See Red, Turn Blue?

The Congress of New Urbanism’s Urban Transect.  What people don’t realize?  As they move outward on the transect, it moves with them.

Since the presidential election earlier this month there have been tons of positions taken on how and why the Republicans lost in a political climate that was favorable to them.  All sorts of perspectives have been tossed about: Mitt Romney’s affluent background and comments about the 47%; the GOP alienated Latino voters with their immigration policy; the GOP has a broader minority voter problem.  All have played a role in the GOP defeat.
But lately a few perspectives have arisen that might reflect broader systemic problems that the Republican Party must contend with.  This article in the Atlantic Cities, which was really just a summary of this article in the blog Fueled By Randomness both touch on an interesting phenomenon – as population density increases, voting for the GOP generally declines.  Put another way, the Democratic Party is increasingly becoming the party of large metro areas, and the Republican Party is increasingly becoming the party of small metros and rural areas.
This wasn’t always the case.  It seems that throughout much of the ‘80s and ‘90s I would hear that the emptying of cities and related suburban expansion within metro areas would ensure GOP control at the state and national levels.  Republicans had long enjoyed an advantage in rural areas, and with a suburban demographic that shared many values with the exurban and rural base, they were ready to build on that foundation.  Republicans rode this wave over the next two decades.
But as suburban expansion continued, its makeup shifted and communities started shifting with it.   Some communities began diversifying their housing stock to meet shifting demands.  Commercial strips began housing stores that catered to the various ethnic groups that were moving in nearby.  The new suburbanites were asking for government services – improved and expanded infrastructure, maybe transit access – that were inconsequential to the early arrivers. 
The urbanization of the suburbs had begun, and the seeds of the GOP’s current dilemma had been planted. 
A lot of this began taking place at the local level around the country in the ‘90s, maybe even earlier in some states, and accelerated in the last decade.  Two counties come to mind that seem to exemplify this transition – DuPage County, Illinois and Oakland County, Michigan.  The affluent suburban counties of Chicago (DuPage) and Detroit (Oakland) had been solidly – if moderate – Republican counties through the middle of the last decade.  However, they began to shift slightly leftward in recent years, as their suburban development matured. 
This has had an impact on state politics where this has occurred, and ultimately on national politics.  Prior to 1990, for example, Illinois state politics could be described as being one-third solidly Democratic in Cook County (where Chicago is located), one-third moderately Republican in the suburban collar counties (including DuPage mentioned above), and one-third solidly Republican throughout much of Downstate Illinois (islands of small Democratic cities like Rockford or Peoria notwithstanding).  The overall result on a state level was a slight Republican advantage that could deliver Illinois electorally to the GOP.  Since then, the Cook County and Downstate political profile has stayed the same, but the suburban Chicago profile shifted from slightly Republican to slightly Democratic, and that has since turned the state from red to blue.  The state legislature has held a Democratic majority now for years, and the last Republican to win Illinois in a presidential election was George H.W. Bush in 1988.
So the question in my mind is, how is metro area growth and diversification impacting national politics?  I did a little research to assess this.  Using 2010 U.S. Census population data for states and metro areas with over 500,000 people, I wanted to see if there was any relationship between a state’s population within large metros and voting for President Barack Obama.  Knowing that many metro areas cross state boundaries, I assigned populations accordingly.
The results are fascinating, and you can see them below:
