This is the fifth in a series of state by state voter registration diaries that I have decided to do since the beginning of this month (the states I have done diaries for before this one were for the states of Florida, Iowa, Nevada, and North Carolina).
By devoting a diary to a single state at a time, I hope that I can give more than just a simple report of the raw numbers but provide a more in-depth exploration of that information so as to provide some commentary regarding what those numbers may mean for the Presidential contest in that state this year. In that vein I will try to provide more of a long term history of the voter registration numbers for each of the states, both over the past year and just as importantly in comparison to the numbers from the 2004 election with accompanying raw vote totals from the election results in those states that year.
That having been said onto the numbers for today's state Colorado.
Before getting into the numbers a disclaimer is appropriate. The Colorado Secretary of State's office has released its voter registration numbers this year in two different, and non-compatible formats. For the first six months of this year, it released a "total registered voter" number(as it had done for years in the past) for the various political parties that lumped together indiscrimantely both "active voters" and "inactive voters" without ever separating out how many active voters or inactive voters subsets there were for each political party. Because of this failure to separate out active voters from inactive voters, I could only do a comparison between the parties based on their total voters number. Of course this has the disadvantage of including within it voters who are probably unlikely to vote; generally States classify someone as an "inactive voter" if the state receives word (usually from receipt of change of address forms sent to the Postal Service) that the person in question has moved outside the county. While these "inactive voters" are technically eligible to vote in the upcoming election, the receipt of such information makes the chances of that happening very slim. Further complicating this problem is that beginning in June of this year the Colorado Secretary of State started to put out information showing how many "active voters" there were for each political party and how many "inactive voters" there were for that party. Thus, for the first time useful information was made available to do a true comparison of how many actual, confirmed Democratic voters (i.e. "active Dem voters") there were as compared to actual, confirmed Republican voters. But because the information for the first half of this year did not include that breakdown the only way to do a trend analysis that walks through the entire year was to still look at the total registered voter numbers (meaning the number that indiscriminatly included active and inactive voters together).
Seeking to make the best of both worlds, I have decided to give two sets of numbers in this diary: One using the total number of voters (thereby making a trend analysis possible), and then next only compare amongst active registered voters even though there is not enough information in that regard to do any real trend analysis.
Colorado
TOTAL REGISTERED VOTERS
Dems Reps Ind.
1/25/08 880,761 1,011,152 998,939
4/9/08 900,823 1,017,738 1,013,548
6/30/08 932,503 1,019,446 1,021,979
7/31/08 946,277 1,024,504 1,022,376
9/2/08 955,428 1,029,062 1,022,575
9/30/08 997,146 1,045,457 1,046,244
10/22/08 1,051,110 1,063,190 1,069,492
As of January, 2008, Democrats made up 30.33% of the total registered electorate, Republicans made up 34.82%, and Independents made up 34.40%. As of October 22, 2008, Democrats make up 32.81%, Republicans 33.19%, and Indies 33.38%. So at the beginning of the year there was a 4.49% advantage of Republicans over Democrats in Colorado, and now (or at least the last publically available data shows) Republicans hold a measly .38% advantage, a decline of 4.11% in the span of 10 months! Indeed, the only party whose percentage of the electorate is expanding is the Democratic party, both Indies and Republicans share of the total registered electorate has dwindled over the past 10 months, with the steepest decline going to the Republicans who have lost 1.63% of their share (Indies lost 1.02% of their share).
These numbers however do not tell the real story as they include what are referred to as inactive voters. Unfortunately, the Colorado Sectretary of State site did not start breaking down its registration numbers between active and inactive voters, and more particularly the numbers of active and inactive voters by party until this June. So all I have so far are two months worth of information detailing this more relevant group of voters, with nothing before that (including what the break down was in prior elections). Nonetheless, since active voters totals are the real ones to watch, here is the partial list:
Active Registered Voters
Dems Reps Ind
6/30/08 717,495 785,115 671,244
7/31/08 737,565 795,949 681,639
UPDATE --- The Colorado Secretary of State has backtracked and is now refusing to show the breakdown between active and inactive voters by party for August's numbers and going forward. Given that he is a Republican one can only surmise that the last three months the active voter numbers must have been really, really bad for Republicans in Colorado. Hat tip for all the hard work by Obama staff and volunteers.
