Election Geography 2008

For the past eight years our politics has been riven by the red versus blue state narrative. While the popular media cast red versus blue as a culture war rooted in the ‘60s, subsequent research shows our divisions have much to do with geography. As Obama and McCain distance themselves from partisan stereotypes, many hope the upcoming election will break this pattern, but recent primary results should give us pause. (We should note that explaining overall election results is different than explaining geographic patterns. For instance, all women voters could vote the same and since women voters are a majority of the electorate, that would explain how their candidate won. But since women are fairly evenly distributed across the population, no geographic pattern would emerge.)

Our political geography has been deciphered by several studies by the Metropolitan Institute at Virginia Tech, political scientist James Gimpel with The Christian Science Monitor’s Patchwork Nation website, and journalist Bill Bishop in his book titled, The Big Sort. All these studies show how the basic divisions plays out among urban, rural, and suburban communities. The best way to examine this phenomenon is with census demographic data by county.

The following table shows how presidential voting in 2000 and 2004 broke down by county characteristics. The relevant county data include population per sq. mi., median family income, share of married households, share of female heads-of-household, as well as shares of white and black households.


Regression analysis confirms that population density and marriage status explain most of the differences in voting patterns. One might guess that race was a more significant factor, but female heads-of-household and black households were very highly correlated—at .81, where 1.0 is perfect correlation—and female heads-of-household dominated the racial factor.


Fast forward to 2008 and this is where it gets interesting. We apply this same methodology to recent hotly contested Democratic primaries and what we discover about how different communities voted may surprise those banking on a new post-partisan geography.

The following table displays the county profiles of three state primaries in Pennsylvania, North Carolina and Indiana compared to the national profile. Amazingly, these 259 counties offer almost a perfect demographic sample for the total set of the nation’s counties, so these three primaries taken together offer a good proxy for the national profile.


Comparing the primary results for these three states to the election results for the same counties in 2000 and 2004, yields the following results.


We see that the voters in these three states’ counties voted in a distinct red vs. blue pattern. Counties that voted for Obama align closely with those who voted for Gore and Kerry and those that voted for Clinton align almost perfectly with Bush. But remember, all these voters were Democrats! So partisanship has been taken out of the equation and what we’re left with is political preference based upon lifestyle, economic, and community interests.

Regression results are a bit more mixed for these votes because of how identity groups voted. For example, black households and female heads-of-household were even more highly correlated (.9), but black women tended to vote for Obama and white women tended to vote for Clinton. In general, exit polls confirmed that urban, black and college-educated voters favored Obama while older women, suburban and rural, working class whites favored Clinton.

Unless something else changes, the upcoming presidential campaign’s increased ideological rhetoric will likely push voters toward their communal red vs. blue comfort zones. It’s doubtful the personal strengths and campaign strategies of McCain and Obama will be enough to overcome this. Rather, campaign incentives to win at any cost will probably seek to exploit it.

Can Wikis Save Democracy?

Recent polls show President Bush's approval ratings are languishing at a historic low. The only thing worse is the all-time low approval ratings for that wild and crazy gang over on Capitol Hill. Their approval ratings are 10 points lower! Only 16% of those polled say the country is moving in the right direction.
Have we lost all faith in our democracy? Have we decided that no matter who runs things they're going to run them into the ground? The dissatisfaction with our government institutions has reached a nadir (and I don't mean Ralph Nader either). Perhaps there's some audacity for hope (and I don't mean Barack's either).

Our problems of creating a government by the people and for the people may be solved by social network power and a little idea called a wiki. You know, like Wikipedia.

What is a Wiki?

A wiki is an Internet-based technology that enables mass collaboration among peers. It's a collection of web pages designed to enable anyone who accesses it to contribute or modify content, using a simplified markup language. Wikis are often used to create collaborative websites and to power community websites. Some defining features:
• A wiki invites all users to edit any page or to create new pages within the wiki Web site, using only a plain-vanilla Web browser without any extra add-ons.
• Wiki promotes meaningful topic associations between different pages by making page link creation almost intuitively easy and showing whether an intended target page exists or not.
• A wiki is not a carefully-crafted site for casual visitors. Instead, it seeks to involve the visitor in an ongoing process of creation and collaboration that constantly changes the website landscape.

Advantages of Wikis

1. Low cost organization – free software and hosting
2. Enables mass collaboration – ground-up creation process
3. Open participation  sense of ownership and control over product
4. Easy to make and correct mistakes
5. Converges solutions: thesisantithesissynthesis
6. Solves collective action problems by reducing costs and raising benefits of participation
7. Integrates ideas across many levels and issues
8. Promotes public goods public commons
9. Many-to-many network
10. Favors populism over elitism by offsetting organizational power, money and fame.
11. Power and control resides with the users, i.e., not the elites
12. Saves history of changes, reversible

Why a Policy Wiki?

