Category Archives: Electoral Reform

More Unsolved Riddles

A recent article on iPolitics by Alan Freeman under the headline “Governing is a lot harder than you think, Mr. O’Leary”  :: :: provoked the following comment from me:

Mr. Freeman makes an extraordinarily important point, worth noting not only by Mr. O’Leary, but by all of us who would over-simplify political discussion. All governments, however small, are complex organisms comprised of PEOPLE who know, and are learning, who think, and feel, and act within their means. They are not machines to be manipulated from the top, although they can be led. But changing their direction takes time. They are inherently conservative, much more easily stalled by negative forces than motivated by positive ones. The culture of anecdotally-fed carping negativity that permeates so much of our contemporary political discourse fastens a huge drag on our government organizations and their leadership. We indulge ourselves in strident expressions of polarized over-simplified opinions, and expect our public servants, elected or not, to sort out the mess. Then we yell at them from all sides no matter what they do. Stephen Leacock advised us to think of complex issues of social justice as unsolved riddles, to be addressed, not by the simple-minded application of ideology or formulas derived from theory, but by groping our collective humane way forward, guided by Knowledge, Imagination, and Compassion. Anyone got a better idea?

Recent discourse on Electoral Reform and Immigration illustrate this concern very clearly. Discourse on electoral reform became so poisoned that to persist towards change could only make things worse. The government therefore made a decision to postpone. As Mackenzie King said, more or less, when confronted about a broken “promise”: “In politics you do what you can, not what you want.” The voice of experience. The Liberal Party “promise” on electoral reform was perhaps the voice of inexperience. But since when is it a sin to be inexperienced?

Immigration policy is another kind of unsolved riddle. When we engage with the peoples of the world for the purpose of trade and investment, which we yearn to do, then we engage with them also as people. From a practical and moral perspective we cannot have one without the other. We could, of course, adopt the role of brutal exploiters, as colonial powers did and do, but that is not for us, not for the whole us. Deep down we know, in our heart of hearts, that if the money is global, then so are the people, and so is the land. We’ll struggle with that reality, and with its cost, but we will keep trying not to ignore it. So will the Americans, decent people that they are, as they begin to see what the alternative looks like.


Concerning Electoral Reform: First-Two-Past-the-Post

Our newly-minted prime minister, Mr. Trudeau, has promised that 2015 will be the last election conducted on the old first-past-the-post system. I am sure I do not have to explain why I support that resolve: it is abundantly evident that first-past-the-post is a very bad system now that we have four credible parties, five in Québec, and would be just as bad if we only considered three. Reform is long overdue. But what kind of reform?

I would like to propose two general principles:

(1) That the system chosen should be simple enough for the vast majority of electors to understand and use. People should be absolutely clear about how to vote and what use is being made of their votes. This is the great strength of first-past-the-post. Everyone understands a simple race. Unfortunately the job that we are electing people to do, and their lines of accountability nowadays, are not simple at all. The simplicity of the present system becomes much more questionable, once we get beyond election day, which we do very quickly, in one day, as a matter of fact. Fifteen-hundred days, or thereabouts, lie ahead.

(2) That the direct connection between the person of the MP and the geographic constituency should be maintained. I believe the voters should appoint their representatives directly, and not through the intermediation of a party. Parties now choose candidates, but they do not decide who gets the job. It would be a long step in a bad direction if they did.

Regardless of the technicalities (and I recognize there are many) I think there could be problems — perceptual if not practical — if someone were messing around with the votes after the polls close, beyond simply counting them. All forms of preferential voting or ranking have this disadvantage. Of course, reassignment of ranked votes can take place in a perfectly honest, mechanical way, but would people believe it? And what happens if a substantial number of people don’t trust the voting system, or even understand it?


This system, I think, would work exactly as the present one does, from the point of view of voters. I am not suggesting that each voter would vote for two candidates. One voter, one vote. One constituency, two MP’s.

For voters in cities, the doubling of the size of constituencies would make no apparent difference. Candidates may not live in their urban constituencies now. The change would be apparent in rural constituencies, but I am not sure it is hugely significant. I think I would be just as well represented, perhaps even better, by two members from different parties, even if the area and population they are representing is twice as big.

People in huge northern constituencies might object to this idea, quite validly. It won’t strain the system if those constituencies stay the way they are, and won’t hurt the rest of us if they too have two MP’s. I have lived in the north, and know the difficulties. I would not want do anything to increase them, or give people reason to believe they had been increased. 338 MP’s; 350? 360? What difference does it make?

I recognize that larger constituencies would increase the costs of campaigning, thus building in an advantage for those candidates and their parties who are well organized and do a good job. Is it really a disadvantage, to us voters, to see how well these people can organize to do a large, important job? Are we really worse off if that job is made even larger?

I will perform detailed analysis on the recent election, and up-date the analysis for 2011, as soon as I have time. Not too long I hope. To see the earlier, preliminary analysis of 2011, click on the category “Electoral Reform” and it will pop up right below this one.

