Apartments go boom!

With all the chatter about apartment construction in the last couple months, I wanted to see whether the current level of activity is really an aberration or just a way to sell newspapers.  There certainly are a lot of proposals floating about, but after the severe downturn of the last few years, it’s hard to know what’s normal.  Besides, is it a coincidence that the paper that says the most about the new construction just happens to have a subscribers only online map of it?

But what exactly is the current level of activity?  If 8500 units were under construction or proposed as of September, what does that mean in terms of eventual places to live?  Presumably almost all of what is under construction will be finished, but much of what is proposed will never see the light of day.  I thought it safe to compare the number proposed to building permit data, although probably a bit more is proposed than actually gets permitted.  From F&C’s 8500, I thought it safe to subtract 2000 considering my count of 1,732 units that were under construction in 2011 just in Minneapolis (most of which would have been permitted in 2010).

So how rare is it for 6500 multifamily units to be permitted in the metro area?  Met Council data going back to 1970 gives us a hint:

Permit rainbow!

Well, not very rare.  10 of the last 40 years saw 6500 units permitted.*  In fact, in the 70s the average year saw 6100 multifamily units permitted, helped out by the massive years 1970-1972 that permitted over 10,000 multifamily units each.

I think the story here is more likely that not nearly enough rental units were constructed in the last 20 years.  As I’ve noted before, the 90s were a disastrous decade for dense development.  When multifamily heated up again, it was the condo craze, leaving little room for renters.  But while the ownership housing stock was increased, sometimes at the expense of rental housing, the number of renter households was increasing faster than the number of owner households.

In other words, it’s true that there is more multifamily rental units being proposed and built than in recent years, but don’t think of it like a speculator-driven bubble.  Instead it is more likely to be a “new normal,” where the market is providing a supply in reaction to demand.  That’s good news for people who want an energy-efficient, walkable, low-maintenance place to live.

Zooming In

There’s more news buried in this building permit data, and I’m going to finish up with a long digression on it.  Check out this table of the top 10 metro area cities for total residential building permits issued between 1970 and 2010:

City DTQ DUP MF3 MF5 SFD TH Total
Minneapolis 98 854 21507 6543 4238 2807 36047
Plymouth 44 118 9172 240 13925 3273 26772
St. Paul 146 652 16033 3234 4537 1172 25774
Eden Prairie 32 1112 6498 95 12132 4430 24299
Eagan 75 704 6886 68 12430 3710 23873
Woodbury 107 442 4836 1127 12490 4123 23125
Maple Grove 157 565 3646 692 14496 3473 23029
Brooklyn Park 16 834 2750 475 12820 3504 20399
Burnsville 0 366 8431 394 7472 3305 19968
Coon Rapids 2 736 4760 48 9106 3462 18114

DTQ=Duplex, Triplex, Fourplex

DUP=Duplex

MF3=Multifamily (3 units or more)

MF5=Multifamily (5 units or more)

SFD=Single-Family Detached

TH=Townhome

Minneapolis has a comfortable lead, appearing to have added more housing units in the 40-year period than any other municipality (assuming the same rate of actual construction resulting from permits across all the municipalities and years).  This lead seems to have primarily resulted from the 70s and 00s, in the latter of which Minneapolis added significantly more housing units than anywhere else in the Twin Cities.

Seas of purple and green

So how could Minneapolis add tens of thousands of housing units in the last 40 years, while simultaneously losing more than 50,000 residents?  Some of the reasons for a similar effect in the 60s are also valid for later decades; the entirety of the drop occurred in the 70s, when a great deal of (edit: Froggie points out in the comments that most of the clearance for freeway construction had been wrapped up by the 70s) freeway construction and some slum clearance was still underway.

