The Effect of Redistricting on House Elections, Revisited

by Eric McGhee on October 15, 2012 · 12 comments

in Blogs

The Cook Political Report has come out with its “PVI” numbers of each of the new districts, which are measures of district partisanship based on the presidential vote.  They use this measure to argue for an important redistricting effect this cycle in favor of Republicans.  David Wasserman has the details:

Unfortunately for Democrats, this year’s index tells a dire story of what can happen when a party suffers an ugly election cycle right before redistricting. Democrats…lost…more than 680 state legislative seats – empowering Republicans to draw ten-year maps in four times as many districts as Democrats. As a result, thanks to effective GOP cartography, the number of “strong” Republican seats has jumped from 182 to 190 and the number of “strong” Democratic seats has fallen from 150 to 146. Meanwhile, the number of “swing” seats has fallen below 100 for the first time, from 104 to 99. (emphasis in original)

The Cook report is a venerable organization whose analysis is rightly respected, and I have no doubt that their presidential vote numbers by district are the most accurate available.  But the notion of big redistricting effects this year is Zombie Politics, and the poor corpse should be put to rest.

First, contrary to the quote above, a party that aims to gain seats through redistricting wants fewer “strong” seats, not more.  Since one only needs 50%+1 votes to win (in a two-party race), every vote beyond that threshold is “wasted” in the sense that it could be used to bolster the party’s chances in some other district.  More strong seats means bigger winning margins, which means more wasted votes.

Second, the idea that the 2010 elections dramatically increased Republican opportunities to gerrymander presumes they can actually control the process in the states they won.  This is not necessarily the case.  Here’s what we said on this subject a couple weeks ago:

[A] redistricting party doesn’t have free rein to do as it pleases.  Districts must be equal in population (the allowed deviation between districts is vanishingly small), the Voting Rights Act constrains activity in many states, and some states also have specific rules that limit the options.  (Along these lines, Florida voters recently imposed boundaries on what the legislature could draw, and Texas’s lines ended up being drawn by the courts.)  Moreover, a party’s desire for more districts can often conflict with its own incumbents’ desire for safe reelection.  And a redistricting party must work with the state it has, not the one it wants.  Republicans and Democrats may not live in close enough quarters to permit the sort of seat-maximizing gerrymander a party would otherwise want.

Finally, counting districts as Democratic or Republican using only the presidential vote assumes everyone who votes for Obama will vote Democratic for Congress, and ditto for Romney and the Republicans.  But even in today’s polarized environment, incumbency matters.  A moderately Democratic PVI with a Republican incumbent is probably a Republican win.  What happens when we pair redistricting with incumbency?  A close look at the Cook Report’s 25 incumbents most hurt by redistricting reveals that almost half of them (11) are Republicans.  And the Cook predictions for those races split evenly between the parties at 9 a piece, with the rest as toss-ups.

That’s why we feel the best approach incorporates both district partisanship and incumbency, and uses past election results as a guide for how important each is likely to be.  When we ran those numbers, we found redistricting was a wash.  We also found in the same analysis that Democrats would get a much more substantial seat gain if there were no incumbents running and all seats were open.  The structural advantage for Republicans this year stems from incumbency, not redistricting.

 

{ 12 comments }

Stephen October 15, 2012 at 3:54 am

Nonsense. The idea that we should see whether redistricting benefited Republicans by looking whether they GAINED seats is terrible – they’ve just maxed our their gains in 2010 (the most seats they’ve had in 70 years). The PVI of districts a far better indicator.

First the Congressional races are highly correlated with the Presidential vote (around 80%). So PVI a reasonably good metric. Incumbency will be a big part of that 20%.

Second, no you don’t want a 50%+1 district because: it is vulnerable to tides e.g a seat Obama got 50%+1 is one he’s set to lose. You want to maximize not just representatives but conservative/liberal representatives respectively AND not have to spend money on races so you can divert them to competitive ones. PVI of 5-10 is ideal aka likely/solid territory.
Sure, you don’t want seats that are too strong but you want them to lean such that they still favour your party even when your opponents win the Presidential race. Too many close seats means too much risk (pay a lot of money and risk control in a bad Presidential year).

Third, whilst the restrictions you mention make it harder to gerrymander, it doesn’t prevent it. The fact that the median seat is R+3 is indicative. Republicans have maximized their seats so their version of seat-maximizing gerrymandering is merely to maintain their seats Florida is a great example: 2R-1D Congressional delegation in a swing state seems to (even accounting for demographic distributions) been an exceedingly good result for the GOP. Did they gain seats? No, but they’ve made it much harder to lose their seats by making each seat they hold more Republican (which PVI captures).

The fact that in an election that President Obama is at worst about tied and and best a couple of points ahead and yet the GOP is slated to stay pretty close the largest House margin they’ve had in 70 years is indicative of more than incumbency.

Eric McGhee October 15, 2012 at 12:45 pm

Stephen:

The median district was +2R in 2008 and +3R in 2004. It’s +3R now. Where’s the big redistricting change?

