Mexico: Tribunal orders partial recount

The Mexican federal Electoral Tribunal–the independent court of last resort in election disputes–ordered a partial recount of the presidential ballots. The seven magistrates unanimously rejected the calls of Andrés Manuel López Obrador, who trailed by about half a percent in the preliminary count, for a recount of all ballots. López Obrador, in turn, rejected the ruling and promises to continue civil disobedience.

The Tribunal’s order calls for the re-opening of ballots boxes from 11,839 polling places (about 9% of the total) where arithmetic errors have been found in the count reports filed with the Federal Electoral Institute (IFE) on election night.

See also the LA Times story, which includes some discussion of the background of election procedures in Mexico. For example:

Recounts must be based on evidence specific to a poll station, said Justice Alfonsina Navarro, not broad suspicion.

Chief Magistrate Leonel Castillo, arguing against a full recount, said Mexicans had already counted the vote in a system that gives ordinary citizens the job of running the national election.

Mexican polling stations are operated by trained volunteers, and the votes are counted in front of political party representatives before the results are marked on tally sheets and the ballot boxes sealed.

“They are citizens — not permanent members of state institutions — who are chosen randomly among their own neighbors to count the votes,” Castillo said during a nationally televised broadcast of Saturday’s session. “They verify, instant by instant, step by step, moment by moment. They’re the witnesses.”

The partial recount will start Wednesday and last about five days. If substantial discrepancies are found, then the Tribunal will have to make a further decision as to whether to allow a more complete recount or annul the election. If the partial recount does not turn up serious errors in this sample, then the Tribunal will certify the election, which it must do by 6 September in order for the apparent victory of Felipe Calderón to be official.

A somewhat extended version of this planting appears at PoliBlog.

Interview with Mexico Watch

Mexico Watch reports on political and economic developments in that country for business investors and others with an interest in Mexico. The most recent issue includes an interview [PDF; begins on page 3] with your Orchardist about the election. With the permission of the publisher, F&V is able to present some excerpts:

Mexico Watch: Given the experience of other presidential democracies in Latin America, and the specifics of the Mexican system, how viable is Calderón’s pledge to form a “coalition” government? What forms could this take and what institutional barriers might impede it?

Matthew Shugart: My answers assume a Calderón presidency, but of course that is not a sure thing. We have to wait for the Federal Electoral Tribunal (TEPJF).

In any event, we have to ask, coalition with whom? The experience of other presidential systems suggests that a coalition of the president’s party with that of the runner-up (in this case the PRD) is highly unlikely. The very definition of a presidential system is that the president is solely responsible for organizing and directing the executive branch. Thus the second-place party has little to gain and much to lose from being a junior partner in a coalition for which the president’s party will get most of the credit – or blame. And it is blame that it will count on, for the PAN taking blame for whatever goes wrong over the next six years increases the Partido de la Revolución Democrática’s (PRD) chances of picking up the top prize in 2012.

A coalition with the PRI is more likely. Probably not a formal coalition, but a working arrangement. The PRI is badly weakened and should be willing to cooperate without demanding too high a price.


MW: What is the likelihood of electoral reform to break the three-party stalemate, and what might a reformed system look like?

MS: I do not like the use of the word, “stalemate,” or the notion that three-party politics needs to be “broken.” There is no majority party in Mexico, and the idea that we should engineer one with crafty electoral-system design is simply the wrong way to approach the situation. While we certainly could create an electoral system that would give one party a majority of deputies despite its not having a majority of votes, I don’t see such a system being legitimate. That is, these parties – yes, even the PRI – represent real constituencies of real Mexicans. Until such time as one of them can convince a majority of Mexicans to vote for it, the parties will have to learn to bargain with one another. That’s democracy, and thus something to celebrate!

Of course, one could make a very convincing case that the electoral system for the presidency should be changed to require a runoff. As a student of presidential elections, I do not see anything inherently wrong with presidents being elected with less than 50 percent of the votes, but when they are elected with much less than 50 percent and also a tiny margin, the case for a runoff is strong. But let’s recall that constitutional amendments are unlikely to pass without the cooperation of the PRD. Would that party agree to a majority-runoff format? Could the PRD expect to win a nationwide majority? Would either the PAN or the PRD want to put the PRI in a “kingmaker” position in a second round? Interesting questions to ponder! Maybe they could agree to a plurality of less than 50 percent remaining sufficient, but only if some stipulated margin over the runner-up has been achieved. If the margin requirement were not met, then there would be a runoff.

In short, Mexico has a divided electorate, but despite the tensions of this election, I would not say a deeply divided one. Compromise is possible – likely, in fact. The existing electoral system for congress works well, as does the existing balance of powers between the executive and legislature. The troubles Fox had with Congress were more a result of three parties sort of feeling their way in the new competitive environment than of anything structural. If I were asked what one thing to change about Mexico’s institutions, it would be to allow legislators (maybe the president, too!) to be reelected. Otherwise, do no harm! Mexico has come a long way in a short time, and its democracy is arguably healthier than most in the region.