Later decades fared better.  The 90s saw a population increase; looking at population by sector makes it clear that the mediocre performance of the 00s was almost entirely a product of the foreclosure crisis:

Sector 1980 1990 2000 2010
Downtown 19155 21824 24977 31034
North 61278 64001 67674 59970
Northeast 37507 36515 36913 36255
South 137551 136333 142150 139854
Southwest 83728 79912 78292 77989
University 29615 29798 32612 37476
Citywide 368834 368383 382618 382578

The two sectors with the most foreclosure activities were also the only two with significant population decline.  In the case of North, two decades of steady growth were wiped out.

The 80s are the mystery for me.  Seven or eight thousand units were constructed in Minneapolis, which should have resulted in some population growth.  Instead the most population growth occurred in North, not in the Downtown and University neighborhoods that saw the most units added.  I don’t have demolition permit data, so I don’t know if an unusually high number of units were demolished.  Household size may also have been a factor, since many of the units added were likely smaller than any units lost.

Regardless of what happened in the 80s, the census data seems to suggest that, barring any new freeway construction or popular predatory lending practices, Minneapolis should see steady population growth in this decade.  Wandering back to the main topic of this post, the return to historic levels of multifamily rental construction, a greater proportion of which tends to occur in central cities, is another indicator that the chatter may soon be about how Minneapolis and St Paul are leading the metro in population growth.

*Until 2004, semi-detached units with more than two units were counted in the multifamily category.  In 2004, they were moved into the Duplex category.

Traveling in Moderation, part I: U of W/M

City and Lakes

For the last three years I’ve traveled to Madison over the Thanksgiving weekend to accompany my girlfriend on a visit to her grandmother.  Grandma Dee was born and raised in Madison, and has proven to be an excellent source for the history and culture of the city (beer and football, mostly).  In the course of these travels, I’ve accumulated some observations about Madison that I’d like to share.

This may be the inaugural post of an occasional series documenting my various Upper Midwestern excursions.  I travel fairly often but thanks to a combination of full-time employment and neurotic antipathy toward air transportation, my travel is mostly limited to Minnesota and neighboring states.  Madison is a particularly suitable city to kick off this series since it has implemented a number of experimental streetscaping techniques.  I’m going to start off with something more basic, though:

Why does UW feel so much more urban than the U of M?

Don't fence me out

The Twin Cities metro is around six times larger than the Madison metro, but somehow the UW campus feels urban in a way that the U of M doesn’t.  Madison’s main shopping street is State Street, which gradually accumulates more and more academic function until it terminates at the University’s Bascom Mall.  This side-by-side, close-knit nature is in contrast to the U of M, which literally fences itself off from Dinkytown.  Only a handful of University uses penetrate the half-mile perimeter trench that is University Ave between 11th and 17th Aves, and while everyone thinks of Dinkytown as the University Neighborhood, it doesn’t look terribly different from any other Minneapolis neighborhood if the streets happen to be deserted of the maroon-clad denizens.  The West Bank and St Paul campuses are a bit more integrated with their surrounding neighborhoods, in that they’re only separated by a broad lawn or parking lot rather than an actual fence.  Probably the area that is most integrated with its surroundings is Stadium Village, which is gradually being annexed by the University.  There you’ll find a few commercial buildings sharing a block with the University’s IT department, for example, in a coziness that wouldn’t be out of place in Madison but which the U of M apparently finds uncomfortable, as evidenced by their decades-long effort to demolish the neighborhood.

College kids getting high

But it’s not just proximity to the city that makes UW feel urban – even when you can’t see any building without a UW logo on it, you often still feel like you’re in a city.  The reason is right above you – buildings on the UW campus are tall.  UW has a cool interactive campus map tool where you can click on any University building and there will be a tiny little sketch of it, which gives you a sense of the heights of campus buildings (bing works too).  I encourage you to look around on those mapping sites, because the best confirmation I could find for my perception was Emporis, which lists 71% of UW buildings as being more than 6 stories as opposed to only 16% of U of M buildings (including St Paul).  The caveat?  Emporis only lists 34 UW buildings, but they list 102 U of M buildings.  So it may give a truer picture of the U of M campus than the UW campus.

Too close for comfort

Besides height, it seems like UW’s buildings have narrower setbacks, which reinforces the street wall and gives a more urban feel.  This first came to my attention with the Pres House apartments, only 10 feet from their namesake church, but neither of those are official campus buildings.  Still, there are plenty of buildings on the UW campus that are 30′ apart – too many to list here.  They would likely no longer be standing if they were on their western counterpart campus; the U of M tore down Wesbrook this summer for the crime of standing 35′ from Northrup.  And many of the close-standing UW buildings aren’t as ancient as Wesbrook, suggesting the UW administration doesn’t think an urban campus is a bad thing.

Or were they just drunk when they signed off on the site plan?  What accounts for the differences between the campuses?  Why does the U of M seek out a simple, park-like atmosphere while UW is content with the complex geometry of an urban campus?  I have no idea, but  wild guess is that geography was a prime contributor – UW’s location very near to Madison’s downtown and smack in the line of a primary growth axis for that city both restrained campus expansion (UW is now about a third of the area of the U of M, though they were likely originally about the same size) and allowed denser buildings to fit in with the surroundings.  The U of M’s more suburban location allowed for easier campus expansion and required more suburban building styles to match its streetcar suburb neighbors.

But I’d like to throw out a wilder guess:  I’ve noticed development throughout SE Wisconsin that seems denser that comparable developments around the Twin Cities.  Buildings seem taller, closer together and more fancifully adorned – while most of this is within a suburban context; by which I mean what in the Twin Cities would be a football field-sized parking lot is a soccer field-sized parking lot in SE Wisconsin.  (A small distinction, maybe, but I’ll take what I can get.)  Could this be the influence of that nearby modern megalith, Chicago?

On the other hand, maybe I’m just reading too much into the sheen that often accompanies new sights.  Maybe a Madisonian visiting the U of M would make similar observations.  Maybe I was just thrown off-balance by the presence of hills.  In that case, expect a couple more posts of unreliable observations, including one touching on a bike facility that makes a cameo in one of the above pictures.

St Louis, Stop Spacing, and the Future of the 50s

Stop spacing in NW Minneapolis

Riding the bus is slow.  It is sometimes vein-bulgingly, pencil-snappingly slow.

Since I prefer riding transit, I choose where to live based on the quality of the transit service, and so a few years ago I moved to Kingfield because of the 18 line, one of the most frequent in town.  It didn’t take long for me to move away because despite the 18′s frequency, it takes forever to get anywhere – specifically it takes 27 minutes at rush hour to travel the 3.2 miles between 7th St and 38th St.  At about 7 mph, that’s not much faster than walking (well, it’s twice as fast as walking, but counting wait times and assuming typical delays, it’s usually only 10-15 minutes faster and it’s not uncommon that it’s slower).

In an effort to speed things up, St Louis is eliminating bus stops.  They predict their effort “will help keep buses on time, while saving fuel and maintenance expenses.”  Based on the blog entry, it appears their spacing standards didn’t change, they’re just enforcing them for the first time.  Here are the standards:

Local Service
Stops located at major intersections, major traffic generators, and where bus routes or rail lines cross
Stops located in high populated areas every 1/8 to 1/4 mile apart
Stops located in lower populated areas every 1/4 to 1/2 miles apart
Express Service – Limited Stop
Express routes over local service in high density areas should be located approximately 1/3 to 1/2 mile apart

It’s interesting that their policy recommends closer spacing in denser areas.  While it’s logical to include more stops to serve more people, when actual stops are on demand you risk less by allowing people to stop more often in low density areas.  In addition, there is a limit to how far people will actually walk, and as Jarrett Walker mentions in that link suburban areas tend to have less connected street networks that require even more walking.

