Colorado state capitol
Two of the states that are going to be watched most closely on election night are Colorado and Arkansas, both of which have key Senate and gubernatorial races. The Democratic Senate incumbents, Mark Udall and Mark Pryor, are trailing in both elections, although Udall (in Colorado) has a better shot at pulling it off. That's not only because he faces a narrower margin against Republican opponent Cory Gardner than Pryor does against Tom Cotton, but also because, as I've discussed many times this year, polls in
Colorado have a recent track record of understating Democratic chances. Almost all polls in the 2010 Senate race put Republican Ken Buck ahead of Michael Bennet (who went on to win narrowly), and most 2012 presidential polls saw a much closer race than Barack Obama's eventual victory.
The current close race in Colorado comes as a bit of a late-breaking surprise, given the state's blue trend over the last decade. More than anything, it may simply have to do with Gardner being a skillful candidate appealing to swing voters, optimistic-sounding and with great bamboozlement skills, having caught Udall rather flat-footed. On the other hand, the 2014 Arkansas race has been seen as a huge problem for Pryor for years ahead of time, given the state's lurch to the right—a state that gave Bill Clinton huge margins two decades ago became Barack Obama's sixth-worst state.
How did we get to that point? If you go back to the 1980s, both of these states were light-red states, slightly to the right of the national median. They both had about the same levels of diversity, although most of the non-white residents of Colorado were Latino and most of the non-white residents of Arkansas were African-American. What's happened since then is demographic change: The populations of the states were reshaped, in different ways. Colorado became much more diverse and educated—traits that are generally correlated with better Democratic performance—and while those things happened in Arkansas too, it happened at a slower rate. As we'll see in closer detail, those contrasts are especially sharp when you zoom in on the suburbs of Denver versus the suburbs of Little Rock.
We'll dig into the charts below the fold:
For both Colorado and Arkansas, I've put together two series of charts. Both zoom in on the state's key counties (the ones that, individually, make up 2 percent or more of the state's votes). The charts also look at the "balance" of the state, or all the rural counties that individually are too small but together make up a significant part of the state. The first set of charts tracks the presidential vote in each county from 2012 back to 1984; "PVI" is Partisan Voting Index, which reflects how far that county deviated from the national two-way vote median that particular year (it's based on only the Democratic and GOP votes to smooth out the comparison between years when there were and weren't significant third-party challenges).
The second set of charts tracks the changes in the state's racial and education composition over the 2010, 2000, and 1990 censuses. The "race" column describes the percentages of the non-Hispanic, African-American, Asian, and Hispanic populations (in that order). The "education" column describes the percentage of persons over the age of 25 who have a bachelor's degree or better. (If that seems vaguely familiar, I used this same technique a year ago when previewing Virginia before the 2013 gubernatorial election, comparing the dramatic differences in demographic change in northern Virginia versus the rest of the state).
County |
2012 % |
2012 PVI |
