Last week, I attended a constituency (precinct in the USA) meeting to select a candidate for one of the political parties for the coming federal election in Canada. It was very instructive. There were three candidates seeking the nomination in a riding with about 85,000 voters. Party membership totalled about 1,800 in the riding. The vast majority of those members were recruited by the three candidates. If each candidate recruited about 500 new members, and there were already 300 existing members on the rolls, one key task was getting those supporters signed up to the ballot box on nomination voting day. It was sort of a dry run for the election.

About 50% of those recruited were expected to vote. That meant that one key to winning was first signing up members. If, to win, you needed 50% of the members that vote, then a winning candidate would have to obtain 450 votes. If the candidate signed up 500 new members and he or she could get 65% out to vote instead of an average of 50%, then that candidate would have 325 votes and would need to pick up about 125 additional voters.

Thus, signing up members was critical, but not sufficient. Getting the vote out was critical but also not sufficient. The ballot was a preferential one. That means, voters marked their preferences on the ballot as “1” and “2” indicating which of the other two candidates they favoured if their choice did not get past the first ballot and was lowest in ranking. If they marked an “x” or a check mark for only one candidate, they effectively lost their vote in the runoff ballot, though it counted in the initial vote. If they made two x’s, the ballot was considered invalid.

Thus, a third requirement entered the fray. A candidate’s supporters had to be taught how to mark their ballots. Not just how technically, but how substantively and strategically. For if one of the candidates signed up more than the average of 500 – say 750 – then that candidate had a larger base to draw from. If that candidate could also get out 65% of his supporters to vote, then he or she would have an excellent and sure chance of winning, for that candidate would get about 490 votes. However, if the leading candidate could only deliver 50% of his or her supporters, the candidate would only have 375 votes and could not cross the finish line on the first ballot.

Then the second preferences came into play. If the other two candidates made a deal instructing each of their supporters to vote for the other one if he or she ran third, that means that the candidate who ran second on the first ballot would win the vote, provided, of course, that supporters followed the instructions of the candidate on how to mark their preferential ballot.

But other elements enter the possibility of winning a nomination. It was noticeable that in a multicultural riding, the bulk of the voters for each candidate came from the candidate’s own ethnic group. Further, it was also clear that one of the candidates was in a better position to deliver his supporters to the polls from the central institution of that ethnic group. Thus, an alliance not only between the candidate with the second and third most votes on the first ballot, but between and among ethnic groups counted, not only between a candidate’s own ethnic group and that of another, but between members of ethnic groups that did not have a candidate on the ballot. Which candidate could best appeal to other ethnic group members as well as to his or her own supporters in his or her own ethnic group?

The strategy gets even more complicated. Members of an ethnic group not represented by one of the ethnic groups were unlikely to turn out to vote even if they were members of the party. What if deals could be made across constituencies? The voters in an adjoining riding might come primarily from two or three other ethnic groups than the ones represented in your riding. If you could get a candidate in the next riding to recruit and get out the vote of this fourth or fifth ethnic group, and you as a candidate in turn agreed to recruit and get out the vote of members of one’s own ethnic group who lived in that adjoining riding, then it would be possible to pick up 50-150 additional voters from outside of your own ethnic group.

Some might call this a betrayal of democracy for the following reasons:

1. First and foremost, are democracies not premised on the responsibility of individual citizens rather than communitarian allegiances and cross-communitarian alliances?
2. Second, nomination victory does not seem to depend on which candidate puts forward the best policies, but which candidate can recruit and turn out to the polls their own supporters. [As I observed, virtually the only ones who turned out to hear the speeches of the three candidates were core supporters, about 150 of the total of 1,800 members of the party in that constituency. Few people were present to be persuaded by arguments and positions.]
3. Since it is generally believed that about 90% of votes cast in the general election depend on the leader and on the national campaign of the party, of 85,000 constituents, of hopefully two-thirds or about 56,000 voters, in a two-way race – and most are 3 or 4-way races – local candidate influences only about 3,000 votes. This can be a matter of victory or defeat, so the quality of the local candidate does count, especially in close races.
4. Finally, if the selection of the local candidate really depends on only about 500 votes from a constituency of 85,000, it means that very small numbers, perhaps .5%, yes, ½ of one percent, of the voting population in a riding selects the local party candidate.

However, this is how a democratic polity operates on the ground. The selection of local candidates on the ground in each constituency, especially in ridings in contention, can be influenced by selecting a candidate from a riding in whom the party had great confidence and, then, backing that candidate with manpower and financial resources to recruit new members, get the members out to vote, get them to vote in the way needed according to the number on the ballot, and helping forge cross-constituency alliances. Then, the central party can play a decisive role in the selection of candidates.

My attention above focused on ethnic groups, a dominant feature of Canadian politics on the ground level. It has the great advantage of facilitating the entry of representatives from minorities into the political process. In the American election, they may be Hispanic, black or Native American. The system also encourages cross-ethnic group alliances.

There are other communities than ethnic or religious ones. There are youth. There are older constituents. There are rural and there are urban voters. There are women versus male voters. Gender orientation is relevant. There is also education – those having a post-secondary education and those who do not. And then, especially in the United States, there is the issue of race. As you follow the results of an election as they rolled in for the midterms last night and this morning, all of these factors were at work in various different combinations.

One tentative conclusion was that Trump’s alienation of educated women voters in suburbs served as an important stimulus that permitted a number of victories across the nation in suburban areas. On the other hand, Trump’s emphasis in the last weeks of the midterm election on fear of immigrants and foreigners, rather than lauding the success of the economy, did seem to energize his base and get them to turn out in unprecedented numbers in a midterm election.

I have been shifting back and forth between writing this blog and observing the results of the American election on television last evening and this morning. The actual election is but the icing on the cake of a democracy. But the process starts out at the base, a base in which only .5 % of eligible voters in a riding may be need to pick a candidate. I will now turn my attention to my preliminary observations on the American election.

With the help of Alex Zisman