Discussion Paper


Home | Site Map


Board Logo



Health Care Conference
September 24-25, 1996

[1997] Alta.L.R.B.R. DP-002

Probably no other determination is as central to the mandate of labour relations boards as determining the appropriate bargaining unit. The purpose behind the concept of the "appropriate bargaining unit" is easy to state: it is "an effort to inject a public policy component into the initial shaping of the collective bargaining structure, so as to encourage the practice and procedure of collective bargaining and enhance the likelihood of a more viable and harmonious collective bargaining relationship": Hospital for Sick Children [1985] O.L.R.B. Rep. Feb. 266.

It is easier to state that purpose than to meet it because of the number of public policy considerations, often conflicting, that come into play. A lengthy passage from another Ontario case, Kidd Creek Mines Ltd. [1984] O.L.R.B. Rep. March 481, gives a sense of those considerations:

(...) the Board's determination is obviously of immense practical importance, not only for the immediate parties, but for the structure and performance of the collective bargaining system as a whole. The definition of the unit affects the bargaining power of the union and the point of balance it creates with that of the employer. It influences the potential scope and effectiveness of collective bargaining for dealing with different matters, and to some extent, even the substantive issues covered in the collective agreement. And, perhaps most important, the shape of the bargaining unit can profoundly influence the potential for industrial peace or collective bargaining discord. The more disparate are the interests enclosed within the unit, the more difficult it may be for the union to effectively represent the collectivity. Insufficient attention to these special interests generates internal strife, while too much attention to minorities may make it more difficult for a union to formulate a coherent package of proposals or make necessary concessions. On the other hand, there are dangers at the other extreme, as the Board noted in Bestview Holdings Limited (...)

(...) The Board must also strive to create a viable structure for ongoing collective bargaining and, to this end, undue fragmentation must be avoided. Consolidated bargaining offers several advantages over a fragmented structure. A proliferation of small units may result in unnecessary work stoppages. (...) Secondly, each of several units typically becomes a separate seniority district, enclosed by walls which impede the movement of employees between jobs. In addition, broader-based structures may lower the cost and thereby increase the availability of insurance schemes and benefit plans. A multiplicity of bargaining units also inevitably spawns jurisdictional disputes over the assignment of work and entails the cost of negotiating and applying several collective agreements. Finally, the existence of a single bargaining unit facilitates equitable treatment of employees doing similar jobs.

This paper briefly discusses some of the principles that the Alberta Labour Relations Board and other labour boards use to determine appropriate bargaining units in the organized workplace. It overlaps and serves as background to the topics of functional bargaining units in the health care industry and geographic limits in health bargaining units. The aim of the paper is to focus participants on the factors that contribute to a "good" bargaining unit, meaning one that serves the interests of its members well while remaining consistent with public policy. The other papers then elaborate on these principles in the context of the geographical scope of the unit, and the Board's functional bargaining units in the health care industry.

General Principles

  • The case law tells us that an appropriate bargaining unit should have these qualities:
  • it should be drawn so that employees have reasonable access to collective bargaining;
  • it should be large enough to make the unit a viable vehicle for collective bargaining;
  • it should not be drawn in a way that unduly interferes with the employer's operations;
  • it should associate employees having enough of a "community of interest" that the collective bargaining process is not unduly impaired by the conflicting interests that the bargaining agent is called upon to represent. The concept of "community of interest" in turn encompasses a host of reasons why a given group of employees should or should not bargain with their employer as a group;
  • it should promote industrial stability in the long term; and
  • its boundaries should be precise enough to minimize disputes over who is and is not in the bargaining unit.

These objectives in a bargaining unit description often conflict, and the final shape of the unit must be the product of a balancing of the conflicting objectives. Each of these qualities of an appropriate bargaining unit is discussed below.

Access to Collective Bargaining

The appropriate bargaining unit sets the initial constituency within which a trade union must gain employee support for collective representation. The right to collective bargaining set out in labour statutes should not be illusory, so labour boards resist creating such large and diverse bargaining units that they are impossible to organize. The B.C. Board put the proposition this way in one of its leading cases:

It is an absolutely fundamental policy of the Code that the achievement of collective bargaining is to be facilitated for those groups of employees who choose to use this procedure as the means for settling their terms and conditions of employment. (...) If bargaining units are defined too widely, or a number of separate groups are put into one unit, it is unlikely in the department store industry that the employees will agree on union representation. In these circumstances we will not deny collective bargaining to those small pockets of employees who, by reason of their own special needs and interests, have.