The Mexican Electoral Tribunal: This is no Bush v. Gore

Regarding the Electoral Tribunal of the Federal Judiciary (TEPJF), Mexican political scientist José Antonio Crespo notes:

La calificación presidencial del año 2000 fue una prueba facilísima para el TEPJF, fue como pasar el kínder. Ahora, en este 2006, la calificación presidencial será para el TEPJF como su doctorado.

Indeed. (Roughly translated: In 2000, validating the election was as easy as passing kindergarten. In 2006, it will be like defending one’s doctorate.)

So, what is this body that now has the resolution of Mexico’s electoral dispute in its hands? It is a judicial body of last resort, charged with resolving election disputes and nothing else. Its Higher Chamber (Sala Superior) consists of seven magistrates who serve ten-year terms, expiring this October. (There are also five regional Salas of three members each.)

The terms of TEPJF magistrates are non-renewable. The body was established by the 1996 electoral reform (a constitutional amendment), and its members are elected by two-thirds vote of the Senate, from a terna, or list of three names (per vacancy), presented by the Supreme Court. (The Supreme Court justices themselves, since another constitutional reform in 1994, are also elected by a two-thirds vote of the Senate from a terna sent by the President, and serve for fifteen-year, non-renewable terms.)

The first TEPJ Sala Superior (which thus is the current one) was actually required to be elected by three fouths of the Senate, and in fact, all votes on these magistrates were unanimous.

In other words, the upcoming case is no Bush v. Gore.

Additional notes:

El Universal has a short profile of each magistrate. Some are career judges, others are academic law professors.

The TEPJF itself has an English-language page that explains its role and also offers profiles of the magistrates.

It is worth noting that the Tribunal is sometimes referred to as the “TRIFE,” after the name of the tribunal that was in place in the early 1990s. The older acronym, often written Trife, is still used, presumably because “Tepjf” is not pronounceable!

Mexico congressional result and “governability”

With all the international media focus on the disputed outcome of the presidential race in Mexico, attention to the equally important congressional outcome has been minimal. Whichever candidate is ultimately inaugurated–and while it is likely to be the PAN’s Felipe Calderón, a reversal in favor of the PRD’s Andrés Manuel López Obrador can’t be ruled out yet–how effectively will the president be able to govern?

The answer depends on which one is president, because the PAN performed far better than the PRD in the congressional seat-allocation process.

The PAN has won 207 of the 500 seats in the Chamber of Deputies, against 160 for the PRD-led coalition and 119 for the PRI. Two smaller parties obtained representation: Nueva Alianza (9) and Alternativa (5). The votes percentages for the three leading parties split: 33.4, 29.0, 28.2. The PAN’s enormous advantage ratio (41.4% seats/33.4% votes) of 1.24 is a stark reminder that Mexico’s mixed-member system is not MMP. In fact, the PAN’s degree of over-representation is just short of the legal cap of eight percentage points. Mexico’s MMM (parallel) system greatly advantages the party that performs best in the single-seat-district plurality races. With 139 such wins in 300 districts (46.3%) on just over a third of the vote, that was the PAN, by a big margin. (The PRD-led alliance won 99, or 33.0%, and the PRI a paltry 62, or 20.7%.)

(The result is classic three-way competition in plurality SSD races: The PRI undoubtedly lost votes to the PRD, but seats to the PAN.)

In the Senate, the PAN will have 53 seats, the PRD-led coalition and the PRI each will have 37, and one seat will be held by Nueva Alianza. (Votes percentages were: 33.4, 29.7, 28.1.)

If the narrow victory of Felipe Calderón in the preliminary official count is upheld after legal challenges, he will be in a strong position vis-a-vis the congress.

Much commentary that I have seen and heard–that which has even referred to the congressional result, that is–has lamented the supposedly difficult “governability” that Mexico is in for. I have to put the word, “governability,” in quotation marks for two reasons. One, I never know precisely what the concept is supposed to mean, and two, whatever it might mean, it appears to imply a bias in favor of the president. That is, a situation of “low governability” is loosely defined as one in which the president lacks the ability to assure passage of his policy proposals, as though it were completely natural that such proposals should pass. The Fox years have been described as years of “deadlock” in numerous quarters. But, of course, the very nature of presidential democracy is that there are two elected branches (with one of them, as in Mexico, divided into two chambers). If the president has been unable to win a majority in the legislature for his party, we should have no expectation that his proposals should pass, at least not without compromise and alteration. That’s democracy.

So, how would a Calderón presidency compare with that of Fox, based on the performance of the president’s party in congress? Pretty well, we can expect.

Consider that in 2000, the PAN-led coalition managed 223 seats. Superficially, this looks like a stronger position than the PAN will find for itself in the new congress. However, that coalition included the PVEM (green party), which was an unreliable ally and later aligned itself with the PRI. Moreover, the PRI itself had 209 seats. In the Senate, the situation was worse for Fox, as the PRI held the plurality (44 to 41).

So, after the 2000 election, the PRI and PAN were close in seats in both chambers. The PRI could make the calculation that it was likely to make a comeback. Indeed it did, in the 2003 midterm deputies election, increasing its total to 223 against only 155 for the PAN. The PRD was a distant third in both the 2000 and 2003 elections, but its best known national officeholder, Mexico City mayor López Obrador, was considered the front-running presidential contender for 2006 throughout most of Fox’s term. The expectation that it would win the next presidential contest put the PRD in little mood to help Fox and his party. There was little incentive for either major opposition party to work to help Fox succeed, because both parties expected to improve their electoral position in the near future. Helping the PAN and Fox could only hurt their own cause.