A Modest Proposal for Electoral Reform

HOT NEWS: First-Past-the-Post is now unequivocally a lousy way to choose a legislature! Well, maybe the news is not all that hot. I observe I have been scooped many times by many illustrious and well-meaning people stretching back to John Stuart Mill.

Proportional Representation (called “PR” among aficionados) is often said to be more fair, more “representative”, particularly if a majority First-Past party does not receive at least 50% of the vote, as is usual in Canada. In the most recent, 2011 federal election, it did not: The Conservative majority of 54% of the seats was achieved with less than 40% of the vote. Majority government in this country very seldom means majority support, which must be judged a grievous flaw in the system, no matter which party you prefer.

Proportional Representation, however, breaks the traditional bond between the Representative to whom the People’s Authority is delegated and the People-in-the-Place that he or she is solemnly charged to represent. It places the Party between the People and the Representative in the chain of accountability, and that, I would argue, is another grievous flaw. It also, in pure form, unless arbitrary cut-offs such as 5% are employed, leads to a proliferation of small parties with relatively narrow points of view, another kind of imbalance that may not constitute the best interests of the electors as a whole any more than does First-Past.

In pursuit of the elusive Balance proponents of “PR” sometimes advocate mixed systems such as Single Transferable Vote (STV) or Mixed Member Proportional Representation (MMPR), or Preferential Ballot (PB), system, which rapidly become quite complicated and further muddy the accountability waters, these being muddy enough already under contemporary political conditions.

British Columbia held two referenda on the issue, in 2005, which came close to approving a form of STV, and in 2009, which rejected it. Ontario held one in 2007 on a form of MMPR that was also soundly rejected. Regardless of these outcomes, however, First-Past is so widely recognized as fundamentally flawed that the issue must remain alive even in those jurisdictions.

I believe that a well informed and rational voter can quite legitimately reject proposals for change which are unduly complicated and violate traditional principles of a fundamental kind on what “representation” means and how it ought to work. I think we should look for a simpler way to change.

Here’s one, which I will call “First-Two-Past-the-Post”, or “First-Two” for short. Instead of sending just one representative per riding, send two, those who come first and second.

When I tested the idea, riding by riding, on the results of the 2011 Election, I spotted only one major defect. In those provinces, such as Alberta and Saskatchewan, where voters expressed a very strong preference for one party, Two-Past skewed the results too far the other way. I therefore added a simple adjustment to my formula, applicable to any riding where the top candidate received more than 60% of the vote. I am suggesting that in those cases the winning party should be allowed to appoint a second representative, ideally in my view the person who came second in that party’s election to choose a candidate.

Here’s the final result, for the 2011 election, compared with First-Past and Proportional:


First-One-Past Proportional First-Two-Past


124 133








Bloc Québécois









0 2



308 309


The second column is the sum of proportional representations for all provinces and territories, calculated separately. For the third I tallied the formula’s results by province and territory, yielding 616 seats, and divided by 2. The totals in the last two columns are 309 instead of 308 due to rounding-off. I chose not to admit fractional members of Parliament, even though everybody knows they are entirely feasible, perhaps common.

In 2011, under First-Past, the probability that your vote would work positively to send a particular person to Parliament would be 50.4%. Under First-Two-Past, that probability, which is a measure of the power of your preference, rises to 76.6%, an outcome which, to my mind, is a huge advantage.

I caution that these results, while strongly suggestive, are only preliminary, because they come from only one election with its own particular circumstances. I will go back further in the weeks ahead, and publish the results as they emerge.

Under a Two-Past system, retaining the same number of ridings would require doubling their size, which is some disadvantage. I would recommend that we mitigate it by merging the House of Commons and the Senate into one Parliament with what would have been, in the 2011 Election, 413 ridings, distributed according to our customary practice. Under the most recent re-distribution the result would be 443 ridings (338 from the Commons, 105 from the Senate).

I suggest that this reform would make a suitable sesquicentennial project for 2017, with a redistribution based on the 2016 Census.

If Proportional Representation sets the highest standard of fairness to parties, the Two-Past system using my formula cannot truly be said to be grotesquely unfair to any party except the Greens, who after all received nearly 600,000 votes, and in 2008 close to a million. I suggest, however, that by opening up the opportunity to enter Parliament by coming second, a measure of greater fairness to developing parties would follow.

But fairness to parties is not the standard, nor some notion of “perfection” that is certainly unattainable. We should seek a balanced system in which the majority of the voting people are fairly represented. First-Past no longer meets this standard, and cannot, because we now have three strong parties nationally, with a fourth perhaps emerging, and another strong one regionally in Québec. Regardless of what becomes of the Green Party and the Bloc, First-Past will remain a bad system unless either the NDP or the Liberals disappear, which seems unlikely. Labels may shift, but I think right, left and centre are political orientations firmly established among voters and most likely to persist. Only by blatant manipulation of a bad system would we be able to suppress one of them in Parliament.

First-Two-Past-the-Post: a simple, more balanced system consistent with our traditions. I will return to this topic when I have analyzed another election, and continue the process until the case is persuasive, one way or the other. In the meantime, I invite comments.