Electoral Insight – Aboriginal Participation in Elections

Electoral Insight – November 2003

The Effect of Expansion of the Franchise on Turnout

Michael Kinnear
Professor, Department of History, University of Manitoba

In 1960, the Diefenbaker government extended the franchise to Status Indians in Canada, and the country suddenly had several hundred thousand new voters. In other countries, when the franchise has been extended to certain groups, the new potential voters have been slow to take up the right, although eventually they come to vote in much the same proportion as the rest of the population. This article compares several cases of franchise expansion to see whether they can help shed light on trends in voter participation among Aboriginal people in Canada.

The article begins by examining the polling results of the past 23 federal and provincial elections in Manitoba, and shows that there has been a much more serious decline in turnout among First Nations electors than among other voters in Canada. It then contrasts the Manitoba voting trends with four foreign cases where the franchise was extended to groups that had not previously had it and explores whether these examples can be of use in assessing voting trends among First Nations people in Canada.

The four non-Canadian franchise expansions considered are: Germany in 1871, Great Britain in 1918, the United States in the 1960s (Civil Rights era registration of African-Americans), and the United States in 1924 (enfranchisement of Native Americans). The first three followed a similar pattern, namely, an initial reluctance to exercise the vote, followed by a gradual increase in turnout. In contrast, voting in Native American areas followed a pattern similar to that among Manitoba's First Nations voters, where a moderately high turnout in the first few elections after franchise expansion was followed by a disproportionate drop.

Aboriginal voting in Manitoba elections

After the Diefenbaker government extended the franchise to Status Indians in 1960, the first reserve Indians to vote federally were those of the Rice Lake Band near Peterborough, Ontario, in a by-election on October 31, 1960

Before examining the results, a short methodological explanation is in order. Survey data are indispensable in analyzing elections, as a recent study of non-voting behaviour in Canada shows.^{Footnote 1} However, because Aboriginal people form only 3.3 percent of the Canadian population,^{Footnote 2} most surveys are unable to provide an accurate indication of their rates of voting or not voting. The analysis of polling results makes it possible to consider turnout rates for those First Nations populations that are highly concentrated geographically. Moreover, it is possible to gather past voting results and determine what the major trends have been for that group over a long period.

Manitoba has long followed Canadian turnout trends, and Manitoba's rate was within 3 percent of the national average in every federal election in the past half-century except 1997, when flooding during the election led to a 3.8 percent gap. Manitoba has a large and growing First Nations population,^{Footnote 3} and the shifts in both turnout and partisan support show several trends that differ from the rest of the province, and from Canada as a whole.

Table 1 shows the turnout and partisan support on 64 First Nations reserves in the 23 elections held since enfranchisement in 1960. In some elections, there was no separate poll for a particular reserve, generally because of small populations. However, nearly all reserves had a separate poll in most elections. Practically all the on-reserve voters were Aboriginal people.

Table 1 shows that the initial high turnout coincided with support for the Progressive Conservatives in both federal and provincial elections. This was not surprising, since the federal Tories initiated enfranchisement in 1960. Support for the Conservatives then waned among First Nations voters, dropping to 6.0 percent in the 2003 provincial election, and only 2.5 percent in the 2000 federal election.

A more striking observation is the significant decline in First Nations turnout: it dropped from 65 percent to 26 percent between 1962 and 2003. The actual number of voters in 1962 (5,664) was not much lower than in 2003 (7,924), despite the more than threefold increase in the potential First Nations electorate. Thus, while the overall trend in turnout for both Manitoba and Canada has been downward, in the polls located on reserves the decline has been dramatic.

Note: F = Federal election; P = provincial election. The results of the 1963 and 1966 provincial elections are not included due to numerous errors in the official reports.

Some non-Canadian comparisons

Germany

The most famous mass enfranchisement probably took place in the German Empire in 1871, when Otto von Bismarck replaced the three-tiered voting system of the kingdom of Prussia with universal adult male suffrage. As Table 2 shows, the first election held had a turnout of only 51 percent, but this rose gradually to a high of 85 percent in the last election held before the First World War. Some of the increase in votes went to the Social Democrats (SPD), particularly after 1890, when SPD public activity was legalized.

