I came across this Family Friendly Cities blog post a while back, and of course I’m inclined to agree with the general conclusion. You hear a variety of reasons people want to move outward to the suburbs, and while there’s a nuance in the millions of decisions made across the metro (and country), my anecdotal understanding is the ranking follows the referenced ‘best cities’ ranking pretty closely:
- Safety
- Housing Cost (particularly, price per finished square foot and/or yard size)
- Schools
- Amenities for kids (or presence of other kids)
I’m sure there are other items people consider; proximity to at least one household member’s job, preference for suburban shopping, you get the picture. Maybe the order of importance shuffles around. But the gist is there, and I think whether many admit it or not, safety is more at the front of peoples’ minds than they may even be willing to admit. So that’s what I’ll be tackling today (though the links in the list provide some food for thought).
As discussed in the blog post, safety is obviously much more than crime. Crime is important, and definitely baked into the collective consciousness in how Americans view cities, but it’s not the only factor. It’s really surprising that we suffer this huge blind spot in risk assessment. Despite years of improvement in fatality rate for both per mile driven and per person in the United States:
… driving is still the leading cause of injury deaths for many age groups:
Source: CDC
Maybe we ignore it because we all think we’re above-average drivers. Maybe we just really like the utility of driving climate controlled vehicles. It’s easy to get caught up in the many benefits of cars and only consider the direct costs associated with owning/operating them (which is definitely less than the $10,000/year value touted in many urbanist blog posts, at least for most households in this country).
But again, the fact remains that the US is woefully behind other first-world countries when it comes to road safety, for drivers and non-drivers alike. Furthermore, those very countries are outpacing us in the drops in deaths. Some may point out that Minnesota is actually pretty safe for drivers, and I agree, but we’re still deadlier than most of those European countries.
Regardless, blogs usually go as far as that. Cite some national-level numbers and comparative death rates, call it a day. It’s not a stance that will win many over; it’s too easy to wave away blanket-statements saying they “aren’t true for *my* city or neighborhood.”
I’m going to unpack the data comparing some Minneapolis neighborhoods to suburbs to answer the question: Does living in a city and driving much less (or not at all), with all the crime that may come with it, outweigh the safety benefits of avoiding crime?
The Method
As I said earlier, there’s some heavy perception bias and willingness to ignore some data, on both sides. Living in a ‘safe’ suburb doesn’t mean you’ll never experience crime, and living in a city doesn’t exempt you from motor vehicle deaths or injury (or driving to the suburbs). Living in the suburbs doesn’t mean you’ll never ever visit the dangerous city and avoid that mugging – many work or party or catch sports or visit friends in the city from time to time.
I can’t evaluate that level of nuance with the data available, so I’ll only compare rates for driving and crime within each area. Side note, I’m using data from a bit of research I did over a year ago when I had this exact conversation with my father. I actually broke down crime by Minneapolis neighborhood and only included much of Southwest Minneapolis (right) and compared against what I would call “mostly safe” suburbs: Lakeville (my hometown), Burnsville, Prior Lake, Savage, Bloomington, Eden Prairie, Chaska, and Eagan. Go ahead and accuse me of cherry-picking nice Minneapolis neighborhoods with low crime. However:
My chosen Minneapolis zone includes several neighborhoods with concentrated areas of poverty, whereas the suburbs I picked are almost entirely without poverty. Additionally, my Minneapolis neighborhoods have a collective crime rate just 15% below the Minneapolis average. Anyone who knows the metro would say the suburbs I chose lean toward the rich and desirable side, so I’d say the comparison is apt to be in the suburbs’ favor (as a broad point, the median household income for Minneapolis was $45,000, the selected suburb average $75,000). They’re also not overly rural, so they won’t see as many super-high death rates from notoriously dangerous rural roads (though parts of Prior Lake, Lakeville, and Chaska do fall in this category).
I pulled crash rates from MN DPS by city to get a rate per person. This was only 2012 data, and given the small numbers (relative to crime rates) it would have been better to smooth it out using the average of a few years, but again, I don’t feel like it. I used an average of Minneapolis bicycle and pedestrian (non-occupant) death/injury rates and cross-referenced it with city data. I pulled 2 years worth of neighborhood crime data from 2011–2012 for Minneapolis and used CityData for the suburbs.
I wanted to understand the risk of death as well as the risk of death+injury. Deaths are easy: homicide or crash death (no suicides). Injury is more difficult, but I basically include any car-related injuries and rape, robbery, & aggravated assault for criminal injuries. If I were to guess, I’d say the average car-related injury is more serious (medical cost, bodily trauma, etc) than a criminal event, costs more (car damage deductible/etc vs any property loss), but I don’t have the data to back that up.