Thus, in June, active Democratic voters made up 32.85% of the electorate, active Republican voter comprised 35.94%, and Indies made up 30.73%. Now compare that to July's numbers where active Democratic voters made up 33.12% of the electorate, Republicans were 35.75%, and Indies made up 30.61%. In ONE month, Democrats share of active voters increased by .27%, while Republicans share dropped by .19% and Indies dropped by .12%. Interestingly, looking only at the June and July numbers amongst total registered voter numbers would mask this steep drop off among active Republican voters. That said, the total registered numbers do pretty accurately track the expansion (in terms of percetages) of the Democratic share and the contraction of the Indies share of the overall electorate. Specifically, looking only at the total registered numbers for June and July would only reveal a .04% drop off in Republican's share of the electorate as compared to the .19% drop off revealed when looking at the active voter totals (a difference of .15%). On the other hand, the total registered voters numbers between those two months reveal a .26% increase among Democrats (as compared to the .27% increase shown among active voters) and a .2% decrease in share among Indies (as compared to a .12% drop shown in the active voter numbers. The only reason I can give for this discrepancy is that a lot of Republican voters (the aforementioned .15%) were moved over to the inactive voter category during the month of July, which is actually good news for Democratic prospects in Colorado.
Anyway returning to a comparison of the final numbers, it appears that unlike with the total registered voter numbers, the percentage differential between active Democratic and Republican voters in Colorado as of July of this year was 2.63%, or net gain for Republicans of only three one hundreadths of a percent, .03%, when measured against their advantage among total registered voters. This gives me some confidence in the overall trend numbers shown in the total registered voters category as applying equally to the active voter group. The one noticeable difference being in the reversal of positions between the Democratic party and Indies. Among total registered voters Indies make up the second largest group and Democrats are in third, but among active voters Democrats are the second largest group and Indies are third.
So what were the numbers in Colorado in 2004? Well at the beginning of 2004, there were 1,042,296 registered Republicans (or 37.02% of the electorate), 852,910 registered Democrats (or 30.29% of the electorate), and 907,602 Indies (or 32.23% of the electorate). Come November, 2004, the final registration numbers were as follows: 1,118,597 registered Republicans (or 36.10% of the electorate), 942,025 registered Democrats (or 30.40% of the electorate), and 1,024,973 Indies (or 33.08% of the electorate). In other words, while everyone's raw numbers increased, only Indies and Democrats percentage share of the electorate increased. But even then, Indies far and away were the largest movers in the voter registration in Colorado that year.
Come election day in November, 2004, Bush beat Kerry in Colorado by 107,567 votes, or percentage wise 52% to 47%. Exit polls indicated that 37% of those showing up at the polls that day were Republicans, 37% were Democrats, and 26% were Indies. In other words, Republicans performed at close to their share of the electorate, Democrats overperformed significantly while Indies woefully underperformed from their share of the registered electorate. As a final note, exit polls showed Kerry barely edging Bush among Indies (49 to 48 percent), and Bush performing slightly better among Democrats (89 to 11 percent) than Kerry performed among Republicans (93 to 6 percent). Indeed, Bush's larger share of the Democratic vote over Kerry's share of the Republican vote was the difference in who won in Colorado in 2004.
