Aboriginal Electoral Participation in Canada
Elections Canada's stated objectives for 2008–2013 are to maintain trust in the electoral process, increase its accessibility, and strengthen the engagement of key stakeholders. In light of our findings, we offer five policy recommendations aimed at increasing the accessibility of the electoral process for Aboriginal people. The recommendations, which could lead to a higher rate of participation among Aboriginal Canadians, are as follows:
- Target Aboriginals living on reserves
- Focus on young Aboriginals
- Develop and promote programs that increase political resources and sense of civic duty
- Expand voter registration efforts
- Study outlier cases
Below, we identify the logic of each recommendation and point toward actionable means by which it could be implemented.
1. Target Aboriginals living on reserves
According to our estimates, the effect of living on a First Nations reserve is a decrease of 7–8 percentage points in the probability of casting a ballot, holding all else equal. This effect grows to 9–10 percentage points for those Aboriginals living on rural reserves. There are at least three reasons why Aboriginal electors would benefit from a focus on increasing the participation of on-reserve Aboriginals. First, our findings suggest that an important predictor of whether an individual votes is whether others in their household have voted, which we think is evidence of a more general network effect (see Christakis and Fowler 2009). This provides an important opportunity for mobilization. By targeting mobilization efforts to geographically concentrated populations, Elections Canada can aim to harness these network effects. Leveraging such effects can decrease the costs of voting, as electors share information on the voting process. It can also increase the sense of civic duty and social obligation, as the electoral participation of one individual acts as an impetus for the participation of others.
Second, by working closely with band leaders, Elections Canada's mobilization programs can harness local traditions and knowledge to increase effectiveness. In implementing voter education programs and registration initiatives, context unquestionably matters. By focusing first on populations living on reserves in close co-operation with local leaders, Elections Canada can tailor programs to meet identifiable local needs. This is almost certainly more difficult among off-reserve populations, which are often dispersed. Elections Canada should engage in survey research and focused discussions with Aboriginal electors to appreciate how their understanding of community obligation and commitment does or does not involve an obligation toward participation in federal elections. For example, Elections Canada could ask a battery of questions on various senses of obligation to understand if electoral participation differs from obligation towards family, community and other forms of participation.
Finally, because turnout is lower on reserves, there is more potential for growth. Full participation in any geographic unit is unlikely, but it does stand to reason that where turnout is lower there is more potential for marginal growth.
2. Focus on young Aboriginals
Young Aboriginals participate at a rate far below their elders. According to our estimates, an Aboriginal elector aged 25 or under has a 25–26 percentage point lower probability of voting than an Aboriginal in the oldest age category (65 or over). This, we note, is already controlling for the effects of the lower than average income of Aboriginal youth. As with on-reserve Aboriginals, there is great potential for growth.
Since our next recommendation focuses on the importance of political resources, we concentrate here on how Elections Canada can better target youth. First, Elections Canada should reconsider its advertising targeting Aboriginal youth. While it is important to laud all efforts to increase voter turnout, it seems clear from our estimates that Elections Canada's advertising aimed at Aboriginal electors has not been effective, particularly among young people.
Elections provide a limited opportunity for learning which mobilization techniques work and which do not. Because voting only occurs at the end of an electoral campaign, it is hard to know until that time whether a particular advertising campaign has been effective. Accordingly, if Elections Canada wishes to maximize its outreach and mobilization efforts among youth during elections, we recommend that it consider mid-campaign surveys, paired with field experimental rollouts of advertisements to gauge their usefulness.
We likewise recommend experimentation among the full suite of available persuasion tools during by-elections. Indeed, by-elections in geographically dispersed electoral districts – which are most likely to involve on-reserve electors – are an ideal place to test several mobilization techniques within one electoral district, during one election.
In sum, we recommend rigorously testing a wide schedule of approaches. As research in the United States has shown, such testing is the most effective way to identify the right tools for mobilization.
3. Develop and promote programs that increase political resources and sense of civic duty.
No two single factors exercise as much impact on the decision to vote as political resources and a belief that voting is a duty. Alarmingly, both of these are exhibiting a steep decline across Canada, particularly among younger Canadians. Those wishing to increase voter participation need to identify how to halt this decline. There are, unfortunately, no easy answers, as the decline appears attributable to a long-term and large-scale values shift among younger generations (Blais and Rubenson, forthcoming; Howe 2010).
While we cannot point to precise solutions, we do recommend two specific policies going forward. First, Elections Canada should act as a clearinghouse and funding source for the best in civic education. Key to such a role would be piloting or funding the piloting of educational tools that increase the political interest and knowledge of young people (both Aboriginal and non-Aboriginal) in a manner that equips them for voting. This in turn requires our knowing exactly what resources young citizens need. While we suspect it has much more to do with being interested and knowledgeable about the differences between parties and the consequences of voting, rather than the formal functioning of our political system, this is merely a hypothesis.
What we propose is that Elections Canada and partner organizations design and field a number of such civic education programs over a three- to five-year time horizon, both on reserves and in high schools with large numbers of Aboriginal students. By pitting different approaches against each other and then comparing their effects on targeted electors over one or two election cycles, we can gain a systematic understanding of what could increase political resources and sense of civic duty among Aboriginal Canadians.
Such an experimental program of civic education could likewise be adapted to Aboriginal people of all ages. Elections Canada should explore the effects of different community-based civic education programs. Such programs could be run inside or outside of an election cycle, and could be administered by local citizens. If these programs are rolled out and tested in a systematic fashion, we can identify which programs increased political resources and sense of civic duty in a local area, and then had positive downstream effects on voter turnout.
We wish to stress that such programs need not be developed by Elections Canada. Central to acting as a clearinghouse and funding source is a willingness to support locally developed initiatives, provided they are subject to proper program evaluation.
4. Expand voter registration efforts
As our results show, registration exercises a strong and independent effect on the decision to vote, even after controlling for key demographic drivers and the influence of political resources and civic duty. Indeed, the average effect of being registered and recalling receiving a voter information card is a 20 percentage-point increase in the probability of voting. We have two recommendations for increasing registration.
First, Elections Canada should expand its Community Relations Officer Program. This program has been in place since the 38th general election in 2004. It provides important registration and voting information, and also aids in the targeted revision of the National Register of Electors. Where possible, greater resources should be committed to these efforts.
Second, Elections Canada should consider other means of allowing electors to register to vote, whether online or using their telephone. Other jurisdictions have moved in this direction, and we recommend that Elections Canada take preliminary steps to do the same. We likewise urge necessary changes in the legal framework required for such opportunities to be made swiftly. As many rural areas in Canada still lag in access to broadband Internet, we do add a caveat that such a move should also carefully monitor how this policy may have differential benefits according to Internet access.
5. Study outlier cases
Finally, we recommend that Elections Canada commission a series of case studies on locations in which Aboriginal voter turnout is either far above or far below the average. In some cases, these polling divisions will border one another or be in the same electoral district.
Survey research can identify the principal factors across which turnout varies, but it cannot shed similar light upon the on-the-ground differences between places where turnout is high and low. Accordingly, a series of in-depth case studies could identify the crucial factors that lead to unusually high or unusually low turnout. Intuitively, potential candidates would include mobilization by local political and community elites, the professionalism of polling station operations, visits by local candidates, and visits by community relations officers. These are just a few among many possible explanations, and further research is undoubtedly needed.