The potential expansion of the use of project labour agreements (PLAs) in Ontario requires a careful examination based on evidence and past experience. PLAs can introduce tendering practices that restrict labour on major infrastructure projects to workers who are affiliated with one labour union or a set of labour unions, excluding construction workers who wish to affiliate with other unions or not to affiliate at all. The practices can, in turn, reduce competition, leading to higher costs for taxpayers and a loss of fairness and freedom of association for workers.

This research brief does not purport to be an exhaustive analysis of PLAs. Its purpose is to highlight a few key aspects of PLAs in the context of Ontario and to flag some policy concerns associated with introducing them to public sector construction projects. To that end, this brief reviews previous research on PLAs in Ontario and related research on restrictive tendering practices.

A PLA is a contract between a construction owner or proponent and a labour organization, usually a labour union or an alliance of labour unions. 1 1 P.-L. Bilodeau et al., “Project Labour Agreements in Ontario: An Assessment of 25 Years of Negotiating Under the OLRA,” Institute for Construction Employment Research, 2023, https://iceres.org/wp-content/uploads/2024/12/Project-Labour-Agreements-in-Ontario-25-year-assessment.pdf. It governs labour issues for a specific construction project and can cover employment issues as wide-ranging as wage rates, work hours (including lunch breaks), health and safety provisions, and others. Importantly, a PLA also limits the use of strikes by the union and lockouts by the employer.

PLAs and collective agreements share some similarities and deal with many of the same issues, but they differ in some important respects. First, a PLA covers a specific construction project or a specific set of construction projects. A collective agreement is generally not specific to a project but governs the labour relations between a labour union and an employer for a specified period of time, regardless of the number or variety of construction projects taken on by the company during that period.

Second, a PLA is typically a contract between a labour organization and a construction owner—that is, the proponent of the construction project, not the contractor or company hired to complete it. This differs from a collective agreement, which is generally between the labour organization and the workers’ employer, that is, a contractor. In this way, a PLA can be a tool for labour unions to bypass negotiations with contractors and set the working conditions directly with the owner of a construction project. 2 2 See, for example, the quotation of a representative of a building trade union cited in P.-L. Bilodeau et al., “Project Labour Agreements in Ontario,” 33: “when you enter into a project labour agreement, you’re basically, as the building trades, bypassing your contracting partner and going to the client and saying to the client: ‘we will be your partner.’” Because a PLA does not involve negotiation with the contractor, it may be signed well before a contractor is even chosen for a project.

PLAs in Ontario bear some unique features because of their legislative context. Section 163.1 of the Ontario Labour Relations Act (“the Act”) provides the legal basis for PLAs under Ontario law and governs their use in Ontario.

Section 163.1 exists in a part of the Act that lays out the overall framework for the labour relations model for the industrial, commercial, and institutional (ICI) construction sector in Ontario. Under this model, each of the building trade unions negotiates a province-wide collective agreement with the respective employers’ group representative. Once completed, these collective agreements cover all the work in the ICI sector done by the workers affiliated with those unions across the province, for a period of three years, at which point they are renegotiated.

In 1998, the Mike Harris government introduced section 163.1 into the Act to allow for PLAs as an exception to these province-wide collective agreements. According to the government at the time, this new section was designed to “ensure stable labour relations” and thereby attract construction jobs and investment into the province on “economically significant projects.” 3 3 Ontario, Legislative Assembly, Statements by Ministry and Responses (Hansard), 36th Parl, 2nd Sess (June 4, 1998) (Jim Flaherty), https://www.ola.org/en/legislative-business/house-documents/parliament-36/session-2/1998-06-04/hansard#P630_137783. Yet despite being an exception to the province-wide collective agreements, PLAs nevertheless accept the basic framework of this part of the Act. As such, the legislation specifies that the labour organizations that are party to a PLA under this section of the Act may only be those that are bound by a provincial agreement, that is, the building trade unions. 4 4 Labour Relations Act, 1995, SO 1995, c 1, Sch A, para 163.1 (2), https://www.ontario.ca/laws/statute/95l01.