Now, on the other hand, the PAN will have more than a forty-seat margin over the next largest party in the 500-seat Chamber and a 16-seat edge in the 128-seat Senate. While the PRD is unlikely to offer much of a hand of cooperation to a Calderón presidency, the PRI is not likely to be as much an obstacle as it has been. It will want access to policy-making and patronage to sustain itself in the states where it remains strong. It should be willing to bargain, as it is no longer in a position where a return to national plurality status is realistic.

It is worth noting here that on any matters for which the president–either man–wishes to change the constitution, the PRI is not in a pivotal position. Its votes in the Chamber of Deputies remain short of the necessary two thirds when combined with those of either the PAN or the PRD.* That is, any major structrural changes–such as opening up oil extraction to foreign partnerships–the PRD and the PAN would have to agree. That seems like a good thing, given the closeness of the result and the “dinosaurness” of the PRI. Mexico’s two modern, programmatic parties will have to cooperate for anything really major to be done. Which probably means not much will be, which is not a bad thing, given the evident absence of consensus within the Mexican electorate.

So, on legislation (but not constitutional amendments) the Calderón “governability” outlook is actually pretty good. For regular statutory and budgetary politics, 40% of both chambers and a large margin over the next largest party provide a good bargaining situation for Calderón. If he is indeed the next president of Mexico. What if AMLO is instead? Then things are not so bright. A president López Obrador would have his party in command of less than a third of the seats in either chamber. In other words, not even enough to sustain vetoes (which can be overridden by two-thirds votes), let alone much of a base around which to build positive majorities to pass administration programs. He would be highly dependent upon the PRI to accomplish much of anything (just as he would have owed his narrow plurality in the presidential race to voters who cast PRI votes for congress but AMLO votes for pesident). The alternative is that he and his party come to agreement with the PAN. He would have to bargain.

Or would he? There are those who think he would be a “Chávez” and govern without regard for the niceties of checks and balances. The idea of an AMLO presidency is at this point very hypothetical, but let’s think a bit about the validity of these scary scenarios. The short story is I do not find these claims credible. It is not as though presidents can circumvent congress by force of will or even through mobs in the streets. Especially presidents who won barely 35% of the vote. Chávez won a solid majority in 1998–the largest in the then 40-year history of multiparty elections in Venezuela (and in a single-round election). He likely would have won a comparable share of congress had not the old-line parties changed the electoral cycle before the 1998 election to prevent a “coattail” effect (holding the elections separately and long before the Chávez phenomenon had really taken off). Chávez, a cashiered Lt. Colonel, also had elements of the armed forces on his side. The Venezuelan Supreme Court was highly corrupt and politicized and in little position to defend the constitution against Chávez and the obvious popularity of his planned “revolution.” Federalism in Venezuela hardly mattered at all. In fact, de-facto federalism probably was a less significant constraint on the central government in “democratic” Venezuela than in “authoritarian” Mexico.

AMLO would have none of these advantages: A small personal electoral plurality, a poor showing in congress, a quite professional Supreme Court, control of few states, and no prospect of the Mexican army coming to his side. An AMLO presidency would be weak. There are some of us who, regardless of how we might feel about the candidates (and full disclosure here: were I Mexican, I almost surely would have voted for Calderón**), do not think such an outcome would be so bad. After all, who is to say that AMLO was not, in fact, the Condorcet winner, regardless of whether or not he won the plurality? Not me.

I’ll address the Condorcet issue–and the related questions of the value for Mexico in adopting, or not, a two-round presidential election system in the future–at another time.

* I suppose, theoretically, the PRI could make itself pivotal by first making an accord with the Nueva Alianza (9 seats). (The other small party does not have quite enough for its votes, plus the PRI and PAN to equal two thirds of the Chamber.) I suspect this is highly unlikely and even if it were to occur, on the other side of this equation it is worth noting that the PRI will not actually have all the 119 seats I attributed to it above. Some of those are actually PVEM seats as part of the parties’ pre-election alliance. Similarly, the PRD does not have 160 seats, as some of those have been won by its alliance partners, the Worker’s Party and Convergence. The PAN, on the other hand, is just the PAN; it did not offer any nominations on its lists (or districts) to other parties.

** Though I am just as sure I would have given una de tres to the Nueva Alianza.

Mixed-member systems and regionalism

Miguel asks a good question about the extent to which mixed-member systems might encourage regionalism and thus discourage nationwide coalition-building. He was responding to an earlier (and since revised) planting of mine on Mexico, and offering lessons from the Bolivian experience. Regarding Bolivia, Miguel suggested that in 1997-2002 there appeared to be an emergence of separate regional party systems, similar to what appears to have happened in Mexico. He says:

I actually suspect this is a byproduct of MMP systems, since it encourages the localization of candidates (to win SSD seats) — which is part of my skepticism for MMP in new democracies, if it encourages regionalist splits & discourages nation-wide coalition-building.

