September election poll

09.18.2005

Technorati tag:

The most recent Bolivian voter poll by Mercado, Opinión y Apoyo, published in Sunday's La Razón, puts Evo Morales (MAS) ahead in Bolivia's election 9 weeks away. But I think these numbers are soft, principally because Bolivian voters tend to break away from front-runners at the last minute. More importantly, the numbers seem heavily weighted by the La Paz sample and don't include Beni or Pando samples.

This was a good sample from across the country, w/ respondents from the cities of La Paz, El Alto, Cochabamba, and Santa Cruz. As well as rural areas in the departments of La Paz, Cochabamba, and Santa Cruz. But the real story (which not told clearly by the graphic) is the difference across departments. In the end, it'll all come down to electoral math. Because the presidential vote is also a vote for the Senate & half the House of Deputies.

But a look at regional poll figures is instructive. Evo's commanding 28% national lead is driven mainly by his 31% & 49% leads in urban La Paz & El Alto, respectively. In those two cities, Jorge "Tuto" Quiroga (PODEMOS) scores 25% & 14%. While Doria Medina (UN) comes in w/ 19% & 16%. If the election were held today, it seems MAS would get two senate seats and a good chunk of deputies. However, if MAS wins more seats from the single seat districts (SSDs) than it would otherwise get form proportional representation, it won't win any more seats. So. If PODEMOS can put together a good slate of SSD candidates, we could still end up w/ a near tie in La Paz (in terms of seats), w/ a slight advantage for MAS. And even though Quispe (MIP) is polling dead at 1%, he could play spoiler by taking a chunk of SSD seats in the altiplano.

In Cochabamba (his home department), Evo loses to Tuto 28% to 26%, w/ Doria Medina coming in third w/ 12%. Again, this is a statistical tie, but w/ a slight advantage for PODEMOS (meaning it wins 2 senate seats, to 1 for MAS).

Moving east, the trend continues w/ Evo taking only 17%. Here, Doria Medina wins w/ 31%; Tuto places third w/ 26%. The result means UN would take 2 senate seats,
PODEMOS 1, w/ the three parties splitting most of the lower house (but w/ MAS taking a small piece of the pie). Also, notice the difference in rural votes. While Evo does well overall in the rural vote (49%), he does a low 4% in Porongo (Santa Cruz) — tying the MNR's Nagatani. Here's where it matters. Santa Cruz has one of the largest number of lower house seats. Having split La Paz & Cochabamba, Evo needs a large chunk of Santa Cruz seats to have a chance.

Why? Because no doubt Tarija, Beni, and Pando will swing closer to PODEMOS and/or UN, taking probably all of those 9 senate seats & the bulk of their lower house seats. The regional poll graph shows a table for urban respondents in the cities of Oruro, Potosí, Tarija, and Sucre: Tuto wins them all. MAS would have to completely sweep rural Oruro, Chuquisaca, and Potosí to make up the difference. In contrast, Tuto & Doria Medina only have to get a few seats here & there to block Evo's presidential aspirations.

In the end, the legislature will likely elect the president. While there is talk of suspending that in favor of a second-round election between the first & second place candidate, it would require a major constitutional alteration. Not likely to happen. Instead, the legislature will elect a president. Legislators from UN & PODEMOS are more likely to come to an agreement w/ each other, than either of them w/ MAS. To make that clear, Doria Medina (UN) has categorically denied an alliance w/ MAS — and no one expects Evo to embrace Tuto.

Of course, there's still 16% of voters undecided, and another 11% planning to vote null or blank. So as much as a fifth of the electorate is still up for grabs.

Posted by Miguel at 10:34 PM