St Louis block size map

That reminds me – this blog isn’t called Getting Around St Louis.  How does St Louis’ policy compare to Metro Transit in Minneapolis?  Well, for one, it’s hard to find any of Metro Transit’s policy documents.  You can find some information on their website, but only through the magic of google – you can’t seem to navigate to any policy information on their website and it isn’t on their site map.  I’m not sure that using a blog is the best way to publish policy, and St Louis doesn’t seem to have any more policy information on their site, but a blog is a good way to solicit comments (and they’ve flooded in on this issue) and update on the progress of a project as it happens.

A little googling reveals that Metro Transit’s stop spacing policy recommends a stop every eighth of a mile.  (Edit: Commenter Charles linked to the 2030 Transportation Policy Plan, which has this to say about stop spacing:

Recommended Bus Stop Spacing
Bus stops that are close together reduce walking distance and access to transit, but tend to increase bus travel time. This recommended spacing seeks to achieve a balance.

• 6-8 stops per mile for local service
• 1-2 stops per mile for limited stop service

An allowable exception to standards may be central business districts and major traffic generators. These guidelines are goals, not a minimum nor a maximum.

While I admire a policy that defers to real-world conditions, I have a hard time believing there is a situation that would justify a stop less than an eighth of a mile from another stop.  The only possible exception is the disaster that is Lake & Nicollet, but that situation could and should be mitigated by reconfiguring the bus routing and street designs.)

As the map linked at the top shows, many of the east-west streets are spaced every sixteenth of a mile.  Considering most routes in Minneapolis lie no further than a half-mile from the nearest parallel route, I think it’s reasonable to recommend spacing every quarter mile on most routes.  For example, if stop spacing on the 4 and the 18 were reduced to every quarter mile, the maximum someone would have to walk to a stop north of Lake St would be .37 miles.

St Louis actually seems to have larger blocks than Minneapolis, which I’m assuming is mostly a result of more urban renewal.  Larger blocks actually require closer stop spacing as the street network requires more walking.  Minneapolis, on the other hand, has a relatively intact grid network.  This will allow wider stop spacing since the grid network and small blocks shortens walks.  (Note:  I basically added this paragraph because I wanted the maps in this post.  I just screen captured them from the H+T Affordability index, which has the most user-friendly and in-depth demographic mapping I’ve seen.)

Minneapolis block size map

Stop spacing is something I’ve griped about before, and wider stop spacing shouldn’t exactly be considered state of the art.  The Citizens’ League already called for quarter-mile stop spacing on north-south streets and eighth-mile stop spacing on east-west streets… in 1956.  They exempted Downtown (then called the Loop) from their proposal, but that’s actually where Metro Transit has done the most work with stop spacing.  Metro Transit has also looked at stop spacing in their sector restructuring studies, which began around 1998 and I think has been completed in 4 of the 8 sectors.  However, their recommendation of a stop every eighth of a mile is half that of the rest of the world.  It’s time for Metro Transit to join St Louis, the 50s, and the rest of the world and start spacing bus stops every quarter mile.

Feat of feet of street

The brilliant blog Mapping the Strait posted an infographic yesterday comparing the feet of street per resident of 8 American cities.

The metric is supposed to give an indication of the amount of infrastructure per resident, to augment standard persons per area measures of population density.

According to the Design Guidelines for Streets and Sidewalks, Minneapolis has 1,423 miles of roads and vehicle bridges, not counting freeways.  My rough Google Earth measurement of freeways within city limits is 30.3 miles (that includes the part of 62 on the border but does not include highways 55 & 121 because I think they are in the city’s measurement, although that’s just a guess).  That makes for 7,673,424 feet of streets and highways, or 20.1 feet for each of the 382,578 residents counted in the 2010 census.  We’re closer to Detroit, Phoenix or San Antonio than Philadelphia, Los Angeles, or Chicago on this count.

That doesn’t seem to be an unreasonable result to me, although by measuring residents only you ignore the significant market for infrastructure represented by workers.  In that case cities such as San Antonio or Houston that contain most of their employment catchment area in their city limits are going to be more accurately portrayed by this metric.  One of the commenters at Mapping the Straight asked for this metric by area of paved surface – I think using lane feet would be better than centerline feet, but probably less widely available.  Fun to think about anyway.