Analysis
So what to make of this? Looking at 2004, it is clear that the largest movers in terms of registered voters were Independents. Such movement away from the major parties would tend to suggest some level of apathy toward either of the major parties among voters in Colorado that year. Apathetic voters normally don't show up to vote, and true enough come November, 2004, Independents woefully underperformed. This year, on the other hand, Democrats are the largest movers in terms of number of voters registered, with Independents a distant second, and Republicans an even further third. This may suggest that two things are going on in Colorado, either beneficial for Democrats: (1) That new voters are registering as Democrats in Colorado this year in large numbers and/or (2) that there is some party migration going on with moderate Republicans switching to Independents (thus accounting for the relatively larger uptick in Independents numbers as compared to the flatline in Republican numbers) and Democratic-leaning Independents moving over and formally becoming Democrats. In other words, there seems to be a swing towards the Democrats in almost the same way as there was a swing to Independents in 2004 (although the swing this year is much bigger than in 2004). This would suggest a larger number of Democrats showing up at the polls in Colorado come November than Republicans. Where exactly Independents will go is a little more uncertain. My initial guess is that Independents, being who they are, will underperform but perhaps not as poorly as they did in 2004. In this sense, the macro numbers would seem to indicate that Obama should perform as well in Colorado as Bush did in 2004.
Now that is presentation of the registration numbers on a macro-level, lets take a look from more of a micro-level (presented in some measure to see if we can see any evidence for the argument I made above of new voters going to the Democrats and some party migration away from the Republicans and toward the Democrats). I have compiled the trend lines for the registration numbers from the past 7 months for the 8 largest counties (in terms of registered voters) in Colorado. Specifically, Adams, Arapahoe, Boulder, Denver, Douglas, El Paso, Jefferson, and Larimer counties (all of which have over 100,000 active registered voters). Of the 8, 5 are Republican leaning (meaning they have more registered Republican voters than Democrats) ---- Arapahoe, Douglas, El Paso, Jefferson, and Larimer --- and the remaining 3 are Democrat leaning --- Adams, Boulder, and Denver. I have also provided snap shots of these counties registration numbers from 2004 to see if any comparisons or contrasts could be made between then and this year. So lets dig down and see the trend lines in each of these counties over the past 7 months and then compare them to 2004.
Republican leaning counties -- 2008
ARAPAHOE
Reps Dems Ind
1/25/08 109,115 95,275 104,036
6/30/08 109,010 101,782 105,365
7/31/08 110,185 105,081 103,302
9/02/08 110,209 105,172 102,851
9/30/08 111,818 110,748 105,000
10/22/08 114,072 119,381 108,299
DOUGLAS
Reps Dems Ind
1/25/08 80,910 32,070 50,289
6/30/08 82,465 35,175 51,243
7/31/08 82,853 35,671 51,643
9/02/08 83,526 36,020 52,088
9/30/08 85,668 37,579 53,619
10/22/08 87,494 39,493 55,072
EL PASO
Reps Dems Ind
1/25/08 153,532 68,689 108,406
6/30/08 155,558 72,590 112,630
7/31/08 157,047 74,774 113,699
9/02/08 157,998 75,241 112,665
9/30/08 162,786 81,045 118,517
10/22/08 165,881 85,216 120,862
JEFFERSON
Reps Dems Ind
1/25/08 123,975 102,191 115,003
6/30/08 124,768 108,135 116,652
7/31/08 124,225 108,581 116,128
9/02/08 124,397 109,306 116,731
9/30/08 126,179 113,644 120,728
10/22/08 127,226 118,096 123,601
LARIMER
Reps Dems Ind
1/25/08 70,621 50,221 67,846
6/30/08 71,418 53,013 69,496
7/31/08 71,685 53,673 70,241
9/02/08 72,189 54,872 71,395
9/30/08 73,474 57,707 73,403
10/22/08 74,482 60,278 74,659
Republican leaning counties -- 2004
ARAPAHOE
Reps Dems Ind
1/15/04 127,496 96,696 111,355
11/04 133,885 106,690 122,970
DOUGLAS
Reps Dems Ind
1/15/04 72,551 26,292 41,115
11/04 81,382 30,402 46,172
EL PASO
Reps Dems Ind