It should be emphasized, however, that the province-wide collective agreements—and therefore the PLAs that are governed by the same legislative framework—cover only ICI construction work that is done by workers affiliated with the building trade unions. They do not apply to non-unionized workers or workers affiliated with alternative unions. In fact, there is significant labour diversity in Ontario, with many construction workers who are not represented by the building trade unions but by alternative unions or who are not unionized at all. Yet no provision is given in the Act for a PLA with a labour union that falls outside of the provincial collective agreements. Ontario’s framework thus fails to reflect the existing labour diversity within the construction workforce.

The prospect of greater use of PLAs by the public sector requires a careful analysis of both the advantages and the disadvantages associated with them. The rest of the paper provides this analysis, flagging serious reservations associated with increased frequency of their use.

A major concern about PLAs in Ontario is that they can lead to restrictive tendering for government construction projects, similar to what some Ontario municipalities faced prior to the passage of Bill 66.

Historically, a select handful of municipal governments in Ontario were subject to the province-wide collective agreements for the ICI construction sector. This led to a structure of restrictive tendering, in which those municipalities could accept bids for construction projects only from contractors that had a collective agreement with Ontario’s building trade unions. This limited the pool of potential bidders on government projects to those companies associated with the building trade unions, effectively giving a monopoly on the labour to these unions and an oligopoly on bidding to those companies affiliated with them.

In 2019, the Doug Ford government ended this system for Ontario municipalities (with the exception of the City of Toronto) through Bill 66, freeing them from the constraints of the province-wide collective agreements and thereby introducing open tendering. These municipal governments can now accept bids from any contractor, regardless of the union affiliation of their workers.

There are significant financial costs associated with restrictive tendering. By limiting the kinds of companies that can bid on construction projects, these policies restrict competition. Less competition means higher prices for taxpayers. Conversely, policies that promote competition for these projects reduce the price that government pays. Therefore, procurement policies should generally seek to maximize the number of companies that can bid for government infrastructure projects, provided that the companies are qualified to do the work.

Bill 66 ushered in a new era of open tendering for all Ontario municipalities except the City of Toronto, which opted out of the new structure. When the City of Hamilton was deciding whether to move to open tendering following the passage of Bill 66 into law, the municipal analysts cited potential savings of 21 percent for construction projects newly subject to open tendering. 5 5 City of Hamilton, General Issues Committee Revised (June 19, 2019), 101–2, https://pub-hamilton.escribemeetings.com/FileStream.ashx?DocumentId=196165.

A previous Cardus analysis of tendering in the Region of Waterloo showed similar results. In this case, Waterloo had open tendering until 2014, when it became subject to restrictive tendering. It returned to open tendering in 2019 following the passage of Bill 66. This allowed Cardus, in 2021, to compare results in three distinct phases of this municipality’s procurement policy. The findings showed that the municipality received a greater average number of bids per project under open tendering than under restrictive tendering:

• Prior to 2014, under open tendering, the average number of bids per project was 8.14.
• Between 2014 and 2019, under restrictive tendering, the average was 3.68.
• After 2019 and until the time of the analysis, under open tendering once again, the average rose to 5.54, demonstrating a movement back to the numbers during the previous period of open competition. 6 6 B. Dijkema, “Bouncing Back Through Diversity: The Effects of Bill 66 on Construction Competition in the Region of Waterloo,” Cardus, July 2021, https://www.cardus.ca/research/work-economics/reports/bouncing-back-through-diversity-the-effects-of-bill-66-on-construction-competition-in-the-region-of-waterloo/.