Gotta go to work, gotta have a job

The series of tubes spat out a couple of reports this week about employment and access to transit.  Both of them contain useful specifics about Twin Cities employment patterns, but seem to disagree whether our region is keeping up with other regions in terms of accessing jobs through transit.

The first report was put out by the Center for Transit Oriented Development, which not long ago produced a practical guide to increasing walkability in certain Twin Cities neighborhoods.  Their new white paper has the simple title “Transit-Oriented Development and Employment” and may be most useful as a guide to research detailing the relationship between transit and employment geography for those of us who don’t have the time or money to read the scholarly journals.  The paper also contains a brief case study of 3 metros:  Phoenix, Atlanta and the Twin Cities.

Brookings begs to differ

As this graphic shows, Twin Citians on transit can access a pretty high number of jobs relative to some other American cities.  This even though CTOD rates our transit system as medium, despite the fact that it will be barely longer than Phoenix’s 20 mile “small” system even after Central is finished, at which point its 24 miles will be half the 49 miles of Atlanta’s MARTA, also rated “medium.”  They apparently are counting the 12 Northstar trains a day towards our total, even though they provide less than 1% of the system’s total weekday trips (contrast with Hiawatha, which provides 11%).

But the map shows that the Twin Cities’ soon-to-be light rail system will provide access to 19.6% of the region’s jobs.  While that seems low, it’s higher than Atlanta’s 13% of jobs accessible by MARTA, and comparable to the share of jobs accessible by LA’s patchwork of higher-capacity transit systems.  And it’s a good sign considering Hiawatha and Central barely reach outside of the central cities, and even miss employment clusters inside the central cities.

Anyway, this paper proves that transit advocates have discovered the suburbs, only 20 years after Edge City.  Having grown up in the Southdale area, I can say it’s none too soon.  They don’t go into too much detail, but the clusters CTOD identifies aren’t too different from those that Orfield and Luce identified in Region.  Orfield and Luce described their methodology thusly:

Employment centers were defined as contiguous TAZ’s with greater than average numbers of jobs per square mile and total employment exceeding 1,800 jobs.  Large job agglomerations like those in the centers of Minneapolis and St. Paul were divided into multiple employment centers based on job densities in different parts of the larger clusters.

The cool thing about these employment centers is that they mostly cling to highways, a.k.a available right-of-way.  Of course, that springs out of the shitty thing about them, that they sprawled in response to auto-dependence, as a result of constraints on growth by zoning, and without the guidance of regional planning.  But it’s possible that some day, cooler heads will conquer the capitol, and transit may be expanded.  If that ever happens, it would be a good idea follow the advice of the CTOD and aim for suburban job centers.

Which brings us to the other report, from Brookings, which maintains that transit access to jobs in the Twin Cities is average for the USA.  This report has taken a beating in the blogosphere, which I think is not surprising, considering the report looks at transit a little differently than the average transit rider probably looks at it.  That’s because it’s instead supposed to represent how the average American looks at transit, basically from the viewpoint, “how am I going to get to work?”

According to Brookings’ ranking, the Twin Cities has the 39th best transit access to jobs in the nation.  That might surprise some politicians, who don’t care anyway, but seems about right to me.  It’s when I start looking at the composite rankings that up becomes down for me.  First off, Brookings claims that 67% of working-age residents are near a transit stop.  67%?  How can that be when most bus lines don’t extend outside Minneapolis and St Paul?  Well, the fine print reveals that Brookings looked at bus routes that operate during rush hour, when commuter buses extend the web of transit lines by many times their midday size.

The rush hour focus also explains the next Wonderland metric, median waiting time, which is 11.6 minutes for the Twin Cities, 1.5 minutes more than the national average.  This actually also reflects the size of the commuter bus net in Minneapolis-St Paul – if commuter buses were excluded, the median wait would probably actually be shorter at peak, when several routes have 5-7 minute headways.

