I wrote a post a while back covering this and thought I would write a new one about how voter registration has changed among Oregon's counties in recent history. Below I cover the changes since 2001 and 2012.
In the below map each color represents a 5 point increment. Light blue is where Democrats lead Republicans in voter registration by less than 5 points, the next lightest blue by 5-10 points, then 10-15, 15-20, and then over 20 points. The same scheme applies for the shades of pink and browns for Republicans.
In May 2001 there were 748,187 Democrats and 686,342 Republicans in Oregon, out of a total of 1,902,403, meaning Democrats were 39.3% of voters, and Republicans 36.1%. Democrats had advantages on the coast, in the college town counties of Lane and Benton, in Columbia County, and in Multnomah County, where Democrats led Republicans in registration 49-24%. Republicans led in suburban Clackamas and Washington counties and the rest of the Willamette Valley, most of Eastern Oregon, and Southern Oregon as well. The Columbia Gorge counties were split.
Using the same color scheme, this is where we stand today:
As of June 2014, there were 818,217 Democrats, and 649,614 Republicans in Oregon, about 38.6 and 30.6% percent of the registered voters in the state, respectively. Eastern Oregon, most of Southern Oregon, Coos and Columbia counties have Republican gains, the Gorge has seen polarization between the east and west, while Democrats have made substantial gains in the Willamette Valley. The following map uses the same color scheme, but with 3 point increments to show where each party has gained and roughly by how much.
Republicans made substantial gains in most of Eastern and Southern Oregon, smaller ones in Linn and Columbia counties on the west, and fairly small on the coast. In fact, their gain in Clatsop County is really more of a wash, at a relative gain of only 0.03% in the last 13 years.
And the Republican gains are insignificant compared to the Democratic gains, when the actual numbers of voters are concerned. Democrats gained over 70,000 voters since May 2001, now standing at 818,217 registered Democrats as of June 2014, whereas Republicans actually lost over 36,000 voters, now standing at 649,614 registered Republicans. Meanwhile, voters who don't belong to either party have swelled by almost 3% since 2001. Today Democrats stand at about 40.4% to Republican's about 32.2% of the state's registered voters (Edit: oops, that was in 2012! I already said where we were above and both parties have dropped since then).
And much of the Republican "gains" in many counties is more from the loss of Democrats than an actual gain among Republicans. In many counties in Eastern Oregon, Republicans did gain voters, but in many they only saw their advantages grow because they lost fewer voters than Democrats did. It's true on the coast, too, as in Tillamook, Coos, and Curry counties both parties lost voters between 2001 and 2014. See the map below where green counties gained registered voters in the last 13 years, while red ones lost registered voters.
Since May 2012 the shift has been similar but regionally a little more positive for Republicans.
Outside of Central Oregon, where both parties gained voters, in the rest of the east both parties lost voters, with more loss among Democrats than Republicans. On the Coast, in Southern Oregon and in the Upper Valley both parties lost voters in every county. In the Mid Valley Democrats lost voters and Republicans gained barely 150 voters. In the Portland metro area Democrats gained voters in Multnomah, Clackamas, and Washington counties, but lost some in the much smaller Columbia, while Republicans gained in Columbia, gained less than Democrats in Clackamas, and lost more voters than Democrats gained in both Washington and Multnomah counties.
Since 2012 Democrats have lost about 8900 registered voters and Republicans have lost about 9800, while registered voters from neither party have grown. Democratic losses in recent years shouldn't be surprising, as we haven't had a reason to register Democratic since the contested presidential primary of 2008. The shrinking of the Republicans should be more surprising, as they've had seriously contested statewide primaries in 2010 and 2014, and all the action in 2012 was on the Republican side, so people should have had a motivation to register as Republican, given the state's closed primaries.
What real value, though, is there in looking at voter registration? I think it's the case that Democrats don't hold any legislative districts in the state that don't have more registered Democrats than Republicans, but there are a number of Republicans in districts with more Democrats, too, and the margin doesn't seem to correspond too closely with election results. Democrats won the 5th congressional district when it used to have a plurality of registered Republicans, and we still hold it now that it flipped to a plurality of Democrats. Does analyzing small shifts really help that much?
The trend of voter registration doesn't seem to be matching actual election trends, either. While Multnomah and Washington counties have trended Democratic in both registration and in elections, and many other counties like Columbia have for Republicans in both, other counties show inconsistent trends. Coos County has long shown a Republican trend in voter registration, as it loses more Democrats than Republicans, but from 2008-2012 it had a slight Democratic trend in presidential election results, halting a long Republican trend. Malheur County continues to shed more registered Democrats than Republicans as well, despite trending Democratic in presidential elections from 2008 to 2012, from 2004 to 2008, and from 2000 to 2004.
We do know some things. There is significant evidence that people are moving from the rural counties to the metropolitan areas, and the voter registration seems to be showing a shift from the more rural counties to the more urban counties.
The regions within the state that have seen positive net migration are the Portland Metro, Columbia Gorge and Northeast Oregon. This makes sense given these regions have had relatively stronger economies than the rest of the state in recent years. However, the Portland Metro has seen, by far, the largest in-migration from elsewhere in the state.Similarly Republican voters seem to be shifting from more sparsely populated rural counties to more populous and urban counties in Central Oregon, Southern Oregon, the Mid Valley and Clackamas and Columbia counties in the metro area, plus Umatilla County, the biggest county in Eastern Oregon. Democratic votes are shifting more directly from almost all of the more rural counties into almost all of the more urban counties.
What is interesting to note that is even if you exclude Portland from the patterns and look at where people not going to or from Portland move, the pattern is still to the Willamette Valley. Again, this largely shows the pattern of migration to larger cities and the continued urbanization of the country.
When you consider where people are moving to Oregon from it also makes sense. People moving to rural counties tend to be from Southern California and other parts of the Southwest, while people moving to the Portland metro area are also from those areas, but also more likely to come from Northern California or other places like Minneapolis or New York.
There has also been a substantial increase among voters registered with neither of the major parties. In areas where the Democratic registration is collapsing, like Umatilla and Malheur counties, there is a significant growth of unaffiliated voters. in places like Washington and Multnomah counties where the Republican registration is collapsing, they seem to be becoming unaffiliated as well. Then there are urban counties booming like Deschutes and Jackson, where both parties seem to be shrinking relative to the boom of unaffiliated voters. But in many counties there is a real loss of voters, not just from the parties but among all voters. I hope to address that and more in a future post about demographics.
Update: here are the county by county numbers, broken down by the regions I was counting them in above. The first two columns on the left count the change since 2001, the columns on the right count the change since 2012. Numbers in black are gains, numbers in red are losses.