This demonstrates a fundamental truth: greater competition means lower prices. While it is impossible to ascertain the precise effect of open tendering, an analysis of post-2019 tendering processes in the Region of Waterloo suggests that the financial gains from open tendering were significant. This analysis found that, in competitions in which firms affiliated with the carpenters’ building trade union made bids, the winning bid from other competitors not affiliated with the carpenters’ union was an average of 14 percent lower. 7 7 B. Dijkema, “Bouncing Back Through Diversity.” Moreover, in the post-2019 period, 93 percent of bidders in the Region of Waterloo were not affiliated with the union and would therefore have been prevented from bidding under the previous system of restrictive tendering.

It should be noted that the lower prices for construction do not have to come from lower wages for construction workers. As Cardus has noted previously, firms compete on many factors, not even primarily on labour. 8 8 B. Dijkema, “The Facts on Fair and Open Contract Bidding,” Cardus, June 2019, https://www.cardus.ca/research/work-economics/policy-brief/the-facts-on-fair-and-open-contract-bidding/. In fact, competition drives lower prices through many other sources, such as more efficient management practices, greater innovation, better use of technology, and other factors. That said, if governments are concerned about protecting the wages of construction workers, there are ways to do this that do not involve restricting competition to a certain kind of unionized firm. These can include Fair Wage policies, such as the one used by the City of Toronto, 9 9 City of Toronto, Fair Wage Office & Policy, https://www.toronto.ca/business-economy/doing-business-with-the-city/understanding-the-procurement-process/fair-wage-office-policy/. or by preventing non-unionized contractors from reducing wages below those of other firms working on a project. 10 10 P.-L. Bilodeau et al., “Project Labour Agreements in Ontario.” Moreover, the existence of Fair Wage policies in the City of Toronto demonstrates the redundancy of restrictive tendering in supporting wages.

The fact remains that greater competition among firms for government construction projects leads to better use by firms of resources of all kinds. As extensive Cardus research has shown, the resultant savings can be quite significant. It is these sources of savings, borne through competition, that governments must not allow to pass by.

PLAs in Ontario tend to give exclusive rights in providing labour for construction projects to building trade unions, which is similar to what restrictive tendering does.

It is important to recall that the basic framework of a PLA is an agreement with a labour union. In Ontario’s legislative framework, it is an agreement with a labour union that is party to a provincial collective agreement, namely, a building trade union or a set of building trade unions. Thus, the starting point for these PLAs assumes an agreement with a building trade union. As such, a PLA usually contains exclusivity provisions that restrict bids on the project to firms that are party to a provincial collective agreement.

An analysis of Ontario PLAs by the Institute for Construction Employment Research (“the Institute”) has shown that 60 percent of a sample of PLAs in Ontario contain such provisions. 11 11 P.-L. Bilodeau et al., “Project Labour Agreements in Ontario.” This is essentially the same model of restrictive tendering that was ended by Bill 66: mandatory contracting with contractors that are unionized by the building trade unions.

Does this mean that the other 40 percent were subject to open tendering? Unfortunately, no. According to this same analysis, another 16 percent of this sample of PLAs gave preference in the bidding process to contractors with collective agreements with the building trade unions, even though they did not specifically ban other contractors from bidding.

The remaining 24 percent of the PLAs were silent on who was eligible or preferred in the bidding process. But this does not necessarily mean that these PLAs were subject to open and competitive bidding, either. Even if a PLA technically allows a bid from a contractor with an alternative union affiliation, it may still require any employees of that contractor to become members of the signatory labour union during their employment on that construction project. The Institute report, for instance, found several examples of PLAs in Ontario that, even though they allowed contractors to hire from other places than just the building trade union, required those new employees to join the union signatory to the PLA within one week of being hired for the project. 12 12 P.-L. Bilodeau et al., “Project Labour Agreements in Ontario.”

This is basically restrictive tendering, as a contractor is effectively barred from working on construction projects under this kind of PLA if their workers wish to remain affiliated with another union or if they wish to remain unaffiliated.

