Would Redistricting Really Cost the Democrats Seats in CA?

Today’s Washington Post notes that both of the major initiatives to reorganize the way in which Congressional districts are drawn up, one in Ohio and one in California, look headed for defeat. In both states it seems likely that partisan gerrymandering will remain the process by which Congressional boundaries are drawn in those states. A lot of Democrats will be happy about the result, at least in CA. But not me.

The argument made by many Democrats is that the CA redistricting proposal would almost certainly result in fewer Democratic districts, and should thus be opposed. Kevin Drum recently noted one such analysis (in this case by Brad Plumer) of why Democrats would likely lose seats in the proposed redistricting.

But even though Plumer’s reasoning makes theoretical sense in the abstract, I’m not convinced that his conclusion that the Democrats will lose seats is correct, given the reality of current district boundaries in California.

Consider the following table. It shows the number of Democratic and Republican seats in Ohio and California, by degree of competitiveness in 2004.

Ohio clearly shows the telltale signs of partisan gerrymandering. Huge numbers of Democratic voters are packed into a few extremely safe Dem districts (winning margins of 40+ points), while Republican voters are spread around into a large number of districts that have what I will call the “ideal winning margin”, at least from a Republican gerrymander point of view: winning margins in the neighborhood of 10 to 30 points, which leave the seat reasonably reliable, but without wasted partisan votes. That’s why a state that voted roughly 50-50 in the last election sent a delegation of 12R, 6D to Congress.

Now look at California. Interestingly, the effects of the gerrymander in CA look rather similar to OH. The Republicans actually have more districts with the “ideal winning margin” than Democrats. And Democrats have a huge number of seats that are overwhelmingly Democratic – which essentially means that they contain lots of wasted Democratic votes.

This suggests two things to me. First, the gerrymander of 2001 in CA was not a particularly good one from the Democratic party’s point of view. If the Democrats had done a really successful gerrymander, Democratic voters would be spread among more districts that they could win by 10 to 30 points, not concentrated into a number of super-Democratic districts.

But secondly, it suggests to me that non-partisan redistricting probably won’t result in a large net loss in Democratic seats, and might even result in a gain. Why not? Because the redistricting would only result in fewer Democratic seats if Democratic voters in CA end up more intensely concentrated they already are. But given the fact that the 2001 gerrymander has already packed many Democratic voters in CA into a number of super-Democratic seats (7 of those Dem seats were won by more than 60 points in 2004, by the way), it seems highly unlikely that any new scheme could make that problem even worse. If anything, it seems much more likely that non-partisan redistricting will spread Democratic voters out a bit more than they currently are. In other words, there’s pretty much only one way to go from here.

The question of whether or not the Democrats gain or lose seats in CA is actually of secondary importance to me, however. There are two other important reasons that I would vote for the proposal, if I lived in CA.

First, there’s the general principle of the thing. This is what led the liberal advocacy group CalPirg, as well as several leading California newspapers, to endorse the proposition.

Secondly, this seems like a wasted chance for the Democratic party to take a positive stand on principle, as Kos has pointed out. By opposing non-partisan redistricting in CA while supporting it in OH, the Democratic party has missed an easy opportunity to distinguish itself from the Republican party by adhering to a general principle – in this case, that gerrymandered districts distort the democratic process – instead of simply calculating the chances of narrow partisan gain or loss in each state.

Given these latter points, I think that Democrats should support the measure even at the cost of a couple of seats in CA. But it’s not at all clear that that would happen. It would be particularly sad and ironic if Democrats vote against the redistricting proposal because they think it will cost the Democratic party some Congressional seats, when it may well do the opposite.


UPDATE: For more data that shows why defining Congressional districts by county or city boundaries would probably disperse, not concentrate, Democratic voters relative to where they are today, see California Redistricting, Part II.

For PGL’s reply to my argument, see No on Proposition 77.