1/15/04 151,507 70,515 101,039
11/04 164,884 77,485 112,149
JEFFERSON
Reps Dems Ind
1/15/04 135,189 99,434 109,729
11/04 142,925 110,016 122,213
LARIMER
Reps Dems Ind
1/15/04 73,106 46,534 59,443
11/04 78,087 52,288 68,565
As you can see there is a different trajectory in many of these Republican counties than there was in 2004. The stand out is Arapahoe county. Just look at those numbers for this year as compared to 2004. In November, 2004, Republicans outnumbered Democrats in Arapahoe county by over 27,000. At the beginning of this year, Republicans again outnumbered Democrats although by not nearly as much, but by a significant number nonetheless --- nearly 14,000. As of October 22, 2008 (after the close of the registration deadline), Arapahoe is NO LONGER A REPUBLICAN COUNTY!. Democrats now outnumber Republicans in the county by more than 5,000. Even more impressive is the fact that the only group's whose numbers are moving in Arapahoe county are Democrats; Republicans and Independents numbers have essentially stayed flat for the past ten months. This same pattern holds but to a lesser extent with Jefferson county. These are two counties Democrats should be (and apparently are) work hard this election. There exists (as I predicted) now the fact that Jefferson county is teetering on the edge of flipping to being a Democratic county. It did not happen this election cycle (like it did to Arapahoe county), coming just 9,000 short, but Democrats are awfully close to flipping Jefferson come next election cycle (and Larimer looms on the horizon as well).
Another point showing the different trajectory of these Republican counties from where they were in 2004. In that year, the largest mover in 3 of the 5 counties was Independents, in the other 2 the largest mover was Republicans. But come 2008, the largest mover in ALL of these Republican counties are Democrats with Independents in 3 of the 5 (El Paso, Jefferson, and Larimer) coming in second. This would seem to support the idea of party migration by Republicans (the one from whom people are moving away would have the smallest movement, followed by the intermediary/weigh station Independents, and then the party toward which voters are moving toward having the largest movement). However, in the other 2 counties where Republicans are the second largest mover (Arapahoe and Douglas) this would seem to suggest party migration by Democratic-leaning Independents and/or possible new voters. New voters typically do not choose Independent as their first registration. So how does it look in Democratic-leaning counties?
Democratic leaning counties -- 2008
ADAMS
Reps Dems Ind
1/25/08 54,355 67,820 73,458
6/30/08 54,969 71,865 76,910
7/31/08 54,954 72,212 76,820
9/02/08 55,093 72,776 76,711
9/30/08 55,658 75,156 77,431
10/22/08 56,910 81,374 79,357
BOULDER
Reps Dems Ind
1/25/08 43,471 75,214 78,800
6/30/08 43,295 78,141 78,841
7/31/08 43,263 79,999 78,528
9/02/08 43,306 81,522 78,828
9/30/08 43,631 85,563 79,310
10/22/08 44,355 90,289 81,231
DENVER
Reps Dems Ind
1/25/08 70,741 164,012 133,150
6/30/08 70,883 178,527 135,413
7/31/08 70,696 180,812 135,175
9/02/08 70,573 181,802 134,866
9/30/08 70,915 189,602 136,055
10/22/08 71,711 200,875 138,359
Democratic leaning counties -- 2004
ADAMS
Reps Dems Ind
1/15/04 59,438 70,887 70,272
11/04 64,838 77,311 78,826
BOULDER
Reps Dems Ind
1/15/04 49,265 68,168 71,915
11/04 51,924 77,717 83,527
DENVER
Reps Dems Ind
1/15/04 75,941 152,341 106,869
11/04 79,074 169,584 132,115
Again notice the difference from 2004. In that year (just as it was in most of the Republican counties that year) the largest mover was Independents, followed by Democrats, and Republicans. This year, however, Democrats are the largest movers, followed by Independents in 2 of the counties (Adams and Denver), with Republicans coming in third. This would support a party migration theory. One final note, looking at the margins between Democrats and Republicans in these three counties from 2004 and 2008, it would appear that, at least in Denver county, a lot of the uptick in numbers is attributable to new voters not people switching from one party to another.