Nearly all studies of this trend have attributed the rise in turnout to a straightforward movement from not voting to voting socialist. However, a recent study indicates that the SPD recruited voters from other parties, while the non-Socialist parties were engaged in anti-socialist campaigns of their own, and recruited most of the previous non-voters.^{Footnote 5} Thus a major force behind the increased turnout seems to have been the campaigns of the various parties to mobilize non-voters to resist (or support) socialism.

Great Britain

In the United Kingdom, until 1918, only 59 percent of adult males had the vote, despite the Reform Acts of the 19th century. In 1918, the franchise was extended to all adult males and to women over 29. In the election that year, only half of those eligible voted, a significant drop from the December 1910 election, when more than 80 percent voted.

In 1922, turnout rose to 72 percent, then to 76 percent, and to 84 percent in 1950. Since then it has declined, to 59.4 percent in 2001. From the late 1920s to the mid-1950s, elections in Britain were polarized along the lines of socialism versus antisocialism. From the 1970s (polarization was still strong beyond the 1960s) this influence waned, and turnout gradually dropped. As in Germany, the period of high turnout coincided with strong partisan rivalry based on ideological differences.

United States: enfranchisement of African-Americans

A third example of enfranchisement is the Civil Rights voter registration activity of the 1960s. Theoretically, African-Americans were enfranchised by the 15th Amendment to the American Constitution in 1869. However, effective enfranchisement did not happen in the South until the registration campaigns. Tables 4 and 5 compare voter participation in states with varying percentages of African-Americans in both Presidential and Congressional elections.^{Footnote 8}

Table 4 shows that turnout in Presidential elections in 1900 was low in most Southern states (Column C), compared with 74 percent in Northern states (Column A). The differences remained high until after 1960, before the Civil Rights movement gained momentum. The gap dropped to 10 percent in 1980, and 7.5 percent in 2000. These changes resulted from both an increased turnout in Southern states and a decline in the other states. This pattern was similar to both the German and British cases, where a low initial turnout was followed by increased participation.

Table 5 shows how the turnout in 1940 was low, perhaps because the Democrats won an average of 91 percent of the time; voters were unlikely to have much individual impact with so many safe seats for one party. During the 1980s, and even more so in the 1990s, the South became much more competitive, with the Democrats winning only 40.1 percent of Southern Congressional Districts in 2002; Congressional voter turnout increased during that period, although it is still low compared with most democracies.

As in both Germany and Britain, higher turnout appears to have followed voter mobilization campaigns combined with increased partisan rivalry, when the Republicans emerged as the leading, though not the overwhelming, party in the South.

United States: enfranchisement of Native Americans

A fourth enfranchisement to consider is that of Native Americans in the 1920s. The first Presidential election in which this applied was that of 1924. Table 6 shows the rate of turnout in the 20 counties in the U.S. that today have a more than 50 percent Native American population,^{Footnote 9} and compares the vote in those counties with the overall vote in the states where the counties are located.^{Footnote 10}

In the counties concerned, turnout was greater in the 20 majority-Native American counties than in the rest of the same state in both 1940 and 1960. In 1980 and 2000, the state turnout declined – by 10 percent between 1960 and 1980, and a further 5 percent between 1980 and 2000. However, turnout in the 20 counties with a Native American majority declined even more, by 13 percent in 1980 and another 8 percent in 2000. This pattern of decline resembles the changes observed in Manitoba, rather than the British, German or African-American cases.

Conclusion

The data presented in this article do not allow us to determine the causes of the increasing tendency to abstain among Manitoba's on-reserve electors. Nevertheless, the type of comparative analysis conducted here can be instructive. Of particular interest is the similarity observed between the Manitoba case and the enfranchisement of Native Americans. In both cases, an initial increase in turnout was followed by a gradual but steady decline. The contrast between these two cases and the other three examined here (Germany, Britain and African- Americans) is striking. Moreover, it is interesting to note that in the latter three cases, increased turnout coincided with an increased effort by certain political parties to educate and mobilize voters. This pattern supports the conclusion, made elsewhere as well,^{Footnote 12} that mobilization of voters by political parties is an effective way to encourage turnout.

