontario labor laws. The proposed amendments would, among other things, make it easier for employees to decertify their unions. While elimination of the mandatory Rand formula and restrictions on union expenditures were part of discussion papers circulated in advance of the amendments, no changes to union security were ultimately included in the amendment legislation. 19This is a concern that proved ephemeral. As pointed out by Dennis Nolan (in correspondence with the authors), employee rebate claims have been few, the monetary value of individual rebates is usually measured in pennies, and the litigation over such claims has amounted to a "tiny handful of cases". 2~ March 20, 2000 the Supreme Court of Canada heard arguments in a new case involving a challenge to closed shop legislation in the Quebec construction industry (Advance Cutting & Coring Ltd, et al. v. Her Majesty the Queen (Quebec)). The Court reserved judgement and no decision had been issued as of November 2000. Due to a long history of acrimony and violence among rival unions in that industry, Quebec enacted legislation that has the effect of closing the construction industry to nonunion labor. The Quebec case will focus on the issue of compulsory membership and whether the protection of the freedom of association under Canada's Charter of Rights and Freedoms includes the right not to be compelled to join a union, in the particular circumstance of Quebec's legislated industry-wide closed shop. Funding for this case is being provided by various Canadian and American RTW advocates including the National Right to Work Foundation (Cook, 1999; Francis, 1999; Gamble, 1999). John Godard also pointed out to us (in private correspondence) that from 1996 until August 2000, the province of Manitoba had a law allowing individual union members to prevent their union from using dues for political purposes. The law was repealed with no notable employer resistance. 2JAs an aside, it might also be argued that the American treatment of the agency shop as an infringement on personal liberties is a clear endorsement of the business unionism approach that has characterized the development of the American union movement. By contrast, the Canadian treatment recognizes the activism of Canadian unions in the political arena, particularly the historic interdependencies between the NDP and the mainstream Canadian union movement. 22A cluster of studies examines another aspect of the union security issue, whether there is a wage premium enjoyed by union members over their free rider counterparts. For reasons that remain rather speculative, a number of studies have found that in RTW states, free riders' salaries are between 12 and 14 percent lower (e.g., Schumacher, 1999; Budd and Na, 2000) than union members' salaries (leading Budd and Na to conclude that the free rider label is an inaccurate descriptor). Canadian establishment-level data, however, do not find this effect. Reilly (1996) concluded that once the union has organized more than 25 percent of the work force, the within-establishment differences between members and nonmembers disappears. Commenting on the difference between Reilly's Canadian findings, and his own American data, Schumacher posits that "free riding is correlated with weak union bargaining power and lower union earnings." He finds that free riders are clustered where the union power is small, but is unable to determine the direction of causation: Do free riders erode union strength, or do more workers refuse to pay dues to unions that are weak? At present, this avenue of investigation is stymied by a lack of reliable data on both sides of the border. 23In these disputes, the unions complained to the Alberta Labour Relations Board that the companies bargained in bad faith by refusing to allow the unions an agency shop, given that most unionized employers in the relevant industry had been granted this form of union security, The Board rejected both union submissions on the basis that the legislation did not provide an automatic agency shop and an employer's refusal, in and of itself, does not constitute bad faith bargaining, even if the agency shop is an industry norm. See Construction and General Workers Union, Local 1111 and Dynamic Furniture Corp., Alberta Labour Relations Board File GE-03100 (March 29, 2000); and Graphics and Communications International Union, Local 34-M and Southam Inc. Alberta Labour Relations Board File GE-03216 (April 7, 2000).

566


REFERENCES Addison, John T. and W. Stanley Siebert. "Union Security in Britain." Journal of Labor Research 19 (Summer 1998): 495-517. Akyeampong, Ernest B. "Unionization - - A n Update." Statistics Canada, Catalogue no. 75-001-XPE. Perspectives on Labour and Income 9 (Autumn 1999): 45-65. "Non-unionized but Covered by Collective Agreement," Statistics Canada, Catalogue no. 75001 -XPE. Perspectives on Labour and Income 12 (Autumn 2000): 33-8. .

Alberta Economic Development Authority. Final Report: Joint Review Committee, Right-to-Work Study. (McCoy Committee). November 30, 1995. Baer, Douglas, Edward Grabb, and William Johnston. "National Character, Regional Culture, and the Values of Canadians and Americans?' Canadian Review of Sociology and Anthropology 30 (February 1993): 13-37. Bowden, Gary. "Labour Unions in the Public Mind: The Canadian Case." Canadian Review of Sociology and Anthropology 26 (November 1989): 723-43. Brown, William, Simon Deakin, and Paul Ryan. "The Effects of British Industrial Relations Legislation 1979-97." National Institute Economic Review 161 (July 1997): 69-84. Budd, John W. "Canadian Strike Replacement Legislation and Collective Bargaining: Lessons for the United States." Industrial Relations 35 (April 1996): 245-609 Budd, John W. and In-Gang Na. "The Union Membership Wage Premium for Employees Covered by Collective Bargaining Agreements." Journal of Labor Economics 18 (October 2000): 783-807. Bureau of National Affairs. Collective Bargaining Negotiations and Contracts. Washington, D.C.: BNA, 1989. 9 Basic Patterns in Union Contracts. Washington, D.C.: BNA, 1995.

