"That blog isn't called Smart Politics for nothing."
- POLITICO
"Smart Politics is by far the best Minnesota political blog.
I check it every day and encounter much good information
and analysis that I can get nowhere else."
- Steven E. Schier
Dorothy H. and Edward C. Congdon
Professor of Political Science
Carleton College
A Smart Politics analysis of the recently released 2008 Uniform Crime Reports finds that red states across the nation have both higher violent and property crime rates than blue states, across several measures of partisanship.
The average violent crime rate (murder, forcible rape, robbery, aggravated assault) in 2008 for the 28 states that voted for Barack Obama in the 2008 Presidential election was 389 incidents per 100,000 residents. The average violent crime rate for the 22 states that voted for John McCain was 412 incidents per 100,000 residents - or a 5.8 percent higher incidence of violent crime.
For example, 2 of the top 3 states with the highest violent crime rates in the nation in 2008 voted for McCain: South Carolina (#1) and Tennessee (#3). (Nevada was #2).
The difference was even more pronounced for property crimes (burglary, larceny-theft, and motor vehicle theft). Obama states had an average property crime rate of 2,989 incidents per 100,000 residents, with McCain states averaging a rate of 3,228 - or an 8.0 percent higher incidence of property crime.
Eight of the top 11 states with the highest property crime rates voted for McCain: Arizona (#1), South Carolina (#2), Alabama (#4), Tennessee (#6), Georgia (#7), Texas (#8), Arkansas (#10), and Louisiana (#11).
These crime rate findings hold despite the fact that blue states have a higher population of residents in urban areas, which tend to have higher crime rates than rural areas. According to the U.S. Census Bureau's 2000 Census of Population and Housing, Population and Housing Unit Counts, the average statewide percentage of residents living in urban areas in the Obama states was 78.0 percent, compared to a statewide average of just 64.6 percent in the McCain states.
The red state/blue state crime data split also holds true across other measures of statewide partisan groupings.
For example, a Smart Politics analysis of partisan control of state legislatures finds the 27 states with Democratic-controlled legislatures with an average violent crime rate of 390 incidents per 100,000 residents. The average violent crime rate for the 14 states with Republican-controlled legislatures was 11.1 percent higher, at 433 incidents per 100,000 residents. (The rate was lowest among eight states with split partisan control - at 382).
There was also a double-digit percentage difference for property crime rates among the states with Democratic and Republican controlled legislatures. For Democratic-controlled states, the property crime rate was 3,044 incidents per 100,000 residents compared to 3,351 incidents per 100,000 residents for Republican-controlled states - or a 10.1 percent higher rate under GOP legislative control.
The differences in the rate of violent and property crimes between states along partisan lines by control of the governor's office were less stark, but still pointed in the same direction. The 22 states with Republican governors had a 0.4 percent higher violent crime rate in 2008 (400 incidents per 100,000 residents) than the 28 states with Democratic governors (398) as well as a 6.0 percent higher property crime rate (3,196 for GOP states and 3,014 for Democratic states).
So here is the chicken and egg question: are states with high crime rates electing Republicans because the GOP is perceived to be tougher on crime and thus are more likely to take action to fix the state's crime problems, or are Republican policies to combat crime proving less effective than Democratic policies and thus resulting in higher crime rates?
One thing is for certain: 2008 is not an aberration.
Looking back to the 2004 Presidential election, the 19 states that voted for Democrat John Kerry had an average violent crime rate in 2004 of 361 incidents per 100,000 residents. The 31 states that voted for George W. Bush had an average violent crime rate that year of 419 incidents per 100,000 residents - or a 16.3 percent higher rate. Bush states also had an 18.6 percent higher rate of property crimes in 2004 (3,648 incidents per 100,000 residents) than the Kerry states (3,077).
There are, to be sure, many other variables to be considered other than partisanship when examining the different rates of crime between states. For example, red states tend to be less affluent than blue states. The average statewide per capita income in 2008 for the 28 states voting for Obama was 19.4 percent higher ($45,752) than in the 22 states voting for McCain ($38,333).
The per capita income difference was still present, although less pronounced, when grouping states by partisan control of the legislature and the governor's office. States with Democratic-controlled legislatures have an 11.1 percent higher per capita income ($44,470) than states with Republican-controlled legislatures ($40,018). States with Democratic governors had a 2.5 percent higher per capita income in 2008 ($42,955) than those with Republican governors ($41,892).