That does not mean the Board will carve out totally artificial units, based solely on the extent of organization by the union (and sufficiently to give the latter a majority). We will require some reasonably coherent and defensible boundaries around the unit over and above the existing, momentary preference of the employees. (...) However, we will not reject applications for small bargaining units on the basis that a large unit is a more rational structure for hypothetical collective bargaining in the distant future, where the result will be the denial of actual bargaining rights now.

Woodward Stores (Vancouver) Ltd. [1975] 1 Can. L.R.B.R. 114

This approach is especially prevalent in industries that are historically hard to organize. See, e.g. SORWUC v. Canadian Imperial Bank of Commerce [1977] 2 Can.L.R.B.R. 99 (Can.L.R.B.); CUBE v. Canada Trustco Mortgage Company [1977] 2 Can. L.R.B.R. 93 (Ont. L.R.B.). In each of these cases the board found a single branch of a financial institution an appropriate bargaining unit.


Over the long term, however, a bargaining unit must be capable of being a "going concern", a vehicle that is at least somewhat effective in bargaining and representing the employment interests of its members against the employer. A too-small bargaining unit usually lacks the leverage against the employer to make collective bargaining a viable process for setting terms and conditions of employment. Again, perhaps the best illustration of this is in the banking and trust industry, where almost none of the single-branch units certified by Canadian labour boards in the 1970's and 1980's survive today. This is the factor behind the Alberta Board's practice of refusing to certify bargaining units that would leave unorganized a "tag-end" group of employees who could never realistically bargain together: Information Bulletin #9.

Community of Interest

"Community of interest" between employees in the unit means that the employees in the unit share enough employment interests in common that the single bargaining agent can effectively represent all the interests of the collectivity. It does not mean an identity of interests, but a broad similarity of employment interests and an absence of seriously conflicting interests. One of the earliest attempts to catalogue the considerations that come into play in evaluating community of interest appears in the Ontario case of Usarco Ltd. [1967] O.L.R.B. Rep. 526. In deciding that employees at two scrap metal recycling yards should constitute one bargaining unit, the Ontario Board looked at the nature of work employees performed; similarities in their conditions of employment; common skills; geographic proximity; common administration and management; and the interdependence of employees and work processes between the two locations.

The best general discussion of "community of interest" is contained in George W. Adams, Canadian Labour Law, 2d ed. (Canada Law Book, 1993). Adams identifies as significant factors:

  • similarity in the scale and manner of determining earnings; in employment benefits, hours of work and other terms and conditions of employment; in the kind of work performed; and in the qualifications, skills and training of employees. The point behind this factor is that collective bargaining will work best when employees work under similar conditions and so have similar problems and concerns for collective bargaining to address.
  • the frequency of contact or interchange among employees and the geographic proximity of workplaces. Where employees often work in more than one of the employer's operations, they will have an interest in both operations being governed by the same set of rules. Breaking these operations up into several bargaining units tends to create "seniority walls" around each unit so that employees have access to fewer promotions and transfers and fewer jobs to bump to in event of layoff.
  • continuity or integration of production processes. This is more a factor for private-sector employers, especially manufacturers. The idea is not just that an integrated production process tends to create common employment interests among employees, but also that if an integrated production process is broken up into several bargaining units, a dispute in one unit can shut down the entire process.
  • common supervision and determination of labour relations policy. To the extent that supervision and labour relations are administered either centrally or on a decentralized basis, employees have an interest in negotiating with the representatives of management who are capable of dealing with their employment concerns and setting their terms and conditions of employment.
  • relationship to the administrative organization of the employer.
  • collective bargaining history. In cases where the Board is asked to define an appropriate bargaining unit and there has been a history of bargaining, that history becomes extremely important. Often, fashioning the appropriate bargaining unit is an educated guess as to whether the bargaining unit sought will be viable for collective bargaining. Where there is an established history of either successful or unsuccessful bargaining it is not necessary to guess. In a sense, a successful collective bargaining history creates its own community of interest as, the employees in the unit develop relationships to each other and a common allegiance to the group.
  • wishes of the parties. This raises the question of Board policy toward agreed bargaining units. The agreement of the parties immediately affected will often be strong evidence that a group of employees is suitable to bargain together, though the Board is not bound by such an agreement where other factors point to the unit not being appropriate.