Card, David and Richard Freeman. Small Differences That Matter. Chicago: University of Chicago Press, 1993. Cooke, William and David Meyer9"Structural and Market Predictors of Corporate Labor Relations Strategies." Industrial and Labor Relations Review 43 (January 1990): 280-93. Cregan, Christina and Stewart Johnston. "An Industrial Relations Approach to the Free Rider Problem: Young People and Trade Union Membership in the UK." British Journal of Industrial Relations 28 (March 1990): 84-105. Cook, Cathy. "Quebecers Rally for Right to Work." Ottawa Citizen, June 29, 1999. Francis, Diane. "Quebec Labour Laws Will Be Put to Test: Group Argues workers' 'Freedom to Associate' Rights Are Being Undermined." National Post, April 20, 1999. _ _ .

"Quebec Union Requirement to Go to Supreme Court." National Post, April 20, 1999.

Gamble, David. "Workers Win a Day in Court: Quebec Construction Law Challenges." Gazette (Montreal), April 20, 1999. Hunt, Janet C. and Rudolph A. White. "The Effects of Right-to-Work Legislation on Union Outcomes: Additional Evidence." Journal of Labor Research 47 (Winter 1983): 47-63. Kaufman, Bruce and Paula E. Stephan. "The Role of Management Attorneys in Union Organizing Campaigns." Journal of Labor Research 16 (Fall 1995): 439-54. Kenyon, Peter and Phillip Lewis. "Union Membership and the Legal and Institutional Environment: Labour Market Policy in Australia and the United Kingdom." Australian Economic Review 102 (1993): 48-61. Kochan, Thomas, Harry Katz, and Robert McKersie. The Transformation of American Industrial Relations. New York: Basic Books, 1986.


567

Lipset, Seymour Martin. "Trade Unionism in Canada and the United States: A Reply to Bowden." Canadian Review of Sociology and Anthropology 27 (November 1990): 531-36. 9 Continental Divide: The Values and Institutions of the United States and Canada. New York, N.Y.: Routledge, 1989. _

_

_

and Ivan Katchanovski. "The Future of Private Sector Unions in the U.S." Journal of Labor Research 22 (Spring 2001): 229-44.

_

and Noah Meltz. "Canadian and American Attitudes toward Work" Perspectives on Work 1 (1997): 14-19.

Lynk, Michael. "Union Democracy and the Law in Canada." Journal of Labor Research 21 (Winter 2000): 37-63. Maclean 's. "Peering Inward and Outward: The Poll Examines the Canadian Identity and Cross-Border Attitudes." December 20, 1999, p. 48.

MacNeil, Michael, Michael Lynk, and Peter Englemann. Trade Union Law in Canada. Aurora, Ont.: Canada Law Book, 1994. Mainville, D. and C. Olineck. "Unionization: A Retrospective." Statistics Canada Supplement. Catalogue no. 75-001-XPE (Summer 1999). Martinello, Felice and Ronald Meng. "Effects of Labor Legislation and Industry Characteristics of Union Coverage in Canada?' Industrial and Labor Relations Review 46 (October 1992): 176-969 Meltz, Noah. "Labor Movements in Canada and the United States9 In Thomas A. Kochan, ed. Challenges and Choices Facing American Labor. Cambridge, Mass9 MIT Press, 1985, pp. 315-349 . "Interstate vs. lnterprovincial Differences in Union Density." Industrial Relations 28 (Spring 1989): 142-58. _

_ and Anil Verma. "Beyond Union Density: Union Organizing and Certification As Indicators of Union Strength in Canada and the United States." Paper presented at Annual Meeting of the Canadian Industrial Relations Association, Brock University, Ontario, 1996.

Moore, William J. "The Determinants and Effects of Right-to-Work Laws: A Review of the Recent Literature." Journal of Labor Research 19 (Summer 1998): 445-69. Moulton, David. "Ford Windsor 1945." In On Strike: Six Key Labour Struggles in Canada 1919-1949. Toronto: James Lewis & Samuel Publishers, 1974, pp. 129-61. Nevitte, Neil. The Decline of Deference: Canadian Value Change in Cross-National Perspective. Peterborough, Ontario: Broadview Press. 1996. Newman, Peter C. The Canadian Revolution 1985-1995: From Deference to Defiance. Toronto: Viking, 1995. Ontario Labour Relations Board. United Steelworkers of America v. Wal-Mart Canada, decision issued February 10, 1997. Panitcb, Leo and Donald Swartz. The Assault on Trade Union Freedoms: From Wage Controls to Social Contract. Toronto: Garamond Press, 1993. Ponak, Allen and Daphne Gottlieb Taras. "Right to Work Study: Submission to the Alberta Economic Development Authority Joint Review Committee", Appendix 2, also available as Faculty of Management Working Paper WP-95-25, University of Calgary (September 1995)9 Ponak, Allen and Mark Thompson. "Public Sector Collective Bargaining." In Morley Gunderson, Allen Ponak, and Daphne Taras, eds. Union-Management Relations in Canada, 4th ed. Toronto: Pearson, 2001. Ponak, Allen, Yonaton Reshef, and Daphne Gottlieb Taras. "Alberta - - Industrial Relations in a Conservative Climate." In Mark Thompson, Joseph Rose, and Anthony Smith, eds. Regionalism in Canadian

568


Industrial Relations, Research Volume of the Canadian Industrial Relations Association, (forthcoming).

Rand, Justice Ivan C. "Rand Formula." Canadian Law Reports 1958, pp. 1366-73. Minutes of the Rand arbitration hearings are available at the UAW Labor Archives, Wayne State University,Detroit, Michigan. Reilly, Kevin T. "Does Union Membership Matter? The Effect of Establishment Union Density on the Union Wage Differential." Review of Economics and Statistics 78 (August 1996): 547-57. Riddell, W. Craig. "Unionization in Canada and the United States: A Tale of Two Countries." In David Card and Richard Freeman, eds. Small Differences That Matter. Chicago: University of Chicago Press, 1993, pp. 109-47. Robinson, I. "Organizing Labour: The Moral Economy of Canadian-American Union Density Divergence, 1963-1986." Queen's Papers in Industrial Relations 1993 - - 1. Kingston, Ont.: Industrial Relations Centre, Queen's University, 1993. Rose, Joseph B. and Gary N. Chaison. "Linking Union Density and Union Effectiveness: The North American Experience." Industrial Relations 35 (January 1996): 78-105. Schiller, Reuel El. "From Group Rights to Individual Liberties: Post-War Labor Law, Liberalism, and the Waning of Union Strength. Berkelo, Journal of Employment and Labor Law 20 (1999): 1-73. Schumacher, Edward J. "What Explains Wage Differences Between Union Members and Covered Nonmembers?" Southern Economic Journal 65 (January 1999): 493-513. Simpson, Jeffrey. Faultlines: Struggling for a Canadian Vision. Toronto: Harper-Collins, 1993. Taras, Daphne Gottlieb. "Collective Bargaining Regulation in Canada and the United States: Divergent Cultures, Divergent Outcomes." In Bruce E. Kaufman, ed. Government Regulation of the Employment Relationship. Madison, Wisc.: University of Wisconsin Industrial Relations Research Association, 1997, pp. 295-341. _

_ and Allen Ponak. "Union Security in Canada," In Samuel Estreicher, Harry C. Katz, and Bruce E. Kaufman, eds. The Internal Governance and Organizational Effectiveness of Labor Unions. Boston: Kluwer, 2001.

Thompson, Mark. "The Management of Industrial Relations." In Morley Gunderson, Allen Ponak and Daphne G. Taras, eds. Union-Management Relations in Canada, 4th ed. Toronto: Pearson, 2001. Troy, Leo. "Convergence in International Unionism, Etc.: The Case of Canada and the U.S." British Journal of Industrial Relations 30 (March 1992): 1-43. United Food and Commercial Workers Union. Submission to the Alberta Economic Development Authori~ on Right to Work. Calgary, Alberta: UFCW, 1995. Weiler, Paul C. Reconcilable Differences. Toronto: Carswell, 1980. "Promises to Keep: Securing Workers' Rights to Self-Organization under the NLRA." Harvard Law Review 96 (1983): 1769-827. Widenor, Marcus R. "Diverging Patterns: Labor in the Pacific Northwest Wood Products Industry." Industrial Relations 34 (July 1995): 441-63. Wood, Stephen and John Godard. "The Statutory Union Recognition Procedure in the Employment Relations Bill: A Comparative Analysis." British Journal of Industrial Relations 37 (June 1999): 213-22. Woods Task Force. Canadian Industrial Relations: The Report of the Task Force on Labour Relations. Privy Council Office, December 1968.