Tuesday, September 28, 2010

Malapportionment in Venezuela

Much has been made of the electoral engineering in Venezuela that led to a large discrepancy between seats and votes.  In the Venezuelan case, this is due largely to malapportionment (some people's votes count more than others because of how the district lines are drawn).*  In 2009, district lines were redrawn so that, for example, 20,000 people in Amazonas have one representative, while Caracas has one representative for 350,000 voters (see also Matthew Shugart's blog post last year on the reduction of proportionality).  Thus, the opposition votes in the latter are worth far less than the pro-Chávez votes in the former.

Nationwide, this led to the PSUV winning about 60 percent of seats with 48 percent of votes (a difference of +12).  Meanwhile, the opposition received 47 percent of votes and about 38 percent of seats (a difference of -9). (Note: these numbers are approximate and we're still waiting for final results, but they're good enough for illustration).

If democracy means "one person, one vote," then this is clearly a problem.  Unfortunately, it is a problem in many countries widely considered to be democracies.  This does not excuse Venezuela, but rather puts it in comparative perspective.

Take the 2001 legislative elections in Great Britain, for example.  Labour won 40.7 percent of the votes, yet 62.5 percent of the seats, which is a whopping +21.8 percent difference (though this occurred for different reasons than the Venezuelan case).  Labour "took" those seats from Conservatives and Liberal Democrats (who got over 10 percent fewer seats than votes).  Seat and vote discrepancies have long been an issue in Great Britain.  Rein Taagepera and Matthew Shugart note how the 1929 British parliamentary election crushed the Liberal Party because of electoral rules, a defeat that led it to be a minority party for decades.**

What should we take from this?  The Venezuelan opposition has every right to complain, as the rules are set in the PSUV's favor and some votes are more valuable than others.  Electoral rules are generally crafted and re-crafted to benefit those who are making those rules.  Unfortunately, doing something about it will be difficult, because of course the rules themselves make it harder to obtain the necessary votes.  But it is a common problem even in established democracies.

*Many people are using the term "gerrymander" for Venezuela, but gerrymandering refers to changing district lines while keeping the number of voters in each district about the same.  That is not the case in Venezuela.

**Rein Taagepera and Matthew Soberg Shugart, Seats and Votes: The Effects and Determinants of Electoral Systems (New Haven: Yale University Press, 1989): 2.  I note this book as well because it is the classic on the topic.

29 comments:

boz 12:23 PM  

I'm going to have to disagree on the gerrymandering point. District lines within some states were drawn to heavily favor the government.

In Carabobo, for example, the opposition won the state 54-46, but only won 3 of the 10 seats.

This occurred because there were five districts drawn, four districts with one seat and one district with three seats. The districts were drawn in a way so the opposition won one of the single member seats by an enormous margin and lost the other six district level seats.

The only way that is more gerrymandering is if a district is shaped like a salamander drawn by a guy named Gerry.

There were also three state level seats. They won 2 of 3 state level seats for their state level win, giving them 3 and the PSUV 7.

Greg Weeks 12:34 PM  

My point is that redrawing the lines may either be malapportionment or gerrymandering, but not automatically the latter. Both are intended to redraw lines to benefit a particular group. Changing the district magnitude can exacerbate either.

boz 1:07 PM  

Ok, my point is that both occurred in this election. The difference between Amazonas and Caracas is a strong example of malapportionment. The redrawn district lines within Carabobo is an example of gerrymandering.

Tambopaxi 2:59 PM  

In any case, Greg's point still holds, that is, that in cases, the votes of some people (almost always PSUV sympathizers)weighed more than those of opposition in obtaining seats...

Tambopaxi 3:00 PM  

Make that, "...in some cases, the votes of some people...".

leftside 7:26 PM  

The difference between Amazonas and Caracas is a strong example of malapportionment.

So any attempt to give the traditionally underrepresented areas (rural) and people (indigenous) is somehow fundamentally bad policy? But (as Greg points out) many countries do the same thing and the US promotes this very practice in many cases abroad. Not to mention the US Senate, where my California vote gets diluted to the hundredth degree versus the voter in North Dakota or Alaska.

The redrawn district lines within Carabobo is an example of gerrymandering.

Boz, even your likely source for this nonsense about Carabobo (Quico at Caracas Chronicles) admits that this is hardly a clear-cut example of Gerrymandering. For that, the US still takes the cake. (Pro-opposition blogger) Quico says:

Actually, I disagree that "Chavista master-minds did everything within their reach to gerrymander this election as much as possible". I think they were almost prim about it, as you can see in my post today.
Think of all those Toss-Up districts in the Andes. If they'd REALLY wanted to be aggressive about it, they could've lumped in La Grita and Ejido with the West of Barinas into a Massive, four-member chavista constituency... same with Caracas, etc....

The funny thing about that map is that it leaves many more competitive districts than was strictly necessary. These guys really could've taken some lessons from the Texas legislature: they leave ZERO competitive seats after the Gerrymander...that's how the big boys do it.


Indeed, this so-called "gerrymandering" left 25 "toss-up" seats, using 2007 referendum data (ie. the final votes were within 5% of each other). A true gerrymandering effort tries to eliminate all close contests. Quico shows that it would not have taken much for the opposition to win these new districts, by picking up a few thousand votes here and there. They did not and so they lost big time.

Also, what your proclamation of gerrymandering ignores is that the districts were gerrymandered BEFORE. Districts were split up to ensure the dominance of the 2 major pre-Chavez parties. Just take a look at the pre and post apportionment maps. They are without a doubt cleaner and make more geographic sense now. Before they had all the weird salamander shapes. Now they seem to make sense.

And of course, it is the opposition's own fault for not playing the game the last few years through their bone-headed boycotting of the legislature. It is a little much to claim they were shut out of the process when they were responsible for that. Plus, the opposition did have 1 of the 5 CNE redistricting slots.

leftside 7:26 PM  
This comment has been removed by the author.
Justin Delacour 11:07 PM  

So any attempt to give the traditionally underrepresented areas (rural) and people (indigenous) is somehow fundamentally bad policy?

This is one of the few times that I'm inclined to agree with Greg on something pertaining to Venezuela. Malapportionment is something to be avoided, wherever it may occur. The U.S. Senate is indeed an example of malapportionment, and so too is Venezuela's National Assembly. The problem is that, whenever the apportionment of seats doesn't generally reflect how people have voted, people's faith in the fairness of the country's political institutions will tend to decline. I think that's a problem to be avoided, no matter if happens in the United States or in Venezuela.

Anonymous,  11:16 PM  

Leftside--

Please remember in the US existing states created the national govt. Federalism is built in to the original document giving states representation in the legislative branch (Senate). The fact that each state gets an equal number of senators can never be changed. The unfairness of this misrepresentation (electoral college too) was the price of small states agreeing to join the union.

In Venezuela, the 1999 Chavista constitution directly attacked the idea of federalism. The senate was abolished. Hence the gerrymandering is not based on preserving the equality of historic and territorial states that demand representation but rather as a way to win national elections.

What confuses me is that in most countries the rural vote tends to be traditional and conservative. Apparently one sees the greatest levels of support for Chavez in the countryside. What explains this reversal?

Anonymous,  11:21 PM  

I tend to agree with Justin. The examples of malapportionment in the EU, US, Spain etc... leave a bitter taste in the minds of voters. Whether they are historic compromises in a federal system or more recent attempts to gain or perpetuate power, it goes against the basic idea of equal citizenship.

Justin Delacour 11:52 PM  

Federalism is built in to the original document giving states representation in the legislative branch (Senate). The fact that each state gets an equal number of senators can never be changed.

Just because our federal system is deeply institutionalized doesn't make it optimal. Our federal system doesn't even clearly establish which levels of government are responsible for what in the event of a natural disaster (such as Hurricane Katrina). As for our bicameral legislative branch, my sense is that unicameral legislatures make for more efficient governance. The founders' intention in creating two legislative bodies was not to bring about democracy but rather to stifle its full development. As James Madison himself said, the point was to create so many "checks and balances" that the people could not use government as a means of taking away property from the "minority of the opulent" and redistributing it. In other words, our system is designed to preserve gross social inequities, and I don't think there's anything particularly democratic about that.

Tambopaxi 6:05 AM  

Seems to me that there's an article in the Venezuelan Constitution that mandates proportional representation or something asi.

Is this true? If so, to my mind, proportional would mean (in my mind, anyway) one man, one vote, i.e., that all votes would be of equal value, which clearly did not happen on September 26...

boz 7:47 AM  

To partially answer your question Tambopaxi:

Article 186 of the Venezuelan constitution:
The National Assembly shall consist of Deputies elected in each of the federal entities by universal, direct, personalized and secret ballot with proportional representation, using a constituency base of 1.1 % of the total population of the country.

Article 294:
Electoral Power organs shall guarantee the equality, reliability, impartiality, transparency and efficiency of electoral processes, as well as implementation of the personalization of suffrage and proportional representation.

boz 7:52 AM  

One more: Article 63
Suffrage is a right. lt. shall be exercised through free, universal, direct and secret elections. The law shall guarantee the principle of personalization of suffrage and proportional representation.

So the right to proportional representation in voting is in the constitution three times.

Justin Delacour 10:40 AM  

The problem is that, unless you have a specific set of rules about how seats will be apportioned, the Venezuelan courts will interpret the term "proportional representation" however they please.

leftside 1:06 PM  

Is this true? If so, to my mind, proportional would mean (in my mind, anyway) one man, one vote, i.e., that all votes would be of equal value, which clearly did not happen on September 26...

According to someone who's actually studied "malapportionment" in the region, Venezuela is one of three countries that specifically does not have a formal constitutional provision guaranteeing the “one citizen - one vote” principle (along with Colombia and Uruguay). Yet every country in South America, (with the exception of Peru) displays high levels of similar malapportionment despite the fact that their constitutions formally prescribe the equality of each citizen’s vote. So this idea that Venezuela is somehow violating their Constitution is erroneous. If anything, it is all the countries who's Constitutions specifically mention "one vote one person" that are in violation.

Also, this idea that the system put in place only helped Chavez's party (the PSUV) is patently false. For example, the PSUV received at least 40% of the votes in Zulia, Anzoátegui, Nueva Esparta, and Táchira, but the opposition won a total of 27 seats in those states compared to seven for the PSUV. (20% of the seats, with more than 40% of the vote).

The opposition's losses occured because they did not win enough votes in the majority of electoral districts (circuitos). It is not rocket science. The only "skewing" is that each State is required to have 3 legislators - and that the indigenous are guaranteed 3 slots. After that, population determines the number of legislators each state has - based on a standard formula. So yes, a few states with small populations are "malapportioned." But this is to serve a legitimate public purpose (of giving a voice to the traditionally voiceless) - not just to pad PSUV's totals. Like some have noted, it is not written in stone that rural areas vote socialist.

leftside 1:07 PM  

Here's the citation for my first paragraph above.

Justin Delacour 2:41 PM  
This comment has been removed by the author.
Justin Delacour 2:43 PM  

Leftside:

The part of your point that I agree with is that Venezuela is invariably singled out on these sorts of issues. For people like Boz, malapportionment is not an issue unless it happens in Venezuela. Such selective indignation got old a long time ago.

That said, it's awful hard to justify a system of apportionment that has just granted the PSUV roughly 60% of legislative seats when it won only 46% of the national vote. I think the rules of apportionment are short-sighted, and I don't think you're really going to be able to sustain the argument that the rules are in no way affected by the interests of the ruling party.

The danger for the PSUV is that such rules could erode its reputation over the long term in the eyes of many Venezuelans. In short, I don't think these rules will serve the Venezuelan left's interests over the long term.

leftside 10:57 PM  

Justin, fair enough, you are not a believer in malapportionment. I think in an ideal world I'd agree with you. However in a very unequal world - the only one capitalism knows how to create - giving some systematic advantages to traditionally disadvantaged communities does not strike me as anti-democratic. It seems to me to be a progressive idea (if applied correctly), consistent with a pluralist approach to democracy.

Low populations and loss of population is associated with underdevelopment. Minorities always have it a bit rough in representative democracy. Why not tilt the electoral system a bit in those areas favor? Isn't it better to have a structural solution rather than band-aid, hand-out solution?

To make sure we are clear, am I correct in assuming you have a problem with setting aside seats for women in Afghanistan as well? Or ensuring that the Shias are represented in Lebanon? Or that those areas far from Oslo are not forgotten in Norway?

Justin Delacour 11:41 PM  

However in a very unequal world - the only one capitalism knows how to create - giving some systematic advantages to traditionally disadvantaged communities does not strike me as anti-democratic.

This is beginning to sound like a post-hoc rationalization for apportioning more seats to states where the PSUV is strongest. This is the first time I've heard a leftist claim that rural areas are entitled to disproportionately more legislative seats than non-rural areas.

First off, I'm not really so sure that people who live in Venezuela's rural areas are so much more disadvantaged than other Venezuelans, per se. They're generally poorer, but they also live in safer environments. By your logic, rural people are somehow entitled to disproportionately more political representation than Caracas barrio dwellers. I don't understand that logic.

Minorities always have it a bit rough in representative democracy.

How does living in a rural area automatically qualify one as a "minority"? I've never once heard the term "minority" defined in such a way.

To make sure we are clear, am I correct in assuming you have a problem with setting aside seats for women in Afghanistan as well?Or ensuring that the Shias are represented in Lebanon? Or that those areas far from Oslo are not forgotten in Norway?

The problem is that you're now attempting to equate rural status with the categories of gender, ethnicity and religion. You're using a completely unprecedented definition of what the term "minority" has always implied. That's why your argument sounds to me like a post-hoc rationalization for apportioning more seats to states where the PSUV is strongest.

I don't have any problem with guaranteeing Venezuela's indigenous a few seats, but the notion that rural states should be overrepresented to such a degree strikes me as problematic.

leftside 4:54 PM  

Justin, to be clear what we are talking about in Venezuela, they do two things that could be called malapportionment:

1) Giving 3 automatic seats to indigenous groups. You say you don't have a problem with this.

2) Guaranteeing every State has a minimum of three seats in the Assembly. This is designed to ensure that States with low or declining populations have a bare minimum of representation at the national level. Perhaps I over-generalized a bit by saying the intention was to give a rural voters a bigger say. That is part of it, but it is also about States rights. But instead of the US Senate, which says all States get equal representation no matter the population, Venezuela has a formula that ensures proportional representation after the 3 seat minimum is met. It seem entirely reasonable and related to a legitimate public purpose to me. It is basically the same thing Norway does (giving the hinterlands some minimum level of representation) - and they can do no wrong with me :) (my family is Norwegian).

Justin Delacour 6:05 PM  

It is basically the same thing Norway does (giving the hinterlands some minimum level of representation) - and they can do no wrong with me :) (my family is Norwegian).

But I doubt that you ever see a situation in which a Norwegian party wins 46% of the vote and gets almost 60% of the seats in the parliament. If the tables were turned and the party that was overrepresented was one you didn't like, I doubt you'd be so keen on it. And that's really the issue for me. A principled position is not one in which it's okay for my side to set up rules in its favor but not for the other side to do the same thing. If there's a problem in how seats are apportioned, then it's a problem no matter whose side it priveleges. Such are the norms of democracy, in my view.

leftside 8:01 PM  

Justin, as Greg showed, the results in British elections are often far more "skewed" than this one was in Venezuela.

Honestly, I urge you and everyone take a look at the actual Venezuelan vote results. It is not some grand unfathomable conspiracy that allowed the PSUV to gain more seats than their vote totals would seem to predict. In fact, it is totally meaningless to look at nationwide totals and come to some conclusion on National Assembly elections as everyone is doing. You have to look race by race, district (circuito) by district. The PSUV won a "disproportionate" nummber of seats for a very simple reason - because most of the close races happened to go their way. A few thousand votes the other way in these couple dozen races would have swung the proportion to the opposition. But the PSUV won more races, pure and simple.

And again, no one has shown how the rules were specifically written to benefit the PSUV. Gerrymandering appears LESS than before. Giving underpopulated rural districts is done all over. Guaranteeing minorities like indigenous populations a seat does not seem controvertial. So what exactly was done with ill intent?

Justin Delacour 9:23 PM  

Justin, as Greg showed, the results in British elections are often far more "skewed" than this one was in Venezuela.

Actually, that's debatable. What happens in England is that Labor and the Torries effectively divy up a lot of lib-dem votes between them because the U.K. doesn't have proportional representation. When a smaller centrist party's votes are effectively divied up between the larger parties to its left and right, that's probably not as controversial as what you're seeing in Venezuela. If you were to have run-off elections in U.K. voting districts, the results wouldn't be all that different from what they end up being with the country's current system.

Honestly, I urge you and everyone take a look at the actual Venezuelan vote results. It is not some grand unfathomable conspiracy that allowed the PSUV to gain more seats than their vote totals would seem to predict.

No, it's not a grand unfathomable conspiracy. It's just a very questionable set of electoral rules.

In fact, it is totally meaningless to look at nationwide totals and come to some conclusion on National Assembly elections as everyone is doing. You have to look race by race, district (circuito) by district.

No, it's not meaningless at all. The whole reason that some countries developed proportional representation systems is to avoid exactly the problem of malapportionment. In Sweden, for example, the apportionment of seats in parliament almost perfectly matches how the national electorate has voted. The only exception is that parties that don't get past the four-percent threshold don't get seats.

The PSUV won a "disproportionate" nummber of seats for a very simple reason - because most of the close races happened to go their way. A few thousand votes the other way in these couple dozen races would have swung the proportion to the opposition. But the PSUV won more races, pure and simple.

No, actually, it's not that simple. The way seats are apportioned in each state is actually more convoluted than that. I don't completely understand the process, to be honest. There's all this stuff about "safe seats" that I don't fully understand yet.