The commuter bus focus also explains the dismal but average showing of the Twin Cities in the last metric: percent of jobs that can be reached by transit in 90 minutes.  Because our bus system focuses on the downtowns, and so many commuter buses travel long distances to get there, but skip over jobs that might be along their route, it makes sense that so few jobs are reachable.  So while cities with smaller transit systems may rate the same in this metric, the Twin Cities’ comparatively larger bus system does no better because it is so narrowly focused.  (As a Downtown resident, I ain’t complaining.)

The Brookings reportadd  an economic layer to this already complex cake by considering the income level of the people with access to transit as well as the wage level of the jobs accessible by transit.  I wonder then why they didn’t find a way to factor in the non-peak transit coverage crucial to non-9 to 5ers.  Transit access to jobs is hugely important, and suburban job centers may be the next big transit growth market, but non-peak travel seems more important to transit’s core customers and is also less expensive to provide.

At least it’s something to think about while waiting for the bus.

Huh?

What have we learned?

Urban Decay

Yet another downside to municipal fragmentation is the loss of institutional memory.  Many are realizing that urban decay is not a process intrinsic only to central cities due to their inability to adapt to the automobile, but rather a byproduct of the American slash-n-burn style of city-building that can strike anywhere, but at a specific time, often about a half-century after greenfield development.  Unfortunately, as urban decay hits the suburbs, these fragments of cities are less able to learn from the experience of their older siblings what will combat and what will hasten the process of decay.

So when I came across the Strib’s article on the impending redevelopment of Brookdale Center I couldn’t help but think of Minneapolis’ earlier efforts to redevelop the commercial district at Lake & Nicollet.  The moribund Brookdale is probably in a more extreme situation than the struggling but alive Lake & Nicollet of the 70s.  The connection in my mind is the use of TIF to subsidize a developer to build a low-intensity, single use development of the sort that, in all likelihood, will be redeveloped in at least the same time frame as the structure it’s replacing, if not sooner.  Here’s a statement from the very study looking at redevelopment options for the mall area, 2003′s Brooklyn Center Opportunity Site:

Modern retail development often becomes obsolescent in the matter of a few decades…

So what do they go and build?  A modern retail development.  You gotta wonder if Brooklyn Center knew who they

Modern retail development

were hiring when they commissioned the study – Calthorpe and Associates is run by one of the founders of the Congress for New Urbanism.  After the completion of the study and a plan a few years later, the city actually included an 8 point refutation of their principles in their comprehensive plan, with the brilliant recommendation of increasing highway-oriented development and reducing open space.

What is likely to be built is the exact opposite of the design principles enumerated in the Opportunity Site Master Plan & Development Guidelines (although the plan actually applied to a site across Bass Lake Road from Brookdale, and I don’t know if there was ever any move to extend it to the Brookdale site).  Not only do we get a big box Wal-Mart, with its auto-dependent acres of parking and low-intensity land use, but accessory retail uses are scattered throughout the site, making future infill much more difficult.  To be fair, it is possible the planned smattering of smaller stores will never come to be, as a local retail real estate consultant notes in the Strib article:

“The challenge for the developers in Brookdale is, what are the stores that would see an opportunity to be at the Brookdale site that don’t already have a location that serves that area?”

In addition, complementary stores would have to stock items that are unavailable at Wal-Mart, or that are appreciably better or cheaper than at the retail giant.

Plans that came to naught

What could Brooklyn Center have done differently?  They already had a policy framework (in the Opportunity Site Master Plan) to encourage mixed-use, pedestrian-friendly development, but they followed the old suburban course, waiting for a developer to come along and proposed a PUD along the lines of the Master Plan.  If they had looked at the success of older cities in guiding development with zoning districts or overlays, they could have had zoning in place that would have discouraged the Wal-Mart style of rapidly-obsolete shopping strip.  Maybe Wal-Mart would have just moved up the street to a less restrictive city, but maybe they would have come up with a plan more like their proposal for Washington DC.

