See also: The Changing Workplaces Review
The Labour Relations Act, 1995 (LRA) sets out the means by which workers can organize into unions and establish bargaining rights through certification.
Generally, this process involves the union making an application to the Ontario Labour Relations Board (OLRB) and demonstrating at least 40% support from the appropriate bargaining unit of workers (typically in a single workplace), followed by an OLRB-supervised secret ballot vote. The LRA requires that this vote normally be conducted within five working days of the application. Unfair labour practices (such as interference or opinions on unionization expressed by employers having undue influence on employees, or intimidation by either employers or unions) in the course of this process are prohibited.
Between 1950 and 1995, all sectors covered by the LRA operated under a card-based certification system. This model allowed for automatic certification by the OLRB if more than a certain percentage of employees in the proposed bargaining unit signed membership cards. For most of this period, the threshold for card certification was 55%; however, from 1970 to 1975, the threshold was 65%.
For most of this period, the practice was for the OLRB to set a “terminal date” following the certification application. Up to the terminal date, employees could submit petitions indicating opposition to, or support of, the union. So, for example, a worker who had signed a union card would have the opportunity to withdraw his or her support by signing a petition opposing the union. If there was sufficient overlap between the names on a petition and the union membership cards, the OLRB generally ordered a vote. The OLRB would hold a hearing to determine whether a petition was signed voluntarily.
Frequently, the OLRB found that petitioners were unable to prove the petitions were voluntary. The workers’ signatures, both on union membership cards and on any petition opposing the union, were confidential within the OLRB process.
In 1993, Bill 40 amended the Labour Relations Act, requiring the OLRB to consider employee support as of the “certification application date.” The OLRB was expressly prohibited from considering post-application membership evidence either for or against the union. This basically eliminated petitions in all cases and eliminated votes where the union met the threshold of 55% support.
In 1995, Bill 7 eliminated card-based certification and introduced the mandatory vote model. Certification occurs where a majority of ballots cast are in favour of the union.
In 2005, the Bill 40 model of card-based certification was essentially restored in the construction industry. The OLRB must determine an application for certification without a vote in the construction industry “as of the date the application is filed” and based only on the material filed by the union and the employer’s response.
Note that recommending changes to the construction industry provisions of the LRA is beyond our mandate.
Card-based certification is available in Quebec, New Brunswick and Prince Edward Island, and the federal government has introduced a bill to bring it back. The threshold required to achieve certification based on membership cards ranges from over 50% to 60%. In Manitoba, a bill has been introduced to end card-based certification. In all other Canadian jurisdictions, certification votes are generally required.
We are releasing to the public, concurrent with this Interim Report, a list of research projects commissioned for this Review, including a report on collective bargaining. This report assessed the research literature on the topic of certification models and noted that: “These studies consistently find that the presence of [a mandatory vote certification model rather than a card-based certification] procedure is associated with a statistically significant reduction in certification application activity, including success rates.” Two aspects of the vote model identified by these studies as inhibiting certification are the greater opportunity for delay and the related greater opportunity for employer unfair labour practices to occur, as compared with card-based procedures.
An issue was raised that employees should be able to “sign” membership cards online and not be required to sign paper cards.
Under the current OLRB rules, and LRA section 9.2, there is a requirement that membership cards be in writing, signed by the employee, and dated; this would appear to preclude, or at least not to contemplate, electronic membership evidence. There is no current legal requirement for someone to witness a person signing a card, although most unions follow this practice.
A union is required to file a membership declaration with the OLRB confirming to the OLRB that it has made inquiries and can declare that each card was signed by the people whose name is on the card. The person signing the declaration should be able to trace information back to someone who saw the person sign the card.
Mailed membership, where there is no witness to the signing, is permitted by the OLRB provided there is compliance with certain safeguards. For example, a union official would be expected to check with the person who mailed the card to confirm that the person signed the card and mailed it, and also, to disclose this to the OLRB in the declaration.
The appropriate model for certification is a polarizing issue. In the course of our consultations, unions have strongly favoured card-based certification. They heavily criticize the current vote-based system as unfair because they claim there is extensive direct and indirect employer interference or undue influence in the process, particularly between the application and the holding of the vote, resulting in employee support for the union dissolving. In this era, where union coverage has declined so significantly in the private sector, it is argued that it is necessary to make it easier for employees to have access to collective bargaining and to remove measures that continue to ensure the ongoing decline.