This is not the only accepted list of factors in evaluating community of interest. The British Columbia Labour Relations Board, for example, looks to the factors of:

  • similarity in skills, interests, duties and working conditions;
  • the physical and administrative structure of the employer;
  • functional integration; and
  • geography.

See Island Medical Laboratories (1993) 19 C.L.R.B.R.(2d) 161. See also the Canada Board's list of factors in Canadian Broadcasting Corporation [1979] 2 Can.L.R.B.R. 41 .

Community of interest requires not just that the employees sought to be included in a bargaining unit share a reasonable degree of common goals and interests. It requires that all employees who share those goals and interests to a high degree be included in the unit. In that sense, labour boards are not bound by the extent of a union's organizing campaign and will resist applications that appear to gerrymander the bargaining unit to fit the pockets of employees that have expressed support for the union. The B.C. Board set out an extreme hypothetical example of a unit that is under-inclusive on community of interest grounds in Island Medical Laboratories, supra at 187:

(...) at the initial stage of certification, the design of the bargaining unit must ensure the viability of collective bargaining. The Board would not put into a single bargaining unit employees whose communities of interest directly conflict; further, no bargaining unit would be created that cuts across a particular classification, where all members are in the same physical location, resulting in half of the employees in that classification in the bargaining unit and the other half out of the bargaining unit. Both these situations would not be conducive to the settlement of collective bargaining disputes.

Industrial Stability

Labour boards resist creating unduly fragmented bargaining structures on the basis that they are unstable and do not contribute to industrial peace through effective collective bargaining. The B.C. Labour Relations Board states the principle this way:

(...) the second purpose in establishing an appropriate bargaining unit is to foster industrial peace and stability through collective bargaining. The bargaining unit sets the framework for actual bargaining between the parties. That structure has to be conducive to the orderly resolution of collective-bargaining disputes by the parties. If the bargaining unit fails to relate to the specific organization and structure of the employer, efficient and stable collective bargaining will be undermined. (...)

It is axiomatic in labour relations that a proliferation of bargaining units increases the potential for industrial instability. Multiple bargaining units per se raise a serious concern about industrial stability. Instead of one strike, there may be several strikes. Each union may potentially whipsaw the employer by trying to leapfrog the last set of negotiations.

Island Medical Laboratories Ltd. (1993) 19 C.L.R.B.R. (2d) 161.

One might add that even in an industry where disputes may not be settled by strike or lockout, fragmented bargaining units tend to spread the uncertainty associated with the collective bargaining process over a greater period of time. In a very fragmented bargaining structure, bargaining may be almost continuous as one segment or the other of the workforce is usually in an open period.

A fragmented bargaining structure can commit an employer to bargaining with so many bargaining agents that collective bargaining becomes an undue drain on the resources of the employer; or it can subject the employer to too many different sets of rules governing things like job posting, seniority and scheduling, for the employer to efficiently administer its personnel. See, for example, the recent case of HSAA et al. v. Chinook Regional Health Authority et al. (Unreported Alta. L.R.B. Decision, July 16, 1996), where the Board observed that maintaining all existing bargaining rights of the new regional health authority unchanged would result in it becoming bound to 25 certificates and a "multitude" of collective agreements.


An appropriate bargaining unit is one whose boundaries are stable and relatively easy to ascertain. Labour boards resist unit descriptions that are vague. As the Alberta Board observed in Re: City of Edmonton Bargaining Units [1993] Alta. L.R.B.R. 362 at 390:

The units must be capable of co-existing in harmony. Unit descriptions which, by ambiguity or poor design, generate constant disputes about who falls where, are inappropriate. They frustrate, not foster the bargaining relationships they authorize.