% of state |
2008 % |
2008 PVI |
% of state |
2004 % |
2004 PVI |
% of state |
Statewide |
51/46 |
D+1 |
100.0 |
54/45 |
D+1 |
100.0 |
47/52 |
R+1 |
100.0 |
Jefferson |
51/46 |
D+1 |
12.1 |
54/45 |
D+1 |
12.3 |
47/52 |
R+1 |
12.7 |
Denver |
74/24 |
D+23 |
11.7 |
75/23 |
D+23 |
11.3 |
70/29 |
D+22 |
11.2 |
El Paso |
39/59 |
R+12 |
11.3 |
40/59 |
R+13 |
11.4 |
32/67 |
R+16 |
11.3 |
Arapahoe |
54/44 |
D+3 |
11.1 |
56/43 |
D+3 |
11.1 |
47/51 |
R+1 |
10.9 |
Larimer |
51/46 |
D+1 |
7.0 |
54/44 |
D+1 |
6.9 |
47/52 |
R+1 |
6.9 |
Boulder |
70/28 |
D+19 |
7.0 |
72/26 |
D+20 |
7.2 |
66/32 |
D+18 |
7.5 |
Adams |
57/40 |
D+7 |
6.9 |
58/40 |
D+6 |
6.7 |
51/48 |
D+2 |
6.4 |
Douglas |
36/62 |
R+15 |
6.5 |
41/58 |
R+12 |
6.3 |
33/67 |
R+16 |
5.7 |
Weld |
42/55 |
R+8 |
4.5 |
45/53 |
R+8 |
4.4 |
36/63 |
R+12 |
4.1 |
Pueblo |
56/42 |
D+5 |
3.0 |
57/41 |
D+4 |
3.0 |
53/46 |
D+4 |
3.2 |
Mesa |
33/65 |
R+19 |
2.8 |
34/64 |
R+19 |
2.9 |
32/67 |
R+17 |
2.9 |
Balance |
44/54 |
R+7 |
16.1 |
47/51 |
R+6 |
16.5 |
42/57 |
R+7 |
17.2 |
County |
2000 % |
2000 PVI |
% of state |
1996 % |
1996 PVI |
% of state |
1992 % |
1992 PVI |
% of state |
Statewide |
42/51 |
R+5 |
100.0 |
44/46 |
R+5 |
100.0 |
40/36 |
R+1 |
100.0 |
Jefferson |
43/51 |
R+5 |
13.5 |
43/48 |
R+8 |
13.9 |
36/37 |
R+4 |
14.2 |
Denver |
62/31 |
D+16 |
11.4 |
62/30 |
D+13 |
12.9 |
56/25 |
D+13 |
13.9 |
El Paso |
31/64 |
R+18 |
11.5 |
32/59 |
R+19 |
11.5 |
27/51 |
R+19 |
10.7 |
Arapahoe |
43/51 |
R+4 |
10.9 |
42/51 |
R+10 |
10.8 |
36/39 |
R+5 |
11.7 |
Larimer |
39/53 |
R+8 |
6.8 |
42/47 |
R+8 |
6.5 |
38/36 |
R+2 |
6.4 |
Boulder |
50/36 |
D+8 |
8.0 |
52/35 |
D+5 |
8.0 |
51/26 |
D+12 |
8.1 |
Adams |
50/44 |
D+3 |
6.2 |
51/39 |
D+2 |
6.2 |
44/30 |
D+6 |
6.6 |
Douglas |
31/65 |
R+18 |
5.0 |
31/62 |
R+21 |
3.4 |
25/46 |
R+19 |
2.6 |
Weld |
36/58 |
R+12 |
3.7 |
40/50 |
R+10 |
3.5 |
36/39 |
R+6 |
3.4 |
Pueblo |
54/42 |
D+6 |
3.1 |
57/35 |
D+8 |
3.3 |
54/29 |
D+12 |
3.6 |
Mesa |
30/63 |
R+18 |
2.9 |
37/53 |
R+14 |
3.1 |
34/41 |
R+8 |
2.8 |
Balance |
36/57 |
R+12 |
17.0 |
40/48 |
R+9 |
16.9 |
36/36 |
R+4 |
16.0 |
County |
1988 % |
1988 PVI |
% of state |
1984 % |
1984 PVI |
% of state |
Statewide |
45/53 |
R+0 |
100.0 |
35/63 |
R+5 |
100.0 |
Jefferson |
42/56 |
R+4 |
14.3 |
30/69 |
R+11 |
13.9 |
Denver |
61/37 |
D+16 |
15.3 |
50/48 |
D+10 |
17.0 |
El Paso |
29/70 |
R+17 |
10.1 |
24/75 |
R+17 |
9.1 |
Arapahoe |
38/60 |
R+7 |
11.6 |
27/72 |
R+14 |
11.5 |
Larimer |
43/55 |
R+2 |
6.1 |
32/67 |
R+8 |
5.8 |
Boulder |
53/45 |
D+8 |
7.8 |
43/55 |
D+3 |
7.5 |
Adams |
53/46 |
D+7 |
6.9 |
39/60 |
R+2 |
7.1 |
Douglas |
28/70 |
R+17 |
1.8 |
20/79 |
R+21 |
1.2 |
Weld |
43/55 |
R+2 |
3.5 |
30/69 |
R+10 |
3.5 |
Pueblo |
61/38 |
D+16 |
3.9 |
52/47 |
D+12 |
4.0 |
Mesa |
39/60 |
R+7 |
2.7 |
29/70 |
R+11 |
2.6 |
Balance |
43/56 |
R+3 |
16.0 |
31/67 |
R+9 |
16.8 |
Let's start with Colorado, which actually has followed a very similar path as Virginia in the last three decades, starting as a light-red state, moving closer in 2004, and then crossing over quickly into the Democratic column in 2008. Unlike Virginia, though, Colorado had an earlier flirtation with the Democratic side; Bill Clinton narrowly won in 1992, although Colorado still clocked in at R+1. (With a strong-enough nationwide performance setting the overall bar, a Democratic candidate can still win states while staying in the R+ column. For instance, Barack Obama won Florida in 2012 at R+1.) That was thanks in large part, though, to a particularly strong Ross Perot performance in 1992—it was one of the few states where Perot topped 20 percent, and he seemed to eat a lot into George H.W. Bush's center-right vote share.
Also, notably, Colorado kind of liked Mike Dukakis, performing right at the nation's median that year, although, of course, that was nowhere near enough to win. Instead, there was a lot of falloff, PVI-wise, in 1996 and 2000. That seems partly because of some backlash to Democratic environmental policies in that period (a common thread in a lot of western states those years), but mostly because Ralph Nader performed particularly well in Colorado those years, and his vote share came lopsidedly out of the Democratic column, unlike Perot, who drew pretty evenly from both sides.
If you drill down to individual counties, though, you can see that most of the movement in the Democratic direction happened in Denver and the other counties in its metropolitan area. Probably the most striking move is in Arapahoe County, which went from R+17 in 1984 to D+3 in 2012, but there are moves almost as large in Denver proper (which is its own county), Boulder, and Jefferson counties. Contrast that with El Paso County (Colorado Springs), which moved left but much more slowly, and the smaller cities in Pueblo County and Mesa County (Grand Junction), which actually moved appreciably to the right over the decades. (Pueblo, in fact, used to be the state's Democratic stronghold—not very helpful since it's only 3 or 4 percent of the state's population—but now is only a bit to the left of the state's median.) The "balance," the state's rural counties, also stayed pretty much stationary, moving only from R+9 in 1984 to R+7 today.
One other striking detail is that Denver and its adjacent counties are actually a smaller percentage of the state's population than they used to be, but the shift in voting preferences was sharp enough to overcome that decrease in relative size. Denver itself provided 17 percent of the state's votes in 1984 but only 12 percent now, and even suburban Jefferson County fell from 14 percent to 12 percent in that same time.
Most of the shift wasn't to rural areas, though, but mostly to exurban Douglas County, located between Colorado Springs and Denver. This county was mostly empty in the 1980s (providing 1.2 percent of the state's votes in 1984), but now is brimming with McMansions (6.5 percent of the state's votes in 2012). Unfortunately, it's still a pretty right-leaning county, but if it can undergo the same shift as closer-in Jefferson and Arapahoe counties as it too becomes more diverse, that could help push Colorado from swing state to fully blue. (If you want an East Coast analogy, think Loudoun County in northern Virginia, an affluent exurban county that went from dark-red to swingy in the last few decades as it also went from mostly white to diverse.)
County |
2010 Races |
2010 Educ. |
% of state |
2000 Races |
2000 Educ. |
% of state |
1990 Races |
1990 Educ. |
% of state |
Statewide |
70/4/3/21 |
35.9 |
100 |
74/4/2/17 |
32.7 |
100 |
81/4/2/13 |
27.0 |
100 |
Jefferson |
80/1/3/14 |
39.3 |
10.6 |
85/1/2/10 |
36.5 |
12.3 |
90/1/2/7 |
30.7 |
13.3 |
Denver |
52/10/3/32 |
40.1 |
11.9 |
52/11/3/32 |
34.5 |
12.9 |
61/12/2/23 |
29.0 |
14.2 |
El Paso |
72/6/3/15 |
34.9 |
12.4 |
76/6/2/11 |
31.8 |
12.0 |
81/7/2/9 |
