This piece is part of our Forum of Federations Series on National Unity and Canadian Federalism.

By Jared Wesley

October 22, 2026

With the dust now settling on the outcome of the Alberta referendum, among the many takeaways we can all agree on is this: the most overused phrases in Alberta politics in 2026 were “send a message” and “gain leverage”.

Separatists used these tactics to persuade federalists that they could safely vote against remaining in Canada without sparking a constitutional crisis. Prominent pro-separation billboards stopped well short of calling for outright independence, urging Albertans to “Send Ottawa a Message!” instead.

Voting to have another referendum, they reasoned, would signal Alberta’s frustration to Ottawa and give the premier power to renegotiate better terms within Confederation.

Premier Danielle Smith relied on much the same rhetoric in putting four constitutional reform proposals to Albertans. A strong vote, she suggested, would “prove a point” about Alberta alienation and strengthen her hand in dealing with the federal government.

Smith and the separatists employed the same strategy, and neither reached their objective.  The separatists lost by a 3 to 1 margin on the question of whether to remain in Canada, and, according to unofficial results, the Smith government failed to secure Albertans’ endorsement for any of its constitutional plans.

There are lessons in these ill-fated campaigns. If governments ask citizens whether they support a particular constitutional reform, voters should — and often do — assume that their ballots will be interpreted as an endorsement or rejection of that change. If they are instead asked to participate in a protest vote, their enthusiasm wanes. And so does the “leverage” leaders were hoping to generate as part of the exercise. Let’s have a closer look at the strategy and why it fails.

Bad Bluffing

Someone voting out of conviction for a particular outcome casts exactly the same ballot as someone seeking to “send a message.” The final tally cannot distinguish between the two of them, which — on the surface — appears to favour the side calling for leverage by expanding the universe of available voters. In this case, the separatists could attract frustrated federalists to their side without fully recruiting them to the cause. But the result would produce ambiguity when bargaining power depends heavily on certainty. And at least some voters appear to have seen through the ruse.

The work of the late American economist, Nobel laureate and game theory scholar Thomas Schelling demonstrates that threats can be powerful only when they are credible. A negotiator can sometimes gain advantage by restricting their own freedom to manoeuvre — effectively telling the other side: I can’t back down, so you’ll have to give. But that commitment has to be believable beyond being a bluff.

In other words, you don’t strengthen your bargaining position by declaring you don’t really mean it. Consider how these moves sounded to Ottawa and the rest of Canada. First, separatists threatened to commence a process that could break up the country. Then, they openly explained that some of their supporters had no intention of leaving. That’s the political equivalent of threatening to quit your job while showing your boss your empty bank account.

The referendum result made the problem worse for the separatists. More than three-quarters of participating Albertans rejected even commencing the separation process. Of the 25 per cent who voted for option 2, public polling suggests half to two-thirds did so simply to “send a message.”

The outcome was so lopsided, the premier committed to never holding another referendum on the issue, removing all doubt about the weakness of the separatist threat in Alberta (at least in the short term).

Smith’s own strategy ran into a related problem.

Losing Two-Level Games

Harvard political scientist Robert Putnam’s classic theory of “two-level games” explains how political leaders negotiate simultaneously at two tables: with other governments and, in the case of democracies, with constituents back home. Again, domestic constraints can sometimes strengthen a leader’s hand, with a premier telling her counterparts: I might be willing to compromise, but my voters won’t let me.

Calling a referendum seems like a good way to create such a constraint. Public endorsement can convert what might otherwise look like a hollow threat into a democratic mandate for change, even when the demand is unlikely to be met. Smith’s predecessor, Jason Kenney, used this strategy when he convened a referendum on removing the equalization principle from the constitution.

Kenney himself admitted that the vote wouldn’t spark constitutional reform, but that it was meant to generate “leverage” to pursue other changes to fiscal federalism in Canada. Surveys indicate most Albertans didn’t understand the game they were being asked to play, and many — including separatists — felt jilted by the experience when nothing came of the “yes” vote.

Nonetheless, Smith pursued a similar approach in asking Albertans to endorse her calls on the federal government to devolve constitutional authority to the provinces. Like Kenney, Smith publicly downplayed the prospect of Ottawa agreeing to allow provinces to appoint higher court judges, abolish the Senate, opt-out of national programs with compensation, or allow provincial laws to have precedence over federal ones.

Her goal was political, not constitutional: obtain public backing for dramatic reforms to secure concessions on more reasonable ones.

As with the separatists’ calls to “send a message,” Smith’s strategy was laid out in full view. And as with equalization, this gave Ottawa reason to dismiss the referendum results as manufactured consent.

In the end, the public rejected all four constitutional reform proposals. Rather than revealing an Alberta united behind its leader, the referendum exposed a province divided over the premier’s signature demands and tactical agenda. This has weakened, not strengthened, her bargaining power with the rest of Canada.

That is the paradox of trying to manufacture leverage through direct democracy: a referendum does not just reveal your strength — it can also reveal your weakness.

Lessons for Leaders

There is a larger lesson here for leaders across Canada. Referendums can have legitimate democratic purposes. They can be used to ratify agreements or constitutional accords, or decide fundamental policy questions. They can give citizens direct authority to reject or accept deals that governments have struck, explained, and defended in good faith.

But voters and other governments can tell the difference between using a referendum to decide something and using it as a way of enlisting voters for a whole other political purpose. “Send a message” or “prove a point” is an invitation to voters to answer a question that’s not printed on the ballot. “Give me leverage” may seem like a reasonable appeal, but it is an obviously empty threat to negotiating partners. “Help me manage my party” is an even weaker message. Each introduces ambiguity into a process whose value rests heavily on clarity.

Leverage is relational, not relative. It does not exist simply because one side announces it possesses it. It depends on credible threats that the party is willing and able to impose. When those threats are obviously artificial, the strategy falls apart. When voters reject your premise outright, they are, ironically, sending you a message.

Successful intergovernmental negotiation is less theatrical. Governments identify achievable objectives. They find allies and understand their needs. They put several issues on the table, trading off their low priorities for higher ones. They build consensus and bank gains so they can return to the table another day. None of that fits neatly on a billboard, of course. And it occurs largely behind closed doors, not from the ends of megaphones, employing the sort of elite accommodation that political scientists call consociationalism — a mode of politicking that has held deeply-divided societies together for centuries.

But, unlike the winner-take-all outcomes of referendums, this alternative approach produces nuanced results that can then be taken back to cabinets, legislatures, or — in very rare constitutional cases — the public in the form of a ratification referendum. But the hard work of consensus building comes first.

Alberta’s experiment with direct democracy and constitutional cosplay is a cautionary tale for populist leaders who see referendums as a bargaining tool and the country’s supreme law as a plaything.

If your strategy depends on threatening something you have already admitted you can’t deliver, don’t be surprised when voters and the people sitting across the bargaining table refuse to play the game.

Jared Wesley is a Professor of Political Science at the University of Alberta.