So, assuming you live in one of those (slightly safer than average) Minneapolis neighborhoods and never drive, you’re basically just as likely to die in a year as a person living in a suburb driving the average amount. Drive the average rate in Minneapolis and the suburbs beat you out. If you’re looking to avoid injury, even going car-free is still technically more dangerous than living in those suburbs.
Could someone live in Minneapolis and rarely, if ever drive? Yes. 19.7% of our households don’t have cars (though many may borrow one to get to work). 42% of our residents work in the city (with another 10% working in nearby St Paul). 55% work within 5 miles of their home.
Missing Pieces
It’s important to point out what was left out. Crime rates are dropping much faster in core cities relative to suburbs, nationwide. I suspect this is true in MSP. Teen suicide rates are inversely correlated with population density (which has been hitting home in Lakeville recently). There are long-term health effects from car-dependent lifestyles on children and adults. There are positive health/risk outcomes with quality cycling infrastructure bringing more users. Vehicle particulate emissions cause 55,000 premature deaths per year nationwide (exacerbated where highways run through urban areas).
Point being, urban areas could make up the gap with better street designs that we know are relatively inexpensive and have huge returns on safety. Retrofitting completely auto-dependent areas in suburbs will be much more challenging. Suburban crime is holding flat to rising while cities’ crime rates are dropping quickly. Better transit and vehicle electrification will make urban air quality better. If you’re taking the 30 year view, it is my opinion that Minneapolis/St Paul living is a far healthier and less risky environment than a typical MSP suburb.
great post! the nuance that you point out at the end shows how complex this can get.
i’d really like to see core cities improve their vehicular safety records. i think this would have positive impacts on crime rates too, as more people would be out on the street where we belong.
Thanks. I think I tend to do throw a bunch of other considerations at the end of posts more than I’d like to, but they need to be brought up sometimes. Nuance is important – there’s a great big world of super complex interactions. We can’t boil anything down to “density = crime” “roads = pollution” “cars = safe/unsafe” etc.
And I agree on your second point. I think a point I didn’t articulate on injury rates shows what some may already know – Minneapolis actually has more injury accidents per capita than more auto-dependent areas, but the death rate is lower. More accidents, less severe. That’s great for car occupants inside 50 years of safety engineering. But “more accidents” can still be very severe for non-occupants.
Thank you for presenting the data as per 100,000 residents. To many report only work with raw numbers which can distort the conclusions.
Have you taken the following into consideration?
http://www.mndaily.com/2004/04/22/nuisance-crimes-drop-during-strike
I am trying to understand the relationship between crime and public transportation but want to ignore anecdotal data. Have there been other strikes or interuptions in the US and a notable change in crimes (better or worse) that you are aware of?
I think, in general, there’s a big misconception that transit = crime. This article goes over some of the studies that mostly debunk that notion. http://www.citylab.com/crime/2014/12/the-myth-that-mass-transit-attracts-crime-persists-in-atlanta/383609/
This is a hugely problematic article. Firstly, it appears that the drop around bus stops was primarily in “nuisance crimes.” These are laws that ban things like loitering and which are pretty much used to target and harass men of color in particular; the Minneapolis City Council is looking at dropping these laws from the books for this reason. When waiting for a bus is technically a crime, people are less likely to commit the crime of waiting for the bus. The other drop in crime given is “drug dealing”–keep in mind that the MPD arrests black people 12 times more often than white people for pot-related offenses, despite identical usage rates. Both of these supposed “drops” in crime are highly suspect because of how tied they are to racial profiling.
Note that actual crime–assault, burglary, and bike thefts–increased around the city during the strike.
Pimps/prostitutes are repeatedly mentioned, but the article does not actually share that there was an in/decrease during that time–the article just repeatedly implies/explicitly says that bus stops are a haven for drug dealers and sex solicitation. Given the community meetings I’ve attended and what I’ve heard and read, as well as my own experience waiting for/on the bus, busses aren’t a popular place for paid sex (cars, on the other hand…). I suppose people who are selling/soliciting sex might end up at the same high-traffic corners that are also bus stops, but to link prostitution to busses makes about as much sense as linking traffic lights.
Ack. I meant to say that when waiting for the bus is a crime, when there is a bus strike, people are less likely to commit the crime of waiting for the bus.
While I found this post interesting, when I consider safety in where to live I do not think of traffic. I personally do not drive and as a female pedestrian I am more likely to consider areas I feel comfortable walking alone. Also for safety reasons I rule out any apartment that is ground level or with no secure entry.
Thanks, Casey. I regret that I didn’t take some time to call out that extremely important side to things. As a male, I have an admittedly very different view of safety when I walk around than a female does. Even the level of data discussed above misses that (major) nuance – many apologies.