25.8 |
12.1 |
Arapahoe |
63/10/5/18 |
37.9 |
11.4 |
74/7/4/12 |
37.0 |
11.3 |
85/6/3/6 |
35.2 |
11.9 |
Larimer |
84/1/2/11 |
42.5 |
6.0 |
88/1/2/8 |
39.5 |
5.8 |
91/1/1/7 |
32.3 |
5.7 |
Boulder |
79/1/4/13 |
57.0 |
5.9 |
84/1/3/10 |
52.4 |
6.8 |
89/1/2/7 |
42.1 |
6.8 |
Adams |
53/3/3/38 |
20.3 |
8.8 |
63/3/3/28 |
17.4 |
8.5 |
75/3/2/19 |
13.0 |
8.0 |
Douglas |
85/1/4/7 |
54.4 |
5.7 |
90/1/2/5 |
51.9 |
4.1 |
95/1/1/3 |
40.7 |
1.8 |
Weld |
68/1/1/28 |
25.4 |
5.0 |
70/0/1/27 |
21.6 |
4.2 |
77/0/1/21 |
18.4 |
4.0 |
Pueblo |
54/2/1/41 |
20.9 |
3.2 |
58/2/1/38 |
18.3 |
3.3 |
61/2/0/36 |
14.0 |
3.7 |
Mesa |
83/1/1/13 |
25.4 |
2.9 |
87/0/1/10 |
22.0 |
2.7 |
90/0/1/8 |
17.4 |
2.8 |
Balance |
76/1/3/19 |
30.0 |
16.3 |
80/1/0/16 |
25.6 |
16.1 |
84/1/0/14 |
20.2 |
15.6 |
Looking at the changes in Colorado's demographics, you can see the correlation with political change especially sharply in Arapahoe County (east of Denver and home to the mega-suburb of Aurora). It became much more diverse faster than the rest of the state, moving from 85 percent non-Hispanic white in 1990 to 63 percent in 2010. Adams County, just north of Denver, also moved at a similar rate, going from 75 percent to 53 percent in the same period. (Interestingly, Adams County, which is much more blue-collar, shifted the least in terms of its vote preference of the Denver metro area counties, staying at D+7 in both 1988 and 2012. That suggests you need a kind of synergy between diversity and education, as we've seen not just in Arapahoe and Jefferson counties but also in a lot of the more affluent suburbs of the East Coast cities, to see the more rapid shift in the Democratic direction.)
While Denver proper and Jefferson County didn't become diverse as rapidly (Denver, for instance, went from 61 percent white in 1990 to 52 percent white in 2000 and then stayed at 52 percent white in 2010, pretty suggestive of gentrification at work), where the areas really shot up was in terms of education. Denver went from 29 percent college-educated in 1990 to 40 percent in 2010, Jefferson County went from 31 to 39 percent in that same period, and Boulder County (home of the state's main university) did even better, going from 42 percent to 57 percent, one of the highest percentages anywhere in the country.
Contrast those with the smaller Pueblo and Mesa counties, which as I mentioned above, are two counties where the Democratic vote share noticeably declined. Those counties didn't diversify as rapidly (Pueblo—which, along with Colorado Springs, was the scene of the Colorado state Senate recall debacle—starts with an already-high Hispanic baseline, but didn't change much more, going from 61 percent white in 1990 to 54 percent now. Mesa went from 90 percent white to 83 percent in that same period.). And they remain some of the most poorly educated parts of the state, even less educated than the rural "balance" of the state, with Pueblo going from 14 percent college-educated in 1990 to 21 percent now, and Mesa going from 17 percent in 1990 to 25 percent now. That again illustrates how correlated Democratic strength in the 21st century is with the overlap of diversity and education.
