Mark Gersh is president of NCEC Services Inc., and a CBS News Consultant. Redistricting Journal is a regular Hotsheet series on redistricting battles across the country.
Congressional redistricting has moved at a slow pace this year. We are more than half way through the calendar year, and far less than half of the states have completed the process, or are even on the verge of enacting a plan. So far, it looks like overall a draw between the two parties, but there's a long way to go.
Fourteen states with two or more congressional districts are expected to yield final maps soon (43 of the 50 states are required to pass maps, while seven states were allocated one at-large district). In addition, legal challenges are a strong possibility in , , and Wisconsin, and can't be ruled out in other states. The North Carolina map, which is , must await Justice Department or judicial clearance, even after the state legislature finalizes the plan.
In order to evaluate what has materialized to date, I have reviewed 21 states (including the 7 one district states). Currently, these 21 states contain 121 districts (43 Democratic and 78 Republican). Collectively, the 21 states will lose five districts as a result of the 2010 census: Illinois, Iowa, Louisiana, Michigan and Missouri each lose one.
Examining the outcome probabilities - and it is early in the cycle so recruiting has slowly evolved on both sides - Democrats so far would lose 2.6 seats and Republicans 2.4. Superficially, the two parties have fought to a draw, with 319 districts yet to be drawn. Cumulatively, only 27 percent of the districts have been constructed, or are close to completion.
This initial analysis is somewhat illusory. Democrats need to win 24 seats in order to regain the majority. So gains, not the status quo, must be garnered by Democrats. Secondly, the Republicans have asserted their redistricting strategy is to protect the 63 seats gained last year, as much as winning more Democratic held seats.
In order to truly assess the viability of that strategy, one must estimate the number of marginal districts that have been created to date. We define marginality as the ability of either party to win a district. Analyzing the data of past elections such as the 2008 presidential contest is the focal point of the marginal designation. Demographic changes are also a factor, but not as important tin 2012, as they will become in future years.
In the 21 states that have been included in this analysis, 32 districts are safe or almost safe for Democrats (in Arkansas 4, Mike Ross, D-Ark., has also decided to retire and that opens a potentially hot race). Two other races: Iowa 3 (with two incumbents combined in one district) and Oklahoma 2 (Boren is retiring, and the seat is jeopardy) are toss-ups.
Democrats are likely to lose the reconstructed Indiana 2nd district, where Rep. Joe Donnelly is running for the U.S. Senate. If Democrats win one or two of the toss-ups, Democrats would claim a total of 34 to 35 districts.
The story becomes more intriguing in Illinois. Redistricting has , and the Democrats start as outright favorites in three districts. If Democrats win four seats in Illinois, they would be at 38 to 39 seats, with eight other currently held Republican districts rated as marginal.
Who wins those races? It's impossible to tell at this juncture. Moreover, Democrats might face a fight in a few districts we have accorded to them -- in Arkansas, Iowa and Oregon.
In a wave election, for example 2006 or 2010, one party might win 80 percent of the marginal districts. In a non-wave election, and the initial generic congressional polls lean in that direction, it appears as if neither party has forged an advantage during the incipient stages of redistricting. That is my bottom line assessment to date.
Overall, Illinois and North Carolina redistricting may offset each other, and we await the next big dominos to fall: California is out and Texas is awaiting legal review and approval... stay tuned, we'll have a look at those soon.
More of Mark Gersh's redistricting coverage in Hotsheet: