Good morning. Thank you very much for the invitation to speak.
I'm appearing as a researcher. I've worked in labour relations law and labour employment law for about 20 years as a practitioner, and now as a teacher and a scholar, and I also serve as an arbitrator. I'm appearing to offer personal observations on what the research can tell us about the issues you're dealing with today and not to present my personal views or to take a partisan position on this bill.
I'll begin with some observations on certification procedures. A secret ballot vote is normally understood as safeguarding employee choice. The reasons for this are fairly obvious. It provides an opportunity for quiet reflection and private decision-making, but voting procedures, research shows, can also pose risks to employee free choice. Specifically they tend to increase the opportunity for—and the effectiveness of—coercive employer tactics such as retaliations, dismissals, or threats of dismissals.
Not all employers are inclined to do this, but there's good evidence that some employers are, for economic reasons or because they are opposed to unionization. Experience suggests that a significant number of employers will do this from time to time. There is also good empirical evidence from research done in Canada that those kinds of tactics can be effective in reducing employee support for unionization. It's also well known that it's challenging for labour relations boards to provide timely and effective remedies in the face of those kinds of tactics. The reasons for this are multiple. It takes time to fairly adjudicate allegations of unfair labour practices. Labour boards need to be able to expedite hearings, to provide interim remedies, and so on.
Now why might it be the case that a secret ballot vote procedure would provide an opportunity for more effective employer opposition to unionization and in particular for unfair tactics in certification campaigns?
First of all there's more time, so an employer who's inclined to engage in those kinds of tactics has more time to mount such a campaign, and there's less opportunity for those who are sympathetic to unionization to respond to the effects of such a campaign on those whose support is wavering.
There's a very good study by Chris Riddell, an economist who looked at something of a natural experiment that took place in British Columbia when the labour relations regime there moved from card check to vote and then back to card check. In B.C. there was a relatively short time period within which the vote needed to take place; it was a ten-day time period. Nonetheless the study found that the success rate of unionization in the private sector dropped about 20% and in the public sector barely moved at all, suggesting that the economic incentives that affect private sector employers had a lot to do with it.
Riddell then went on to study the effect of unfair labour practices on the level of support for unionization and found that during the period in which votes were permitted, unfair labour practices were more than twice as effective as they were under the card check regime. He also found they accounted for at least 25% of the drop in union success rates. He notes that the rate of applications by unions declined significantly during this period as unions probably tended to pick their stronger cases for certification and to leave others behind.
All of this suggests that if you're going to consider moving away from card check to a vote, part of the package might usefully be a short time period within which to implement the vote, effective interim remedies for unfair labour practices, expedited unfair labour practice procedures, and the availability of remedial certification—none of which seem to be addressed in the bill as it currently stands.
The only other observation I would offer with respect to the certification provisions is that in most jurisdictions, the failure to vote is treated as a decision effectively to let others decide the outcome rather than as counting against certification of the union.
The decertification procedures outlined in the proposed bill focus on voting, which is quite normal in this kind of a process, but it is anomalous to base the outcome on what percentage of employees do not vote in favour of continued union representation. As others have observed, this effectively counts the decision of those who abstain from voting as a decision against continued union representation.
Among the provinces, all legislation but that of Nova Scotia expressly require that a majority positively vote that it no longer supports the union before the labour board can decertify. No jurisdiction requires that a board decertify if a majority of employees do not vote in favour of the union, whether the majority be determined on the basis of ballots cast or on the basis of the bargaining unit as a whole. Most jurisdictions measure this on the basis of ballots cast, but some jurisdictions require that those wishing to decertify show that the union no longer represents an absolute majority of the bargaining unit. Why? Well, because a positive showing that the union lacks support is generally thought to be necessary to undo an earlier positive showing that it had support.
Counting abstentions as being effectively against continued union representation is a potentially unreliable measure. It perhaps also puts too much emphasis on the ability of the union to get out the vote. The public sector provisions of the—