First off, Mexico does not have MMP. But we might expect that MMM would have more localizing tendencies than MMP (because the districts actually matter to the overall outcome in terms of the balance of seats held by parties). Such tendencies could generate party systems that are distinct in different regions, possibly with two major parties in one region that are not the same two parties that are most competitive in some other region–national multipartism, regional bipartism.

Does MMP (or MMM) discourage nationwide coalition building and emphasize regionalism? As always in comparison of electoral systems, one needs to ask, relative to what? If the allocation of PR seats to parties is national (as it is in Mexico, though not Bolivia) then it obviously encourages parties to think nationally to a degree that a system of just single-seat districts would not. In Bolivia, it is possible that the adoption of MMP increased both personalism and localism because the antecedent electoral system was pure closed-list PR. In fact, on the interparty dimension–how seats are awarded to parties–Bolivia changed little. The old system was PR with each province as a self-contained multi-seat district. The new one is also PR, but of the MMP variety. There is no national allocation.

What there is, with the change, is a series of individual races by plurality. With MMP, that means a change primarily on the intraparty dimension, in that some legislators are being elected on their own “nominal” votes instead of by party ranking. As a result parties may care about the qualities of the candidates being nominated to a greater degree than on the lists, and those candidates might seek to develop personal connections to the districts.

So, should we expect MMP in Bolivia to be more localizing and personalizing than what went before? Probably, and because there is no provision for overhang seats (the addition of seats to compensate for some party’s getting more seats in the SSD competition than its proportional share entitles it to), there should be more premium on parties’ putting effort into the SSDs than if winning these had no impact on the partisan balance of the legislature. In other words, the MMP system provides some new incentives on both the interparty and intraparty dimensions for parties and candidates to exert effort in winning local races. Given that some parties may have an edge over others in any given geographic region either programmatically or in terms of the types of candidates they can recruit, then this should promote more “localization (to win SSD seats)” than the system that went before it.

As for the expectation that I allude to above that MMM would generate greater emphasis than MMP on the single-seat districts (and hence localization and at least local bi-polarization) the only problem with that is that while the rank order of the parties differs across Mexican states, in only a minority of the states is the third party far behind. So, Mexico would seem not to support the hypothesis that MMM (or mixed-member systems more generally) promote local 2-party politics.* The continued strength of the third party (whichever it may be) in districts that it has no chance to win, even under MMM, is partly a result of the single vote. That is, every party that hopes to win PR seats has a strong incentive to nominate 300 candidates, whether or not viable. Even with this consideration in mind, however, we have to ask what the mixed-member system adopted in Mexico is being compared to. Does it promote localization or nationalization, relative to what?

For Mexico, where at one time the system was all SSDs (and all dominated by one party), the addition of a nationwide PR allocation clearly has helped nationalize politics to a degree that could hardly have happened under a pure SSD system.

In assessing the impact of adopting MMP or MMM (or any other electoral system), we have to ask what variables are changing relative to the former system. And the different starting points, as much as the differences in the systems adopted, affect our expectations about whether mixed-member rules would increase or decrease the extent of regional electoral competition in any given polity.

* I realize that Miguel is responding to the original post, where I employed the “slip of the keyboard” and referred to regional two-party systems, when what I meant was simply that the second party is the PRI almost everywhere and thus local competition is usually PRD vs. PRI or PAN vs. PRI, but rarely what it was in the presidential election: PAN vs. PRD.

Mexico’s 3-party politics: Regional differences

Extended and somewhat revised since initial planting.

With all due caveats about making inferences to individual behavior from aggregate statistics, I thought it would be interesting to look at the presidential and deputies votes in the Mexican states for patterns. This is barefoot empiricism, folks!

The presidential race is nationwide plurality. The deputies are elected in 300 individual plurality races, and the votes cast in those districts are aggregated nationwide to determine another 200 seats by proportional representation (in parallel fashion, other than caps on maximum permitted over-representation that were not triggered in this election).

We would expect that three-party politics would result in some ticket-splitting, with some supporters of the trailing presidential candidate (Madrazo of the PRI) choosing between the two leading presidential candidates, but sticking with the PRI for deputies. An indicator that is readily available is the state-level votes for the parties for president and chamber.

The PRI won the plurality of votes for president in no states. (Please pause for a moment and think just how historic that is!!) However, the PRI did retain a plurality for deputies in six states: Campeche, Durango, Hidalgo, Quintana Roo, Sinaloa, and Veracruz. (If you do not know Mexican states, check out the Clickable map of Mexico.)

Of the states in which the PRI won the chamber plurality, which of the PAN and PRD presidential candidates won the most votes? The states split 4-2, with AMLO of the PRD winning Campeche, Hidalgo, Quintana Roo, and Veracruz, while Calderón of the PAN won Durango and Sinaloa.

This breakdown really demonstrates the extent to which Mexico does not really have three-party politics, but two distinct regional party systems with different pairs of leading parties. (In an earlier draft, I said two distinct “two-party” systems, but that is not really right, as the party that came in third in the presidential vote has at least 20% in eighteen states.)

