Polarization Growing

In lecterns in New Hampshire, town hall meetings in Iowa, and television studios inside the Beltway, candidates, pundits, and just about everyone else bemoan the growing polarization of the American electorate. As the political tectonic plates slowly shift, the seismic shocks cause the fault lines between red and blue to rent themselves even farther apart, while swing voters scramble so as not to fall into the widening gap. Surely, it can’t be as bad as all that!

For once, the chattering classes may be on to something, at least to a point. Earthquake metaphors aside, elections are growing increasingly lopsided in congressional districts, and states generally have some significant partisan tilt as well. New data from FairVote shows how this tilt is growing decidedly more pronounced in both states and counties, with a direct link to a decline in competitive congressional elections and battleground states in presidential elections.

FairVote has found that residential battlegrounds have dropped sharply in the past generation – from 24 states representing 345 electoral votes in 1976 to 21 states representing 272 electoral votes in 1988 to just 13 states representing 159 electoral votes in 2004. No one is redrawing state lines; rather, we have completed a century-long partisan realignment between the major parties and, at the federal levels, those new patterns keep hardening.

The new data from FairVote suggests that there is much more at work here than who is drawing lines and to what end. While congressional districts are drawn with particular intentions in mind, the borders of counties are fixed. Just like state borders in the presidential race, they are not subject to the whims of a gerrymander, independent or otherwise. One might expect that within a static geographical area, partisanship would remain relatively steady, as the lines are not shifting from one election to the next – or, if it changed, to shift in one direction.

What we have found, however, is the exact opposite. Even within the unmoving boundaries of California counties (a state that perennially contemplates independent redistricting), competitiveness has dropped significantly over the past 16 years. In the 1992 presidential election, for example, fully 19 counties were won by 5% or less. By the 2004 election, that number had plummeted to 3. What’s more, nearly half of Democratic-leaning counties in 1992 became at least 3% more Democratic by 2004, while nearly nine in ten of Republican counties became at least 3% more Republican.

No party hacks connived in smoke-filled rooms to redraw county borders, and yet county partisanship has risen dramatically all on its own. California’s counties are only mirroring the downward trend of competition in congressional districts and states. Any way you slice a piece of geography, that area will tend to grow in partisanship without anyone’s help, which will also tend to result in less competitive elections and foregone conclusions realized over and over again. This is the very difficulty faced by non-swing states in presidential elections that, because of winner-take-all contests for a slate of electoral votes, are so easily ignored, as there is no mystery as to which way they will vote.

What this means is that there is no easy fix – and quite possibly no fix at all – within winner-take-all district elections. If we truly want competitive choice, elections where the participation of all voters matters, we need to establish a national popular vote for president and some methods of proportional voting for legislative elections. The punditry can point to redistricting all they want; of the many values of independent redistricting, having a substantial impact on competitive elections and the number of representatives willing to cooperate with the opposition party are not among them.