How Big a Swing Would House Republicans Need in 2010?
Andrew Gelman tries to look ahead to the 2010 midterms by first looking back at the Democrats’ share of the House vote over time:
From this picture, it looks possible but unlikely that there will be a 6% swing toward the Republicans (which is what it would take for them to bring their average district vote from 44% to 50%). Historically speaking, a 6% swing is a lot. The biggest shifts in the past few decades appear to be 1946-48, 1956-58, and 1972-74 (in favor of the Democrats) and 1964-66 and 1992-194 (for the Republicans). I don’t know if any of these would quite be enough to swing the House majority. A more likely outcome, if the Republicans indeed improve in next year’s election, is for them to make some gains but still be in the minority.
It is worth noting, as you can see from the chart, that in recent years there have been several occasions in the late-1990s in which Democrats won a narrow majority of House votes but the GOP controlled the majority of House seats. The drawing of district lines thus seems to favor Republicans. Since the lines will be redone after 2010, the electoral landscape may come to be more favorable for the Democrats once the census/redistricting process is done.
16 Responses to “How Big a Swing Would House Republicans Need in 2010?”
Register
RSSLogin
Topic Cloud
ArchivesBlog RollAbout Matt YglesiasContact Me
Donate
July 30th, 2009 at 2:29 pm
“as you can see from the chart”
Sadly, no chart to see. Nice tagging, Matt.
July 30th, 2009 at 2:38 pm
It is worth noting, as you can see from the chart, that in recent years there have been several occasions in the late-1990s in which Democrats won a narrow majority of House votes but the GOP controlled the majority of House seats. The drawing of district lines thus seems to favor Republicans.
Besides there being no chart to see, the is an incredibly unsophisticated analysis. Let’s just take 2008, for example. Democrats won 54.6% of the 2-party vote, but they won 59.1% of the House seats. Using the 2008 data point, the drawing of district lines seems to favor Democrats, not Republicans.
I don’t think you can generalize one way or another based on the data presented in the chart.
July 30th, 2009 at 2:49 pm
Also, Andrew Gelman has his data wrong. The Republicans won the House popular vote in 1998, for example.
July 30th, 2009 at 2:53 pm
I see the chart.
July 30th, 2009 at 3:05 pm
Al, that’s because districts aren’t dartboards, and they’re winner-take-all. It is more likely that a district votes at the national average than it is to vote 50% more Republican or Democratic than the national average. More than 10% of all districts are going to be within 5 points of the national average. So using D+5 to mean “Democrats get 5% more votes than the national average in this district”, then, if 40% of all districts are D+5 or greater, 40% are R+5 or greater, and the other 20% are between R+5 and D+5, then the Democrats would win 60% of the seats when they win 55% of the two-party vote.
Other example: Ronald Reagan, 1984. Reagan won 59.2% of the two-party vote (58.8% overall), and 49 out of 51 states+DC. A seemingly-small nationwide advantage (say 5-10%) can translate into winning a LOT of 50%+1 situations. The Democrats in the House won the popular vote, but lost the House equivalent to the Electoral College. That means the system was slightly biased against them (Republican state legislators wanted to create lots of narrowly-Republican districts, and Democrats wanted to do the same. But the Republicans had more control over the redistricting process in 2000 than the Democrats. Like how Bush was helped in 2000 by having lots of small safe GOP states, but that did nothing for McCain when he lost by a solid margin).
July 30th, 2009 at 3:06 pm
First, the 1990s elections were before the last round of redistricting.
Second, most of the redistricting following the 2000 Census is getting pretty old at this point. My understanding is that it appears through population shifts the Republicans are currently vulnerable to losing a disproportionate number of seats if the Democrats have the overall majority vote.
July 30th, 2009 at 3:16 pm
The Republicans are going to loose seats. Bad as the Democrats are, everyone knows the Republicans are worse.
The interesting question is not whether but when the Chinese pull the plug on their financing the deficit. Hopefully – and I’m serious – they will do so as a deliberate attack on our fisc. The alternative scenario – much worse – would be that internal pressures force the Chinese willy nilly to cut their financing us.
That is going to render moot much of what passes for politics in Washington.
July 30th, 2009 at 3:27 pm
There have been several occasions in the late-1990s in which Democrats won a narrow majority of House votes but the GOP controlled the majority of House seats. The drawing of district lines thus seems to favor Republicans.
I think this is mistaken. The important factor here isn’t that district lines favor Republicans; it’s the power of incumbency. The Dems won majorities in the late 90s but didn’t win a majority of House seats because both parties had a lot of safe seats. Presumably, the Dems racked up big majorities in their safe districts but fell slightly short (45-49%) in a lot of swing districts. The power of incumbency makes it easier for an incumbent who was barely elected in a swing district to get re-elected. In 1996 and 1998 and 2000, that benefited Republicans–their incumbents held on in swing districts. Looking at 2010, this should benefit Democrats. An incumbent like Tom Perriello, who barely got elected in a swing district, now has all the advantages of incumbency: name recognition, a chance to prove his competence, Congressional offices that provide services to constituents, a less-formidable opponent (last time he was challenging an incumbent) and of course fundraising resources. If the Democrats win 50% nationally, Perriello stands a good chance of being re-elected. And the Democrats have a lot of incumbents like him. Given that most of the swing districts are currently held by moderate Dems, the Republicans probably need a sizable national majority (say, 53%) to retake the House.
That said, the District lines do somewhat favor Republicans. This is partly because of the Voting Rights Act and majority-minority districts (http://www.fivethirtyeight.com/2009/06/elections-without-section-5.html). The Democrats have packed their supporters into relatively few extremely blue districts, whereas the Republicans have spread their supporters more evenly across districts. The most Democratic district has a PVI of something like D+43, while the most Republican district has a PVI of about R+27. That means there are more districts on the R+ side of the PVI scale than on the D+ side. Both parties have a bunch of completely safe districts, but the Republicans have more districts where they have a significant but not overwhelming advantage (say, +2 to R+8 on either side of the PVI scale). Hopefully redistricting will change this, but don’t count on it because the Democrats haven’t shown any interest in either (a) ending the “majority-minority” provisions of the Voting Rights Act or (b) generally drawing districts in a way that would make incumbents slightly less safe while making the party stronger overall.
July 30th, 2009 at 4:16 pm
Since the lines will be redone after 2010, the electoral landscape may come to be more favorable for the Democrats once the census/redistricting process is done.
===========================================================
Perhaps, but you have to factor in the fact that with populations shifts, about 8 seats are projected to shift from mostly blue states to red ones. Texas will gain three seats.
http://www.ballot-access.org/2009/04/14/non-partisan-group-estimates-number-of-electoral-votes-in-each-state-in-2012/
After the 2010 census has been held, the number of seats held by each state in the U.S. House of Representatives will change. The National Conference of State Legislatures recently estimated what the 2010 reapportionment will mean for each state.
Eight states are expected to lose one seat each, in the U.S. House and in the Electoral College. They are Iowa, Louisiana, Massachusetts, Michigan, New Jersey, New York, Ohio, and Pennsylvania.
States that will gain will be Texas (3 seats), and one each for Arizona, Florida, Georgia, Nevada and Utah.
July 30th, 2009 at 4:25 pm
States that will gain will be Texas (3 seats), and one each for Arizona, Florida, Georgia, Nevada and Utah.
Those have all been Red States lately, but the trend in a lot of them is blueward. Florida and Nevada went for Obama. Arizona and Texas are seeing a lot more Democratic-trending Hispanic voters.
And likewise, while it may be blue states that are losing representation, I wouldn’t put it past the state governments to gerrymander things so that it’s a Republican seat that goes away.
I’m not necessarily saying that’s what will happen, just that it could happen. And long-term, extra seats in places like Texas and Arizona may go Blue.
July 30th, 2009 at 4:41 pm
To simplify Adam’s point even further, states don’t make voters red or blue, voters make states red or blue.
July 30th, 2009 at 4:46 pm
Perhaps, but you have to factor in the fact that with populations shifts, about 8 seats are projected to shift from mostly blue states to red ones. Texas will gain three seats.
This is true. But given where the population gains in Texas have been, it’s pretty likely that these new districts are going to be in the liberal (by comparison) metro areas. And if you know anything about Tom DeLay’s antics, you’d be aware that Texas has already been gerrymandered to be about as favorable to Republicans as possible. Florida is just about as bad. Georgia will probably be able to carve another Republican district though since much of the population growth is all in Newt’s suburbs, which while getting bluer aren’t there yet.
July 30th, 2009 at 6:54 pm
DTM is right. If a voter from Pennsylvania moves to Florida, and then that gets reflected in the redistricting, it’s really just a wash whether that voter’s a Democrat or a Republican, unless somehow the Floridian environment turns the voter from a Democrat to a Republican.
I suppose it can make a difference if a big block of Democrats worth a Congressional district or two were to get dispersed amongst a number of Red states in such a way that they don’t form a majority in a new district, but I don’t see any structural reason why we’d need to worry about that.
People shifting around from one part of the country to another doesn’t really make much of a difference. Really what you want to look at in terms of large-scale voter shifts are the demographics that bring in new voters (and “retire” older ones). Latinos have relatively high fertility rates, and since the GOP seems determined to stick to its “white resentment” strategy, those new voters should mostly be Democrats. Ditto for immigrants in general; you might say that entrepreneurial-minded, socially-conservative immigrants would be natural Republicans, but the Republicans don’t seem to want them as voters.
I suppose the new seat in Utah will probably go Republican, though, so there is that.
July 31st, 2009 at 5:09 am
Not many Republican seats you can do away with in MA and NY
July 31st, 2009 at 9:11 am
In the late 1990s, Democrats didn’t even bother to contest “safe” Republican seats in many parts of the country. In the state of Florida, with about 6% of House seats in the 1990s, many incumbent Republicans won with 0 votes tallied because of state laws. Ditto the state of Louisiana, where Republicans had a lock on several seats and there was no election in November.
July 31st, 2009 at 11:54 am
The GOP seems to be preparing for any disadvantage stemming from the 2010 census by harping on ACORN. Just how far do you think they’ll push it when the time comes?