The four states that split their chamber/presidential votes PRI/PRD are located in the Yucatan peninsula or the Gulf Coast, except for Hidalgo (which lies between coastal Veracruz and the Mexico City area). On the other hand, the two states that split PRI/PAN are in the north: Sinaloa is along the Pacific coast south of the border state of Sonora, and Durango is Sinaloa’s neighbor to the east (and immediately south of the border state of Chihuahua).

Continue reading

Mexico: Calderón leads official count

The Mexican Federal Electoral Institute (IFE) has completed its official count of the 2 July election, and the result confirms the preliminary count: PAN candidate Felipe Calderón leads PRD candidate Andrés Manuel López Obrador by about 0.6%.

We now have multiple pieces of evidence that Calderón won in a squeaker: Most of the exit polls (though none were released on election night), the IFE’s own “quick count” sample on election night, the preliminary summation of district-level polling reports (the PREP), and now the official result.

This does not mean that Calderón is the President-Elect of Mexico, however. Under Mexican law, López Obrador (“AMLO”) has a right to bring evidence of fraud to the TRIFE TEPJF, which is the high court for election appeals. (Mexico has the very good sense, unlike its neighbor to the north, not only to have an independent professional election-administration agency, the IFE, but also to separate the process of election-dispute adjudication from the regular court system.) The TEPJF has invalidated elections before, although never a presidential election. (It has never had the opportunity; it did not exist in 1988 when there was a close and probably fraudulent election, and was not needed in 2000.)

AMLO alleges irregularities, and that is his right. If he has a case, the TEPJF will adjudicate it. Democracy can wait. It is better to get it right than to rush to stop legal challenges and possibly install the wrong candidate, as has happened in the very recent past in the neighborhood (as indeed in Mexico in 1988, probably).

The graph below shows the progressive erosion over time in AMLO’s lead in the official count.


It is worth noting that to call the count that took place on 5 July and through to the early morning hours today a “recount” is actually a misnomer. True, in the literal sense that the ballots were individually tallied on election night to produce polling-place reports (actas), and that these actas were then reviewed and some ballot boxes reopened and counted again, it was a “recount.”

However, it was not a recount in the sense that that term is understood in the USA. It was legally mandated and was the only official count. All that preceded it was preliminary. In US states, on the other hand, a recount is a procedure that a trailing candidate is entitled to demand in a close result after the official certification of a final result. Mexico’s “recount” was the official final result, and now the process of contesting (and perhaps obtaining from the TEPJF a full recount) begins.

Why did AMLO have such a lead, and why did it vanish? Does this buttress his claims of fraud? That was clearly his intent. It was the PRD’s explicit strategy to make it look as though he was ahead and was then “robbed” of his lead in the wee hours of the morning. The campaign stalled the verification of actas in districts in which it knew AMLO was far behind Calderón. In that way, Calderón’s votes were systematically added to the official count at a slower rate than were AMLO’s. But the final result almost perfectly confirms the PREP and the quick count.

The two graphs below show results of two of the three statistical sampling methods used to generate the quick count on election night. These were based on samples of just over 7,000 polling places around the country. The IFE was to delcare a probable winner at 23:00 on election night if all three models showed a lead for one candidate outside the margin of error of ±0.3%.


(Click the image to open a larger version in a new window.)

As the quick count came to a conclusion, this model showed Calderón ahead, but with the barest of overlaps in the error bands. This is the model deemed by the IFE to be its most “conservative” (statistically, not politically!).


The second model–deemed the “classic” model–appears to show the confidence intervals just barely separated. A third, Bayesian, model (which I am not showing) has a result that is somewhat in between the others. Calderón appears to be safely ahead, but then the confidence intervals just barely touch each other at the very end of the process.

We now have muliple indicators that Calderón won, with the final official result showing it as 35.88% to 35.31% (the PRI’s Roberto Madrazo is at 22.27%). These multiple indicators do not guarantee that AMLO was defeated–all could be repeating the same errors–but it looks more and more likely that current President Vicente Fox will be succeeded by a new president of his own party.

AMLO has called a demonstration in Mexico City’s Zócalo (central square) for Saturday to press his demands for a recount “voto por voto.” Manuel Camacho Solís, one of AMLO’s top campaign officials, denies that the PRD seeks to anull the election (as the PAN has been charging).

I am not in the business of making predictions, but I suspect that AMLO “has” to do this–given his and his party’s history and the closeness of an election that he had been expected to win until polls tightened this spring–but that he has a weak case legally and little stomach for a major fight in the streets. Public opinion is likely to turn against him and the party if they press the matter too far, much as public opinion turned decisively towards him when the Fox Administration (through Fox’s originally expected successor as PAN candidate, then-Minister of the Interior Santiago Creel) and the PRI tried to have him barred from running over an alleged incident of corruption in AMLO’s administration of Mexico City. Unless the PAN has become Mexico’s new dominant party–and, with just over a third of the votes and a drop of around five percentage points compared to 2000, there is little evidence of that–the PRD stands an excellent chance of winning in 2012. It is not that far away! In the meantime, they have held on to the Mexico City mayorship in Sunday’s election. Mexico’s various states have elections on different cycles, and several are coming up where the PRD could retain or gain governorships. Additionally, the party won just under a third of the lower-chamber seats on Sunday and may well gain more in the 2009 midterm election (as opposition parties usually do). In short, unless the party’s case is much better than what it now seems, it will likely lose and return to its role as a loyal opposition, ready to fight another day.