Employer stakeholders strongly oppose card-based certification asserting that the secret ballot vote is the most democratic.
It is argued that in the current online world most transactions that previously were done in writing are now done electronically and that, not accommodating electronic transactions hinders organizing as it generally requires one-on-one personal contact. It is argued that if safeguards can be found for mailed membership evidence to assure the Board of its authenticity, similar devices or other electronic methods can be found to reassure the OLRB that electronic evidence is genuine.
On the other hand, it is argued that anything other than an actual signature raises doubt about the authenticity of the commitment and permits fraud or irregularities. Since current technology does not easily permit an actual electronic signature, it is argued that this should not be permitted.
Currently, when a union files an application for certification the employer must, in its response to the application, provide a list of employees in its proposed bargaining unit. Unions are not entitled to get a list of employees before the application is filed. Other than information they can ascertain from employees, unions do not have any other access to employer information, including the number of employees or employee contact information, such as addresses or emails. Unions have no right to campaign inside a workplace or to have access to employees inside a workplace; employees can engage in organizing activity during lunches and breaks but not during actual work time. Employers know how many employees there are, where they work, and their contact information.
It does not appear that any other Canadian jurisdiction requires that employers provide employee lists in the context of certification campaigns. In the United States, where an election in the workplaces is ordered under the NLRA, unions receive address and email information for the employees on the voters list.
Unions have argued that it is often difficult for them to know how many employees are in a workplace and where they work, particularly if it is a large employer spread out over a large geographic area, such as a university or a large manufacturing plant with staggered shifts. Some unions say that they have spent large amounts of time and resources organizing, only to find out at the time of the application that there were many more employees than they knew about. Unions claim the practice of having workers assigned to a client by temporary help agencies compounds this problem, as the workforce fluctuates and turns over.
Unions also argue that whatever system is used for employees to express their choice, card-based or secret ballot, the union should be able to easily communicate with employees and therefore have some access to voters. It is said that the lack of practical methods of communication with employees impedes the right to organize and freedom of association.
Labour groups have proposed that where an organizing campaign is under way and the union meets the threshold of, for example, 20%, the union could apply to the OLRB for an order requiring the employer to provide a list of employees in the union’s proposed bargaining unit. The lists could include names and job information, such as which department employees work in, and/or personal information, such as home addresses or other contact information.
Regarding the proposal that they might be compelled to provide a union with employee information, employers have raised concerns about:
Secret ballot certification votes are generally cast in person at the workplace. The OLRB conducts the vote, and both the employer and the union are entitled to have representatives present to act as “scrutineers”.
Currently, the LRA does not indicate where or how representation votes must be conducted. The OLRB determines this pursuant to its general powers. The OLRB may already have the power under the current legislation to experiment with different voting techniques. Generally, it is has been acknowledged that a long interval between a vote order and the holding of the vote is usually detrimental to the union. At present, the OLRB normally conducts a vote within five working days.
So far in Canada, there is very limited experience with alternative methods of voting in certification applications. The Canada Industrial Relations Board, for example, has conducted several certification votes using telephone and internet methods. The practice of off-site voting also appears uncommon. Some labour boards (including the OLRB) have used mail-in ballots, but it appears that this method of voting is used only occasionally (as in cases involving a geographically dispersed workforce). Some unions use private sector providers to conduct secret ballot ratification votes via the internet or telephone.
Labour groups have told us that having employees vote at the workplace allows for undue influence by the employer. For example, it is said that a ballot box placed outside a supervisor’s office, or another non-neutral location, can discourage employees from freely expressing their will. There is a concern, particularly in smaller workplaces, that it is quite possible for the employer to deduce, or for employees to believe that the employer can deduce, who is sympathetic to the union. This, in turn, leads to ballots being cast against the union. It is said to be hard for employees to vote for the union while having to line up and vote in the employer’s workplace. In short, employees are said to be intimidated by the fact of, and the circumstances surrounding, the holding of the vote in the employer’s premises.
