Thursday, October 16, 2008

Post-Electoral Numerology

This election has had the lowest voter turnout in Canadian history  What does that mean?  Among other things, it means that all of the major parties had fewer votes cast for them in this election than in the election of 2006--even those who increased their vote share.  Of the Federally-funded parties, only the Green Party saw an increase in absolute number of votes.

These two charts display the change of total votes received by each party.  First, as a difference from the election of 2006, and second as a trend over the past three elections.  If you are a Liberal strategist, that trend line should have you sweating profusely, and possibly muttering darkly about the end times.  That's a decline of 1.35 Million votes or 27% in four and a half years.  (Exercise for the reader: if these trends continue, when does the Liberal vote drop below that of the NDP?)  But really, nobody--except perhaps the Greens--should be taking solace in these numbers, if for no other reason than that the Quarterly Allowance payments under section 435.01 of the Canada Elections Act will decline commensurately.

The Conservatives can, of course, look forward to having the fact that only 22.2% of eligible voters supported them raised every time they try to claim to have a mandate to govern.  Jack Layton's concession speech on election night didn't include this number--but probably only because it wasn't yet known at that point.  He certainly emphasized the Conservative's minority status, and how this imposed a requirement to seek the consent of parliament.  It would be consistent with his usual practice to at least rhetorically acknowledge a similar requirement for the consent of the populace.

Many people are pointing at the results of this election as evidence of the need for electoral reform--typically with some kind of proportional representation system being held up as the way to re-engage the electorate.  It might work.  I'm not convinced.  The argument seems to turn on whether the resulting parliament would be more appealing to the current non-voter, which seems impossible to predict.  In any case, pure proportional representation reduces the voter's choice to selecting a party--which isn't all that helpful if none of the parties actually represent the voter's interests.  I think, if I had absolute freedom to construct an electoral system, I'd prefer to outlaw parties and put the emphasis on choosing specific representatives, personally.  For that matter, various reality TV shows demonstrate that it would probably be feasible to have direct voting on specific issues on a regular basis, which could render the representative moot.

However, before getting too side-tracked by a consideration of theoretical electoral systems, there are other questions to ask about how this one worked, particularly with respect to strategic voting.  Two questions I was asking myself on election night were: A) did the Bloc Québécois prevent a Conservative majority; and B) did the Green party split the "pro-environment" vote in any meaningful way?  To explore these questions, consider the following table.

Actual No BQ NDP + Green Liberal + Green Liberal + NDP Liberal + NDP + Green

Conservative 143 156 138 130 114 91

Liberal 76 110 72 98 155 179

NDP 37 40 46 33 - -

BQ 50 - 50 45 37 36

GRN 0 0 - - 0 -

Other 2 2 2 2 2 2

On the left are the actual seat numbers from this election, as of this morning.  The remaining columns are hypothetical results derived from the reported vote counts under various hypotheses.  

The first column--labelled "No BQ"--awards every seat won by the Bloc to which ever candidate came second in that riding.  One can argue about how accurate this is as a model of what the vote would be if the BQ didn't run.  Effectively, it assumes that the second choice of Bloc voters is not distributed significantly differently than the first choice of other voters, within each riding.  In fact polling data suggests that Bloc voters have a preference for the NDP as a second choice.  So this analysis may overestimate the Conservative seat count for this hypothesis, and underestimate the NDP.  But with that caveat, the result is that the Conservatives would pick up enough seats to reach a majority.  So, yes, there is some justification for a claim that the BQ prevented a Conservative majority.

The next two columns show the seat counts if all of the Green party votes were rolled in with either the NDP or the Liberals.  While there are a number of ridings that swing in this analysis, it isn't enough to substantially change the outcome.  In both of these scenarios, the Conservatives still win a minority government, with a gain in seats relative to the previous election.

By contrast, the last two columns demonstrate that if the NDP vote was rolled into the Liberal party (with or without the Green vote) it would be enough to convert the Conservative minority into a Liberal majority.  This is not at all surprising, since it amounts to introducing a two-party scenario into a first-past-the-post vote count.  The Conservatives are a lot more likely to win their seats outright that the other parties--they received more than 50% of the votes cast in 55% of the ridings they won, where the other parties did so in more like 20% of ridings they won.  But with only 37.6% of the popular vote, geographic concentration works against you; excess votes in a riding you have already won are of no use in winning other seats.

It's also not desirable, I would argue, to reduce the entire spectrum of possible electoral choice to Conservative vs. not-Conservative.  In fact, if it is necessary to do so in order to prevent the Conservatives from winning, this would seem to support the legitimacy of the Conservative victory.  Their mandate, however thin, does represent a functional coalition of votes that agree to support that platform; whereas the coalition one would have to assemble to defeat it contains some rather substantial policy differences.

Which suggests that there is little more to be gained by leveraging strategic voting against the Conservatives.  For another party to take power, without collapsing into a two-party system, they will have to pull votes from the Conservatives, or from the growing pool of non-voters.  


1 comment:

Peter Jene said...

Really fantastic your post. I look forward to future posts.

In common usage, horoscope often refers to an astrologer's interpretation, usually through a system of Sun sign astrology or based on calendar significance of an event, as in Chinese astrology.

Thanks for sharing,

uk horoscopes