Ultimately, the Institute report found that “96% of the agreements from the sample provide for some form of mandatory union membership.” 13 13 P.-L. Bilodeau et al., “Project Labour Agreements in Ontario.” In this sense, it may be a misnomer to describe PLAs in Ontario as “open.” The current PLA landscape in Ontario may be better analyzed in terms of those that are more restrictive, that is, those that accept bids only from contractors subject to provincial collective agreements—and those that are somewhat less restrictive, that is, those that require workers to become members of a building trade union once they are hired.

That 96 percent of sampled PLAs require a form of mandatory union membership should not come as a surprise. Recall that, in the context of the Act, PLAs are exceptions to the provincial collective agreements, and the labour signatories must be building trade unions. The Institute for Construction Employment Research report sums up the basic arrangement of PLAs this way:

PLAs mandate that work on a project continues no matter what conflicts arise on a project (e.g., grievances) or offsite (e.g., a general area strike by one or more trades). In return for these and often other concessions by labour organizations, PLAs typically require that most skilled workers on the job site are selected through the union referral system. 14 14 P.-L. Bilodeau et al., “Project Labour Agreements in Ontario,” 9.

Put differently, the PLA model is to “requir[e] contractors’ workers to join specified unions for a major project.” 15 15 P.-L. Bilodeau et al., “Project Labour Agreements in Ontario,” 29. If the basic deal is labour-relations stability in exchange for exclusive labour supply by a set of labour unions, then it should not be unexpected that 96 percent of sampled PLAs end up requiring workers to join that set of unions. Rather, mandatory union membership is inherent in the Ontario model.

While PLAs are not exactly the same as the restrictive tendering regimes that existed prior to 2019 and that continue to exist in the City of Toronto, governments should be wary of inadvertently re-introducing the restrictive tendering that Bill 66 mostly did away with. The scale of potential cost increases could be quite substantial, especially given the size of “economically significant” projects that would likely be subject to PLAs. As an illustration of the scale of the potential costs involved, the Montreal Economic Institute has used the foregoing figures to estimate that the Ottawa Hospital’s Civic Campus Redevelopment Project will see cost overruns of between $168 million and $525 million, owing to the restrictive PLA in place. 16 16 M. Ouellette and M. L. Shaw, “How to Reduce Construction Costs in Ontario: Modernizing the Construction Industry,” Montreal Economic Institute, February 2022, https://www.iedm.org/how-to-reduce-construction-costs-in-ontario-modernizing-the-construction-industry/.

As suggested by the Institute for Construction Employment Research report, among the most significant benefits that proponents of PLAs cite is the certainty they provide for labour relations, because PLAs generally limit labour disruptions from strikes and lockouts. This is particularly relevant for large projects with long timelines that are likely to span one or more renegotiations of provincial collective agreements. This was a primary motivation for the introduction of section 163.1 into the Act in the first place, as outlined above. That said, the existence of these provincial agreements already provided a substantial degree of certainty in labour relations. Aside from reducing strikes and lockouts, PLAs can also be used to harmonize work hours for members of different construction trades who would otherwise have asynchronous work hours as parties of different collective agreements. 17 17 P.-L. Bilodeau et al., “Project Labour Agreements in Ontario.” For complex projects, such as those “economically significant” ones contemplated by section 163.1 of the Act, this harmonization could lead to important efficiencies and contribute to the timely completion of a construction project.