Half a page of scribbled plans

I’m writing under the assumption that Brooklyn Center wants to move away from auto-dependent commercial strips.  They have every reason to do so.  From the 2000 census to the 2005-09 ACS data, single-occupancy vehicular travel declined only very slightly and public transportation use increased at a similarly minuscule rate.  But in the same time frame, poverty increased dramatically in the suburb, from 7.4% to 12.9% of individuals.  This suggests that an increasing number of Brooklyn Centrists could benefit from the affordability of transit and improved opportunities for walking.

Brooklyn Center is only one tiny part of the region, and an adjustment of regional priorities would result in better development in inner ring suburbs.  A map from the Calthorpe planning effort shows as a third-tier regional center, while distant Maple Grove ranks as a second-tier center.   Why is our region prioritizing development in distant greenfields over vast acreage closer to the city?  These priorities have consequences, exemplified in the Bottineau Transitway’s decision to bypass existing transit centers, such as Brookdale, in the hopes that new transit centers will spring up on the fringe.

Maple Grove is sitting pretty

Until we agree to focus development in existing areas instead of on the edge of town, the municipal cogs that make up the regional machine will continue to spin freely, leaving minor cogs like Brooklyn Center to make their own mistakes.

Hit by nice Berg, census reeling

Portland Model City?

Steve Berg gets my nomination for King of Urbanists in the Twin Cities.  A talented writer, I consider him the most eloquent Minnesotan activist for safer, more inclusive streets, smart density, and mixing uses.

He’s been writing lately about the 2010 census results (2 more census articles than either of the local newspapers, by the way), and while I agree with his conclusion – municipalities in the Twin Cities need to do a better job of encouraging dense, transit-oriented growth as well as transit for the growth to orient to – I’ve been a bit irked about his decision to compare us to the same three cities of Denver, Seattle and Portland.

Portland annexation map

Portland does a great job encouraging growth along transit lines in developed areas, but it also has a dirty secret:  The greenfield area around Powell Butte was a significant contributor to the city’s growth.  As Portland’s annexation map makes clear, it has annexed land as recently as the early 90s, and plans to eventually annex the entirety of its urban growth boundary.  That means that Portland has as much in common with Forest Lake as it does with Minneapolis.

The population growth in the Powell Butte area accounted for a greater share of the city’s growth than the downtown area – although downtown had a higher growth rate and is a smaller area.  Still, it’s not really fair to ask a city that has been built out for decades to grow as fast as a city that still has a greenfield advantage.

Denver is an even worse comparison, since its population was boosted by massive redevelopments of Air Force bases.  The Lowry and Stapleton developments added a cumulative 16,664 residents to the Mile High City, way more than Downtown Denver’s 9,815 added residents.  Those three areas account for more than half of the 45,000 residents that moved into Denver in the oughts – other areas of the city grew as well, but there were also substantial sections that declined, specifically the Highland area across the river from Downtown.  It doesn’t seem to me that Denver’s census change pattern deviates all that much from MSP, except that it grew a lot more:

Denver Population Change 2000-2010

Mpls-StP Population Change 2000-2010

These maps are from Data Pointed and I’m pretty sure they’re not to scale.

Edit:  Data Pointed apparently doesn’t like hosting images for my blog so for now you’ll have to find the maps yourself on that site.  I’ll maybe screen print the NY Times maps or grab them from Transport Politic this weekend – I live to serve.

Seattle, however, is a more fair comparison to Minneapolis-St Paul.  I wrote a few months ago about how it contains more recently-built suburban areas than Minneapolis, but not necessarily more than St Paul.  Still, it hasn’t annexed any land since the 50s, so there isn’t any greenfield development in the city proper.