County |
2012 % |
2012 PVI |
% of state |
2008 % |
2008 PVI |
% of state |
2004 % |
2004 PVI |
% of state |
Statewide |
37/61 |
R+14 |
100.0 |
39/59 |
R+14 |
100.0 |
45/54 |
R+4 |
100.0 |
Pulaski |
55/43 |
D+4 |
14.9 |
55/44 |
D+2 |
14.8 |
55/44 |
D+7 |
14.6 |
Benton |
29/69 |
R+23 |
7.4 |
31/67 |
R+22 |
7.0 |
30/68 |
R+18 |
6.5 |
Washington |
40/56 |
R+10 |
6.6 |
42/56 |
R+10 |
6.3 |
43/56 |
R+5 |
6.1 |
Saline |
27/70 |
R+24 |
4.4 |
28/69 |
R+25 |
4.1 |
36/63 |
R+12 |
3.7 |
Sebastian |
30/67 |
R+21 |
4.0 |
32/66 |
R+21 |
4.0 |
37/62 |
R+11 |
4.1 |
Faulkner |
33/64 |
R+18 |
3.9 |
36/62 |
R+17 |
3.8 |
40/59 |
R+8 |
3.5 |
Garland |
34/64 |
R+17 |
3.8 |
36/61 |
R+16 |
4.0 |
45/54 |
R+3 |
3.8 |
Craighead |
33/64 |
R+18 |
3.0 |
36/61 |
R+16 |
2.8 |
46/53 |
R+2 |
2.8 |
Jefferson |
64/35 |
D+13 |
2.6 |
62/36 |
D+10 |
2.7 |
65/34 |
D+17 |
2.9 |
White |
22/75 |
R+30 |
2.5 |
25/72 |
R+28 |
2.5 |
35/64 |
R+14 |
2.5 |
Lonoke |
23/74 |
R+28 |
2.3 |
25/73 |
R+28 |
2.2 |
34/65 |
R+15 |
2.1 |
Balance |
34/63 |
R+17 |
44.6 |
37/61 |
R+16 |
45.8 |
45/54 |
R+3 |
47.4 |
County |
2000 % |
2000 PVI |
% of state |
1996 % |
1996 PVI |
% of state |
1992 % |
1992 PVI |
% of state |
Statewide |
46/51 |
R+3 |
100.0 |
54/37 |
D+5 |
100.0 |
53/35 |
D+7 |
100.0 |
Pulaski |
54/44 |
D+5 |
13.8 |
59/35 |
D+8 |
14.4 |
58/35 |
D+9 |
14.4 |
Benton |
32/65 |
R+17 |
5.8 |
38/52 |
R+13 |
5.2 |
36/49 |
R+11 |
4.6 |
Washington |
42/55 |
R+7 |
5.6 |
46/44 |
R+4 |
5.0 |
46/42 |
R+1 |
5.0 |
Saline |
39/57 |
R+10 |
3.5 |
49/41 |
R+0 |
3.2 |
49/39 |
D+2 |
2.7 |
Sebastian |
39/58 |
R+10 |
4.4 |
44/57 |
R+6 |
4.0 |
42/42 |
R+4 |
4.2 |
Faulkner |
41/55 |
R+8 |
3.2 |
50/42 |
R+1 |
2.7 |
52/38 |
D+4 |
2.7 |
Garland |
44/53 |
R+5 |
3.9 |
53/38 |
D+4 |
4.1 |
52/36 |
D+6 |
3.8 |
Craighead |
49/48 |
D+0 |
2.7 |
54/37 |
D+4 |
2.8 |
54/36 |
D+7 |
2.7 |
Jefferson |
65/32 |
D+17 |
3.0 |
71/23 |
D+21 |
3.1 |
69/24 |
D+21 |
3.3 |
White |
38/59 |
R+11 |
2.4 |
49/41 |
R+1 |
2.4 |
49/40 |
D+2 |
2.3 |
Lonoke |
38/59 |
R+11 |
1.9 |
50/40 |
D+1 |
1.8 |
50/40 |
D+3 |
1.7 |
Balance |
47/51 |
R+2 |
49.8 |
55/34 |
D+7 |
51.3 |
55/33 |
D+9 |
52.6 |
County |
1988 % |
1988 PVI |
% of state |
1984 % |
1984 PVI |
% of state |
Statewide |
42/56 |
R+3 |
100.0 |
38/60 |
R+2 |
100.0 |
Pulaski |
44/55 |
R+2 |
15.5 |
41/58 |
D+0 |
15.1 |
Benton |
28/71 |
R+18 |
4.1 |
23/76 |
R+18 |
3.6 |
Washington |
34/64 |
R+11 |
4.4 |
31/68 |
R+10 |
4.1 |
Saline |
40/59 |
R+6 |
2.5 |
36/61 |
R+3 |
2.2 |
Sebastian |
28/71 |
R+18 |
4.2 |
24/75 |
R+17 |
4.2 |
Faulkner |
40/58 |
R+5 |
2.2 |
38/61 |
R+3 |
2.2 |
Garland |
36/61 |
R+9 |
3.8 |
34/62 |
R+6 |
3.8 |
Craighead |
42/55 |
R+3 |
2.6 |
36/63 |
R+4 |
2.5 |
Jefferson |
56/42 |
D+11 |
3.6 |
55/44 |
D+15 |
3.7 |
White |
38/61 |
R+8 |
2.2 |
34/65 |
R+6 |
2.2 |
Lonoke |
40/60 |
R+6 |
1.5 |
35/64 |
R+5 |
1.5 |
Balance |
45/54 |
R+0 |
53.4 |
40/59 |
R+0 |
54.9 |
With that in mind, what happens when you have a whole state that hasn't diversified much and remains poorly educated? You've got a state that's going to move rapidly in the GOP's direction, and more specifically, you've got Arkansas. (In a similar instance, you've also got West Virginia.)