The TEPJF has two months to review challenges and declare a winner, and its decision is final. The next president will be inaugurated in December.

Sources: The above is gleaned from reading El Universal (which was also the source of the first graph) and other Mexican publications, from having watched many hours of Televisa, and from having corresponded with various friends in Mexico. The graphs from IFE were released at its public website, although I obtained my copy of the document via a private e-mail from a correspondent.

Mexico: Calderón’s lead precarious

Remember, it’s all preliminary in Mexico’s vote count. While we were out watching fireworks, Mexico was experiencing post-electoral fireworks of its own. It turns out that Andrés Manuel López Obrador was right that up to three million votes had been excluded from the reported preliminary count.

With just over two and a half million additional votes added to the count, Felipe Calderón’s lead has shrunk to around 0.6%, or about 257,000 votes. About 1.55% of the vote has still not been included in the preliminary results.

Subsequent to my original posting of this item, various reports have shown AMLO pulling ahead. Note that, while this is the beginning of the official count, it is senseless to project an outcome of this close a race until the count is complete. The candidates have rather different regional bases, and thus the order in which districts are scrutinized will affect who leads at any given point in the process. Just sit back and wait!

The comments below by Matt and Rici are really good. Don’t miss them.

Mexico: Preliminary results

UPDATE 5 July: Calderon’s lead is now down to about 0.6%, as the PRD turned out to be right in its claim that three million (actually about 2.6 million) votes had been improperly excluded from the preliminary results. Remember, it is all preliminary!!

UPDATE: Rici has some information in a comment below about the preliminary seat breakdown. My own information is very close to his: PAN 209 (of 500) Chamber seats, 52 (of 128) Senate seats. PRI-PVEM 119 and 38, PRD+PT+Conv 159 and 36. So the PAN alone will have over 1/3 of the seats in each house–in fact, it will have over 40%, despite under 35% of the vote. I suspect it has won the presidency, too, but that’s not yet certain.

Please note, this is all preliminary. A final count will not be known till at least Wednesday. But the IFE’s “PREP” (via the mirror site at Universal) shows the following as of around 8:00 a.m., Pacific Time today.

President, with over 96% of actas processed:

    Calderon (PAN), 36.4%
    López Obrador (PRD), 35.4
    Madrazo (PRI), 21.5

There appears to have been a lot of ticket-splitting, and the current PAN vote shares for each of the congressional chambers rests at under 34%, while the PRI is around 28% and the PRD around 29%.

Nonetheless, a correspondent who has looked at the district-level results tells me that the PAN has won substantial pluralities in both chambers. The PAN has emerged as the most national of the parties, and the electoral systems of both chambers, while having a proportional component, are quite favorable to any party that has good regional spread.

The “parallel” feature of the lower-chamber mixed-member system means that a party that can win a lot of the single-seat districts (SSDs) will be substantially over-represented. The PAN may have won close to half the 300 SSDs, and will be close to the cap (whereby no party is allowed to add so many seats off the party lists that its seat share of the whole chamber is greater than its nationwide votes share, plus eight points).

In the Senate, the “limited seats” feature, whereby the plurality party in each state gets two seats and the runner-up gets one (plus a parallel nationwide PR allocation of 32 seats) also greatly favors the party that has the greatest regional distribution of strength.

These features of the system used to benefit the PRI. Now, with that party in third place, it is the PAN that is benefited by the plurality features of the system for congress. The PRD is somewhat over-concentrated (dominating the capital), and while this concentration has no negative impact on it for presidency (single nationwide plurality contest), it means that even if the PRD wins the presidency (still very possible), it will face a congress with up to 40% held by its main opponent and perhaps well under a third in its own hands.

The PREP results will be taken down Monday afternoon, and then we will have to wait for the full final count. This a real test for IFE, as the PRI may have had some opportunities in rural areas to pad the count for its congressional candidates. Not all polling places are monitored by the opposition. Can the PRI pressure poll-workers? I do not know. But it is not out of the question. Presumably the PRI, if it could pull it off, would prefer the more divided government under AMLO to Calderón and a strong PAN plurality. [UPDATE: Apparently the number not monitored is very small, so this concern expressed in this paragraph may not be so valid.]

I think Mexico’s electoral institutions are up to the challenge. But this is a big test.

Surprise, surprise

The exit polling says it’s too close to call (per Televisa).

Update: At 11:00 p.m., Mexico City time, the head of the electoral institute and President Fox indicated the election is too close to call. Official results not expected till at least 5 July.

Mexico: Traders update

Caveat emptor, but for what it is worth, the traders now are selling AMLO shares. The race is tightening. (It will be a while before we know anything about the real race.)

Scroll down or click on “Mexico” above for the link to the trader site, and my election preview.

Modern fraud

In advance of today’s election in Mexico, a correspondent tells me the following:

the PRD … is planning to buy votes (300 to 1000 pesos per vote), guaranteed by the voter showing the photo of the marked ballot with a cell phone provided by the party.