As for the rest of your observations, I would agree with all of them to a point. But why rely on PVI alone when you could include other important factors like incumbency, too? When you run that more complete set of numbers, you get no redistricting effect. Even Wasserman at Cook Report seems to agree–he just doesn’t emphasize that part of his analysis.

Eric

Stephen October 15, 2012 at 8:32 pm

Thanks for replying.

I’m not saying ignore incumbency (indeed, you need it to make sense of most of that 20% deviation from the Presidential results + the effect of relatively unpopular incumbents e.g. Bachmann, Cicilline). Although the Republican incumbents hurt are in the minority states where they didn’t control redistricting – CA, MD, IL. But the crux is that if you offered to put an incumbent of my party in every seat (20%) or offered my a map that maximized my number of strong seats (80%), I’d take the latter.

My main point is that looking at seats gained (and concluding that it’s a wash) isn’t a good metric. The median district is part of it. When you’ve got such a large House advantage, you’ll find it difficult to squeeze more seats out of a map. But Republicans have increased the PVI of seats they already hold. **Overall, they won’t have gained seats, but their incumbents have been made harder to unseat (above and beyond the benefits of incumbency) AND, should they retire or be beaten, they’ve made it easier to hold open seats and unseat Democrats who capture said seats.** This is the big Republican Redistricting Effect.

A better metric is to see who’s maximized their number of strong seats whilst preserving swing seats. In big maps, the Republicans shifted several marginal seats in their direction – yes, you’ve reduced your margin in stronger seats but those seats still remain strong (going from R+15 to R+10 to get a R+0 to R+5 is more efficient than three R+3 districts). On balance, the big states had this effect in FL, MI, NJ, NC, PA, VA,WA, WI vs CA, IL for the Democrats.

Consider this:
Number-(Percentage) of Democrats in seats by 08-PVI: total for 111th/112th (total seats) [Change]
R+10+: 13-(15%) / 2-(.02%) (88) [VA9,MD1,AL2,AL5,LA3,TN6,TN4,MS1,MS4,ID1,TX17][+D UT2,OK2]
R+10: 3-(25%) / 0-(0%) (12) [GA8,LA6,ND]
R+9: 4-(23.5%) / 1-(6%) (17) [SD,WV1,KS2]
R+8: 3-(37.5%) / 0-(0%)(8) [PA10,AR1,IN8]
R+7: 3-(23%) / 1-(8%) (13) [SC5,OH18]
R+6: 10-(38%) / 5-(19%) (26) [NM2,TN8,IN9,AZ1,CO4,FL2][+D NY26]
R+5: 8-(57%) / 2-(14%) (14) [AZ5,AR2,CO3,NY29,VA2,VA5]
R+4: 5-(50%) / 1-(10%) (10) [NY13,OH16,FL24,TX23]
R+3: 4-(36%) / 0-(0%) (11) [KS3,MI1,NY19,PA3]
R+2: 10-(71%) / 2-(14%) (14) [FL8,MI7,NY20,24,NC2,OH6,TX27,WI8]
R+1: 8-(80%) / 4-(40%) (10) [IL8,11,14,NJ3]
Even: 4-(44%) / 2-(22%) (9) [NH1,WA3]
D+1: 7-(78%) / 4-(44%) (9) [FL22,OH1,15]
D+2: 7-(88%) / 5-(63%) (8) [NV3,PA8]
D+3: 11-(92%) / 5-(42%) (12) [IL17,MN8,NH2,NY25,PA7,WI7]
D+4: 7-(88%) / 6-(75%) (8) [PA11]
D+5: 11-(100%) / 10-(91%) (11) [NY9]
D+6: 7-(88%) / 7-(88%) (8)
* D+7: 6-(86%) / 7-(100%) (7) [+D DE]
D+8: 10-(100%) / 10-(100%) (10)
D+9: 8-(100%) / 8-(100%) (8)
D+10: 5-(100%) / 5-(100%) (5)
D+10+: (100%) (117) [.02% 08 Exceptions: LA2 - Scandal, HI1 - 3-way spoiler effect]

There is two big categories of Republican gains there: R-leaning seats that are completing the realignment of Southern Whites (which don’t need improvement) and a big chunk of swing seats – most of which have been improved, and with each improvement have increased the odds of that seat electing a Republican (whether there is a D/R incumbent or open seat); just look at the correlations. It’s not the seat gains that matter but how seats have had their PVI increase e.g. OH: 5 Republican seats gained, I believe 4 (accounting for apportionment) have exchanged residents to maximize strong seats by making strong+ R seats strong, and marginal R seats strong [**]. You’re not going to capture that by looking at net-seat gains.

Eric McGhee October 16, 2012 at 12:14 pm

These are all good and valid points, but they’re already incorporated in our model. If Republicans were effective protecting the gains they had made (rather than expanding their gains), it would still show up in our numbers. See here (http://themonkeycage.org/blog/2012/09/19/is-the-2011-redistricting-hurting-democrats-no-but-republican-incumbents-are/) for details.