Boards also reject unit descriptions that rely on criteria that are too subject to change. This is one reason that the Alberta Board tries to cast its unit descriptions in functional language, i.e. language that describes what functions employees carry out in the workplace, rather than by job classifications or the employer's administrative structure, which can be unilaterally changed.

The Importance of Context

The question of what is the appropriate bargaining unit can come up in several different contexts. It may be an issue in a first certification; a second or subsequent certification in the same employer's workforce; a raid; or a successorship application. Different policy considerations come into play in each of these contexts. Labour boards have acknowledged that there are often several different units that are appropriate in any given application, and whether the Board will accept a unit description that is less than the best available unit depends greatly on what kind of application it is.

The difference in approach to bargaining unit appropriateness is particularly obvious between first certifications and second and subsequent certifications. It must be remembered that in certification applications, Alberta's labour statutes allow the Board to certify a trade union to represent "an" appropriate unit, not necessarily the most appropriate unit. But even so, a unit that may be "an" appropriate unit in one context may be inappropriate in another. In the first certification application for an employer, especially in historically hard-to-organize industries, access to collective bargaining is the paramount consideration (as the Canada Board's acceptance of single-branch bargaining units in the banking industry showed). With second and subsequent certifications, industrial stability assumes increasing importance. One noted commentator writes:

That is the industrial relations problem [of bargaining unit fragmentation]. What should be the legal response? Clearly, one avenue that must be pursued is prevention, especially in the development of labour board policy relating to appropriate bargaining units. I have already described the natural and, I repeat, justifiable tendency of labour boards to fashion bargaining units within which groups of employees have a meaningful ability to choose to engage in collective bargaining. But as industries become largely organized; as employees in this line of work look upon union representation as a matter of course; as their employers accept that as a fact of life; then the statutory policy of fostering collective bargaining need have relatively little weight in unit determination in that setting. Instead, labour boards should address themselves to the long-term structural consequences of such decisions, and take a more critical attitude to the boundaries within which a particular trade union has chosen to organize employees.

Paul Weiler, Reconcilable Differences (Carswell: 1980) at 159

Therefore, labour boards will often grant a first certification for a less-than-optimum bargaining unit as long as a "rational and defensible boundary" can be drawn around the bargaining unit applied for. But as the employer's workforce becomes more thoroughly organized, they may refuse to grant additional certifications for less than the remainder of the workforce, in order to prevent fragmentation. Or they may refuse to grant another certification at all, instead requiring additional employees to brought into the existing certificate by an application to vary the certificate. See Island Medical Laboratories, supra; Woodward Stores (Vancouver) Ltd., supra; and UFCW, Loc. 401 v. Freson Market Ltd. [1995] Alta.L.R.B.R. 491.

Where there is a successful bargaining history, that bargaining history tends to entrench the existing bargaining structure. A history of successful bargaining rebuts any concerns about inappropriateness that a labour board might have had if it were making the determination on a "clean slate". And, as noted before, the mere fact of bargaining together over time contributes to a community of interest among employees in the unit. For these reasons — and to afford certainty in raid applications— boards strongly favour the existing bargaining unit as the appropriate bargaining unit in a raid. The Alberta Board has said in its Information Bulletin #9 that an applicant for certification must show "strong, compelling reasons" for carving out a smaller group from an existing bargaining unit. See also UA, Loc. 496 v. Goodyear Canada Ltd. [1992] Alta. L.R.B.R. 89.

Finally, the issue of appropriateness often arises in successorship applications. The established approach to bargaining unit appropriateness in a successorship— and this is particularly important for regionalization initiatives in organized industries — is that the successor employer takes the bargaining unit as it finds it. The existing bargaining unit presumptively remains appropriate in the hands of the new employer and labour boards will preserve that bargaining unit unless it is convinced that the existing structure has become inappropriate — for example, through a mingling of the predecessor's and successor's workforces. As the Alberta Board wrote in CUPE, Loc. 3203 v. Horizon School District No. 67 [1995] Alta. L.R.B.R. 439:

... the intention of the successorship provisions is to place the successor employer in the same bargaining capacity as its predecessor and the predecessor trade union continues to be the bargaining agent of the employees now working for the successor in a "like" bargaining unit, until the Board declares otherwise. Thus, for successorship applications, the previous bargaining unit remains appropriate until shown otherwise through, for example, proving to the satisfaction of the Board the existence of integration and intermingling. This is not a certification application where the Board must only be satisfied that the unit applied for is an appropriate unit for collective bargaining.