There is no question Seattle has done a better job encouraging growth in the center city than Minneapolis.  If you look at their growth map, you see strong growth in the downtown and around the university, like the Twin Cities and most cities nationwide.  But you also see people moving into areas outside of downtown, such as Ballard, Northgate, and NewHolly – these growth areas were codified in their most recent comprehensive plan as Urban Villages, areas where a dense mix of uses will be encouraged.  It’s a similar concept to Minneapolis’ Activity Centers, but Seattle sets aggressive targets for job and residential growth in these clusters.

Seattle Population Change 2000-2010

So if only one of Berg’s three comparison cities is actually comparable, are there other cities that are more like the Twin Cities, if just so that we’re not adrift in a sea of relativism?  Let’s look to our neighbors, who are of a similar vintage, and who were similar choked off by the upper classes seeking their own municipalities safe from the votes of the teeming, ethnic masses.

Milwaukee, St Louis and Cleveland are of similar size, age and metropolitan structure, and at first glance Minneapolis and St Paul look good in comparison.  St Louis and Cleveland each lost tens of thousands of residents in the last decade, and Milwaukee lost about two thousand – eerily similar to the Twin Cities’ combined losses.  But the three rust belt cities also had population booms in their downtowns – all three had growth rates that surpassed Minneapolis and St Paul, and St Louis beat Minneapolis in absolute increase as well.

Downtown population change

Just for kicks, I’ll throw in this info for the cities Steve Berg likes to compare to the Twin Cities:

Downtown Population Change

You can, of course, find similarities and differences between most cities.  And certainly all of these cities are auto-dependent, Euclidian-zoned (although I think Denver is experimenting with a form-based code) and in the Anglo-American tradition.  And, honestly, Berg’s points hold up in all of them – the USA has a racial ghetto problem, and while it’s less pronounced in cities with smaller minority populations, the Twin Cities is one of several metro areas that have failed to handle this problem.  Denver seems to have the same problem, and I don’t think we should follow Seattle’s lead by exporting the ghetto to a different city (Tacoma, in Seattle’s case; we’ve already gotten a start on sending minorities to the Brooklyns).  Instead we should continue the Met Council’s work on increasing affordable housing opportunities in the suburbs.  Here is some data to back up these assertions:

Census race 2010

Because of the racist nature of American settlement patterns, it’s predictable that cities with greenfield development (Portland, Denver) would have a smaller percentage of minority populations.  Conversely, it may be that the Twin Cities, with relatively small central cities relative to suburbs, have actually done a better job than these “peer” cities of reducing minority concentration, although a large ghetto remains on the Northside and Minneapolis sure suffered for it in the 2010 census.

Steve Berg’s other point, that successful cities develop their transit systems and encourage dense growth around stations, is more supported by census data.  Looking at the percent of metro area growth that occurred downtown, it roughly corresponds with the level of transit investment, although Milwaukee is a major outlier.  Also the metric doesn’t work with metros like Cleveland that lost population, although the fact that the downtown nevertheless grew is a major triumph.

Downtown vs Metro population change 2000-2010

I’m going to put my spreadsheet out there for people to look at and build on.  This rambling entry is not meant to be the final word on anything, so feel free to engage in the discussion by tearing my points to shreds in the comments.  I’m going to add more and more stats to this spreadsheet and maybe eventually I’ll do a another post when I have a more complete picture.

downtown census pop

A note about the data here:  it is always debatable how to define unofficial geographic areas such as downtowns.  As you might expect, I have my own opinion about what constitutes  Downtown Minneapolis and Downtown St Paul, but amazingly I don’t consider myself an expert on the neighborhood geography of other cities.  Therefore I’ve relied on others’ definitions, which I’ve referenced in the spreadsheet.  When I pulled the census data myself, I’ve referenced the census tracts I used, which usually didn’t correspond exactly with the downtown boundaries.  But then life itself is inexact.  As always, feel free to disagree, but if you do I ask you to specify your disagreement in the comments.