The fact redistricting was a wash nationally does not mean it was not successful in certain states. (We have a post on that, too.) It just means that Republican control of state legislatures wasn’t enough to move the needle.

Art October 15, 2012 at 9:49 am

The median Congressional district for the 2012 House election has a PVI of R + 3, meaning that the 218th most Republican district is 3% more than Republican than the country as a whole in Presidential voting. An unbiased set of national districts would have the 218th most Republican district with a PVI of even, so ithe actual districts clearly favor Republicans. In fact, 221 of the 435 districts have a PVI of R+3 or more, and 232 districts (53.3%) have a PVI of R+2 or more. So, on balance, the recent round of redistricting gave Republicans an advantage for the next 10 years in House elections; the Democrats would have to obtain approximately 53% of the two-party House vote to have better than a 50% chance of winning control of the House.

Dan October 15, 2012 at 11:22 am

No – there’s a difference between a districting system favoring one party and a system that favors one party due to partisan gerrymandering. You need more evidence to show that such a GOP advantage was caused by redistricting. In fact, territorial districts in general favor the GOP due to residency patterns of the two parties. A districting plan devoid all nefarious party manipulations, if drawn only to consider geographic compactness and equal population, would still be biased toward the GOP. Majority-dem areas tend to be much more Democratic than majority-GOP areas tend to be Republican.

Craig October 16, 2012 at 11:16 pm

You need 51.5% of the two-party vote to beat a 3% structural disadvantage, not 53%. 51.5 – 48.5 = 3

Also, R+3 is not exactly historically huge, given that it was previously . . . R+2. As alluded to n the post, gerrymandering is not the issue, it’s the fact that Democratic voters are more likely to live in urban districts and thus be “packed.”

Todd October 15, 2012 at 12:55 pm

“a party that aims to gain seats through redistricting wants fewer “strong” seats, not more. Since one only needs 50%+1 votes to win (in a two-party race), every vote beyond that threshold is “wasted” in the sense that it could be used to bolster the party’s chances in some other district.”

It’s always a balance to create safe seats while not hurting others, but the parties have gotten very, very good at it over the last few years. The way the above argument is framed seems absurdly naive. I’ve lived through two redistrictings in North Carolina, one Dem and one Rep, and the idea that what I was watching was the creation of “fewer ‘strong’ seats” is insane. Creating as many strong seats as possible – and punishing particularly troublesome members of the other party while you’re at it – is clearly what drove both processes.

Eric McGhee October 15, 2012 at 1:42 pm

It may seem naive, but it’s a simple mathematical necessity if one wants to gain seats through redistricting. That doesn’t mean that every seat a gerrymandering party draws for itself is 50%+1, but they definitely want to bring down the margins in their own seats.

Take your own state of NC, which is one of the few to show significant redistricting effects. Every Republican incumbent has a more competitive district this time around. Every Dem has either a safer district or a district that now favors Republicans. That’s how it’s done.

Bill October 15, 2012 at 6:16 pm

Second, the idea that the 2010 elections dramatically increased Republican opportunities to gerrymander presumes they can actually control the process in the states they won.

The post would be more interesting if you picked up on this point more. If it were true that the states where Rs picked up state legislature seats were overwhelmingly states where the state legislature is roughly irrelevant to re-districting, then that would be a pretty interesting fact. Just waving around the fact that, in some states, the legislature doesn’t affect redistricting much is pretty unconvincing. The measure you really want is something like what was the net percentage pickup of the things-which-decide-redistricting (these things are presumably often control of state legislature houses, governorships where the governor has a veto over redistricting, and etc)

Without this fact (the fact, if it is a fact, that R pickups were all in legislature-irrelevant states), the rest of the analysis is unconvincing. The Rs have consultants with better GIS skills than you have, right (or at least as good) and better polling information? So, it is not at all plausible that they did not cash out their greater control of redistricting in some way which they think benefits them. Maybe median PVI doesn’t capture whatever it is that they did, but that does not establish that they didn’t succeed in whatever it is that they wanted to do.

Is the crude, first-order strategy “Make the other guys’ seats super-safe and ours only kinda safe” really all there is to redistricting? There is nothing more subtle than that? It seems kind of 19th century. Plus boring.

Eric McGhee October 15, 2012 at 6:50 pm

What notion of gerrymandering did you have in mind, besides “redistricting is going to make one side win more seats than it would have otherwise?”

And if you think we don’t have enough information to recognize gerrymanders where they occur (certainly a possibility), then why not pile on Cook Report, too? We bring more info to bear on the question than they do.

Total October 17, 2012 at 10:26 am

Is the crude, first-order strategy “Make the other guys’ seats super-safe and ours only kinda safe” really all there is to redistricting? There is nothing more subtle than that? It seems kind of 19th century. Plus boring.

You find the attempt to win elections efficiently boring?

Comments on this entry are closed.

Previous post:

Next post: