The Democrats Have a Concentration Problem
Saturday, July 31, 2010
Will Republicans gain the net 40 seats they need for a majority in the House? Several factors will certainly help.
One of the asymmetries of American politics is the Republican advantage in the House of Representatives. Evidence: in 2004, George W. Bush, winning 51% of the national popular vote, carried 255 of the 435 congressional districts, while John Kerry carried only 180. In 2008, Barack Obama, winning 53% of the national popular vote, carried 242 of the 435 congressional districts, while John McCain carried 193.
To try to get an idea of how many districts Republican and Democratic candidates would carry if they got the same percentage of the national vote, I used this handy website that shows the percentage of vote for the Republican and Democratic candidates in each of the 435 House districts in 2000, 2004, and 2008.
First, I estimated the number of districts Bush would have carried in 2004 if he would have won 53% of the vote, by assuming that his percentage rose 2% in every district and Kerry’s percentage fell by 2%. Result: Bush would have carried 15 more districts than he actually did. Second, I estimated the number of districts Obama would have carried in 2008 if he had won 51% of the vote, by assuming that his percentage fell 2% in every district and McCain’s percentage rose by 2%. Result: he would have lost 18 districts that in fact he won.
Using those estimates, I constructed the following matrix, showing the number of congressional districts carried by each president actually and hypothetically:
Republican with 53% of the vote: 270 Democrat with 53% of the vote: 242
Republican with 51% of the vote: 255 Democrat with 51% of the vote: 224
The results are rather striking. Had Bush won the same percentage of the vote that Obama did, he would have carried 270 districts to only 165 for his Democratic opponent. That’s a big, big majority. No party has held as many as 270 seats in the House in the last 30 years. In contrast, if Obama had won the same percentage of the vote that Bush did, he would have carried only 224 districts, not very much more than the 211 his Republican opponent would have carried.
Why do Republicans have what appears to be a structural advantage in House seats?
Look at it another way. With 53% of the popular vote, the Republican carries 270 districts while the Democrat carries only 242. That’s a difference—a really significant difference—of 28 seats. With 51% of the popular vote, the Republican carries 255 districts while the Democrat carries only 224. That’s a difference—and even bigger difference—of 31 seats. And, one must add, that a party that holds 255 House seats has a pretty solid majority in that body, while a party that holds only 224 seats has only a tenuous majority. House Speaker Nancy Pelosi would have had a hard time prevailing on cap-and-trade and healthcare if her party had only 224 House seats. She probably would not have brought those measures forward, at least in the form that passed, with only that level of support.
The question arises: Why do Republicans have what appears to be a structural advantage in House seats? One answer, which might occur immediately to Democrats, is redistricting. In the redistricting cycle following the 2000 census (which involved post-2002 redistricting in Georgia and Texas), Republicans had an advantage. They passed partisan districting plans in Texas, Florida, Pennsylvania, Michigan, and Georgia, among the ten largest states. In contrast, Democrats passed partisan districting plans only in Georgia (which was later overturned by the Republicans) and North Carolina among the ten largest states, and also passed a partisan plan in Maryland which netted them two seats in an eight-seat delegation.
But redistricting can’t explain a structural advantage for Republicans of something like 30 seats. The more powerful explanation is that Democrats have too many seats that are heavily Democratic—or, to put it another way, Democratic voters are heavily concentrated in too few House seats. The reasons are mainly demographic: areas with large black, Hispanic, and (to use Joel Kotkin’s term) gentry liberal populations are more heavily Democratic than any correspondingly sized part of the country is heavily Republican.
Democrats have too many seats that are heavily Democratic—or, to put it another way, Democratic voters are heavily concentrated in too few House seats.
Aggravating this tendency is the prevailing interpretation of the Voting Rights Act, which seems to require maximizing the number of majority-minority districts. To translate this jargon into English, that means maximizing the number of majority-black and/or Hispanic districts. That results in packing Democratic voters into a relatively small number of districts and draining Democratic voters away from adjacent districts. In several Southern states, this has resulted in legislative alliances over redistricting between Republicans and black Democrats, with white Democrats on the outside.
The results of these two factors—of which demography is by far the greater—is that in 2004 John Kerry won 80% or more of the vote in 19 congressional districts, while the number of congressional districts in which George W. Bush won 80% or more was zero. Similarly and even more starkly, in 2008 Barack Obama won 80% or more of the vote in 28 congressional districts, while the number of congressional districts in which John McCain won 80% or more was zero.
The following table shows the number of districts carried by Democratic and Republican presidential candidates by the indicated percentages (rounded off) in 2004 and 2008.
In the current Congress, some 69 districts voted 70% or more for Obama, 67 of which are represented by Democrats and two of which (HI-1, LA-2) are represented by Republicans elected in unusual circumstances. Only 10 districts, all in the South, voted 70% or more for McCain.
In the 2010 elections, we can expect that Democratic percentages will hold up well in heavily black and gentry liberal congressional districts, as they did in similar areas in the Massachusetts special Senate election. Democratic percentages will probably sag somewhat in Hispanic areas, as they did in the very few identifiable Hispanic constituencies (Lawrence, Chelsea) in Massachusetts, which would leave most such districts heavily Democratic but might help Republicans in certain heavily Hispanic districts in Texas and Florida. All of which suggests that Democratic percentages might decline by more than the national average in the 366 districts in which Barack Obama did not win 70% or more of the vote in 2008.
This gives Republicans a structural advantage in addition to the historically unique 45%-41% advantage they currently hold on the generic ballot question—which party’s candidate will you vote for in elections to the House of Representatives? All of which does not guarantee that Republicans will gain the net 40 seats they need for a majority in the House. But it certainly helps.
Democrats have regarded their current 255-178 majority in the House as solid enough to enable them to pass controversial measures like the 2009 stimulus package, the Senate healthcare bill, and the cap-and-trade legislation, which conspicuously lacks a majority in the Senate. But that majority, to use a seismic metaphor familiar to residents of Speaker Nancy Pelosi’s San Francisco district (85% for both John Kerry and Obama), may be built on sand and fill land, capable of being dislodged in a tremor that looks, on the basis of current polling metrics, to be historically significant. With 53% of the popular vote, Obama carried 242 districts; with a hypothetical 51% he would have carried 224; with support at his current job approval rating of 46%, he would carry only about 162, which is to say 28 less than the 180 Kerry carried with 48% of the vote, and 31 less than the 193 which McCain carried with 46% of the vote. At the 46% level of support, a Democrat is carrying far, far less than a majority of House districts.
Historic metrics, as I argued in my July 28 Washington Examiner column, bode ill for Democrats in this year’s House elections. These metrics bode even worse.
Michael Barone is a resident fellow at the American Enterprise Institute.
FURTHER READING: Barone recently detailed the tea party as “More Anti-Democrat than Anti-Incumbent” and examined “What 1946 Can Tell Us About 2010.” He has asked “Why Do Parties Last Longer in Britain,” and noted “The Coastal Conundrum.” Elsewhere, he forecasts that “House Democrats Head for a Thumping at the Polls” and discusses “Rising Speculation about Bombing Iran’s Nukes.”
Image by Rob Green/Bergman Group.