Too often, disputes over bargaining unit appropriateness are driven by the tactical considerations of the day. It is hoped that this paper, and this conference, will help to inject into the looming debate over bargaining unit appropriateness in a regionalized health care system, some more thought about the first principles of how bargaining units should be structured. In preparing for the conference, then, the organizers invite participants to consider the principles discussed here and apply them to what they know to be happening in Alberta's health care industry.

Appropriate Bargaining Units Questions to Ponder

  1. What size bargaining unit becomes viable for collective bargaining? Is viability better measured in numbers of employees or in context of the collective bargaining structure affecting the unit? Contrast - a two employee unit in the construction registration system, a small unit in health care, and a small retail unit.
  2. What are the factors existing in the health care industry which may impact employees reasonable access to collective bargaining:

    -in new certifications?

    -in raid situations?

  3. What does or should community of interest include in today's healthcare environment? Have any new factors emerged in the regionalization process.
  4. What factors exist which impact the trade unions' ability to represent employees at the worksite?
  5. What impact is regionalization having on collective bargaining? How do bargaining unit structures affect that?
  6. Are multi-union certificates appropriate in healthcare in Alberta? Why or why not?
  7. How much does provincial bargaining mitigate the burden to employers of having many smaller bargaining units? Is provincial bargaining likely to remain a strong factor in health care labour relations?


1. George W. Adams, Canadian Labour Law, 2d ed. (Canada Law Book, 1993), pp. 7-2 ff.

2. Paul Weiler, Reconcilable Differences (Carswell, 1980), c.

3. Alberta Labour Relations Board, Information Bulletin #9.

4. James Dorsey, Canada Labour Relations Board: Federal Law and Practice (Carswell, 1983), pp. 123-130.

5. A.W.R. Carrothers et al., Collective Bargaining Law in Canada, 2d edition (Butterworths, 1986), c. 14

6. Claude Foisy et al., Canada Labour Relations Board Policies and Procedures (Butterworths, 1986), pp. 59-71

7. J. Sack and M. Mitchell, Ontario Labour Relations Board Law and Practice (Butterworths, 1985) pp. 134-153.

8. Re: City of Edmonton Bargaining Units [1993] Alta. L.R.B.R. 362

9. UFCW, Loc. 401 v. Freson Market Ltd. [1995] Alta.L.R.B.R. 491.

10. UA, Loc. 496 v. Goodyear Canada Ltd. [1992] Alta. L.R.B.R. 89.

11. CUPE, Loc. 3203 v. Horizon School District No. 67 [1995] Alta. L.R.B.R. 439

12. HSAA et al. v. Chinook Regional Health Authority et al. (Unreported Alta. L.R.B. Decision, July 16, 1996)

13. Insurance Corporation of British Columbia [1974] 1 Can. L.R.B.R. 403 (B.C.L.R.B.)

14. Woodward Stores (Vancouver) Ltd. [1975] 1 Can. L.R.B.R. 114 (B.C.L.R.B.)

15. Island Medical Laboratories (1993) 19 C.L.R.B.R.(2d) 161 (B.C.L.R.B.)

16. Usarco Ltd. [1967] O.L.R.B. Rep. 526.

17. CUBE v. Canada Trustco Mortgage Company [1977] 2 Can. L.R.B.R. 93 (Ont. L.R.B.).

18. Kidd Creek Mines Ltd. [1984] O.L.R.B. Rep. March 481

19. Hospital for Sick Children [1985] O.L.R.B. Rep. Feb. 266.

20. SORWUC v. Canadian Imperial Bank of Commerce [1977] 2 Can.L.R.B.R. 99 (Can.L.R.B.)

21. Canadian Broadcasting Corporation [1979] 2 Can.L.R.B.R. 41 (Can.L.R.B.)


| Home | Disclaimer |Contact Us |
Copyright ©1996 Alberta Labour Relations Board