A wiki enables mass collaboration among peers by reducing the costs of collaboration and providing the necessary incentives for participation. This changes the costs and benefits of collaboration and facilitates collective action. This is especially significant for the provision of public goods.

Government policy design, implementation and adaptation is a public good that requires mass collaboration of citizens, experts, NGOs, government agencies and those involved in the political process. This collaboration is costly, requiring subsidies from a variety of sources including philanthropic foundations, research and educational institutions, ideological or partisan organizations such as unions, political parties or business organizations, politicians, and bureaucrats. The process is top-down and suffers many disadvantages of institutional dynamics and citizen participation is minimal.

A policy wiki would be bottom-up, virtually cost-free and administered with a minimal of effort. It would be a network of dynamic intelligence integrating many different issues and levels of analysis. A policy wiki would not be a forum for partisan propagandizing. It could engage the public in a much more direct and effective way than traditional methods of political activism or writing to one’s congressperson. The wiki would link local, state and federal levels of policy so that users could quickly locate the specific issue they wish to address. In a sense, the wiki would be like a local to national community bulletin board that constantly informs and adapts. Most critical, a policy wiki would be productive by converging on solutions rather than splintering policy debates into various opposing camps.

Two Parties Too Few?

In a Townhall article today Michael Medved argues against the idea of third parties in our two-party political system (article here).

The reader comments posted reveal how little people understand the two-party system and what trade-offs are involved compared to other possible party systems. First, there is the context. The USA is not Switzerland, France or Germany. National scale matters. The USA is also much more ethnically and culturally diverse than nations like Germany, France, Japan, India or China. The plurality of interests that make up American society means possible political differences are significantly greater and this has implications for how the political system arrives at single national policy outcomes. (Think about the red-blue divide we currently experience? Would a multi-colored rainbow divide be better?)

A second major assumption is what we wish a party system to do - what is the main objective. Public comments suggest that the primary objective is to validate individual voters preferences and reward their participation. But is this really the primary objective? Is not the primary objective to govern a free democratic society in a way that strengthens the nation to insure those constitutional principles under which we choose to live together? Our elections are not about making voters happy - it's about finding candidates that can represent and lead the entire nation, from sea to shining sea.

The third caveat is how the party system is a function of the electoral rules. We have a two-party system by design with an electoral system based on single-district, simple plurality, winner-take-all rules. This means the election strategy is to capture the middle where one can win a simple plurality of the votes. This election strategy forces candidates and parties towards the center where only two parties can be left or right of each other. A third party would have to frog leap over one of the two parties to capture the necessary votes to prevail. To make third parties viable over the long term would require changing the electoral rules.

Under these three conditions, a two-party system provides the best of all worlds or, if one chooses the glass half-empty attitude, the least of all evils among alternative electoral systems. A two-party system forces a highly diverse polity to converge on an acceptable policy. It forces the extremes to move toward compromise at the center. In contrast, a multi-party system allows interests to diverge into many small camps where compromise becomes impossible. Then the government must be stitched together from many small coalitions that have failed to compromise over policy. It's a recipe for chronic instability, as one can witness in countries with proportional representation and multi-party coalition governments.

So, what does this mean for third parties? The system is not static, and one of the two major parties can fail and be replaced by a third party. If Obama loses in November one might expect the Democratic party to split and dissolve, where the centrist Democrats would repudiate the left wing radicals and vice versa. If the Republicans lose one could expect the same split unless the party recalibrates its ideological identity. But absent an abject failure and collapse of one of the two parties, there is no real need or benefit from a third-party movement in American politics. There are many other ways to fiddle with electoral processes to introduce more choice, like instant run-offs or weighted preferences, but the point is still to converge on one candidate with a national mandate. The party system doesn't exist for the happiness and satisfaction of we voters. Time we got over it.

No We Can't?

Dan Henninger cites the example of the WTC 9/11 site in NYC to show how dysfunctional our self-validating democratic politics has become (article here). He writes:

Given a choice between unity and politics, we chose the indulgent pleasures of politics.

...as a case study of system malfunction, the Port Authority report on unbuilt Ground Zero is a warning shot to our acrimonious national politics. A can-do tradition is losing ground to can't-possibly-do. Barack Obama's appeal rests heavily on the belief that he'll bring back can-do. He's one man. The answer lies deeper, with a people who have to choose between politics that moves its system forward or a politics that just wants to have fun.

This is what this blog and the companion website are all about. The self-indulgence of our politics has cost us dearly and will continue to until we, the voters, choose otherwise. There are two additional resources I recommend to readers to delve past the nonsense about how we vote: Bill Bishop's book The Big Sort and this new website I've added to my blogroll: New Geography.
Drop your biases and explore...