People tend to think of Arkansas as having once been a Democratic stronghold, and it was if you go back far enough, to the pre-Civil Rights Act period. However, people tend to look at the gaudy numbers that Bill Clinton put up in Arkansas in his two elections and assume that was always the case. That had a lot to do with his favorite son status, though, and his unique appeal to the state. Arkansas didn't respond favorably to northern Democrats pre-Clinton, though better than other Deep South states did. Michael Dukakis clocked in at R+3 in Arkansas, and Walter Mondale was a similar R+2. (If you go even further back, George McGovern did far worse here than Obama or even Mondale, with an appalling 30 percent of the vote, though that's still good for R+7, against the backdrop of terrible nationwide numbers.)
The main place in Arkansas that has moved in the Democratic direction—part of the trend of cities, in general, regardless of location, moving toward the Democrats—is Pulaski County, the state's most populous county and the location of Little Rock. It had an even PVI in 1984, but is D+4 now, one of the few large counties in the state with an Obama majority. The only other one is Jefferson County, with the mostly black city of Pine Bluff, but it's been stable (D+15 then, D+13 now, and losing population).
To really understand what's happening in Arkansas, you need to look two other places: one is Benton County in the state's northwest corner, which has surged to become the state's second most populous county in recent years (thanks largely to its most famous resident, Walmart headquarters), going from 3.6 percent of the state's vote in 1984 to 7.4 percent now. The northwest was always the most heavily Republican corner of the state, though, going back centuries (see nearby Sebastian County, location of Fort Smith, which hasn't grown but has stayed similarly dark-red); even as it's become a larger share of the state, population-wise, it's only gotten a little redder, moving from R+18 in 1984 to R+23 now.
What's most striking, though, is the four exurban counties that surround Little Rock: Saline, Faulkner, White, and Lonoke counties. These counties, politically, used to be close to the state's median. For instance, they gave wide margins to Bill Clinton in his two elections, unlike the northwestern counties, which were so staunchly GOP that they even resisted the Big Dog's charms.
However, two things are happening in these counties that amplify their importance; one, they're the fastest-growing parts of the state (together, they were 8.1 percent of the state's votes in 1984, when they were basically rural, and now, they're 13.1 percent of the state's votes, through a combination of former Little Rock residents white-flighting out of town and former rural residents moving closer to where jobs are). The other is that they've shifted to become the most hard-right places in the state, with White County the most extreme example, going from R+6 in 1984 to R+30 now. That's an even more abrupt shift than the state's rural "balance," which went from "only" R+0 to R+17 in that same period. Also worth noting—the rural counties are a significantly smaller part of the state than they used to be, going from 55 percent of the state's votes in 1984 to 45 percent now (with movement to both the exurban counties and Benton County instead).
County |
2010 Races |
2010 Educ. |
% of state |
2000 Races |
2000 Educ. |
% of state |
1990 Races |
1990 Educ. |
% of state |
Statewide |
75/15/1/6 |
19.1 |
100 |
79/16/1/3 |
16.7 |
100 |
82/16/1/1 |
13.3 |
100 |
Pulaski |
58/35/2/6 |
30.7 |
13.1 |
63/32/1/2 |
28.1 |
13.5 |
72/26/1/1 |
23.5 |
14.9 |
Benton |
77/1/3/15 |
25.9 |
7.6 |
87/0/1/9 |
20.3 |
5.7 |
97/0/0/1 |
14.4 |
4.1 |
Washington |
74/3/2/15 |
27.4 |
7.0 |
85/2/2/8 |
24.5 |
5.9 |
95/1/1/1 |
20.0 |
4.8 |
Saline |
89/5/1/4 |
22.6 |
3.4 |
95/2/1/1 |
16.4 |
3.1 |
97/2/0/1 |
11.9 |
2.7 |
Sebastian |
73/6/4/12 |
17.9 |
4.3 |
80/6/3/7 |
16.6 |
4.3 |
88/6/3/1 |
14.6 |
4.2 |
Faulkner |
82/10/1/4 |
26.4 |
3.9 |
87/8/1/2 |
25.2 |
3.2 |
91/8/0/1 |
17.9 |
2.6 |
Garland |
84/8/1/5 |
20.2 |
3.3 |
87/8/0/3 |
18.0 |
3.3 |
90/8/0/1 |
14.2 |
3.1 |
Craighead |
80/13/1/4 |
23.1 |
3.3 |
88/8/1/2 |
20.9 |
3.1 |
93/5/1/1 |
16.4 |
2.9 |
Jefferson |
41/55/1/2 |
16.5 |
2.7 |
48/49/1/1 |
15.7 |
3.2 |
56/43/0/0 |
14.6 |
3.6 |
White |
90/4/1/4 |
17.3 |
2.6 |
93/4/0/2 |
15.5 |
2.5 |
