Monday, May 31, 2010

Colombian election

Antanas Mockus received far fewer votes than expected yesterday (21.5 percent to Juan Manuel Santos' 46.6 percent) though perhaps just about the percentage we would have expected six weeks ago.  Santos is looking to scoop up the votes necessary to get over 50 percent (and to retain the ones he has) while Santistas in the legislature are telling Mockus to conceded to avoid the cost of a second round.

Matthew Shugart notes that Mockus' chances "dimmed dramatically," but I wonder precisely how dramatically?  Put differently, what is the highest percentage differential that a candidate has overcome from a first round?  The difference here is 25.1 percent, which is higher even than what Mockus received, and so it seems inconceivable that Santos would lose in a second round.

Boz and Steven Taylor also have quick rundowns, while Colombia Reports notes how the traditional parties lost out yet again.

20 comments:

Justin Delacour 4:12 PM  

Antanas Mockus received far fewer votes than expected yesterday (21.5 percent to Juan Manuel Santos' 46.6 percent) though perhaps just about the percentage we would have expected six weeks ago.

Doesn't this seem a little fishy, though? One normally doesn't see such a dramatic shift away from what the pre-election polls had been indicating.

My guess is that the paramilitary strongholds in the Colombian countryside are delivering some suspiciously lopsided vote counts in favor of Santos.

Steven Taylor 9:21 PM  

@Justin: Actually, that would not seem to be the case. Santos has outperformed expectations across the country, including in Bogotá, which should have been the cornerstone of Mockus' electoral strength. There is little doubt that Santos has the traditional get-out-the-vote capacity that Mockus doesn't, and that is part of the explanation. Another part is that accurate polling in Colombia is difficult. Further, the numbers were trending to Santos before polling stopped a week ago.

Still, I am surprised by the margins.

And btw, the areas of the country dominated by paramilitary groups in terms of politics are not large enough to produce the results we saw from yesterday's vote, even if they manage to perpetrate a remarkable amount of fraud.

Justin Delacour 2:14 PM  

accurate polling in Colombia is difficult

Funny how we never hear this from U.S. observers when a poll comes out suggesting 80% popular support for Uribe. The only time we hear questions about the accuracy of Colombian polling is when an Uribe-backed candidate outperforms the pollsters' projections to such a degree as to seem suspicious.

The sudden about-turn on whether Colombian polling is reliable strikes me as a bit odd.

In a country where the military has been known to kill civilians by the thousands and dress them up as guerrillas to pad their battle kill numbers, one could be excused for having doubts as to whether the country's legal and institutional framework is really up to the task of guaranteeing a genuinely "free and fair" electoral process. A little skepticism might be in order.

Steven Taylor 3:11 PM  

I think that a little skepticism is warranted, and in some parts of the country more than others. Indeed, in some locales, maybe more than a little.

The sudden about-turn on whether Colombian polling is reliable strikes me as a bit odd.

Perhaps, although I am not sure as to exactly what you are referring. I can't be held responsible for all persons in all situations who may have made claims about Colombian polling. I have never made any claims here or anywhere about the significance (or accuracy) of Uribe's popularity polls.

I will say that Santos' showing tends to indicate that Uribe's popularity is, indeed, quite high.

I do know something about Colombian elections, however, and based on what I have seen, the notion that the paramilitary groups were able to swing the numbers this radically simply doesn't make sense.

For what it's worth, I think that a better showing for Mockus (and, indeed, a win in 3 weeks) would be better for Colombia. For that matter, it would be better for me as a guy who studies Colombian parties and elections, so it isn't as if I am shilling for Uribe or Santos.

Justin Delacour 3:53 PM  

it isn't as if I am shilling for Uribe or Santos.

I don't think you're shilling for Uribe or Santos. I think, rather, that you --like most U.S. observers of Latin American politics-- have an unconscious interest in deflecting criticisms of Colombia's political institutions. The reason that most U.S. observers of Latin American politics have certain interests in deflecting such criticisms is that Colombia is a U.S. ally.

Steven Taylor 4:32 PM  

And upon what do you predicate said assumption, save for the fact that I am a US political scientist?

For that matter, why can't it be that it is actually possible that I have at least a passing clue about what I am talking about?

I would note that you are basing your initial assertions based on rather vague impressions.

Justin Delacour 4:51 PM  

I would note that you are basing your initial assertions based on rather vague impressions.

My assertions are based on the contradictions implicit in your own points.

When pressed, you write:

I think that a little skepticism (about Colombia's electoral process) is warranted, and in some parts of the country more than others. Indeed, in some locales, maybe more than a little.

But notice that you never actually volunteer such skepticism in your posts about Colombian elections. I think that is telling.

You say "accurate polling in Colombia is difficult" when such polling raises questions about the country's election results, but it would seem to me that you're very selective about when to be skeptical of a Colombian poll and when not to be.

Steven Taylor 9:34 PM  

Justin,

I remember why I frequently avoid engaging you, as you really don't want to have a conversation, you want to indict the person with whom you are having interaction.

Let's deal with some basics:

1. It is not unreasonable to state that polling in any developed state might be difficult, especially when trying to capture the opinions of rural voters. This is not an especially controversial position to take.

That does not mean, by the way, that the polling is utterly unreliable, just that there is more room for error. Again, this is not all that controversial a position to take.

The trends in the polls were going to Santos a week ago when polling stopped. As such, a Santos win is not a shock. Further, the polls got the rank-order right.

The gap was a surprise, however. Still, one thing that was always possible was that Santos would have a better get-out-the-vote mechanism that would Mockus. Santos has more money and is from a well-connected political family and had the ability to tap into old-school Liberal networks.

The pro-Uribe sentiments in the country (which are quite real) clearly also are relevant here.

Heck, polling in the US (and, recently, the UK, for that matter) does not produce perfect predictions.

2. Again, let's return not to impressions, but to facts. You contended originally: "My guess is that the paramilitary strongholds in the Colombian countryside are delivering some suspiciously lopsided vote counts in favor of Santos." However, the vote tallies don't support this hypothesis. For example, Mockus did only slightly better in Bogota than he did nationally. Bogota is not a "paramilitary stronghold." Do you even know which departments constitute such? For that matter, wouldn't such vote totals be obvious vis-a-vis the other parts of the country? There is no such evidence to my knowledge.

And yes, I have more confidence in Colombian institutions than do you, although I am quite aware of their shortcomings. Of course, I have been studying them rather intently now for almost twenty years, so I have some grounds for my positions. Not, by the way, that I expect you to agree with me. Evidence for you arguments, however, are always welcome.

Justin Delacour 12:26 AM  

It is not unreasonable to state that polling in any developed state might be difficult, especially when trying to capture the opinions of rural voters.

No, it is not unreasonable in the least. What is unreasonable is when people like yourself only make an issue of polling problems when it suits your theses (and then downplay such problems when they don't suit your theses). If the accuracy of Colombian polls is suspect now, then the accuracy of Colombian polls is no less suspect at any other time in recent Colombian history. You can't say, on the one hand, that a poll showing 80% support for Uribe is reliable and then turn around and say "accurate polling in Colombia is difficult" whenever a poll result doesn't suit your thesis.

So, if we're gonna go over some "basics," I think it's important to point out the basic fact that your position on Colombia's polling is inconsistent.

Heck, polling in the US (and, recently, the UK, for that matter) does not produce perfect predictions.

The point is deceptive. Polling can never produce perfect predictions, but the shift from a projected 35-35 split to a 46-21 blowout in a matter of two weeks is odd. And it's especially odd when nobody has offered a logically coherent explanation as to how the methods of Colombian pollsters could have been so biased against Santos. (It's not as if the polling methods of Colombian pollsters have ever been known for bias against establishment candidates and in favor of outsider candidates.)

Again, let's return not to impressions, but to facts. You contended originally: "My guess is that the paramilitary strongholds in the Colombian countryside are delivering some suspiciously lopsided vote counts in favor of Santos." However, the vote tallies don't support this hypothesis.

