Questioning conventional wisdom on voter turnout

This is a holiday, you say. I should be relaxing with family, or cooking, or having a nap.

Not today. Today, I have been poking around on social media looking for gossip on the election. And I got to thinking about voter turnout.

We have a lot of moral panic about voter turnout. Turnout is dropping, we hear every year. People are increasingly disengaged, and those who lead our country are chosen by a smaller and smaller number of its people.

But, like any political scientists on a holiday, I got to thinking. So I looked up the historical data from Elections Canada, which comes in a handy chart.

If you map voter turnout based on this information, you get something that looks roughly like this:

voter turnout

That drop in recent years does look concerning. (Don’t worry about that plunge in 1898; Elections Canada says it was a referendum of some sort. There’s probably an interesting story there.)

But I got to wondering about how much of our population might be on the list of electors. And if you divide the number of people on the list by the population, you get this:

percent on voters list

There are some pretty startling increases over time on that graph. I’m guessing the jump in the late teens and early ’20s is the result of the fact that women got the vote around that time. But we’ve got more and more of our population on the list of electors from about the 1960s.

So what does that mean for turnout numbers? Well, it seems that turnout numbers are based on a denominator of how many people are on the list of electors. So if we want to roughly control for changes to the percentage of people on the voter’s list, we can compare the population to the number of ballots cast. That looks like this:

ballots by population

Based on this graph, there doesn’t seem to be much of a decline at all. From about 1940 onward, between about 40% and 51% of the population has voted. Since the early ’70s, it’s been 44% to 51%. Our last three elections have been 50%, 44%, and 47%, respectively.

Of course, I’m entirely glossing over some pretty significant demographic changes here. One of them, I’d think, would be big changes in and since the post-war period around what percentage of the population was old enough to vote. Those could have a big effect. But I do find these numbers to be particularly curious.

Now, I’ve spent about 5 minutes on this, and my expertise isn’t in quantitative political science. I’m hoping that some of you political science folks out there have actually done more work on this, or at least know of people who have. I’d certainly appreciate any links to relevant works you could share in the comments section. (And let me know if you see any errors in the graphs or calculations I’ve got above. I slapped them together pretty quickly.)

In the mean time, I’m left wondering: why are we constantly hearing about the moral decay of declining turnout, when as many of us are voting as have in the last several decades? It seems like our actual level of participation may be masked by how good we’ve gotten at getting people signed up to vote.

If the percent of the population voting doesn’t have a clear downward trend over the last few decades, how do our concerns about or perceptions of turnout change? Could we be asking better questions about changes in democratic engagement that go beyond the ballot box itself?

Date of election/referendum Population NumberĀ of electors on list Total ballots cast Voter turnout Percent on voters list Ballots by population
7 August – 20 September 1867Footnote 1 3,230,000 361,028 268,387 73.1 11.2% 8.3%
20 July – 12 October 1872 3,689,000 426,974 318,329 70.3 11.6% 8.6%
22 January 1874 3,689,000 432,410 324,006 69.6 11.7% 8.8%
17 September 1878 3,689,000 715,279 534,029 69.1 19.4% 14.5%
20 June 1882 4,325,000 663,873 508,496 70.3 15.3% 11.8%
22 February 1887 4,325,000 948,222 724,517 70.1 21.9% 16.8%
5 March 1891 4,833,000 1,113,140 778,495 64.4 23.0% 16.1%
23 June 1896 4,833,000 1,358,328 912,992 62.9 28.1% 18.9%
29 September 1898Footnote 2 4,833,000 1,236,419 551,405 44.6 25.6% 11.4%
07-Nov-00 4,833,000 1,167,402 958,497 77.4 24.2% 19.8%
03-Nov-04 5,371,000 1,385,440 1,036,878 71.6 25.8% 19.3%
26-Oct-08 5,371,000 1,463,591 1,180,820 70.3 27.2% 22.0%
21-Sep-11 7,204,527 1,820,742 1,314,953 70.2 25.3% 18.3%
17-Dec-17 7,591,971 2,093,799 1,892,741 75 27.6% 24.9%
06-Dec-21 8,760,211 4,435,310 3,139,306 67.7 50.6% 35.8%
29-Oct-25 8,776,352 4,608,636 3,168,412 66.4 52.5% 36.1%
14-Sep-26 8,887,952 4,665,381 3,273,062 67.7 52.5% 36.8%
28-Jul-30 8,887,952 5,153,971 3,922,481 73.5 58.0% 44.1%
14-Oct-35 10,367,063 5,918,207 4,452,675 74.2 57.1% 43.0%
26-Mar-40 10,429,169 6,588,888 4,672,531 69.9 63.2% 44.8%
27 April 1942Footnote 2 11,494,627 6,502,234 4,638,847 71.3 56.6% 40.4%
11-Jun-45 11,494,627 6,952,445 5,305,193 75.3 60.5% 46.2%
27-Jun-49 11,823,649 7,893,629 5,903,572 73.8 66.8% 49.9%
10-Aug-53 14,003,704 8,401,691 5,701,963 67.5 60.0% 40.7%
10-Jun-57 16,073,970 8,902,125 6,680,690 74.1 55.4% 41.6%
31-Mar-58 16,073,970 9,131,200 7,357,139 79.4 56.8% 45.8%
18-Jun-62 18,238,247 9,700,325 7,772,656 79 53.2% 42.6%
08-Apr-63 18,238,247 9,910,757 7,958,636 79.2 54.3% 43.6%
08-Nov-65 18,238,247 10,274,904 7,796,728 74.8 56.3% 42.7%
25-Jun-68 20,014,880 10,860,888 8,217,916 75.7 54.3% 41.1%
30-Oct-72 21,568,311 13,000,778 9,974,661 76.7 60.3% 46.2%
08-Jul-74 21,568,311 13,620,353 9,671,002 71 63.1% 44.8%
22-May-79 22,992,604 15,233,653 11,541,000 75.7 66.3% 50.2%
18-Feb-80 22,992,604 15,890,416 11,015,514 69.3 69.1% 47.9%
04-Sep-84 24,343,181 16,774,941 12,638,424 75.3 68.9% 51.9%
21-Nov-88 25,309,331 17,639,001 13,281,191 75.3 69.7% 52.5%
26 October 1992Footnote 2 Footnote 3 20,400,896 13,725,966 9,855,978 71.8 67.3% 48.3%
25-Oct-93 27,296,859 19,906,796 13,863,135 69.6 72.9% 50.8%
02-Jun-97 27,296,859 19,663,478 13,174,698 67 72.0% 48.3%
27-Nov-00 28,846,761 21,243,473 12,997,185 61.2 73.6% 45.1%
28-Jun-04 30,007,094 22,466,621 13,683,570 60.9 74.9% 45.6%
23-Jan-06 30,007,094 23,054,615 14,908,703 64.7 76.8% 49.7%
14-Oct-08 31,612,897 23,677,639 13,929,093 58.8 74.9% 44.1%
02-May-11 31,612,897 24,257,592 14,823,408 61.1 76.7% 46.9%
Elections Canada Numbers My additions
http://www.elections.ca/content.aspx?dir=turn&document=index&lang=e&section=ele

 

One thought on “Questioning conventional wisdom on voter turnout

  1. Fascinating! It seems so obvious and yet I’ve never seen or heard your point made anywhere else. Down the google hole I go.

Comments are closed.