05 July 2006

Mexico's Election: A Cautionary Tale

Mexico is in the midst of recounting its Presidential election. The first time around PAN candidate Felipe Calderon (the right wing party of incumbent Vincente Fox in Mexico) was ahead by 1% of the vote. Andres Manuel Lopez Obrador, of the left wing PRD is leading by a little more than 2% of the vote with 78% counted in the recount at as I write (not that you could discover that from CNN or the Denver Post). UPDATE: Latest figures Obrador leads by 1.78% at 84.13% recounted.

Equally important, neither candidate has even 37% of the vote at this hour. The candidate of Mexico's Institutional Revolutionary Party, or PRI, Mr. Madrazo, has about 22% of the vote, and a candidate from two other parties have almost 3% of the vote.

At the end of the first count, which is now being recounted, the vote total was as follows:

Calderon (PAN) 14,027,214 (36.38%)
Lopez Obrador (PRD) - 13,624,506 (35.34%)
Madrazo (PRI) - 8,318,886 (21.57%)
Mercado Castro (PASC) - 1,085,966 (2.81%)
Campa Cifrián (New Alliance) - 384,317 (0.99%)
Write In - 281,145 (0.72%)
Blank - 827,317 (2.14%)


No matter who wins in the recount, we won't know which of the two front runners the 41 million Mexican voters really preferred, because there won't be a runoff election.

Thus, even if PRI voters would have decisively favored the loser, that won't matter. PASC voters almost certainly would have favored the PRD candidate. In this race, you see a spoiler effect similar to that of Nader in the Bush v. Gore election, and of Perot, in President Clinton's two elections, with the PRI and PASC and New Alliance together having a much bigger impact than either of those two third party candidates.

I don't know Mexican politics well enough to know if PRI voters would favor Felipe Calderon, or Lopez Obrador, although I suspect that would lean a little to the left. But, the likelihood that PRI voters are evenly divided between the two as a second choice, whomever that choice is, is unlikely, and in a vote as a close as this one, even a slight preference among PRI voters would have made a huge difference.

It could happen in the United States. In fact, there is a chance that no candidate in the three way Democratic Party primary in Colorado's 7th Congressional District will receive a majority, and it is likely that no candidate in the six way Republican Party primary in Colorado's 5th Congressional District will receive a majority.

This is somewhat tolerable at the primary stage, as all candidates from the same party share many views, indeed, primaries were created to make non-majority wins take place at this less important juncture, to prevent two strong members of the same party from splitting their party's vote and losing to a less popular member of the other party. A serious third party impact on the general election in 2006 in Colorado is unlikely this year. But, in a three way race is quite a likely possibility in the Connecticut Senate race, if Ned Lamont wins the Democratic primary and Joe Lieberman runs as an independent.

If it does happen, the quirks of the election laws may have more to do with the result, than the popularity of the winning candidate among the front runners in the eyes of all voters.

But, it shouldn't happen in the United States. We don't need the complexity of instant runoff voting, which requires voters to make speculative second choices. We don't even need to require voters to consider a second choice in every election. Very races are even close, and many have only two candidates anyway. But, when neither candidate receives a majority of the vote, we should have a runoff. Indeed, if we adopted such a system, primary elections, manyh of which are pro forma in any case, and much of the elaborate political party nominating process, would also be unnecessary. Everyone could run in the general election, and in the few races where no one received a majority at that point in time, there would be a runoff election which would be closely watched and have high turnout. In this case, at least, the French are closer to right, than the Americans.

(As a footnote, the Congressional elections for the lower house in Mexico were roughly as follows, making a big gain for the PRD which now holds about 20% of the seats:

Here is how it looks right now, based on district-by-district counts:

PAN: 140
PRI: 101
PRD: 169

Proportional allocation:
PAN: 69
PRI: 52
PRD: 58
*To be determined: 21

Total:
PAN: 209
PRI: 101
PRD: 169
*To be determined: 21

*This is because I don't know how they handle rounding and minor parties in allocation of proportional seats.


The existence of proportional representation voting at this level is one of the main reasons that voter turnout is higher in Mexico than in the U.S. and is also one of the main reasons that third parties survive in Mexico, but not the U.S.

It isn't clear if the recount will affect this part of the election as well.)