One thought I did have about traffic, I prefer to walk on side streets than on busy streets like Lake or Lyndale. Much less harassment, exhaust smells and less cars to possible hit me. : ) Give me a nice neighborhood street with garden yards and friendly neighbors to give the Minneapolis nod hello over blocks with no sunlight and buildings right against the sidewalk.
Even cities where all the buildings are against the sidewalk have quiet side streets… quiet side streets are a very good thing.
Nice post. Since homicide where the perp and victim do not know each other is quite rare, it would be interesting to see the results if you could factor it out.
Great read and I agree it makes sense to take a wholistic approach to safety in making decisions on where to buy a house. Another nuance that came to my mind in reading this was that the true risk of death by homicide in both the cities and suburbs is probably lower than how it shows up on the chart, since in some homicides the victim and assailant know each other. Not to suggest that we don’t know people who might end up killing us, but it seems less likely.
Awesome post! I choose to live in the city and to not drive because all the research I’ve done over many years indicates both as safer, healthier, and cheaper for me AND my community in both the short and long term. I realize that our current infrastructure design and priorities increase the risks of biking/walking, but I take steps to mitigate that personally as well as do help engineers and planners understand UX/BP for those outside cars to improve design.
Curious if you have data or are making assumptions about the percent of households without cars who borrow one to get to work? I would be very surprised if this were the case. I know a number of carless households and not one of them borrows a car for work. On the other hand, I know a number of households with cars where individuals commute via bike/foot/transit and where the car goes unused for weeks at a time.
The zero-car households borrowing cars is part anecdote, part assumption based on larger data trends. The major source I used to inform my thought: http://www.brookings.edu/~/media/research/files/papers/2011/8/18-transportation-tomer-puentes/0818_transportation_tomer
While many carless households may have good job/amenity transit access, invariably there are some who work in a suburban location with terrible or no transit access. I used the term “many” perhaps a bit loosely – “some” is a better term. I should have included the caveat that some may also be carpooling.
Related, this post shows that while Chicago has a similar number of car-free households, an even larger number of *workers* don’t have access to a car (1 car households with 2+ workers, etc). I would guess Minneapolis has a similar situation. http://danielkayhertz.com/2015/01/20/how-many-carless-workers-are-there-in-chicago/
Scanned the first article, read the second. I didn’t see either of them talking about borrowing cars/carpooling, though I am sure it happens. My thought is that here carlessness is geographically concentrated. I believe 40% of households in NoMi are carless–additionally, I know that my own social circles skew towards carlessness at a higher rate than would be expected. This makes it seem likely that access to a car clusters both geographically and socially, as does carlessness.
Anecdotally, I worked for a few years 90 minutes or 2-3 busses away from home. In the mornings, I carpooled part-way with a friend, then caught the bus for the remainder (the time saved was minimal with traffic; the benefit was social). In the evenings, I always bussed the full 1.5 hour home. In my experience, few adults are willing to commit to carpooling, especially regularly. I can’t think of anyone I know in Minneapolis, besides me during that brief period, who has carpooled even infrequently. The allure of the automobile is a highly individualistic promise of “convenience” and being able to hit “the open road” on a whim. Additionally, factor in unpredictable schedules, salary work, changing shifts, other errands, etc., and I would expect carpooling to fall far far far below other carless options, especially in the absence of dedicated HOV lanes and major congestion.
Just a data point, 9% of Minneapolis residents carpool to work (2011 ACS 1-year estimate). I would guess most of those people are spouses or roommates who work near each other. But I have to believe some are carpooling with other families in a car-share type situation.
As to the Brookings piece, page 3 says:
“These metro areas’ commuting statistics reinforce this locational advantage—59.7 percent of zero-vehicle households in the city use mass transportation as their commuting mode. This dwarfs the 25.4 percent transit share for zero-vehicle households in the suburbs. This implies that three-quarters of zero-vehicle households in the suburbs need an alternative mode to get to work. Invariably these households borrow someone else’s car and drive alone (31.1 percent share) or carpool with someone else (30.2 percent share)”
This is, of course, talking about people in the suburbs borrowing cars in specific, but I have to believe the 40% of zero-vehicle households in cities not using mass transpo has *some* share of similar behavior (beyond the share of folks walking, biking, working from home, etc). That assumption may be false, but just explaining where it came from.
Thanks for pulling that info for me!
Do you have info on how carpooling is defined? (Frequency, totality of ride (i.e. is a drop-off at a bus stop counted?), length of trip, etc.) I assume the 9% is in a sample that would add up to over 100% because of multimodal choices.
I appreciate the tangentially related conversation!
Yeah, I bet car-borrowing or sharing is more common than car pooling. I generally bike or take transit to work, but I can only do that because I have the option of the car as a backup when the site, timing, or childcare shifts make it necessary.
And when I lived way out in Big Lake, I carpooled to the park and ride, but it was purest chance that I lived “near” (in exurb terms) a coworker with my same shift.