Red State vs. Blue State Crime Rates by Different Partisan Groupings
Partisan grouping |
Dem |
Rep |
% Difference |
2008 Pres. Vote: violent |
389.2 |
411.8 |
+5.8 |
2008 Pres. Vote: property |
2,988.8 |
3,228.0 |
+8.0 |
2004 Pres. Vote: violent |
360.5 |
419.4 |
+16.3 |
2004 Pres. Vote: property |
3,076.6 |
3,647.5 |
+18.6 |
Legislature: violent |
390.0 |
433.3 |
+11.1 |
Legislature: property |
3,043.9 |
3,350.6 |
+10.1 |
Governor: violent |
398.4 |
400.0 |
+0.4 |
Governor: property |
3,014.1 |
3,195.7 |
+6.0 |
Follow Smart Politics on Twitter.
Comments
It is clear to me that the red states have citizens that take personal initiative and self reliance very seriously. They have a criminal element that is not afraid to go to work. As such they "pull themselves up by their boot straps and "getter done"....
Where as the blue states have a more socialist attitude. As such the criminal element is not as motivated to "getter done" They sit at home and wait for the mail..... (lite snark)
Posted by: being goode | September 16, 2009 6:41 AM
Interesting anaylsis that flies in the face of a frequently emailed mostly false urban legend. The email based on it has been updated for the 2008 election and used as fodder for letters and blog posts.
Despite Hamline law professor Joseph Olson repeated objections to the emailed item falsely attributed to him, there doesn't seem to be any stopping this one. Nonetheless, thanks for adding another fact-based analysis for us legend watchers.
The earlier version of the tale was published as a LTE in a Jackson County, MN newspaper in 2008, as I noted on Bluestem Prairie at the time of publication.
Posted by: Sally Jo Sorensen | September 16, 2009 7:33 AM
It is good that you point out this is a chicken-and-egg problem. Only an idiot would claim to know the answer to that question, so I can assume we will be hearing lots of talk show hosts who claim to know. In the meantime I find the relationship between voting and mean income at least as interesting. According to some rhetoric, all the states with people earning income should be red as their harder working, more self sufficient folks vote for good old red-state lower-taxes-at-any-cost politicians. Your numbers seem to suggest otherwise.
Posted by: Gary Lee | September 16, 2009 9:58 AM
"There are, to be sure, many other variables to be considered other than partisanship when examining the different rates of crime between states."
Gosh, d'ya think?
Seriously - you (pl?) omitted a VERY key bit of context in selecting your sample set for this "analysis". One that skews the numbers beyond any usefulness in comparing party affiliation and crime in any meaningful way, and makes any conclusions drawn (as you've drawn them) downright irresponsible.
Care to guess what it is?
Because I'm going to post about this tomorrow.
Posted by: Mitch Berg | September 16, 2009 10:49 AM
I lied. I posted the first of three parts of my response today:
http://www.shotinthedark.info/wp/?p=5425
Part II tomorrow, Part III Friday.
Posted by: Mitch Berg | September 16, 2009 5:49 PM
> I posted the first of three parts of my response today:
Mitch at Shot in the Dark does raise a legitimate point that the UCR cautions against making some comparisons, particularly from community-to-community, but statewide comparisons are also mentioned. However, the UCR isn't suggesting that no comparisons can be made. For example, the number of murders in a given state (one of the four components to the violent crime measure) is a fairly known quantity, and therefore, one can express with confidence that the homicide rate in, say, Nevada, is much higher than that of North Dakota.
Moreover, almost all of the data presented in this article (there are a few states used as illustrations), is aggregate data -- combining states into general categories along dichotomous lines ('red' and 'blue'). The process of aggregating, statewide, and, in this case, across many states, tends to reduce the statistical significance that variations local law enforcement agencies may have in the reporting of crime in their communities on the UCR data.
Posted by: Eric Ostermeier | September 17, 2009 12:21 AM
Eric,
Your point about aggregate data is a valid one.
So my question is not just why use aggregate data...
...but why choose the aggregation that you chose?
I ran Part II this morning. Part III is tomorrow.
Posted by: Mitch Berg | September 17, 2009 9:43 AM