Alberta’s Referendums:

Not grassroots democracy

On October 19, 2026, Albertans are going to the polls to vote on ten questions: five on immigration, four on constitutional changes, and one on separatism. The Alberta government is describing this as an exercise of grassroots democracy. Government policy would be guided directly by Albertans. But this is a mirage, everything about the referendum is government-led.

Before I dig into the analysis, some definitions are in order. Representative democracy involves citizens electing representatives to the legislature to make decisions on their behalf. In contrast, direct democracy involves giving citizens greater control of public policies and representatives through the use of the following devices:

· Referendums: gives people the opportunity to vote on a particular issue or proposed law

· Citizens’ initiatives: gives citizens the right through a petition to have a proposition put to a vote by the electorate for approval

· Recall elections: allow citizens to remove representatives from office before the end of their term of office

Direct democracy is very popular among populists – especially in Alberta. For example, Social Credit in the 1930s, the Reform Party in the 1980s, the Wildrose Party in the 2010s, and today with the UCP. Jason Kenney passed legislation introducing citizen initiatives and recall in 2021, and held referendums on equalization and daylight savings time in conjunction with the 2021 municipal elections in Alberta. In Spring 2025, Danielle Smith lowered the threshold making it easier for citizen initiatives.

Direct democracy is also popular among the public. Here is data from a series of polls by Janet Brown.

Unfortunately, there are many problems with direct democracy devices. Recall and citizen initiative campaigns can be used to mobilize volunteers and gather data that can be used for other purposes. Yes, by law, the personal information on petitions is supposed to be destroyed after the verification process, but as we have seen with the voters list data breach that may not actually occur. And there is nothing against maintaining the list of volunteers who gathered the signatures for fundraising or future election campaigns.

Referendums also take complex issues and make them binary choices (yes or no, or Option 1 vs Option 2). For example, there are hundreds of thousands of Albertans who do not want to separate, but who do desire substantial reform in Alberta’s place in Canada. But that option is not included in the separation referendum question.

Governments choose which questions go on the ballot and which ones do not. The citizen-initiated question on separatism that was promoted by Stay Free Alberta was blocked by the courts, but the UCP put a separation question on the ballot anyway. Even before Elections Alberta determined that the coal petition did not meet the threshold, the UCP had decided that it would not put the question on the October 19 referendum. There are many important public policy issues – health care, education, taxes, renewable energy, oil royalties, etc - that are not put to the public in a referendum.

Governments write the questions in a way that aligns with their interests. All of the questions – including the one on separatism – were drafted by the government. It needs to be pointed out that both separatists and federalists are unhappy with the wording of the separation question. And all of the 10 questions being asked on October 19 are poorly worded, repetitive, contradictory, require substantial background knowledge, and are complex. For example, the immigration questions asks whether Alberta should restrict social services to those with Alberta-approved immigration status, whether there should be a 12-month waiting period before accessing social services, and whether there should be a reasonable fee before accessing social services. These three questions are both repetitive and contradictory. Further, one of the constitutional questions asks to “allow provinces to opt out of federal programs intruding on provincial jurisdiction such as health, education, and social services without losing any of the associated federal funding for use in their own provincial social programs.” The Federal Spending Power is a complex issue with a long history in Canada. In my graduate public policy course, I spend two weeks dissecting the topic. But the UCP government assumes that ordinary Albertans will have done extensive research on the political, legal, and economic aspects prior to going to the polling station.

The vast majority of public discussion on the referendum questions focuses solely on separation. Are Albertans prepared to vote on all ten questions? How many Albertans will be surprised when they get 10 different coloured ballots when they are at the polling station? Will they vote on all 10? Will they vote all “yes” or all “no” on the first nine questions based on whether they support the UCP government?

Governments also determine the policies that emerge from the results of the referendum questions. For example, Albertans voted narrowly against adopting permanent daylight savings time in 2021. Nevertheless, in 2026, the Smith Government decided to do it anyway. In 2021, Albertans also voted in favour of removing equalization from the constitution, but the Kenney government never pursued it with the rest of Canada. With the October 19 referendum, one of the questions on immigration states “Assuming that all citizens and permanent residents continue to qualify for public healthcare and education as they do now, do you support the government of Alberta charging a reasonable fee or premium to individuals with a non‑permanent immigration status living in Alberta for their and their family’s use of the healthcare and education systems?” But what is a “reasonable fee?” The government will decide, not the public.

So, the UCP government, and many of its supporters, argue that October 19 is an exercise in grassroots democracy. But that is a mirage. Every aspect of the referendum – which questions go on the ballot, how many questions, how the questions were written, and how the results will be interpreted – is government-controlled.