Look, Leftside, I'm every bit as left-wing as you are, and I still sympathize with Chavismo, but I'm just not gonna drink the Kool Aid on this one. For you to put the term "disproportionate" in quotes suggests that you're just not willing to recognize the basic fact that seats are probably not well apportioned in Venezuela.

And again, no one has shown how the rules were specifically written to benefit the PSUV. Gerrymandering appears LESS than before. Giving underpopulated rural districts is done all over.

But when was the decision made to guarantee all rural states three seats? Was that in the 1999 constitution, or was that a more recent change? I'm curious.

Justin Delacour 10:23 PM  

The Venezuelan opposition has every right to complain, as the rules are set in the PSUV's favor and some votes are more valuable than others.

I agree with the point about the opposition having every right to complain, but, after reading up a bit, I'm going to have to agree with Leftside that the primary bias in Venezuela's electoral system is not by conscious design. The CNE changed a few electoral rules that probably swung a few seats to the PSUV in this election, but the primary source of the PSUV's disproportionate number of seats is the 1999 constitution itself, which gave rural states a level of representation in the National Assembly that is vastly disproportionate to their population size. This favors Chavismo because Chavismo dominates in most rural states, but there doesn't seem to be any evidence that this bias was designed to favor Chavismo. Nevertheless, it's clear that rural areas are heavily overrepresented, and the opposition has every right to be upset about it.

MSS 7:02 PM  

The UK 2001 comparison needs a huge caveat: the UK system has minimal malapportionment, but is FPTP and thus not intended to be proportional. The Venezuelan constitution, on the other hand, is generally interpreted as requiring proportional representation. Yet the system obviously has produced grossly disproportional results.

Also, I agree with boz that some of this is gerrymandering. Given that compensation within this mixed-member "proportional" (MMP) system is carried out state-by-state, not nationwide, there is plenty of opportunity to draw district lines in smaller states with partisan bias. And, as boz also notes, not all districts in the nominal tier (where candidates run as individuals, rather than on the "compensatory" lists) are single-member. If magnitudes of districts were chosen for partisan advantage, I would call that a form of gerrymandering (or more accurately, from past Irish magnitude-rigging experience, we might call it tullymandering).

leftside 2:40 AM  

Now that we've all educated ourselves on the intricacies of Venezuela's elections laws, it seems the most knowledgeable of the election critics are not blaming gerrymandering - like Boz and our host did. They are blaming this arcane change in calculating the way the 2 separate votes are combined.

Here's a pretty good article explaining the issue. I finally think I fully understand the system. It's modeled on Germany’s, which is designed to help guarantee:

that small parties could be represented in the legislature even if they did not win any directly elected district representatives, as long as they got over a certain percentage of the statewide party list vote.

The rule change some are citing as "fraud" is actually the result of fixing a loophole in the old system. Parties could split or run as an unofficial alliance, and get awarded extra seats as a result of the gimmick. After the opposition did it first, the Chavistas followed (MVR birthing the UVE). In a supreme tale of hypocrisy, the opposition actually SUED (unsuccessfully) to stop the practice. The arcane 2009 rule change now blamed for the "disproportionate" results was meant to overturn the very policy the opposition sued as Unconstitutional.

The issue of giving States a minimum # of seats was put in place in the 1999 Constitution - primarily as I now understand it, because the Senate was being dissolved. I imagine if we dissolved our Senate we'd have to give something to the low-population States too (because of the enormous power they'd have lost). Still low population states have more power in the US legislature, than in Venezuela. A unicameral system sort of makes this deference to low-population States necessary - again - if we agree that rural areas require some institutionalized bias.

To address your point on urban vs rural poor Justin, I'd argue that poor urban barrio dwellers can be better served through means other than electoral. In Latin American rural areas, the poor are usually the vast majority (60-80%). In cities, the poor may account for more like 25-40%. So putting them on equal footing, actually benefits the urban wealthy. Especially in a country where the capital is so dominant (Caracas), tilting the legislative scales a bit in favor of the hinterlands is simply good policy. The idea is that the rural areas need development more than the cities.

leftside 3:59 AM  

The idea of including a mechanism to "over-represent" the rural areas and low-population States is not a Chavez invention. It was built into the system before. The 1999 Constitution added the 3 indigenous seats and compensated for the loss of the US style Senate by giving States some minimum representation (3 seats). But 3 seats of 165 is FAR less power than was granted through the old US-style bicameral system before 1999.

  © Blogger templates The Professional Template by Ourblogtemplates.com 2008

Back to TOP