Actually, Steven, you haven't genuninely contested my guess that the paramilitary strongholds in the Colombian countryside are delivering some suspiciously lopsided vote counts in favor of Santos. (I haven't said one measly word about Bogota, by the way, but, now that you mention it, I suspect the level of paramilitary infiltration in a massive barrio like Ciudad Bolivar is much more extensive than you choose to believe). Nor have you presented one iota of evidence to suggest that the paramilitary strongholds in the Colombian countryside exhibit anything approaching a political climate conducive to "free and fair" voting. What you contest is that vote counts from the Colombian countryside could fully account for the size of the vote spread. On the latter point, you may very well be correct, but until we get some real clarification as to whether or not there are some significant shenanigans going on, I'm going to remain suspicious.

And yes, I have more confidence in Colombian institutions than do you, although I am quite aware of their shortcomings.

I happen to think that a country with the level of institutionalized tolerance for something so macabre as the "false positives" scandal cannot be accurately described as a democracy. Perhaps you disagree, but I doubt very seriously that you would be as soft on an official enemy like Venezuela if it were to exhibit anything even remotely close to the level institutionalized tolerance of such atrocities.

Steven Taylor 11:52 AM  

Actually, Steven, you haven't genuninely contested my guess that the paramilitary strongholds in the Colombian countryside are delivering some suspiciously lopsided vote counts in favor of Santos.

But the numbers from such locales, even if given massively over to Santos, would not have produced the national results we have seen. It's basic mathematics.

(I haven't said one measly word about Bogota, by the way, but, now that you mention it, I suspect the level of paramilitary infiltration in a massive barrio like Ciudad Bolivar is much more extensive than you choose to believe)

No, you haven't. But to explain the outcomes of the election, one needs to do so.

Again, evoking basic math, one would have to account for the fact that in Bogotá, Mockus only did marginally better than he did nationally. Again, it is impossible (even using your Ciudad Bolivar guess) to account for those numbers. And no, Ciudad Bolivar is not known as a paramilitary stronghold, although it is an area of potential manipulation by traditional political means.

Indeed, the team of international electoral observers that I was part of during the March congressional election did some work in Ciudad Bolivar.

But again, you don't want to discuss or even debate, you want to guess and then impugn and accuse. It always strikes me an odd. It is possible, you know, to have a real conversation (and even disagree).

Justin Delacour 1:29 PM  

The problem, steven, is that Colombia is so lawless that it's difficult to tell what sorts of electoral shenanigans are going on, where they're going on, and how. I've been to ciudad bolivar. It's a totally lawless place and therefore not the kind of environment where people have real guarantees that they can exercise their political rights freely. An election observer won't necessarily figure that all out by simply showing up there on election day.

The problem with your whole analysis is that you ignore the level of lawlessness and vigilantism that is deeply emb edded within colombian politics. When a sizable chunk of medellin is known to be controlled by officially illegal paramilitary militias, you can't tell me that's an environtment where the freedom of speech and assembly thrive. When a nuevo herald reporter receives 24 death threats in the wake of the president's denunciation of him, you can't tell me that's a country with thriving press freedoms either. Given these sorts of realities, I remain doubtful that colombia's legal and insitutional framework is really up to the task of guaranteeing free and fair elections. You haven't given me any reason to believe otherwise.

Steven Taylor 5:50 PM  

I suppose, then, it begs the question of why you commented on this in the first place. Why not just say that you don't really believe any of the numbers at all? Why even question polls v. election returns if, in the end, you don't think any of it is actually valid?

Justin Delacour 7:54 PM  

I suppose, then, it begs the question of why you commented on this in the first place.

You know why I commented on the post. I commented because the tendency of U.S. political scientists to place such faith in Colombia's political institutions strikes me as problematic. You've yet to offer anything in the way of a coherent rejoinder to that central point. You've yet to explain to me how the level of institutionalized tolerance of state and paramilitary atrocities is conducive to democracy. You've yet to explain to me how a country with such a level of impunity for politically-motivated violence (or threats thereof) can guarantee its citizenry of the most basic political rights (such as the freedom of speech, the freedom of assembly and the freedom of the press).

