Correction to “A Tale of Two Centuries? Expanding Constitutional Labour Rights in Canada and Their Impact on Legislation, Conflict and Wages”

Abstract

Weinberg, B. R. 2025. “A Tale of Two Centuries? Expanding Constitutional Labour Rights in Canada and Their Impact on Legislation, Conflict and Wages.” British Journal of Industrial Relations 63, no. 2: 268–286. https://doi.org/10.1111/bjir.12857.

The article “A Tale of Two Centuries? Expanding Constitutional Labour Rights in Canada and Their Impact on Legislation, Conflict and Wages” sought to investigate how the different legal regimes governing public sector dispute resolution in Canada associate with the incidence and duration of conflict, as well as wage settlements. These legal regimes included where the right to strike for union members is removed by compulsory arbitration legislation, circumscribed through essential service designation legislation, uninhibited as in the private sector, and when the union has a choice between two of the aforementioned regimes. The article contributed to the literature by (1) correcting the classification of public sector working groups by prior studies to the various legal regimes, (2) updating the analysis to the near present to incorporate further legislative and economic events, and (3) to investigate the impact of changes to the legal environment, namely the constitutionalizing of the rights to collective bargaining and to strike in Canada.

The article produced several principal findings. First, in contrast to prior studies, it found that the legal regimes that placed restrictions on the right to strike did not statistically significantly associate with a decline in the likelihood that the parties could directly negotiate a contract settlement. This was particularly noteworthy given the concern of a chilling effect that is theorized to occur when the right to strike is substituted with mandatory arbitration. The analysis of contract settlements indicated that the legal regime affects the way in which a contract is settled when an impasse occurs, but it does not significantly associate with the parties’ ability to directly settle an agreement. Second, the article found that the constitutionalizing of labour rights in Canada may have impacted the bargaining power of public sector workers. Wage settlements where the right to strike was unrestricted were significantly higher in the era affirming and expanding constitutional rights than prior to it, suggesting that perhaps those union members had greater bargaining power. However, where the right to strike was restricted (though not removed), it appears that bargaining power may have eroded as wage settlements were significantly lower for the most recent period examined.

In seeking to build upon this research, it was discovered that a coding error in the program performing the analysis resulted in the errant inclusion of numerous collective bargaining settlements that should have been excluded. The contract dataset from which the sample is drawn includes all contracts with more than 500 employees in provincial jurisdictions and more than 150 employees in the Federal jurisdiction. As is stated in the article, any contracts covering less than 500 employees in the Federal jurisdiction were supposed to be excluded in order to keep the size of units consistent across jurisdictions.

The contracts that should have been excluded are only a small fraction of the total agreements in the reported samples for each of the periods of analysis (less than 2% in either). Thus, this means that their exclusion does not substantively change the majority of the findings of the article, including those outlined above (although the level of significance was reduced for wage settlements under the right-to-strike regime post-Dunmore). The variation in the estimates between the samples including and excluding the contracts with less than 500 employees is generally less than a hundredth if at all (or, rather, less than a percent or percentage point, given their interpretation, depending on the analysis), particularly for the previous contract outcomes and political party in power variables. These differences did result in changes to the level of statistical significance associated with some estimates. These changes are detailed in the table below (differences where there was no change in significance level are not reported).

For the main variables of interest on legislation governing dispute resolution, the impact is confined to the choice of procedure regime given that this legal structure is almost exclusively found in the Federal jurisdiction. More specifically, the choice of procedure legal structure does statistically significantly associate with settlements in arbitration as compared to settlements under the right-to-strike legal structure. This should perhaps not be too surprising a result given that unions have the ability to unilaterally choose arbitration as the final step at an impasse under the choice of procedure model. Lastly, the associations of the choice of procedure model with contract outcomes are affected most for the analysis concerning the post-Dunmore period (2001−2019). This is likely due to the fact that the already small number of settlements under the choice of procedure regime is further reduced with the correct sample. It may also be because no occupational groups moved to or from the choice of procedure legal structure during this period and thus there is no variation. For these reasons, the estimates associated with the choice of procedure regime for the post-Dunmore period should be interpreted with caution.

We apologize for this error.