States won by Obama
States won by Romney
State
2010 Pop.
2010 MSA Pop.
MSA Pop. %
RI
1,052,567
1,052,567
100.0%
NY
19,378,102
19,000,193
98.0%
MD
5,773,552
5,014,359
86.9%
CA
37,253,956
31,313,417
84.1%
CT
3,574,097
3,000,267
83.9%
AZ
6,392,017
5,182,456
81.1%
MA
6,547,629
4,826,978
73.7%
IL
12,830,632
9,290,273
72.4%
PA
12,702,379
9,179,122
72.3%
NV
2,700,551
1,951,269
72.3%
VA
8,001,024
5,680,663
71.0%
OH
11,536,504
8,184,448
70.9%
HI
1,360,301
963,607
70.8%
TX
25,145,561
17,752,786
70.6%
FL
18,801,310
13,261,655
70.5%
NJ
8,791,894
6,044,011
68.7%
MN
5,303,925
3,490,538
65.8%
CO
5,029,196
3,259,823
64.8%
DE
897,934
538,479
60.0%
OK
3,751,351
2,190,465
58.4%
TN
6,346,105
3,693,914
58.2%
WA
6,724,540
3,884,293
57.8%
GA
9,687,653
5,508,536
56.9%
MI
9,883,640
5,059,992
51.2%
OR
3,831,074
1,789,580
46.7%
KS
2,853,118
1,324,745
46.4%
NH
1,316,470
609,519
46.3%
LA
4,533,372
1,970,248
43.5%
NM
2,059,179
887,077
43.1%
NE
1,826,341
753,965
41.3%
UT
2,763,885
1,124,197
40.7%
NC
9,535,483
3,757,634
39.4%
ID
1,567,582
616,561
39.3%
SC
4,625,364
1,814,804
39.2%
ME
1,328,361
514,098
38.7%
IN
6,483,802
2,486,638
38.4%
WI
5,686,986
2,124,501
37.4%
MO
5,988,927
2,126,691
35.5%
AR
2,915,918
709,901
24.3%
KY
4,339,367
1,031,130
23.8%
AL
4,779,736
1,128,047
23.6%
IA
3,046,355
692,778
22.7%
MS
2,967,297
539,057
18.2%
WV
1,852,994
77,462
4.2%
AK
710,231
0
0.0%
MT
989,415
0
0.0%
ND
672,591
0
0.0%
SD
814,180
0
0.0%
VT
625,741
0
0.0%
WY
563,626
0
0.0%
A few observations.  First, while the 102 metros with more than 500,000 residents make up 63 percent of the nation’s population, on a state-by-state basis the average percentage is 48 percent, due to the fact that several states have no large metro at all.  Of the 19 states that have 60% or more of their population in large metros, 17 voted for Obama (90 percent).  Of the 19 states that have 40% or less of their population in large metros, 15 voted for Romney (nearly 80 percent).  Of the 12 states with between 40-60 percent of their population in large metros, seven voted for Romney and five voted for Obama.
There appear to be some regional variations, but the overall trend seems clear.  In the Northeast, the most urbanized part of our nation with 65 percent of residents living in large metros, all nine states voted for Obama.  In the Midwest the President won six of 12 states, yet won all states with a large metro population percentage above 50 percent (Illinois, Ohio, Minnesota and Michigan).  In the West Obama won seven of 13 states, and won seven of the nine with large metro population percentages greater than 40 percent (California, Nevada, Hawaii, Colorado, Washington, Oregon and New Mexico).  Even in the South, where the President won just four of 16 states, he won four of the five with a large metro population percentage greater than 60 percent (Maryland, Delaware, Virginia and Florida).  True, each of these states could be considered just nominally Southern by many, but the trend holds.
What could it mean in future elections?  Arizona and Texas are the outliers here; 81 percent of Arizonans and 71 percent of Texans are in large metros, and they both went strongly for Mitt Romney.  One explanation could be that both states have large concentrations of young, unregistered voters, largely minority, who have yet to alter the current political dynamic.  Another explanation could be that the low density nature of Arizona and Texas metros may extend the viability of Republican control for the forseeable future.  Oklahoma, Tennessee and Georgia are next in line as state with large metro populations that still voted Republican in 2012, but it remains to be seen if they experience any transition as well.  Other influences, like migration patterns, culture, and religion, might accelerate or stall any shift. 
Other questions require further investigation.  Has this always been the case?  A historical analysis could address that.  Also, a more extensive exploration of the impact of density might reveal some interesting patterns.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s