Thanks to the establishment some years ago of a world-model Federal Electoral Institute (IFE) that is genuinely independent and professional, Mexican electoral authorities have made great strides in reducing vote-buying and distortions in the count. Nonetheless, the IFE refused to rule against the cell-phone-camera trick.

Mexico’s election

Even though I am co-editing a volume on Mexican elections and lawmaking (which I had once dreamed of having completed before the 2006 election), I have not addressed this Sunday’s election here in some time.

Courtesy of boz’s presentation of the data, I took a look at a series of late polls. Averaging of the nine polls of the week of 23 June gives us the following:

    López Obrador (PRD): 33.6%
    Calderón (PAN/Fox’s party): 32.7
    Madrazo (PRI-asaurus): 24.2

I am not sure of the margin of error on each of these polls, but I know that for at least some of them, it is around +/- 3 percentage points. Applying that margin to each candidate, we get average ranges of:

    López Obrador: 31-37
    Calderón: 30-36
    Madrazo: 21-27

The polls thus pretty clearly agree that the race is too close to call between Calderón and López Obrador (“AMLO”), and that Madrazo is in third place. However, at least two of the polls that boz reports (Mitofsky and Zogby) do not clearly separate second and third place. In fact, Zogby does not clearly separate first and third place. Zogby has the race as Calderón 30, AMLO 27, Madrazo 24. Assuming +/- 3 MOE, that is no different from a three-way dead heat of three candidates at 27%. Zogby also has the largest undecided or “other” (there are two very minor candidates running) of any of these polls (19%), which presumably means they are more conservative in assigning “leaners” to one candidate or the other.

Other polls have a very large gap between first and third, even if none can separate AMLO and Calderón with certainty. For instance Universal has it AMLO 36, Calderón 34, Madrazo 26. Milenio and Marketing Político have Madrazo on 22% and GEA-ISA has him on 20%, but all of these put only two points between the leaders.

While it is hard to concoct a believable scenario in which Madrazo wins, the fact that he has gained a bit in many polls late in the campaign and that his third place is not decisive in several of them is bad news in at least two respects. First, it reduces the pressure on weaker supporters of the PRI to defect, because they may not be convinced that their candidate has no chance. Second, it increases the chance that the candidate who does win might have lost had more PRI voters defected to whichever of the other two is their second choice. My hunch is that Madrazo being in a strong third place (and perhaps second) helps AMLO, on the grounds that AMLO’s image as the rabble-rouser makes him less likely to pick up late PRI defectors than the “conservative” (in almost any sense of that word) Calderón. More PRI voters sticking with Madrazo means fewer votes shifting late to Calderón.

It is possible that the PRI vote is under-sampled, given its strength in rural areas, and out of this possibility come the scenarios (unlikely though they may be) in which Madrazo ekes out a surprising win. But I do think this is indeed unlikely, because I still expect few undecideds to go for Madrazo in the end and many soft supporters of the PRI ultimately to vote tactically, and thus to cancel out any under-sampled PRI vote. The fact that a federation of formerly PRI-affiliated unions endorsed AMLO recently throws further cold water on the chances of a late recovery by the PRI.

I would not bet on this election, as it really could go either way between Calderón and AMLO.

The presidential election is decided by plurality, so there will be no second round even if, as seems likely, the winner has well under 40% of the vote and a very small margin.

The other thing to watch is the congressional race. In 2000, Mexican voters showed some, albeit limited, willingness to split tickets. Thus the PRI in congress could slightly outpoll its presidential result. There is no direct translation of national vote shares to seats in the 500-seat Chamber of Deputies; contrary to some otherwise reputable sources, Mexico’s mixed-member electoral system is not MMP, but parallel.* It is also a rare mixed-member system with a single vote: The voter votes for a candidate in his or her single-seat district, where the plurality prevails, and this vote is also pooled to the separate contest for list seats (closed regional lists).

Because of the single vote for both candidate and list, every party has a strong incentive to run a candidate in every district, even if it clearly can’t win a given district. If it does not have a “face” in a district, it has no way to obtain list votes in that district. And, of course, because of the single vote, there is no incentive for you if you are a voter to defect from your first choice party’s candidate, even if he or she can’t win your district. Thus many districts will be won by narrow margins and much less than 50%, and this will benefit whichever party has the greatest regional spread and harm any party that is regionally concentrated. This aspect of the system is much more likely to give a (small) seat bonus to the PRI than to the PAN, and to harm the PRD.

The system also puts a potential premium on parties’ having attractive candidates, given that the candidates are what the voter must vote for. However, there is not much evidence that voters care much about candidates (there is some, however). The development of a “personal vote” is hampered by the ban on consecutive terms in all elections in Mexico. No incumbents are running for reelection to their seats.

It is very possible that the result could be López Obrador as president with not even one third of congressional seats–the share needed to sustain a veto. If Calderón were to win, he, too, could be short of one third.

In the 128-seat Senate, all of which is also elected Sunday–there is no chance that any party will have a majority.**

Coalition-building will be even more important to Sunday’s winner than it was to Fox–and likely more difficult.

Of course, La Profesora Abstraída has been covering the campaign all along.

