Preliminary Honduran Election Analysis, Part Two

05/25/2011 02:55 pm ET

With the Supreme Electoral Tribunal (TSE) nearing completion of its first pass through Honduras' election results, a more subtle (albeit still incomplete) analysis has become possible. What is certain is that the National Party won an unprecedented victory. What remains in question is precisely why. Answering this question requires a closer examination at voter participation trends in previous elections and inferential analysis of what took place in 2009. Below, I first present the results, before offering a hypothesis to explain them.

The Results

Porfirio Lobo and the National Party won a landslide victory at every level of government. In the presidential elections, the National Party took upwards of 55 percent of votes cast, while the Liberal Party--long the numerically dominant party in Honduras--could not even muster 40 percent. In the Congress, the National Party won over 70 seats of a total of 128. And, in the mayoral races, the National Party has carried more than 200 of 298 municipalities. The final tallies may differ in a handful of races, but the general trend--a National Party routing of the Liberal Party--will hold.

To put this trouncing in perspective: since political liberalization and civilian elections in 1981, no presidential candidate has received 54 percent of valid votes (not to speak of total votes cast). Furthermore, consider the last time that the National Party won the elections (only the second time since 1981). The winning presidential candidate, Ricardo Maduro, won with less than 50 percent of votes cast (52 percent of valid votes), and the National Party obtained neither an outright majority in Congress (it got 61 seats) nor among mayoralties (it won 148 of 298 municipalities).

In fact, after both the 2001 and 2005 elections, the winning party won only a plurality in Congress, forcing it to form coalitions to pass legislation. For the next four years, the National Party will not face this obstacle. This could further marginalize the three smaller parties (Democratic Unification party, UD, the Christian Democratic Party of Honduras, DC, and the Innovation and Unity Party, PINU). Moreover, the UD's very existence stands in question, given sharp internal divisions about whether to participate in the elections (the party decided to participate only a week before the election) and the party's predictably miserable showing.

The municipal results are equally stark. In the 2005 elections, the Liberal Party won 167 (56 percent) of the countries' municipalities, even though Manuel Zelaya received under 50 percent of valid votes cast nationally. As of inauguration day in January, the Liberal Party will have lost roughly 80 municipalities, a sea change for this small country. In these elections, even many previously "safe" municipalities for the Liberal Party, like El Paraiso (department of El Paraiso) and Trinidad (department of Santa Barbara), went to the National Party.

Explaining the Landslide

So, how can one explain these results?

The first part of the explanation is what Hondurans refer to as a voto de castigo (punishment vote) against the Liberal Party. Voters blamed the Liberals for letting their internal squabbling culminate in a coup, the subsequent internal instability, and Honduras' international isolation. Those without firm party affiliations--a rising group, given massive party de-alignment since 2001 in Honduras--swung toward the National Party, providing the Nationalists with a major advantage.

But the voto de castigo alone seems unlikely to have delivered such an unprecedented victory for the National Party. The other part of the story, as I began to discuss in my last post, was turnout. As I have noted before, turnout has been declining in Honduras since 2001. There is now a bit more data on 2009 turnout levels (though results are still tentative), which permits more analysis than was possible last week. According to CNN, the TSE has quietly reduced its reported turnout from 61 to 56 percent, still a bit higher than the 55 percent level in 2005. Nonetheless, the TSE has not explained the discrepancy with Hagamos Democracia's statistically-representative quick count, which estimated turnout at 47.6 percent, with a margin of error of less than two percent.

Furthermore, the TSE has acknowledged cases of ballot-stuffing and other irregularities in 9 percent of the scrutinized actas (records from each polling station), which are currently under investigation. These cases would not be sufficient to swing the election, but they would affect turnout results.

Finally, candidates who have been keeping vigil over the counting process have raised concerns over the repeated turning off and restarting of the computing system. With observers (their own neutrality in question) gone since early last week, no independent sources have been present to gauge the transparency of the counting process, which suggests that we will simply never know whether turnout was greater or lower than in 2005.

Meanwhile, the TSE is trying to use an accounting trick to make turnout appear much higher than in previous elections. Since 1.2 million Hondurans lived outside the country, they have argued (and the pro-Micheletti media has parroted) that the true number of eligible voters should be 3.4 million, not 4.6 million. This attempt to change the denominator is a cynical accounting trick. While it is true that virtually all emigrants will not vote (despite absentee voting stations in the U.S.), similar numbers of Honduran emigrants were included in the voter registry in previous elections. Given that all analysts of this election have been primarily interested in comparing voter turnout with previous elections, no good justification exists for using the lower denominator that the TSE has promoted.

This still leaves unanswered the question of why, in addition to the voto de castigo, the National Party swept these elections. My hypothesis is that, though 2009's voter turnout figures may have been roughly comparable to 2005 figures, who turned out in the two elections differed substantially.

In the 2005 elections, as the Latin American Public Opinion Project's reports (LAPOP) have shown, the five most frequently-given explanations for not voting were, in ascending order: not being in the registry, illness, not liking the candidates, lack of interest, and not having identification to be able to vote. Demographic variables also had an impact. Multiple-regression analysis revealed that the following variables increased the odds of voting: paying attention to the news, identifying with a party, perceiving higher levels of corruption (a counter-intuitive result), perceiving that one's economic condition had improved in the past four years, and being Catholic.

In this election, disgust with the removal of Manuel Zelaya and fear of potential violence were almost certainly two principal predictors of abstention. And, whereas Liberal Party members were more likely to vote than National Party members in 2005, this pattern likely reversed in 2009. With the split in the Liberal Party and the coup, Liberals were less likely to vote, while Nationalists--united around their candidate, Porfirio Lobo, and against Manuel Zelaya's call for a boycott--turned out in droves. This hypothesis is consistent with journalists' election-day reports of less turnout in lower-income Zelaya strongholds in Tegucigalpa. It is also squares with my own pre-election interviews in the department of El Paraíso, a traditional Liberal Party stronghold, where many Liberals expressed ambivalence about voting, while Nationalists more often shared their excitement. To Nationalists, suffrage this year was a more patriotic act than ever before. They were also thrilled for their vote to count as a rejection of Manuel Zelaya.

A full analysis of this hypothesis may have to wait until the next round of LAPOP surveys (scheduled for next year), which should incorporate these additional factors in analyzing voter turnout. For now, when it comes to voter turnout and explaining the National Party landslide, inference is the order of the day.

(Copied with permission from www.americasquarterly.org.)