Sunday, October 09, 2011

Metrics of Democracy's Sickness

Version 0.6

Metrics of Democracy's Sickness

This is in response to John Dean's second installment on the destructive gaming of American democracy at

This is an important topic, but I really feel like you [John Dean] are focusing too much on some of the relatively minor fronts in the larger war. There are at least two other fronts where I think much larger battles are being fought--and lost--by the proponents of democracy. The first front involves gerrymandering, and the second front involves biasing the judicial system.

(1) Why do 40% of the voters ignore elections? That's far more than the roughly 30% that vote for the winning presidential candidates. Mostly because they believe their votes don't count and can't affect the results--and they are right. Their districts have been gerrymandered and their votes have been effectively precounted and negated before they are ever cast.

A metric for the degree of gerrymandering should be easy to calculate, and yet I've never seen a good one published anywhere. The effectiveness of partisan gerrymandering is in the degree to which you can waste or squander your opponents' votes. Equivalently, that means you want to distribute your own votes to maximize their impact. These effects must ultimately be measured by considering the anti-democratic results of the elections. The degree to which the electoral and policy outcomes differ from the preferences of the actual voters should be something that can be mathematically assessed and compared.

I can easily cite two examples of abusive gerrymandering, and both of them involve the neo-GOP politicans gaming the system. The earlier example is in Texas, where the Republicans forced early redistricting in 2003. My very own district was held by Lloyd Doggett at that time, but that district was stretched to Houston to include enough Republican voters to tilt it safely to the GOP, and Doggett was basically obliged to move to a different district with a much higher and newly concentrated percentage of Democratic voters--where more of those votes were now a meaningless excess. There was no massive shift in the voting demographics of Texas or in my original district, but in one fell swoop the Republicans were able to capture a large number of House seats from the Democratic Party.

The second example is in Pennsylvania. On a statewide basis the state voted for Obama in 2008, but the Democratic voters are highly concentrated in certain districts while many other districts have narrow (but safe) Republican majorities. I'm not certain how much of this situation was the result of deliberate gerrymandering, but the current situation is highly imbalanced. Therefore the Republican legislature is trying to change the rules regarding their electors. The math shows how Obama could win most of the votes in Pennsylvania while the gerrymandering would allow his opponent to capture most of the electoral votes.

I spent a while trying to define a good metric for the harm of gerrymandering. The best reference I was able to find was this paper on the geometric assessment of gerrymandering, but it doesn't consider the voters at all. It barely mentions them, so you wind up feeling like 'No harm, no foul.' It barely acknowledges the negation of voters. Based on that paper and my own struggles, I believe a good metric will consider the geometry, but not just the districts' shapes. The geometry of the voters must also be considered. In addition, to assess the harm, the electoral outcomes and even the legislative outcomes should be assessed. Finally a useful purpose for polling? Well, that would be nice, but...

Just to clarify the harms of extreme gerrymandering, I'm going to construct a little example here. We have two imaginary states, A and B, each of which has 10 equal districts of 100 voters. So as to include the most extreme case of the harm, we'll assume that each state's 1,000 voters are divided into 694 consistently blue voters and 306 consistently red voters. You can easily imagine the associated political parties, eh?

State A uses nonpartisan redistricting with the intention of producing the fairest and most representative possible outcome. The voters are somewhat unevenly distributed, with concentrations of blue and red voters. In this situation, it is not difficult to draw the lines to that 7 districts will be blue and 3 will be red. There is some threat of 'dictatorship of the majority', especially if party discipline is strong, but that is a known danger and we even have some established responses, such as some parts of the Bill of Rights and the judicial system.

The situation in State B is similar, but the red party is in charge of the redistricting and they are allowed to use perfect gerrymandering. They draw the districts so that their 306 voters are perfectly placed in 6 districts with 51 red voters each. The result of that election is that less than 31% of the voters would then control 60% of the legislature. If the red party has good party discipline, we can wind up with a fake democracy that is actually a strong dictatorship by a small minority.

I was going to say that the real situation in America is not that extreme, but now I find myself wondering. The actual percentage of Republican voters is only around 25%, and yet their party discipline is so strong that they were able to cripple the Senate with even fewer than 40 Republican Senators. The Democratic Party is fundamentally weak on party discipline, and the Republicans could pretty much always count on getting one or two Democrats to join them. Since gaining that 40th Senator, I feel the Republicans have gone way beyond crippling the Senate and right to the edge of destroying it as any sort of democratic institution. Then you have to consider that the Senate is not even apportioned fairly in relation to the population... Throw in the gerrymandering and various forms of disenfranchisement, and maybe the amazing thing is that the federal legislators EVER pay attention to the actual voters in this reputed democracy?

(2) The second front involves biasing the judiciary by appointing judges based on age and political opinions rather than based upon their judicial qualifications. I actually think this one would be easier to measure. There are only a few categories of data that have to be assessed. The easiest one is the age of judges who are nominated to the federal judiciary. If you think the neo-GOP is biasing things on a political basis, then the obvious prediction is that their judicial appointees will be significantly younger, the longer and better to thwart the voters' will when they pick the 'wrong' president. The other metric would be ideological consistency of the judges in their decisions. The prediction here would be that the judges appointed by Democratic presidents would be less ideologically consistent in their rulings.

If these are two of the main fronts in the war against democracy, then I think America is in a whole lot of trouble...

No comments:


About Me

My photo
As a blogger from before there were blogs, I've concluded what I write is of little interest to the reading public. My current approach is to treat these blogs as notes, with the maturity indicated by the version number. If reader comments show interest, I will probably add some flesh to the skeletons...