By what actual standards does Colombia qualify as a democracy, in your mind? The standards upon which you base your assessments are not clear to me because you've yet to explain them.

Steven Taylor 11:10 PM  

You know why I commented on the post.

Well, I thought that you were actually commenting on the disparity between the polls and actual vote. I did not see a massive critique of US political science in the following sentence: "My guess is that the paramilitary strongholds in the Colombian countryside are delivering some suspiciously lopsided vote counts in favor of Santos."

Indeed, that strikes me as a fairly narrow, empirical claim not a universal critique of all US social science concerning Colombia. My frustration is that you have never actually address this claim. Really, the claim itself makes no sense if what you believe is that Colombia is so utterly under the control of violence that none of its election outcomes are valid in the first place.

BTW, you will be shocked, I suppose, to find that I don't think that you have provided any coherent rejoinders to anything I have said, either.

In regards to my views on Colombian democracy, I would suggest you read my book on the subject, as it is a better forum for a more complicated discussion that a blog comment form. However, I suspect you wouldn't find it very persuasive, so I guess we can leave it at that for the moment.

It is likely fair to say that you don't take me especially seriously, and nor do I you (although I have made, or more than one occasion, a good faith effort at engaging you). I am willing, by the way (as I think I have demonstrated, although you appear to disagree) to discuss these matters, but so it goes.

Justin Delacour 1:40 PM  

What is of interest, steven, is not what I think of you nor what you think of me but rather what we have to say about the issues under discussion. I wish you would at least try to answer a few basic questions that I have posed instead of trying to turn this into some sort of pissing match. We might be able to understand each other's points a little better if you would simply lay out the basic standards by which you evaluate the state of Colombian democracy.

Steven Taylor 3:32 PM  

I wish you would at least try to answer a few basic questions that I have posed instead of trying to turn this into some sort of pissing match.

The funny thing is, this is exactly what I thought I did. Your first comments made a specific point, which I then attempted to address in a perfectly reasonable fashion.

Quite frankly, Justin, you turn almost every conversation I have ever read on this blog in which you are a participant into a pissing match.

Justin Delacour 4:27 PM  

Steven:

Your initial response was a simple knee-jerk defense of colombia's political institutions. Your point didn't even address my speculationabout paramilitary strongholds. When I ask you for specifics about whether you think the political conditions exist for genuinely free and fair elections, you just throw your hands up in disgust and start telling me how awful I am. You're the one doing the pissing. A pissing match is a convenient way to evade pertinent questions.

Steven Taylor 11:48 AM  

You don't suppose that automatically treating someone's response as "a knee-jerk response" might not be part of the problem?

Basically you are admitting what is quite clear, and as I said above: you didn't take me seriously from the get-go, so what's the point of discourse?

Justin Delacour 4:25 PM  

Steven:

The reason I view your initial response as having been a "knee-jerk" response is that your response does not logically follow from that to which it purports to respond. You first claim to refute my speculation about paramilitary strongholds, but, in reality, you say nothing about how the electoral process functions in such places. Your "response" deals with another matter altogether.

When a response does not logically follow from that to which it purports to respond, it is often the product of a reflexive rather than logical thought process.

Does that necessarily mean that everything you say is "knee-jerk"? No.

Unfortunately, though, you're now just evading basic questions. We can't discuss the standards by which we evaluate the state of Colombian democracy if you refuse to tell me what your standards are.

MSS 11:25 AM  

The biggest margin I know of that a second-place candidate ever has overcome in a two-round presidential election would be the case of Portugal 1986. Freitas do Amaral won something like 47 or 48%, with Soares at around 25%. Soares won the second round, by winning virtually every vote of the Communist candidate, who had placed a not-too-distant third. On the other hand, the forces of the right had all rallied behind Freitas in the first round and had hardly any more votes to get.

I know of no other case even close to that. And clearly Mockus is not in the sort of position that Soares was in, in terms of ideological positioning relative to the rest of the field.

  © Blogger templates The Professional Template by Ourblogtemplates.com 2008

Back to TOP