96/3/0/1 |
10.9 |
2.3 |
Lonoke |
88/6/1/3 |
16.6 |
2.3 |
90/6/0/2 |
14.6 |
2.0 |
90/9/0/1 |
10.0 |
1.7 |
Balance |
77/16/1/4 |
13.2 |
46.5 |
79/16/0/3 |
11.7 |
50.2 |
81/17/0/1 |
9.5 |
52.9 |
Now let's look at the demographic changes. Arkansas started in 1990 at the same level of diversity as Colorado (82 percent non-Hispanic white in Arkansas, 81 in Colorado), but in 2010, Arkansas looks notably less diverse in comparison (75 percent in Arkansas, 70 in Colorado). Most of the increased diversity in Arkansas, though, is concentrated in a few places: Little Rock, which went from 72 percent white in 1990 to 58 percent now (with the gains coming from a mix of black and Hispanic residents), Jefferson County (where, given its overall population decline, the white residents seem to be moving out), and, interestingly, Benton County, which went from 97 percent white in 1990 to 77 percent now, with almost all the non-white growth coming from Hispanics. You'd think that might move Benton County a bit in the Democratic direction, but there seems to be little effort by local politicians to engage that segment of the vote.
Contrast that with the four exurban counties. Lonoke County barely moved, in terms of its racial composition; it was 90 percent white in 1990, and now it's 88 percent white. Aptly named White County was 96 percent white in 1990, and now it's 90 percent. Bear in mind, too, that these are the fastest-growing parts of the state.
Turning to statewide education trends, the percentage of college-educated residents rose from 13 percent in 1990 to 19 percent in 2010. That rate of growth isn't much slower than Colorado's, though it starts out from a much lower baseline. What's missing in Arkansas, unlike Colorado, seems to be that synergy between diversity and education. Arkansas' most diverse county (Jefferson), for instance, saw its college-educated share rise from only 15 to 17 percent during that same period. The counties seeing above-average growth in college-educated persons, however, are the ones that are red and getting redder but are still mostly white: Benton County (which went from 14 percent to 26 percent in that period) and the exurban counties that are seeing more middle-class growth (Saline County, for instance, which went from 12 percent to 23 percent; compare that with more blue-collar Lonoke, which only went from 10 to 17).
State |
2010 |
2000 |
1990 |
CO |
12/5/16/4/62 |
11/7/18/4/61 |
10/10/15/4/61 |
AR |
39/7/4/1/45 |
43/9/4/1/43 |
41/11/3/1/39 |
Finally, I wanted to look at one more chart, not based on census data, since the agency doesn't track religion, but on information from another organization, the Association of Religion Data Archives, which collects this type of data by contacting churches during census years and getting membership data. Keep in mind these numbers don't reflect self-reporting and thus might miss people who might consider themselves religious but aren't regular members of a particular congregation, so take these numbers with a lot of salt. I'm not drilling down to the county level on this data, but it adds an additional wrinkle—religion—that doesn't get talked about often in demographic analysis. But it's one that should get talked about. The heavily evangelical Southern states' moves from the Democratic to Republican columns coincide just as much with the rise of the religious right, and the unprecedented attempts to link those churches' agendas with a political agenda, as much as they coincide with the civil rights era.
As you can see, there's not much change over the decades in the religious composition of these two states. (The five numbers in each box are, from left to right, the percentage of evangelicals, mainline Protestants, Catholics, "others"—which tends to be mostly Mormons, but also Jews and Muslims—and "unclaimed.") Like most of the rest of the country, both have moved in the "unclaimed" direction in the last few decades, mostly at the expense of the mainline Protestant churches. But you can see disparities that point to major differences between Colorado and Arkansas: Colorado is above-average in terms of how many "unchurched" residents it has, and thanks to its large Latino population, many of those who do belong to churches are Catholic. Arkansas, on the other hand, is one of the most heavily evangelical states in the country, and that, of course, colors its politics as well.