The fact that turnout has recently declined across Canada, even more so among Aboriginal electors, raises the question of whether the parties are carrying out this important function to the same degree as they once did. The current atmosphere in Canada of reduced voter interest and turnout among the population as a whole^{Footnote 13} could signal a trend to even lower voter participation among First Nations people. Regardless of the exact causes of the decline, however, one thing is certain: democracy loses when a significant and growing segment of the population, in effect, votes with its feet.

Notes

Return to source of ^{Footnote 1} Jon H. Pammett and Lawrence LeDuc, Explaining the Turnout Decline in Canadian Federal Elections: A New Survey of Nonvoters (Elections Canada, 2003). Available on-line at: www.elections.ca.

Return to source of ^{Footnote 2} Or 2.2 percent in 2001, excluding the Métis and Inuit peoples, according to the 2001 Canadian census; see http://www.statcan.ca/english/Pgdb/demo39a.htm.

Return to source of ^{Footnote 3} 13.64 percent including the Métis people, or 8.42 percent not including them, in the 2001 census.

Return to source of ^{Footnote 4} All federal voting figures are derived from the poll-by-poll results in the reports of the Chief Electoral Officer of Canada published after each election; the provincial election data are from the statements of votes published by the Chief Electoral Officer of Manitoba.

Return to source of ^{Footnote 5} Jonathan Sperber, The Kaiser's Voters: Electors and Elections in Imperial Germany (New York: Cambridge University Press, 1997), p. 37.

Return to source of ^{Footnote 6} Based on Sperber, The Kaiser's Voters, p. 38.

Return to source of ^{Footnote 7} Sources: turnout for 1906–1910, Neal Blewett, The Peers, the Parties and the People: The British General Elections of 1910 (Toronto: University of Toronto Press, 1972), p. 277; for 1918–1936, based on Michael Kinnear, The British Voter: An Atlas and Survey Since 1885 (London: Batsford Academic and Educational Ltd., 1981, 2nd ed.), p. 73; for 1950, D. E. Butler, The British Electoral System 1918–1950 (Oxford: Clarendon Press, 1952), p. 172; for 2001, International Institute for Democracy and Electoral Assistance Internet site, www.idea.int/voter_turnout/westeurope/index.html.

Return to source of ^{Footnote 8} County-level data show similar trends: a clear distinction in turnout between counties with significant African-American populations before 1960, and other parts of each state; and no obvious correlation recently. In 2000, the American Census Bureau announced that "African Americans were the only race or ethnic group to defy the trend of declining voter participation in Congressional elections," see www.census.gov/Press-Release/www/2000/cb00-114.html.

Return to source of ^{Footnote 9} The average percentage of American Indians in these counties was 68.86, the highest being 94.2 percent in Shannon, South Dakota.

Return to source of ^{Footnote 10} Electoral data for the United States are derived from R. M. Scammon, America at the Polls, 1920–1964 (Pittsburgh: Pittsburgh University Press, 1965); for 1980, from Scammon, America Votes, 1980 (Washington: Congressional Quarterly, 1981); and for 2000, from the official reports issued by the Secretaries of State for each state.

Return to source of ^{Footnote 11} Wisconsin is not included before 1980, since the only majority-Native American county there, Menominee, was not created until 1961.

Return to source of ^{Footnote 12} See for example, André Blais, Elisabeth Gidengil, Richard Nadeau and Neil Nevitte, Anatomy of a Liberal Victory: Making Sense of the 2000 Canadian Election (Peterborough: Broadview Press, 2002), p. 57.

Return to source of ^{Footnote 13} The average turnout from 1921 to 1984 was 74.3 percent, compared with only 66.0 percent in the past three federal elections. Blais, et al., "The evolving nature of non-voting: Evidence from Canada," presented to the American Political Science Association, 2001, p. 5, suggests that this is due to generational replacement.

Note: 

The opinions expressed are those of the authors; they do not necessarily reflect those of the Chief Electoral Officer of Canada.