* The only senses in which the electoral system is not parallel (MMM) are in two ways that will not be triggered in this election: No party may have an overrepresentation of more than eight percentage points above its nationwide Chamber vote share, nor more than 300 seats. Otherwise, it is a pure parallel system.

** Two senators from each state are elected by the leading party in that state, one from the second party, and another 32 nationwide by proportional representation, in parallel.

Party lists stifling dissent in South Africa–closed party lists, that is

Via the Mail and Guardian, South Africa’s Country Self Assessment Report, prepared before the second African Peer Review Mechanism conference in Soweto, notes that

A wide variety of submissions noted the unintended problems of the nation’s party list electoral system, which stifles dissent and ensures accountability to parties rather than citizens.

It is important to add to this conclusion that the problem is not necessarily party lists, per se, but three more specific factors that, in combination, would be expected indeed to limit accountability of legislators and party leaders–including the executive–to the electorate:

    Closed party lists
    High district magnitudes
    A hegemonic party

Very few successful democracies have all three of these factors, or even the first two.

Closed lists mean that individual legislators depend for their ability to win election and reelection on the rankings given by the party organization, and not on the voters.

High district magnitudes mean long lists, such that a large percentage of legislators are likely invisible to voters.

A hegemonic party–the African National Congress, in this case–means that the lists are longer still (because in most of the large districts most legislators are coming off one list) and also that there is no other party to serve as a credible alternative for voters.

The South African combination of closed lists, exceptionally high magnitudes, and a hegemonic party is unusual, and worrisome for the country’s democratic development.

The report, cited above, notes:

power is concentrated in the presidency and because of his control of the ruling party and ability to appoint [provincial] premiers, directors-general, mayors and party lists, he can end the career of anyone seen to have embarrassed or disagreed with party policy. [my emphasis]

This description sounds very much like standard accounts of Mexico in the decades of hegemony of the PRI. One difference is that South Africa’s “president” is actually a prime minister, who could be removed by the legislative party; however, this is a distinction with little difference in the context of such top-down intra-party authority. Worse for South Africa in this comparison with PRI-era Mexico is that South Africa’s fused party-leader/executive is not subject to a single-term limit, as Mexico’s presidency has been ever since the 1917 constitution.* Also, it is worth noting that, for all the apparent top-down authority in the PRI, Mexico never had a fully closed-list electoral system. In fact, for most of the 20th century most Mexican legislators were elected in single-seat districts and there was much more local and state-level accountability within the party than is conventionally recognized. And much more than appears to be the case in South Africa.

Single-seat districts–for all their other limitations–usually inhibit the kind of stifling central authority that developed in Mexico, and presidents always had to negotiate with state and local leaders to maintain their temporary leadership of the party and government. The ANC could be evolving in a more Priista direction than the PRI itself.

Among established democracies that use PR, a fairly well kept secret is that very few have closed lists. Those that do, like Spain, have many smaller magnitudes as well as more competitive party politics. Others use closed lists alongside single-member districts, like Germany and New Zealand. Most European democracies use flexible lists, in which voters may (or must, in some cases) give preference votes that potentially change the list order. Still others–Finland and Switzerland, for example–use fully open lists in which the rank order of candidates depends solely on voters and not on party organizations.

South Africa has its national legislators elected in nine provincial districts with an average magnitude of 22. And then it has 200 more legislators elected from a national list. By contrast, the largest district using a closed list that I am aware of in a European “pure” PR democracy is the district for Lisbon, Portugal, which has fluctuated over the years in the 40-55 range. But Portugal as a whole has a much smaller average magnitude than even South Africa’s 22 for the regional lists (to say nothing of the 200!), and of course, Portugal does not have a single dominant party, so only a few Portuguese legislators are elected at ranks much lower than around 10th.

Among mixed-member systems, New Zealand has a very high magnitude for its PR tier (50+), but it also has 60+ single-seat districts. One can debate whether the “mixed” nature of the system ameliorates the accountability problem of the closed lists or not, but that is a topic for another thread. Germany’s MMP system is sometimes characterized as having a national PR district, but it does not. It has nationwide compensation on the interparty dimension, but only state-level allocation on the intraparty. And only one state’s PR tier is as big as New Zealand’s, while most are far smaller.

I have already noted the oddity of Ukraine’s 450-seat district with closed lists–a fledgling democracy that appears to have exchanged one severe accountability problem (the 225 deputies formerly elected in single-seat districts, often with little party attachment) for another. But Ukraine has no party close to hegemonic status.

Finally, how refreshing that there is a Self Assessment Report, as part of an AU democracy-strengthenining process, about the limits of South African democracy. Would that the OAS mandated the USA to undergo such a Self Assessment!

* As Alan notes below, the South African PM (“President”) is subject to a two-term limit, or ten years (and it could be up to just short of fifteen years in case of succession between elections). Tenure limits of any length in parliamentary systems are extremely unusual, though this is still a much longer period of time for one party leader to remain chief executive than was the case in Mexico’s PRI (6 years lifetime limit). Perhaps more importantly, it transcends across electoral terms. In Mexico, on the other hand, given their immediate “lame duck” status, presidents probably had to engage in more delicate intra-party negotiations to sustain the kind of control over other politicians of the party (of the sort described in the second quote above for South Africa).