Redistricting 2011: Locking in Advantage

By Fester:

I'm following up on Ron's post about the generic issue environment and political scene being very favorable for Democrats. McClatchey reports that the three non-partisan high prestige CW political predictors are all predicting significant Democratic gains in both the House and the Senate.   That is jibing with what I am thinking too, although I am projecting more in the Senate (net +5 or more for the Dems) than what the experts are willing to go for at this time. 

Chris Bowers at OpenLeft is thinking about the institutions of power and is running on a hopeful and plausible path of speculation for the next couple of years:

If they are further wiped out in the 2008 elections this November, and reduced to a rump party with only 40-43 Senators, 175-180 members of the House, and a 7-8% loss in the Presidential election, we could start witnessing major changes in voting habits, too. Or, even if they don't, we can bury them further in 2010, when they have to defend 21 Senate seats and we only have to defend 12, and when we can draw new maps that will solidify a Democratic majority until at least 2022.

Either way, we are approaching a true, long-term political victory, where we not only realign the public, but where out political opposition comes to accept many of our core policies that they once opposed.

Institutional choice constraints in the form of the decennial redistricting can lock in Democratic power for the next decade against the current Republican Coalition.  But before we look into how this could happen, let us examine the three major goal sets that redistricting battles can accomplish:

1. Maximize Projected Wins:  This goal set seeks to maximize the projected number of seats a given party can win.  The 2004 Texas redistricting led by Tom Delay is an excellent example of the trade-offs that are made to achieve this goal.  The ideal is to concentrate your opponents'  likely voters into a few safe seats and then create as many mildly leaning districts that are favorable to your generic candidates as possible.  Using the Cook Partisan Voter Index as a metric, this goal would see two sets of shifts;  first moderately Republican districts such as an R+7 district would either be packed to become an R+15 districts or it would be reduced to an R+0 or D+1 district.  Secondly, many R+1 or 2 districts and many D+0 to D+3 districts would become D+4 or D+6 districts.    The advantage with this strategy is in most normal years predictable majorities will be re-elected while minimizing and regionalizing the Republicans.  The disadvantage is that in atypical years such as 2006, the slightly lean districts are numerous and VERY vulnerable. 



2. Incumbent Protection:  The goal is for a party or both parties to hold onto their current districts and for a decade long political stalemate to occur.  The procedure is the exact opposite of the Maximization route.  The goal is to create strongly partisan districts so that the only real competition is from the rare primary challenge. The advantage is in a bad year, a party keeps most of their seats.  The disadvantage is in a good year, the party has very few potential pick-ups. 



3. Screw You: All politics are local and there are some long standing intra and inter party personality based feuds in most/all states.  It is not uncommon for boundary lines to be drawn so as to help or hurt a particular individual, often by drawing the naturally reasonably district as ending across the street and down the block from an individual's house. This is a reasonably unpredictable and non-systemic element of redistricting but it is a real factor. 

If the Democrats continue to perform as I and most other analysts are projecting, I think 2011 offers a very unique opportunity to marginalize the current Republican coalition by rendering it an ideologically rump party. This should (thankfully) encourage the GOP to change and restore a reasonable Overton window. At the same time, this opportunity will allow the Democrats to adapt a blended strategy that builds on previous gerrymandering and redistricting that packed and cracked natural Democratic constituencies into super-majority urban districts and competetive/lean GOP suburban districts.

Currently there are 53 districts with a Cook Partisan Voter Index of greater than D+20, including my home district of PA-14 @ D+22 while there are only 10 districts with a PVI of R+20 or greater.  This is a massive pool of Democratic voters which can be spread out a bit and change the dynamics of numerous races while keeping Democratic incumbents very secure.  For instance PA-14 with a D+22 advantage borders PA-4 which is held by Rep. Jason Altmire (D-McCandless) despite it being an R+3 district, and also PA-18 held by Republican Tim Murphy as an R+2 district. 

Adapting a blending strategy, state Democrats could potentially shift portions of Monroeville and the Allegheny River Valley from PA-14 to PA-4 while shifting Fox Chapel and Indiana Townships to PA-14.  Assuming a static shift in population and voting patterns this weakens PA-14 from a D+22 district to a D+18/19 district while making PA-4 a break-even district.  PA-14 is still a very safe seat for the Democrast while PA-4 becomes a little safer.  Another shift would be to add elements of Brookline, Carrick and Overbrook to PA-18 from PA-14 and compensating by shifting a corridor to Bethel Park or Upper St. Clair into PA-14.  Again this is only a shift of a few points which keeps PA-14 safe and makes PA-18 very competitive as it could force Tim Murphy to reside in PA-14 while running for PA-18. 

These types of games can be played in multiple states with particularly large gains plausible in New Jersey, Pennsylvania, New York, Illinois, and California.  The goal would to decrease the number of purley urban and overwhelmingly Democratic districts while building up the number of urban dominated but suburban inclusive districts.