To address this concern, labour groups have proposed that certification votes be held at a neutral location, away from the employer’s premises (e.g., a public library) or that votes be conducted by telephone or other electronic means.
It has been proposed, for example, that the LRA be amended to give the OLRB specific discretion to order that a certification vote take place at a “neutral” site that is not on the employer’s premises, be conducted by telephone or electronic means, or be conducted through a combination of both.
In response, many employers argue that the employer’s premises is the most convenient and cost-efficient location for the vote to take place and where turnout and the opportunity to vote will be the highest. It was also argued that voting at another location would reduce the number of anti-union voters, as only those most committed to voting would take the time to vote and the union would ensure its supporters went to the outside location. On the other hand, some employers supported electronic voting subject to appropriate safeguards.
Regarding telephone and internet voting, some of the issues expressed by employers are:
The OLRB can certify a union without a vote if the employer has contravened the LRA in a way that makes it unlikely that the true wishes of the employees can be ascertained. The provision applies both to cases where the union was unable to attain the 40% membership support needed to trigger a representation vote or where the union loses the vote. The OLRB can certify the union only if “no other remedy would be sufficient to counter the effects of the contravention.” The OLRB may consider the results of a previous representation vote and whether the union has “adequate membership support” for collective bargaining.
This provision has undergone significant legislative change over the years. The original Labour Relations Act enacted in Ontario in 1950 allowed the OLRB to certify a union on a remedial basis where the union had at least 50% membership. The law was criticized on the basis that the effect of the 50% membership requirement was that an employer could prevent a union from acquiring sufficient membership evidence simply by committing severe misconduct early in the organizing campaign. As a result the LRA was amended in 1975 to provide that the OLRB could exercise its discretion to certify without a vote, provided the union had adequate support for collective bargaining without stating a specific percentage figure. The rationale for maintaining the requirement for adequate support was that it was believed little purpose would be served by certifying a union that could not bargain effectively because it would have no realistic chance of getting enough support to mount an effective strike.
While the OLRB stated that there was no “minimum figure” required to constitute adequate membership support, in practice, it rarely exercised its discretion to certify the union unless there was at least 30% support. Critics then, as now, complained that unlawful conduct often paid off because, in the face of unlawful conduct by the employer, employees were discouraged from joining the union and the more egregious the misconduct, the less likely the union would be able to show adequate support.
Further amendments introduced in 1993 removed the requirement of adequate membership support, but these amendments were repealed in 1995. Under the 1995 law, the OLRB had to find that no remedy short of certification, including a second representation vote, was sufficient to counter the effects of the employer’s violation.
In 1998, the LRA was amended to remove the OLRB’s power to grant remedial certification altogether and replaced it with the power to order a second representation vote. Then, in 2005, amendments to the LRA restored the OLRB’s power to order remedial certification, allowing the OLRB to certify the union if “no other remedy would be sufficient to counter the effects of the contravention,” considering the results of a previous representation vote and whether the union has “adequate membership support” for collective bargaining.
The OLRB does not often exercise its discretion to award remedial certification.
Remedial certification is available in seven provinces and the federal jurisdiction. Key features of this remedy vary. In some jurisdictions, for example, remedial certification is available only where the union can establish a minimum level of membership support:
Unions argue that the balance in the present legislation strongly favours a second representation vote over remedial certification without a vote, and that this is harmful and not genuinely remedial because once the employer breaches the LRA in such a way that the true wishes of employees cannot be ascertained, it is virtually impossible to redress the situation and make a second vote meaningful.
In essence, the argument is that ordering a second representation vote where the employer has engaged in serious intimidation and coercion is a useless remedy – like trying to unscramble an egg. In addition, including adequate membership support for bargaining as a consideration is said to reward employers who attack the union early in the organizing process, making it impossible for the union to attract support because of the employer’s threats and coercion.
Some employers have criticized remedial certification (and strongly opposed its reintroduction in 2005), arguing that it threatens the principles of workplace democracy by removing the right of employees to vote on whether they wish to have a union in the workplace. Removing the requirement for adequate support for bargaining is said to merely create a weak unit that cannot accomplish anything substantive for its members.
The LRA provides that parties may apply to the OLRB for a direction that a first collective agreement be resolved by binding interest arbitration. The applicant must demonstrate that collective bargaining has been unsuccessful as a result of:
First contract arbitration now addresses situations in which, following certification, employers refuse to accept the right of their employees to engage in collective bargaining. It ends the immediate dispute and engages the parties in a “trial marriage” through an imposed agreement, aiming both parties to establish mature and enduring bargaining relationships. It may also act as a deterrent to bargaining in bad faith. In addition, it recognizes that first contract negotiations may be particularly difficult and anticipates that negotiations for the renewal of the agreement will likely be made easier.
First contract arbitration was introduced in 1986. In 1993, the law was amended to provide that, in addition to an application to the OLRB, a party could also apply for arbitration to the Minister of Labour where thirty days had elapsed after the parties were in a strike or lock-out position. There were no legislated factors (e.g., related to failure of bargaining) that were required to be considered.
The legislation was amended again in 1995 to restore the previous first contact arbitration provision; “automatic” access to this process (through an application to the Minister) was removed.
In 2000, further amendments made it mandatory for the OLRB to deal with decertification and displacement applications before dealing with or continuing to deal with applications for first contract arbitration. If the decertification or displacement application is granted, the first contract arbitration application must be dismissed. This reversed previous labour relations policy where, if first contract arbitration was justified, the relationship would be allowed the opportunity to take hold before applications for decertification or applications by competing unions would be considered.
Access to first contract arbitration is available in the federal jurisdiction, British Columbia, Manitoba, Newfoundland and Labrador, Nova Scotia, Ontario, Quebec and Saskatchewan. Manitoba has “automatic” access to first contract arbitration, based only on undergoing the conciliation process and the passage of a certain amount of time. All the other jurisdictions require that other substantive conditions be present before first contract arbitration is ordered.
British Columbia has a unique “mediation-intensive” model, introduced in 1993, which treats first contract arbitration as part of the collective bargaining process and not as a remedy. Upon application, a mediator is appointed. If mediation-assisted bargaining does not succeed, the mediator recommends either first contract terms for the parties’ consideration, or a process for settling the agreement, including one or more of: mediation-arbitration, arbitration by an arbitrator or the British Columbia Labour Relations Board, or permitting the parties to engage in a work stoppage.
Labour organizations consistently argue in favour of “automatic” access to first contract arbitration, complaining that the existing provisions are too restrictive. They argue that effective access to first contract arbitration is required as part and parcel of policies needed to reverse the decline in private-sector union density. First contract arbitration would make unionization more attractive to workers in that they would know that a first collective agreement is achievable through arbitration and not through strike action. Accordingly, unions favour the “automatic” access model, requiring only that sufficient time has elapsed since certification and that the conciliation requirement has been met; there would be no other filtering mechanism and no need to find bargaining breakdown or fault.
Alternatively, unions favour broadening the explicit circumstances in which first contract arbitration can be ordered (e.g., where there has been a remedial certification). Labour groups would also recommend repealing the LRA requirement that the OLRB deal with decertification and displacement applications before dealing with applications for first contract arbitration. They argue that because first contract arbitration is remedial in nature and designed to address some employer misconduct or unreasonableness, it is unfair and inappropriate to allow for decertification or an application by a different union when a first collective agreement has not had a chance to operate and there has been no opportunity to stabilize the relationship.
Many employers have historically opposed first contract arbitration, arguing that imposing contracts is contrary to the entire idea of free collective bargaining, adversely impacting employers who engage in tough, but legal, “hard bargaining”. Moreover, employers argue that it creates uncertainty for businesses by putting key decisions into the hands of a third party, and that having automatic first contract arbitration available undermines any need for the union to bargain realistically, since the union can just wait for time to elapse and ask for arbitration.
The successor rights provision of the LRA protects employee and union rights where there is a sale of a business, providing that bargaining rights and collective agreement obligations of the original employer generally flow through to the new successor employer. The term “sale” is very broadly defined and in most situations where the business is transferred, the bargaining rights and all the rights of workers under the collective agreement flow into the relationship with the new employer.
One major exception is that the section has generally not been applied to contracting out or contract tendering situations, in which a lead company contracts out its work to a subcontractor or, more typically, the contract is re-tendered and one subcontractor service provider replaces another. Examples would be where security or cleaning services are moved from one subcontractor to another by re-tendering the contract.
Currently, the bargaining rights and the obligations under the collective agreement normally do not flow from one service provider to the next. This means that the union not only loses its bargaining rights each time a new contractor is selected by the lead company but also (subject to the termination pay provisions of the ESA in the building service provider sector) the employees lose their jobs and their rights and benefits under the collective agreement and may well not be hired by the new employer. Even if the successor subcontractor hires many of the same employees to perform the same work in the same location, the union loses its bargaining rights and the employees lose whatever rights they have under the agreement. If the union can certify again, it has to start bargaining all over again with the new employer.
From 1993 to 1995, this situation changed briefly, as the scope of successor rights was extended to apply to one class of service contracts only, namely, building services contracts. The Labour Relations Act provided that successor rights applied where contracting out and re-tendering occurred with respect to building services (including cleaning services, food services and security services). The Act deemed that a sale of business had occurred where a building services contract was entered into by a lead company or re-tendered. The purpose of this provision was to ensure that bargaining rights and the rights of building cleaners, security staff, and food services employees were continued under successor contractors or subcontractors. This provision was repealed in 1995.
Labour relations legislation in all Canadian jurisdictions protects successor rights where there is a sale of a business. However, such legislation does not extend successor rights to contract service situations. There is one exception: the Canada Labour Code has provided, since 1999, that a successor employer in one specific contract for service situation may not decrease the remuneration of the employees. This provision applies only to “pre-board security screening services” in relation to a “federal work, undertaking or business” and thus generally applies to airport security screening. It also applies to any other service that may be designated by regulation, but there has been no extension of this provision to other sectors.
Labour groups have proposed that successor rights be extended to ensure that employee rights are maintained when a building service contract changes firms. Union stakeholders have argued that employees in the building service sectors are generally vulnerable low-wage workers and that the loss of job security and all other entitlements every time a contract is re-tendered or contracted out has potentially devastating effects on workers. Moreover, they argue that it is extremely difficult for employees to organize and maintain collective bargaining rights in sectors dominated by the practice of contract tendering. Each time the contract for services is awarded to a new contractor, additional resources are expended as unions attempt to re-organize workers and, if successful, the parties have to re-negotiate a new collective agreement and start all over again.
Unions have also suggested that successor rights in the homecare industry would improve continuity of care for patients. A transit union has asked that they be included in the protection of this provision, arguing that contracting out of transit services in some municipalities has resulted in a significant loss of jobs and rights for employees.
Employers appear generally to see all contracting out as a legitimate and necessary means of creating and maintaining efficiencies and argue that it is simply a different situation from a sale of a business, since the lead firm is not permanently divesting itself of a part of its business but is simply having a part of it performed by a specialist contractor more cheaply and/or better than it could itself. They argue that contracting out business services is no different in principle from any other contracting out and that the extension of successor rights to these situations would increase costs and undercut competitiveness and flexibility.
The OLRB has the authority to determine the appropriate bargaining unit with respect to each application for certification. Historically, the most common bargaining unit definition has comprised a single workplace of a specific employer at a particular geographic location. There are separate policies for employers with multiple locations within a municipality. There may be further subdivisions (e.g., separate bargaining units for “office” and “plant” employees). At one time, the Board certified part-time and full-time employees separately, and had various practices for determining multiple appropriate bargaining units in various sectors, such as hospitals, municipalities, universities, newspapers, etc. Over time, a single employer could wind up with many different bargaining units and many sets of collective bargaining with the same or with different unions.
The OLRB has historically taken the position that after it has issued a certificate and the parties have entered into a collective agreement, the certificate is “spent” and the OLRB has no general jurisdiction to reconsider or revise it, except where specifically authorized by the Act. Thus, with minor exceptions, as bargaining units are added over time, the only way to change the configuration of bargaining units now is for parties to voluntarily agree to changes. While the parties are free to expand or to reduce the scope of bargaining units, it is an unfair labour practice to take such issues to impasse (i.e., to make such a dispute the subject of a strike or lock-out). This is an effective bar to changing the bargaining unit structure where one party resists it.
The issue, therefore, is whether there ought to be an explicit power to revise, amend and consolidate bargaining units for the rationalization or modernization of bargaining unit structures in circumstances where the original bargaining structure is no longer appropriate, where bargaining units are overly fragmented, or for other industrial relations reasons.
The power to revise and revamp bargaining units involves not only the issue of the rationalization and modernization of bargaining unit structures, but also the possible tension and interplay between organizing and bargaining in areas of the economy that have been traditionally difficult to organize, such as where employers have many smaller retail locations, in which cases it may only be possible to organize in smaller units. However, a small unit is likely to have little bargaining power; viable, effective and stable bargaining may be possible only where there is a larger unit. If units can be organized on a smaller basis and then consolidated afterwards, this could make collective bargaining in those industries viable.
In some other jurisdictions, including several provinces and the federal jurisdiction, labour relations boards have a general power to amend a bargaining unit or certification order after a union has been certified.
The Labour Relations Act, as it was in 1993 and 1995, included a provision allowing the OLRB, upon application of either party, to consolidate separate bargaining units with respect to the same (or a related) employer represented by the same union at either the same location or in multi-location situations. This provision did not restrict either the type of units to be consolidated (e.g., office and production units) or the timing of the consolidation application. However, it did exclude the possibility of consolidating bargaining units of the same employer that were represented by different unions, which is currently permitted in other jurisdictions.
In Ontario, from 1993 to 1995, in exercising its discretion to consolidate units, the OLRB was required to consider whether the proposed consolidation would: facilitate viable and stable collective bargaining; reduce fragmentation of bargaining units; or cause serious labour relations problems. With respect to manufacturing operations, the OLRB was prohibited from combining bargaining units at geographically separate locations if the employer established that this would interfere with the employer’s ability to continue significantly different methods of operation or production at each location or the employer’s ability to continue to operate these places as viable and independent businesses.
This consolidation provision was repealed in 1995. In addition, bargaining units that had been consolidated were divided back into separate bargaining units, unless the employer and the union agreed in writing that the unit should not be divided.
The Commission on the Reform of Ontario’s Public Services, known as the “Drummond Commission”, included in its final report the recommendation that the Ontario government, “Consider expanding the authority of the OLRB to facilitate the establishment of effective and rationalized bargaining structures that support the delivery of quality and effective public services.” The Drummond Commission made this recommendation as a response to what the Commission described as an overly fragmented collective bargaining structure in Ontario’s broader public sector.
It appears that labour boards have an express, general power to redefine bargaining units (which could include consolidating existing units) in British Columbia, Alberta, New Brunswick, Newfoundland and Labrador, Prince Edward Island, Nova Scotia, and the federal jurisdiction.
The test for applications to redefine bargaining units varies among jurisdictions. For example, the power of the federal labour relations board to consolidate bargaining units was previously quite broad until the Sims Task Force recommended, and Parliament accepted, that bargaining unit reviews should be restricted to situations where there are serious problems with bargaining unit structures, barring which, the employees’ choice of bargaining agent should prevail. The Canada Labour Code was subsequently amended in 1999 to provide that, in order for a review to take place, the Board must now be satisfied that the existing bargaining unit structures are no longer appropriate for collective bargaining.
Even if the corresponding labour legislation does not expressly provide the power to amend a bargaining unit, some labour relations boards may modify the bargaining unit or certification order pursuant to their general powers. The OLRB, however, has maintained consistently that it does not have the jurisdiction to do so.
Unions have told us that they support the introduction of a consolidation provision in the LRA like the one in place between 1993 and 1995. From the labour movement’s perspective, the goal of having a consolidation provision is to ensure that smaller units, once certified, can be combined together into more rational, long-term bargaining structures. At the same time, a structure whereby the OLRB could merge and reconfigure bargaining units, especially where different unions are involved, might be viewed as a heavy-handed, top-down approach that could force change against the wishes of a significant number of employees.
Employers opposed the previous introduction of a consolidation provision, describing its purpose as being simply to boost union bargaining power in situations where the union’s presence is weak. In some cases, however, employers, too, have described the benefits of giving the OLRB the authority to restructure and rationalize bargaining units and have recognized that there is no other effective way to modernize, particularly in circumstances where the existing bargaining structure may be fragmented and antiquated.