Proponents of PLAs may also point to the possibility of squeezing community benefits out of a construction project. These benefits may include increased training and apprenticeship opportunities and greater diversification of the construction workforce. Although this has been absent from the private-sector PLAs seen in Ontario thus far, 18 18 P.-L. Bilodeau et al., “Project Labour Agreements in Ontario.” this has been the stated, though not necessarily exclusive, goal of the public-sector PLA model in British Columbia, known as the B.C. Community Benefits Agreement. 19 19 British Columbia Infrastructure Benefits, “Community Benefits Agreement Between BC Infrastructure Benefits Inc. and Allied Infrastructure and Related Construction Council of British Columbia,” July 17, 2018, amended and restated March 1, 2022, https://bcib.ca/wp-content/uploads/2023/05/Community-Benefits-Agreement.pdf. But this agreement, signed between the Government of British Columbia and the Allied Infrastructure and Related Construction Council of British Columbia, has led to significant problems. A key feature is a requirement that construction workers on designated public-sector infrastructure projects join one of the affiliated building trade unions. Among other issues, this policy led to an Indigenous contractor being prevented from working on a hospital project in his own First Nation, a project that has seen significant cost overruns that are likely exacerbated by these same policies. 20 20 For more details, see R. Nauta, “Benefits for Whom? Assessing British Columbia’s Community Benefits Agreement,” Cardus, September 2024, https://www.cardus.ca/research/work-economics/reports/benefits-for-whom/.

While community benefits may be laudable goals, it is unnecessary to sign a PLA, let alone a PLA with restrictive tendering practices, to achieve them. Alternative methods, such as embedding community benefits clauses within a tender, can accomplish the same goal without the disadvantages of a PLA. 21 21 A. Flootman, “A Framework for Implementing Community Benefits Agreements,” Cardus, July 2022, https://www.cardus.ca/research/work-economics/reports/a-framework-for-implementing-community-benefits-agreements/.

Thus far, PLAs have been used in Ontario almost exclusively by the private sector. The Institute for Construction Employment Research report, discussed above, analyzed a database of forty such private-sector agreements. 22 22 P.-L. Bilodeau et al., “Project Labour Agreements in Ontario.” By contrast, we are aware of only one use of a PLA in the public sector, namely, the Civic Campus Redevelopment Project of the Ottawa Hospital. 23 23 The Ottawa Hospital, “The Ottawa Hospital and Building Trades Unions Sign Historic Partnership Agreement,” January 19, 2023, https://www.ottawahospital.on.ca/en/newsroom/the-ottawa-hospital-and-building-trades-unions-sign-historic-partnership-agreement/. This use of a PLA in a public-sector construction project is a novelty in labour relations in the province. It is unclear whether it will be a one-time use or represents a new movement to greater use of PLAs by public entities.

Although PLAs can in theory incorporate elements of open tendering, such as the right of workers to maintain their pre-existing union affiliation (or non-affiliation), this has overwhelmingly not been the experience so far in Ontario. Of course, it is the right of a private-sector company and its workers to choose to associate with a particular labour union or a set of labour unions. As seen above, there are certain advantages and disadvantages associated with this approach. Private-sector entities can make these decisions in their own best interests.

However, other principles come into play when it concerns the public sector and the use of taxpayer dollars. Foremost among them is fairness for all workers, regardless of how they exercise their freedom of association. Under this freedom, workers should not be excluded from working on government construction projects simply because they have chosen to be represented by an alternative union or not to be represented.

Bill 66 restored this freedom for construction workers in the municipalities, other than Toronto, that were previously subject to restrictive tendering. The danger of expanding PLAs into the public sector is that they would reverse the progress made by Bill 66 in achieving fairness for workers.

PLAs in Ontario have so far largely been restricted to the private sector, although the recent signing of a PLA by the Ottawa Hospital suggests that public-sector entities may make use of them in the future. Governments should be acutely aware of the downsides associated with them. The Ontario experience—in which, according to one sample, nearly all PLAs were subject to some form of mandatory membership of workers in building trade unions—suggests that introducing PLAs to the public sector could undo the progress made by the Ford government’s passage of Bill 66. This could lead to significant cost increases on government procurement and unfairness for workers who wish to remain affiliated with alternative unions or not affiliated at all. Any movement toward the use of public-sector PLAs must address these concerns and would likely require a significant legislative overhaul of section 163.1 of the Act to do so.